GENERAL NEUROSURGICAL CONSIDERATIONS
Complications of Brain and Spine Surgery
Rock et al1 used the American College of Surgeons National Surgical Quality Improvement Program (ACS-NSQIP) database to investigate safety outcomes in patients who underwent either intracranial or neurosurgical spine procedures between 2005 and 2015. Briefly, ACS-NSQIP consists of patient data from 603 medical centers in the United States and collects data from ∼270 perioperative variables. These include demographics, comorbidities, details about the surgical procedure, and outcome data. During the study interval, 137,029 spine and 38,284 intracranial cases were identified. Thirty-day complication rates were significantly higher in patients having cranial (22%) compared with spine surgery (11%; P<0.001). The 3 most common complications in the cranial surgery group were need for reoperation (6.8%), need for blood transfusion (5.6%), and pneumonia (3.2%). In the spine surgery group, the 3 most common complications were need for blood transfusion (4.9%), need for reoperation (3.0%), and urinary tract infection (1.4%). In the entire cohort, the 3 strongest predictors for complications were American Society of Anesthesiologists Physical Status Classification of V, preoperative ventilator dependence, and preoperative need for blood transfusion. Thirty-day mortality rates were higher in the cranial (4.8%) compared with the spine surgery group (0.5%; P<0.001). American Society of Anesthesiologists Physical Status Classification of V, IV, and III were the 3 strongest predictors of mortality. Of note, the authors did not stratify factors that predicted mortality based on surgical type.
The number of surgical procedures performed on the spine is increasing annually.2,3 Although major spine surgery is generally associated with significant risk, it is unclear how anesthetic management contributes to overall risk. Using the Anesthesia Closed Claims Project database, a repository of anesthesia malpractice closed claims,4 Kutteruf et al5 identified 2007 closed claims for surgical procedures performed between 2000 and 2014. During this interval, there were 207 claims associated with spine and 1800 with nonspine procedures. Compared with nonspine procedures, closed claims associated with spine procedures more frequently involved males (62% vs. 46%, P<0.001), nonemergency procedures (2% vs. 14%, P<0.001) and less frequently involved the accusation of substandard care (30% vs. 41%, P=0.001). Median claim payment was similar between spine surgery claims ($338,000; interquartile range [IQR]=$99,750 to $870,858) and nonspine surgery claims ($260,960 [IQR=$68,000 to $672,325]; P=0.12 [data represented in 2016 US dollars]). A summary of injury severities and characteristics is shown in Table 1. Compared with the nonspine surgery group, injuries in the spine surgical group more commonly resulted in severe or disabling injury but less commonly in death. Injuries in the spine surgery group were more commonly associated with cardiovascular events, positioning, patient condition, or surgical procedure. Hemorrhage accounted for 57% of the adverse cardiovascular events in the spine surgery group. Ocular injuries were more common in the spine compared with the nonspine surgery group (36 [17%] vs. 60 [3%]; P<0.0001). Of the 207 spine procedures, 55 (27%) involved the cervical spine. Prone position was more common among cases performed on the thoracolumbar spine (129 [85%]) versus cervical spine cases (11 [20%]; P<0.001]. Ocular injuries, hemorrhage, and positioning-related injuries were more common among thoracolumbar procedures, whereas airway complications were more common among cervical spine procedures. Despite these differences, median claim payment was similar between those having cervical and those having thoracolumbar surgery.
Brown et al6 reported on complications occurring after brain tumor resection in adults at a single high-volume center. During a 5-year period, 4423 intracranial procedures were performed for neoplastic lesions: 567 (13%) biopsies, 1326 (30%) intra-axial mass resections, 1380 (31%) skull base mass resections, and 1150 (26%) pituitary-based mass resections via the transsphenoidal route. A complication within 30 days of surgery occurred in 585 (13%) cases, with 435 (9.8%), 73 (1.7%), and 63 (1.4%) comprising neurological, medical, and wound complications, respectively. The rate of complications was highest among those having skull base surgery (23% of cases), whereas complications occurred in 13%, 6%, and 4% of patients having intra-axial tumor resection, biopsy and transsphenoidal pituitary tumor resection, respectively. Death within 30 days of surgery occurred following 14 (0.3%) procedures, with the highest rates among those having biopsy (1.4%). Mortality rates following skull base, intra-axial resection, and transsphenoidal surgery were 0.2%, 0.1%, and 0%, respectively. The higher mortality rate associated with biopsy was attributed to the grave prognoses associated with mass lesions for which patients were having biopsy rather than tumor resection.
The risk of severe complications following intracranial surgery depends on the specific surgical procedure and patient comorbidities, leading some to question whether subsets of patients may not require intensive care unit (ICU) monitoring and management postoperatively. De Almeida et al7 identified 9 studies that compared postoperative outcomes between patients who had undergone intracranial surgery and either did or did not receive elective postoperative ICU care. These 9 studies included a diverse cohort of 2227 patients of whom 879 (39.5%) did not receive immediate postoperative ICU care. Subsequent transfer to the ICU occurred in only 30 (3%) of the 879 patients who did not receive ICU care. Neurological, medical, and other indications resulted in admission to ICU in 19 (63%), 8 (27%), and 3 (10%), respectively, with nonsurgical and nonmedical indications reported as family anxiety (n=2) and patient preference (n=1). Bypassing postoperative ICU admission resulted in cost savings of between US$871 and US$5224 adjusted for inflation to 2016. Of concern, the authors were unable to describe the characteristics of patients who bypassed the ICU after surgery and had a good outcome. This knowledge would help clinicians identify the cohort for whom transfer to the floor after intracranial surgery might be safe.
Cinotti et al8 identified factors that could be used to predict risk for complications that frequently require >24 hours of ICU management in patients having nonemergent brain tumor resection. These are intracranial hemorrhage, intracranial hypertension, seizures, need for postoperative intubation and mechanical ventilation, impaired consciousness, unmanageable agitation, swallowing disorders, respiratory failure, unexpected severe motor deficit, and death. Eight factors were found to be independently associated with predicting risk for complications: Glasgow Coma Score <15, history of brain tumor surgery, greater tumor size, midline shift ≥3 mm, need for administration of blood products, maximum and minimum systolic blood pressure (SBP), and duration of surgery.
Hypertension during emergence following intracranial surgery is common and can contribute to complications such as surgical site bleeding or adverse cardiac events. Drugs, such as beta-adrenergic receptor antagonists or calcium channel antagonists, are often used to attenuate emergence hypertension. Acupuncture and accupoint stimulation have been shown to reduce anesthetic and analgesic requirements, risk for nausea and vomiting, and anxiety in patients having craniotomy.9,10 In a recent study, 75 patients having supratentorial craniotomy were randomized to receive either electrical accupoint stimulation at 6 points (ie, Hegu, Neiguan, Chize, Lieque, Futu, and Renying) on the right side, or placement of the stimulator over the points but no stimulation.11 Only the acupuncturist was aware of group assignment. In the group assigned to receive stimulation, stimulation was accomplished with a current of 6 to 15 mA, adjusted to slight twitching of regional muscles at alternating frequencies of 2 Hz for 10 seconds and 10 Hz for 5 seconds. Stimulation was started 30 minutes before induction of anesthesia and continued until 5 minutes before the end of surgery. All patients received a standard anesthetic. At 10 minutes following extubation, SBP, mean arterial pressure (MAP), and heart rate (HR) were significantly lower in the accustimulation group (123±12 mm Hg, 95±9 mm Hg, and 83±7 bpm, respectively) compared with the control group (135±14 mm Hg, 102±9 mm Hg, and 87±8 bpm, respectively; P<0.05 for all 3 comparisons). Serum concentrations of epinephrine, norepinephrine, and cortisol were all lower in the accustimulation group 10 minutes after extubation compared with the control group. Postoperative rates of coughing, nausea and vomiting, opioid consumption were also lower in the accustimulation group, and patient satisfaction, assessed with the Quality of Recovery Questionnaire,12 was better with accupoint stimulation. The authors reported no adverse outcomes attributed to accupoint stimulation.
Same Day Discharge for Intracranial Neurosurgery
With better understanding of risk and timing of complications, some patients may be safely discharged from hospital following an intracranial procedure on the day of surgery. Such outpatient intracranial procedures may reduce cost without concerns for adverse effects soon after hospital discharge.13 In a literature review, Sheshadri et al14 identified 7 case series describing same day brain tumor resection including 709 patients of whom 643 (91%) were successfully discharged on the same day and did not require readmission. For those having brain tumor biopsy the authors identified 5 series consisting of 446 patients of whom 411 (92%) were successfully discharged on the same day without readmission. For those having elective cerebral aneurysm clipping, they identified 1 report that consisted of 25 patients with 17 (68%) being successfully discharged on the same day as surgery. The authors describe the criteria used at their institution to assess for eligibility for same day discharge following intracranial surgery. Tumors must be supratentorial, aneurysms unruptured and in the anterior circulation, and expected surgical duration should be <4 hours. The anesthesiologist must consider the patient to be at low-to-moderate risk for medical complications, and the patient should have no cardiac, respiratory or cognitive disorders, and no significant neurological deficits. The patient must also be agreeable to the discharge plan, have an available care provider, and remain close to the hospital immediately after surgery.
Minimally Invasive Neurosurgical Procedures
An alternative option to minimize hospital stay and reduce complications is to consider less invasive treatments for certain conditions. The interested reader is referred to 2 recent review articles that address the use of less invasive technologies to treat neurological disorders.15,16 Jimenez-Ruiz et al16 discuss periprocedural considerations for patients undergoing neurosurgical laser thermal ablation therapy. Briefly, this technology involves placement of a probe that delivers infrared laser light into a region of abnormal tissue resulting in thermal ablation of the tissue. This technique has been used for the treatment of intracranial and spinal tumors, and ablation of epileptic foci.17 Dunn et al15 review deep brain stimulation and focused ultrasound as treatment for various neurological disorders. Deep brain stimulation involves implantation of a stimulating electrode into a deep brain nucleus for the treatment of movement and psychiatric disorders. Alternatively, focused ultrasound, which does not require an incision, can be used to treat many of the disorders that are currently treated with deep brain stimulation. Multiple ultrasound beams are focused on a specific brain region, resulting in an increase in temperature and consequent tissue ablation. Both laser thermal ablation and focused ultrasound are performed during magnetic resonance imaging (MRI) such that magnetic resonance thermometry can be used to assess tissue injury and ablation in real time.
Stereotactic Headframes and Airway Management
Stereotactic headframes are often used to facilitate target localization for a variety of neurosurgical procedures, including intracranial biopsy, mass resection, deep brain stimulator implantation, and laser and focused ultrasound thermal ablation. Stereotactic headframes pose a major limitation if there is a need for urgent or emergent airway management. In one study, 30 physician anesthesia providers (7 residents, 12 fellows, and 11 staff) placed a laryngeal mask airway and intubated via direct and videolaryngoscopy a mannequin with and without a Leksell (Eleka, Stockholm, Sweden) headframe in place.18 Compared with the mannequin without a headframe, a significantly longer time was required to place a laryngeal mask airway (39±15 vs. 25±7 s; P=0.0001) and intubate via direct (59±20 vs. 46±12 s; P=0.0064) and videolaryngoscopy (55±23 vs. 45 ±11 s; P=0.0336) in the mannequin with a headframe. Residents took approximately 29 seconds longer to intubate using videolaryngoscopy compared with fellows (P=0.01) and staff (P=0.03), but times to intubate with direct laryngoscopy and to place a laryngeal mask airway were similar among groups. Simulation of crisis management in neuroanesthesia has a significant role in the education of trainees,19 and management of the airway in the setting of a stereotactic headframe would make an excellent simulation scenario in neuroanesthesia.
Protective Ventilation Strategies
Intubation and mechanical ventilation are required for most intracranial procedures that employ general anesthesia. Mechanical ventilation with lower tidal volume and utilizing nonzero positive end-expired pressure (PEEP) has been shown to reduce the risk for postoperative pulmonary complications compared with ventilation with higher tidal volume.20,21 However, there is potential concern that application of PEEP may increase intracranial pressure (ICP). In a study to investigate this issue, Ruggieri et al22 enrolled 40 adult patients undergoing supratentorial craniotomy for tumor resection and who had >5 mm midline shift on preoperative imaging. Patients received a standardized anesthetic, were positioned supine with 20 to 30 degrees head elevation, and received 0.5 g/kg mannitol upon dural exposure. Patients were randomized to receive either a standard (tidal volume 9 mL/kg without PEEP) or protective (tidal volume 7 mL/kg with 5 cm H2O of PEEP) ventilator strategy first. In either circumstance, respiratory rate was titrated to maintain arterial carbon dioxide tension between 30 and 35 mm Hg; the inspiratory-to-expiratory ratio was 2:1. Following exposure of the dura mater, a subdural catheter was place by the surgeon for ICP measurement. Following this measurement, the alternate ventilation technique was employed and ICP was measured again after 10 minutes. ICP was similar between groups (12.2±7.6 vs. 13.8±9.4 mm Hg in the standard and protective groups, respectively; P=0.4), and operating conditions were assessed by the neurosurgeon as acceptable in all circumstances. Of note, mannitol was administered before measurement of ICP. This may have resulted in decreased cerebral elastance and attenuated any changes that may have resulted from the different ventilation techniques.
The prone position increases intra-abdominal and intrathoracic pressure which can not only affect ventilation and spirometric parameters but also result in hemodynamic changes that may impact ICP and cerebral oxygenation.23–27 To determine the effect of obesity and operating room table design on respiratory mechanics, Ni et al28 stratified 37 patients having lumbar spine surgery in the prone position based on body mass index into 2 groups: normal body mass index (18.5 to 24.9 kg/m2) and obese (body mass index >25 kg/m2). Patients were then randomized to undergo surgery either on a standard flat operating room table or a Jackson table. Dynamic lung compliance was measured as the ratio of tidal volume to the difference between peak inspiratory pressure and PEEP. All patients inhaled 100% oxygen and ventilator parameters were set to deliver a tidal volume of 6 to 8 mL/kg at a rate of 12 to 14 breaths per minute without PEEP. Peak airway pressure increased and dynamic lung compliance decreased with transfer to the prone position in all groups; greater changes occurring in obese patients, especially those obese patients positioned on a standard flat operating table. With all patients inspiring 100% oxygen, arterial partial pressure of oxygen increased in all groups with assumption of the prone position, with greater increases occurring in obese patients, especially those positioned on a flat table. The authors attribute this latter effect to redistribution of pulmonary ventilation resulting in improved ventilation/perfusion mismatch and decreased functional residual capacity.29,30
Lung protective mechanical ventilation refers to delivery of lower tidal volumes with titrated PEEP and use of recruitment maneuvers. Lung protective ventilation strategies have decreased pulmonary complications in critically ill patients and those having abdominal surgery when employed intraoperatively.21,31 To determine if lung protective ventilation improves postoperative lung function in patients undergoing major spine surgery in the prone position, Soh et al32 randomized 78 patients to receive either lung protective or standard ventilation techniques intraoperatively. All patients received a standardized general anesthetic with similar hemodynamic goals, and were ventilated using a volume-controlled mode with an inspiratory-to-expiratory ratio of 1:2 and respiratory rate adjusted to maintain end-expired carbon dioxide tension (ETCO2) between 35 and 40 mm Hg. In the standard ventilation group, patients received a tidal volume of 10 mL/kg with no PEEP. In the lung protective group, the delivered tidal volume was 6 mL/kg with a PEEP of 6 cm H2O, and intermittent but standardized recruitment maneuvers were performed. There were no differences between groups in either forced vital capacity or forced expiratory volume in 1 second on postoperative day 3, both co-primary study endpoints. In addition, there were no differences between groups in rates of leukocytosis, volume of pulmonary secretions, or rates of infiltrates on chest x-ray on postoperative days 0, 1, or 3. There were also no differences in rates of pulmonary complications (pneumonia and pulmonary edema), rates of extrapulmonary complications, or ICU and hospital lengths of stay. Although the authors did not find a difference in any parameter between groups, their assessment was limited to patients with significant lung disease. Thus, these finding may not necessarily be extrapolated to patients with normal lungs.
Hypertonic solutions that consist of a solute that cannot cross the blood-brain barrier are a class of treatment options for elevated ICP. Although sodium bicarbonate, sodium lactate, and urea have been used, most commonly hypertonic solutions of either mannitol or sodium chloride are preferred to treat intracranial hypertension.22,33,34 The effect of mannitol and hypertonic saline were investigated in a study by Ali and colleagues who enrolled 39 patients scheduled to undergo elective supratentorial craniotomy for mass resection and who also had imaging evidence of significant mass effect as indicated by either the presence of ipsilateral lateral ventricular compression, >5 mm midline shift, or contralateral ventricular dilatation.35 All patients received a standard anesthetic with standardization of physiological variables: state entropy=40 to 60, arterial carbon dioxide partial pressure 30 to 35 mm Hg, and MAP, SBP, and HR within 30% of baseline. ICP was measured using an ICP EXPRESS probe (Codman, USA) placed via burr hole stereotactically adjacent to the tumor. Patients then received either mannitol (20%) or hypertonic saline (3%) administered as a 5 mL/kg intravenous bolus over 15 minutes. ICP was recorded for 45 minutes after which craniotomy was performed and the surgeon rated the degree of dural tension using the following scale: 1=relaxed, 2=satisfactory, 3=firm, and 4=bulging. ICP was similar between groups before drug administration. Hypertonic saline resulted in a greater decrease in absolute (median=5 mm Hg, range=1 to 9 mm Hg) and percent (median=38%, range=11% to 47%) ICP after 45 minutes compared with mannitol (absolute change: median=4 mm Hg, range=1 to 7 mm Hg [P<0.05]; percent change: median=30%, range=13% to 38% [P=0.001]). Brain relaxation scores were similar between groups. Physiological variables were also generally similar between the 2 groups except that mannitol was associated with a significant increase in serum lactate concentration and hypertonic saline with a significant decrease in pulse pressure variation over the 45 minutes study period. These findings suggest that, compared with hypertonic saline, mannitol likely results in a more significant decrease in intravascular volume that could cause tissue hypoperfusion. Recently, Lillemae et al36 showed that mannitol, when combined with albumin or Ringers acetate in vitro, may impair coagulation. Although in vivo verification of these findings will be necessary, this effect may be due in part to the development of a dilutional coagulopathy which might not be observed in vivo if mannitol results in diuresis and hemoconcentration.
In patients requiring craniotomy with awake cortical mapping, the primary anesthetic goal is to maintain sedation before and following mapping but facilitate rapid arousal and patient participation in neurological testing during mapping. Propofol is frequently used as a continuous infusion before and following cortical mapping.37 In 13 patients having awake cortical mapping, Soehle et al38 measured plasma concentration and bispectral index (BIS) and estimated effect site concentration by both the Marsh39 and Schnider40 pharmacokinetic models at return to consciousness and when the patient was able to effectively participate in neurological testing during cortical mapping. Before mapping, propofol was titrated to BIS 40 to 60. Return of consciousness occurred 11±3 minutes after discontinuation of propofol when BIS was 77±7. Plasma propofol concentration at this time, measured by liquid chromatography, was 1.2±0.4 μg/mL, whereas estimated effect site concentration was significantly higher based on the Marsh versus Schinder models (1.9±0.4 μg/mL and 1.4±0.4 μg/mL, respectively; P<0.001). Patients were able to participate in neurological testing 23±12 minutes after cessation of propofol when BIS was 92±6. At this point, plasma propofol concentration was 0.8±0.3 μg/mL, whereas the Marsh model again resulted in a higher estimate of effect site concentration (1.3±0.5 μg/mL) compared with the Schinder model (1.0±0.4 μg/mL; P=0.002).
Dexmedetomidine has been used as part of a balanced anesthetic technique, for sedation, and to decrease postoperative pain in patients having a variety of neurosurgical procedures.41–45 Liu and colleagues performed a meta-analysis of randomized controlled trials that measured the effect of dexmedetomidine on opioid requirements and postoperative pain in patients undergoing neurosurgical procedures.46 Eleven studies were identified through March of 2016 and included 674 adult patients having intracranial procedures (supratentorial, infratentorial, and transsphenoidal) randomized either to receive (n=335) or not receive (n=339) dexmedetomidine. Overall, dexmedetomidine was associated with a significant decrease in intraoperative opioid consumption. In the postanesthesia recovery unit, dexmedetomidine was associated with reduced pain scores, rates of shivering, blood pressure, and HR. Time to extubation was also shorter in the dexmedetomidine group. One concern with dexmedetomidine is its short duration of action, and, unfortunately, the authors did not assess for the durability of the beneficial hemodynamic and analgesic effects of dexmedetomidine.
Acetaminophen is a nonopioid analgesic with a good safety profile. There is a paucity of data describing its opioid-sparing effect in patients having craniotomy. Artime and colleagues randomized 100 adults undergoing supratentorial craniotomy to receive either 1 g of acetaminophen or an equivalent volume of normal saline every 6 hours for 24 hours starting after induction of anesthesia but before skin incision.47 All patients received a standardized general anesthetic that included 2 to 3 μg/kg fentanyl and scalp block. Data from 14 patients were excluded due to procedural change, complications, or request to withdraw. Postoperative opioid consumption, time weighted pain scores within the first 24 hours, and rates of complications, such as emesis, pruritus, dizziness, and drowsiness, were similar between groups. Times to extubation and discharge from the postanesthesia recovery unit were also similar. However, the proportion of patients who experienced at least 1 episode of severe pain (visual analog pain score ≥7 on a scale of 0 to 10) was significantly lower in the acetaminophen compared with the control group (50% vs. 76%; P=0.01). When asked to rate satisfaction with pain control regimen on a scale of 1 (unsatisfied) to 10 (most satisfied), patients who received acetaminophen had a higher mean satisfaction scores (8.1±2.3) compared with those who received normal saline (6.9±2.6; P=0.03). Stone and colleagues also recently demonstrated that acetaminophen did not decrease postoperative opioid consumption in patients who underwent craniotomy for cerebral revascularization.48
The traditional transoral approach to the odontoid process may be difficult in patients with limited mouth opening, and is associated with risk for dysphagia and postoperative airway edema. A newer endoscopic transnasal approach to the odontoid process may be associated with a more favorable risk profile. Sexton and colleagues retrospectively identified 22 patients who underwent odontoid surgery either via the transoral (n=17) or transnasal (n=5) approach.49 In those having transoral surgery, no patient presented to the operating room with a tracheostomy in place but 16 of 17 (94%) had a tracheostomy placed for their procedure. Two patients in the transnasal group had a preexisting tracheostomy and all others were orally intubated for the procedure and extubated in the operating room at its conclusion. Median hospital length of stay was shorter in the transnasal group (3 d, range=2 to 23 d) compared with the transoral group (18 d, range=8 to 52 d). Persistent dysphagia requiring feeding via nasal tube upon hospital discharge was more common in the transoral group (9/17; 53%) compared with the transnasal group (1/5; 20%). A larger series will be required to confirm these findings.
In patients with an unstable cervical spine, neck movement during laryngoscopy and intubation could cause secondary spinal cord injury. In this cohort of patients, awake intubation allows for neurological assessment after intubation. The Shikani Optical Stylet (Clarus Medical, Minneapolis, MN) is a flexible fiberoptic stylet that can be used for awake tracheal intubation instead of a standard fiberoptic laryngoscope. Mahrous and Ahmed50 randomized 60 patients with documented or suspected cervical spine instability to receive orotracheal intubation awake with either standard fiberoptic bronchoscopy or the Shinkani Optical Stylet. Successful intubation on the first attempt was similar between groups, whereas successful intubation was achieved faster with the Shikani Optical Stylet (52±7 vs. 103±11 sec in the standard fiberoptic group; P<0.001). Rates of coughing, gagging, and postprocedural sore throat were similar between groups. There were no cases of neurological deterioration following intubation in either group in this series.
Approximately half of patients presenting for spine surgery use opioid medications preoperatively for pain control.51,52 Preoperative opioid use has been associated with delayed rehabilitation and poor outcome following spine surgery.51,53 This association is likely multifactorial and may include greater disease severity or a longer disease course in those requiring opioids preoperatively. Wick and colleagues identified 1836 patients who underwent elective spine surgery at a single center between October 2010 and July 2015.54 Of these, 543 (30%) and 1293 (70%) had cervical and lumbar surgery, respectively, but the rates of fusion versus decompression were not reported. Overall, 1020 (56%) of the entire cohort were receiving opioids before surgery. The authors estimated that a mean daily preoperative opioid dose >48 mg/d (95% confidence interval [CI], 29 to 60 mg/d) oral morphine equivalents was associated with decreased odds of achieving a clinically meaningful improvement in outcome after surgery based on changes in Neck Disability Index and Oswestry Disability Index scores. Unfortunately, the authors did not calculate a cut-off preoperative morphine dose independently for those with cervical versus thoracolumbar disease, as these values would likely differ.
Naloxone, an antagonist of the u-opioid receptor, antagonizes the effects of opioid analgesics, such as morphine. However, low-dose naloxone has been shown to enhance the analgesic effects of morphine, possibly by enhancing encephalin release.55–57 Firouzian and colleagues conducted a prospective and blinded trial where 80 otherwise healthy patients having lumbar discectomy were randomized to receive either low-dose naloxone or placebo.58 All patients received a standard anesthetic, and, postoperatively, intravenous morphine at 1 mg/h and patient-controlled analgesia with morphine set to deliver 0.25 mg boluses with a 15 minutes lockout. To those randomized to receive naloxone, a continuous infusion of 0.25 μg/kg/h was administered for the first 24 hours postoperatively. Pain, nausea, and pruritus were assessed at 1, 6, 12, and 24 hours postoperatively, Naloxone was associated with significantly lower pain scores, nausea, and pruritus throughout the study period. Median total morphine consumption during the first 24 hours was significantly lower in the naloxone group (26 mg, IQR=24 to 28 mg) than in the placebo group (34 mg, IQR=32 to 36 mg). No adverse effects were attributed to naloxone. Further study will be required to better understand patient cohorts that are most likely to benefit from this treatment. For example, the many patients having spine surgery who are utilizing chronic opioids preoperatively may respond differently to this therapy.
Clonidine, when used as a supplement during regional anesthesia, can serve to extend block duration and improve block quality.59 To assess the analgesic benefit of clonidine in patients having spine surgery, Hay et al randomly allocated 225 patients to receive a preincisional field block with 20 mL of 0.25% bupivacaine with or without 150 μg clonidine added to the block solution.60 Overall, 80, 25, 94, and 26 patients underwent lumbar fusion, lumbar laminectomy, lumbar microdiscectomy, and cervical laminectomy, respectively. Postoperative analgesia consisted of acetaminophen 1 g and ketoprofen 50 mg both every 6 hours through postoperative day 3, and 5 mg morphine subcutaneously no more than every 6 hours as needed. Patients were discharged on postoperative day 4; no details about the postdischarge analgesia regimen were disclosed in the manuscript. Postoperative visual analog pain scores (0=no pain, 10=most severe pain) were assessed every 2 hours through hospitalization, and daily until postoperative day 8. The co-primary outcomes were area under the pain versus time curve through day 8, and total morphine consumption through day 3. Clonidine was associated with a significant reduction in area under the pain versus time curve for patients having lumbar fusion, laminectomy, and microdiscectomy, but not cervical laminectomy. Total morphine consumption was decreased only in those having lumbar fusion and lumbar laminectomy. Collectively, pain scores and total morphine consumption were both significantly decreased during the intervals of day 0 to 2, days 3 to 5, and days 6 to 8, suggesting either a durable analgesic sparing effect or benefits from preemptive analgesia in those who received clonidine. No significant adverse effects were attributed to clonidine in this cohort.
Pendi and colleagues performed a meta-analysis of trials that prospectively randomized patients having spine surgery to receive or not receive ketamine in the perioperative period.61 The investigators assessed opioid use and pain scores for up to 48 hours postoperatively. In the 14 identified trials, perioperative ketamine administration varied significantly and consisted of either intraoperative use only (n=4), postoperative use only (n=2), or both intraoperative and postoperative use (n=8), with varied dosing. Ketamine was associated with a significant reduction in opioid consumption and decreased pain scores for up to 24 hours after surgery with insignificant benefits after 24 hours. Ketamine was not associated with an increased rate of adverse events including dysphoria, hallucinations, nausea, or sedation. Of note, the authors did not include in this meta-analysis the prospective trial by Loftus and colleagues where opioid dependent patients having spine surgery were randomized to receive either ketamine or saline placebo intraoperatively. In the study by Loftus and colleagues, patients who received ketamine had a durable decrease in opioid consumption and pain scores at 6 weeks after surgery, without an increase in side effects.62
The Prevention of Delirium and Complications Associated with Surgical Treatment (PODCAST) study was a prospective trial that was designed to assess the utility of ketamine at reducing the risk of postoperative delirium.63 Briefly, PODCAST was a multicenter study conducted in adults above 60 years old having surgery. Patients were randomized to receive either 0.5 mg/kg ketamine, 1.0 mg/kg ketamine, or an equivalent dose of saline intravenously. PODCAST failed to show a reduction in risk for postoperative delirium with ketamine. Mashour and colleagues performed a post hoc analysis of data from the PODCAST trial to determine if ketamine reduced the risk for postoperative depression.64 Depressive symptoms screening was conducted with the Patient Health Questionnaire version 8 (PHQ8) scale for depressive symptoms before surgery and then again on postoperative days 3 and 30. The range of the PHQ8 is 0 to 24 with a score of ≥10 indicating symptoms suggestive of depression. In the entire cohort, 9.6%, 16.6%, and 11.9% had symptoms suggestive of depression preoperatively and on postoperative days 3 and 30, respectively. Interestingly, 64 patients had symptoms of depression preoperatively but only 6 of the 84 patients with symptoms of depression on postoperative day 3 had symptoms of depression preoperatively. This suggests that postoperative depressive symptoms cannot be solely attributed to preoperative symptoms. There was no significant difference in the rates of depression among groups at any time point. The authors hypothesize that the lack of an effect by ketamine on rates of depression could be due to the older patient cohort, a difference between the etiology of postoperative versus bipolar or unipolar depression, or the complexity of the surgical procedure and perioperative course.
In addition to its anesthetic and analgesic effects, ketamine has also been shown to be a rapid and effective treatment for unipolar and bipolar depression.65–67 Carspecken and colleagues prospectively randomized 50 adult patients having electroconvulsive therapy (ECT) to receive either methohexital (1 to 2 mg/kg) or ketamine (1 to 2 mg/kg) intravenously for induction of anesthesia.68 ECT was performed initially with unilateral right-sided electrode placement. If a seizure was not induced, the stimulus intensity was incrementally increased. If maximal stimulus intensity failed to induce a seizure, bilateral stimulation was attempted. If there was still failure to elicit a seizure, induction of anesthesia was conducted with the alternate drug during the following session. Seven of 27 (26%) and 1 of 23 (4%) were switched to bilateral electrode placement in the methohexital and ketamine groups, respectively (P=0.03). Four in the methohexital group were switched to ketamine (and included in the methohexital group for data analysis), whereas no patient in the ketamine group required a switch to methohexital. Depression scores 72 hours after ECT improved in both groups, but there was no difference in the degree of improvement between methohexital and ketamine groups. Postoperative agitation was more common in the methohexital group. The lack of a treatment effect by ketamine in this study may be due to multiple factors including dilution of an effect by ECT, or effects that may have been noticeable at earlier or later time points than 72 hours after ECT. Of note, Sriganesh and colleagues reported on cerebral oximetry changes during ECT prospectively in 41 patients having 82 ECT treatments.69 Baseline cerebral oxygen saturations were 67±6% after induction of anesthesia with thiopental, and increased to a maximum of 81±1% and 79±1% over the course of 135 and 120 seconds after ECT administration in those that did and did not receive pretreatment with atropine, respectively. Cerebral oxygen saturation remained elevated over the 300 seconds measurement period. This increase in cerebral oxygen saturation likely represents cerebral vasodilation in response to seizure activity. The authors were expecting a greater increase in cerebral oxygen saturation in those that received atropine due to its effect on cardiac output. However, they do not provide a hypothesis to account for the lack of an effect by atropine on cerebral oxygen saturation during ECT.
In select patients with acute ischemic stroke (AIS), endovascular management with modern embolectomy devices has been show to improve neurological outcome without increasing risk for complications compared with intravenous thrombolytic therapy.70 However, not all patients with AIS are thought to be candidates for endovascular thrombectomy. For example, endovascular treatment was often not considered in patients who had stroke symptoms for >6 hours, as ischemic tissue was now likely infarcted and risk for hemorrhagic transformation upon restoration of flow was thought to outweigh any benefit from restoration of flow. In patients with an uncertain time of symptom onset, when considering eligibility for endovascular therapy stroke onset time is often considered to be the time at which the patient was last known to be well. With stroke onset time often likely being shorter than that at which the patient was last known to be well, many patients who might be candidates for endovascular management are excluded from an intervention from which they might gain benefit. Patients were enrolled in the Diffusion Weighted Imaging or Computerized Tomographic Perfusion Assessment with Clinical Mismatch in the Triage of Wake-Up and Late Presenting Strokes Undergoing Neurointervention with Trevo (DAWN) trial, if they had neurological deficits of greater severity than would be expected based on the size of their ischemic core as determined by imaging, suggesting ischemic but not infarcted tissue, and were within 6 to 24 hours of being last known neurologically normal.71 Patients were randomized to receive best medical management with (n=107) or without (n=99) mechanical thrombectomy. Mechanical thrombectomy was found to be superior to medical management alone based on the co-primary outcome metrics of utility-weighted modified Rankin Scale (mRS) score and rates of functional dependence (defined as mRS=0 to 2) at 90 days. Rates of symptomatic intracranial hemorrhage and mortality rates at 90 days were similar between groups. Similarly, in the Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke (DEFUSE 3) Trial, patients with greater neurological deficits than would be expected based on imaging and who were last known to be well 6 to 16 hours prior were randomized to receive best medical therapy with (n=92) or without (n=90) mechanical thrombectomy.72 At 90 days there was a shift in mRS score distribution toward better neurological outcome, and a greater fraction of patients with mRS scores 0 to 2 without increased rates of intracranial hemorrhage, death, or serious adverse events, in the cohort that received thrombectomy. Collectively, these findings suggest that comparison of neurological deficits with imaging characteristics may be used to identify a new cohort of patients who might benefit from mechanical thrombectomy.
Prior observational data suggest that general anesthesia for stroke revascularization is associated with worse outcomes than monitored anesthesia care.73,74 Although one could reason that anesthetic technique may modulate outcome, these data are likely biased as patients who are intubated, with or without general anesthesia, likely have greater stroke severity and may not have been candidates for monitored anesthesia care. In 2 recent prospective trials, general anesthesia was not associated with worse outcome in patients with AIS who would be candidates for either anesthetic technique.75,76 A third study, the General or Local Anesthesia in Intra Arterial Therapy (GOLIATH) trial, was conducted at a single center in Denmark.77 Patients who were intubated before entering the angiography suite, those with Glasgow Coma Score <9, those with prestroke mRS>2, those in whom MRI was contraindicated, and those with a baseline estimated infarct volume >70 mL were excluded. One hundred twenty-eight patients with AIS were randomized to receive either general anesthesia or monitored anesthesia care for AIS revascularization. Maintenance of general anesthesia was accomplished with propofol and remifentanil infusions, whereas fentanyl boluses with or without a low-dose propofol infusion was used for sedation in those receiving monitored anesthesia care. Before revascularization, normocapnia was maintained and blood pressure goals were SBP>140 mm Hg and MAP>70 mm Hg. There was no difference in the primary outcome measure, median infarct volume growth in first 48 to 72 hours, between groups (8.2 mL [IQR=2.2 to 38.6 mL] vs. 19.4 mL [IQR=2.4 to 79.0 mL] for general anesthesia vs. monitored anesthesia care, respectively [P=0.1]). Rate of successful reperfusion was greater in the general anesthesia group (77%) compared with the monitored anesthesia care group (60%; P=0.04). Further, there was a significant shift toward more favorable outcome based on mRS in those who received general anesthesia (P=0.04). Four patients randomized to monitored anesthesia care required general anesthesia because of either movement (n=2), vomiting (n=1), or aspiration (n=1). Rates of intracranial hematoma and mortality at 90 days after intervention were similar.
Patients who do not receive general anesthesia for AIS mechanical revascularization may or may not receive sedation. Van de Graaf et al78 performed a post hoc analysis of patient data from the MR CLEAN Trial70 to compare outcomes in those who did not receive general anesthesia but either did or did not receive intraprocedural sedation. Briefly, MR CLEAN was a prospective single center trial where patients with AIS were randomized to receive best medical management with or without mechanical thrombectomy. In this post hoc analysis, 60 (41%) and 86 (59%) received sedation and no sedation, respectively. The rate of prior stroke was higher in those who received no sedation (14% vs. 2% in those who did receive sedation; P=0.01), otherwise baseline characteristics were well matched. All reported workflow time intervals (time from stroke onset to groin puncture, time from stroke onset to reperfusion, and procedure duration) were similar. Among those who did not receive sedation, there was: (1) a lower median mRS at 90 days (3 [IQR=2 to 4] vs. 4 [IQR=3 to 5]; P<0.05); (2) a greater proportion with mRS ≤2 at 90 days (47% vs. 22%; P<0.05); and, (3) a greater proportion with successful reperfusion (62% vs. 38%; P<0.05). Ninety day mortality rates and rates of complications were similar between groups. It is very likely that these data have similar biases as earlier retrospective and post hoc analyses that compared outcome in patients who received general anesthesia versus monitored anesthesia care for mechanical thrombectomy. Specifically, it is likely that some patients required sedation to facilitate the procedure and would not have tolerated the procedure without it. This may have been due to subtle differences in stroke severity or the presence of agitation in those requiring sedation.
Athiraman and colleagues retrospectively reviewed the records of 88 patients who received general anesthesia for mechanical thrombectomy for AIS at a single center to identify potential factors associated with good outcome.79 At the authors’ center, most AIS mechanical thrombectomy procedures are performed with general anesthesia based on proceduralists’ preference. The 88 patients in this study included 25 who were intubated before arrival in the interventional radiology suite. In the entire cohort, mean age was 63±15 years, median National Institutes of Health Stroke Scale (NIHSS) score on admission was 16 (range=4 to 38), 82% of strokes were in the anterior circulation, 23% of patients received intravenous tissue plasminogen activator, and 53% received site-directed intra-arterial tissue plasminogen activator. Nineteen (22%) had a good outcome defined as a mRS<2 at 90 days, and 13 (15%) died before discharge. In those with a good outcome, admission NIHSS score was lower (14±7 vs. 18±8; P=0.04), a higher proportion used beta-blockers before stroke (66% vs. 33%; P=0.01), and more were extubated at the end of the procedure (90% vs. 26%; P<0.001). No patient who was intubated before arrival in the interventional radiology suite had a good outcome. Other demographic parameters, comorbidities, and anesthetic drug choices did not impact outcome. Proportion of patients and duration of time and with SBP<100 mm Hg were lower among those with good outcome, whereas rates and duration with SBP<140, <120, and <110 mm Hg were not associated with rate of good outcome. These observed associations may be due to the small sample size although the authors suggest that further study is required to confirm the relationship of blood pressure thresholds and outcome in patients with AIS. Duration of time with ETCO2<30 mm Hg or minimum ETCO2 also did not modulate outcome in this study. Although longer duration of time with ETCO2>40 mm Hg had a nonsignificant tendency toward higher rates of good outcome, a higher maximum ETCO2 occurred in the group with good outcome (49±8 mm Hg) compared with that with poor outcome (45±7 mm Hg; P=0.02). It is possible the elevated ETCO2 may serve to dilate collateral vessels improving flow to the ischemic penumbra. After correcting for age and admission NIHSS score, higher maximum ETCO2 (odds ratio [OR]=1.14; 95% CI=1.02-1.28; P=0.02) and extubation at the end of the procedure (OR=26; 95% CI=5-144; P<0.0001) were independently associated with good outcome. Excluding those patients who were intubated before arrival and correcting for age and admission NIHSS score, only extubation at the end of the procedure was independently associated with good outcome (OR=13, 95% CI=3 to 69; P=0.002).
Blood pressure is often lower in patients who receive general anesthesia for AIS thrombectomy compared with those who receive monitored anesthesia care, a factor that has been considered a contributor to the poor outcomes associated with general anesthesia in observational or retrospective studies.80 Rasmussen and colleagues performed a post hoc analysis of patient data from the GOLIATH trial to estimate the magnitude of the effect of intraprocedural blood pressure management, if any, on outcome following AIS thrombectomy.81 Hemodynamic management goals before revascularization in GOLIATH were to maintain SBP>140 mm Hg and MAP>70 mm Hg. Average MAP was lower in the general anesthesia group (95±8 mm Hg) compared with the monitored anesthesia care group (101±12 mm Hg; P<0.001). Average SBP was also lower in the general anesthesia group (143±15 vs. 155±20 mm Hg in the monitored anesthesia care group; P<0.001) Blood pressure variability was calculated as the average absolute difference in consecutive SBP, diastolic blood pressure, and MAP and as the difference between baseline MAP and the average of all MAPs during the procedure. Significantly greater blood pressure variability occurred in those who received general anesthesia by all metrics of variability. Despite these differences, a >20% reduction in MAP, duration with MAP<90, <80 or <70 mm Hg, duration with SBP<140 mm Hg, blood pressure variability, or differences in minimum or maximum MAP or SBP were not associated with either a shift in mRS or odds of mRS<2 at 90 days in either the entire cohort or those in whom reperfusion was established. These findings should be interpreted with caution as GOLIATH was not powered to detect differences in the impact of blood pressure on outcome, nor was blood pressure management randomized in this study. Further, the decrease in blood pressure associated with general anesthesia may be offset by a greater decrease in cerebral metabolic rate with general anesthesia compared with monitored anesthesia care. Finally, hypotension was very common in this study cohort; 94% and 62% of patients has at least 1 episode of SBP<140 mm Hg in the general anesthesia group and monitored anesthesia care groups, respectively.
Two recent review articles on AIS will be of interest to the reader.82,83 Bernhardt and colleagues reviewed advances in stroke published in the 2017 literature, including advances in rehabilitation, critical care, emerging therapies, imaging, population studies, and health policy related to AIS.82 Wang and Abramowicz reviewed issues related to anesthesia and endovascular management of AIS, addressing patient selection and preparation, imaging, timing, and anesthetic technique.83
Despite significant advances in the prevention and treatment of cerebral aneurysm rupture and management of rupture-related complications, aneurysmal subarachnoid hemorrhage (SAH) remains a significant cause of permanent neurological deficits and death. Guidelines for the management of patients with aneurysmal SAH have been developed in recent years, but it is unclear how these have affected practice. Gritti and colleagues identified 11 sets of guidelines published before November 2016, and summarized these guidelines in their recent manuscript.84 They also identified data derived from 10 clinical practice surveys and reviewed the effect of guidelines on 5 major management parameters: (1) use of nimodipine; (2) timing of aneurysm treatment; (3) aneurysm treatment modality; (4) use of prophylactic anticonvulsants, and; (5) use of “triple H” therapy (hypertension, hypervolemia, hemodilution) for cerebral vasospasm. From the mid-1980s to present there was: (1) an increase in the use of nimodipine; (2) a move toward earlier treatment of ruptured aneurysms; (3) an increase in coiling and a decreasing in clipping for aneurysm treatment; (4) a decrease in the use of prophylactic anticonvulsants, and; (5) static use of triple H therapy, but with a dramatic decrease in the collective use of this treatment modality in very recent years with induced hypertension being used more frequently without concomitant hypervolemia and hemodilution.85
Dexmedetomidine is a drug frequently used to provide sedation in the ICU. In a rat model of SAH, dexmedetomidine was associated with decreased cerebral edema and improved neurological outcome.86 Okazaki and colleagues retrospectively identified 161 patients with aneurysm SAH who required sedation to facilitate mechanical ventilation.87 Patients were stratified into 3 groups—those who: (1) did not receive dexmedetomidine (n=75); (2) received “low-dose” dexmedetomidine defined as an average daily dose of 0.01 to 0.2 μg/kg/h (n=48), and; (3) received “high dose” dexmedetomidine defined as an average daily dose of >0.2 μg/kg/h (n=38). Good outcome, defined as mRS<2 at discharge, was independently associated with low dose (OR=3.17, 95% CI=1.24-8.53; P=0.02) but not high dose dexmedetomidine use (OR=0.75, 95% CI=0.25-2.16; P=0.59). Compared with the group that did not receive dexmedetomidine, serum lactate concentration was significantly lower in the low-dose dexmedetomidine group, but not the high-dose dexmedetomidine group, at 24 hours and 36 hours after admission. It is possible that dexmedetomidine attenuates the hypercatecholaminergic state commonly associated with SAH and that hypotension, likely associated with the higher dose dexmedetomidine, offsets this effect, accounting for the findings of this study. Alternatively, as this was a retrospective study, selection bias may have played a role affecting choice and dose of sedative drugs. The authors did not report rates of vasospasm or delayed cerebral ischemia (DCI), the functional manifestation of cerebral vasospasm, ICU length of stay, or hospital length of stay stratified by dexmedetomidine group.
Biomarkers that could stratify patients based on risk for poor outcome following SAH would be of significant clinical utility. Some biomarkers involve nonstandard laboratory testing of blood samples, whereas other may require sampling of cerebrospinal fluid.88 Red blood cell distribution width is calculated as the ratio of the range of corpuscular volume to mean corpuscular volume and multiplying by 100%, with normal values in the range of 10.9% to 13.4%. An increase in red cell distribution width occurs in various states of anemia but can also be associated with inflammatory and prothrombotic states, and has been shown to correlate with poor outcome in patients with AIS.89–91 Fontana and colleagues report on the utility of red blood cell distribution width, a parameter often reported with a complete blood count, as a potential biomarker for the stratification of outcome in patients with SAH.92 The records of 270 patients with SAH were reviewed. Elevated red cell distribution width occurred in 177 (66%) of patients on admission, and in 217 (80%) at some point during ICU stay. Increased red cell distribution width during ICU stay, but not at admission, was independently associated with increased odds of poor outcome defined as a Glasgow Outcome Score of 1 to 3 (death, vegetative state, and severe disability) at 90 days. Odds of poor outcome with increased red cell distribution width during ICU stay was 1.618 (95% CI=1.213-2.158; P=0.001). Increased red cell distribution width was also independently predictive of in-hospital mortality (OR=1.464, 95% CI=1.057-2.028; P=0.022) but not DCI (OR not provided).
Significant changes in cerebral blood flow (CBF) occur soon after SAH. Engquist and colleagues characterized these changes using bedside xenon computerized tomography (CT) in 64 patients with SAH who required mechanical ventilation.93 Xenon is an inert lipid-soluble gas that readily passes the blood-brain barrier and is opaque on CT imaging. Therefore, regional and global CBF are proportional to increasing opacity during inhalation of xenon. Imaging was performed within 3 days of SAH and all but 4 patients underwent either clipping (n=11; 17%) or coiling (n=49; 77%) of the ruptured aneurysm. Median global CBF was 35 (IQR=27 to 42) mL/100 g/min. The cerebral cortex was divided into 60 regions of interest. Of 64 patients, 43 (67%) had at least 1 region with CBF<20 mL/100 g/min and 10 (16%) at least 1 region with CBF<10 mL/100 g/min. There was no association of decreased regional CBF with aneurysm location, and no significant difference between hemispheres. Global CBF was higher in younger patients but there was no association with admission Hunt Hess Score. This latter effect may have been due to the development of complications during hospitalization, an effect that may have not been predicted by admission neurological status. One limitation of this study may be that the authors did not determine if differences in global or regional CBF existed between those who underwent clipping versus coiling. In addition, only patients who required mechanical ventilation were included, so those with less severe SAH were likely not included.
In a subsequent study, Engquist and colleagues used xenon CT to characterize the effect of “triple H” therapy in 48 patients with SAH who required mechanical ventilation.94 All patients underwent imaging on days 0 to 3, 4 to 7, and again between days 8 and 12. Sedation was provided with propofol and morphine. If patients developed DCI, blood pressure was increased to maintain SBP>140 mm Hg with dobutamine with or without norepinephrine, and mild hypervolemia was induced with albumin. In those with DCI, there was no significant difference in SBP before and following institution of triple H therapy, and values were also similar in patients without DCI on days 0 to 3 and 5 to 8. In patients with DCI, mean hematocrit decreased from 36.4% (95% CI=34.7-38.0%) to 31.7% (95% CI=30.2%-33.3%; P<0.001) with institution of triple H therapy. Global CBF increased from 29.5 mL/100 g/min (95% CI=24.6-33.9 mL/100 g/min) to 38.4 mL/100 g/min (95% CI=27.0-41.2 mL/100 g/min; P=0.001) in those with DCI following triple H therapy. The cerebral cortex was divided into 60 regions of interest, and the percentage of regions of interest with CBF<20 mL/100 g/min decreased from 26.2% to 8.6% (P=0.019). However, and probably most importantly, regional CBF in the region of interest with the lowest regional CBF increased from 19.6 mL/100 g/min (95% CI=15.0-24.2 mL/100 g/min) to 27.3 mL/100 g/min (95% CI=17.8-34.1 mL/100 g/min; P=0.006) indicating that triple H therapy can improve regional blood flow to the most poorly perfused regions. Unfortunately, the authors did not confirm that the most poorly perfused region corresponded to a region perfused by a spastic artery. In patients without DCI, there was no difference in SBP, hematocrit, global CBF, or regional CBF patterns at the 3 time intervals.
Haegens and colleagues retrospectively identified 1647 patients with aneurysmal SAH of whom 479 (29%) developed signs of cerebral vasospasm indicated by a decrease in Glasgow Coma Score by >1 point or new focal neurological deficit after other potential causes were ruled out.95 The authors then excluded 179 (37%) patients who had infarction already present on CT at the onset of, or very early in the course of, symptomatic vasospasm. The primary goal of the investigation was to determine if induced hypertension in patients with symptomatic vasospasm reduced the incidence of subsequent infarction or poor outcome. Of the remaining 300 patients, hypertension was induced in 201 (67%) by increasing SBP by 20% with further increases titrated to neurological examination findings up to a SBP of 220 mm Hg. Subsequent cerebral infarction was present in 20% and 33% of patients with and without induced hypertension, respectively (P=0.015). Rates of poor clinical outcome (mRS>3 at 3 mo) were 36% and 74% in those with and without induced hypertension, respectively (P<0.05). These differences remained significant even after correcting for demographics, quantity of subarachnoid blood, presence of hydrocephalus, aneurysm location, and treatment. The authors did not compare the characteristics of patients in whom induced hypertension was and was not used; these groups may have differed, potentially contributing to differences in outcome.
TRAUMATIC BRAIN INJURY (TBI)
TBI is major public health problem and source of significant mortality and disability among survivors. Management of patients with TBI is often focused on minimizing secondary injury. During the acute phase following severe TBI, patients require management in an ICU. McCredie and colleagues conducted a survey of trauma medical directors enrolled in the American College of Surgeons Trauma Quality Improvement Program to assess whether either management in a dedicated neurological ICU or use of a standardized management protocol in a general ICU has an impact on mortality among patients with severe TBI.96 Survey data were linked to the Trauma Quality Improvement Program registry to identify patient outcome data from survey respondents. One hundred thirty-four centers responded to the survey. Of these, 50 (37%) were dedicated neurological ICUs with the remaining being a general ICU that also managed patients with TBI. Outcome data were obtained from 9773 patients of whom 4193 (43%) were managed in a dedicated neurological ICU. Rates of tracheostomy, pneumonia, length of ICU or hospital stay, or mortality did not differ based on whether care was provided in a dedicated neurological or general ICU. However, patients managed in either a neurological or general ICU that adhered to a standardized management protocol for patients with severe TBI had a significantly lower risk-adjusted odds of in-hospital mortality (OR=0.77, 95% CI=0.63-0.93; P=0.009) compared with those managed in a general ICU without a standardized protocol.
Hemodynamic instability following TBI can be an important source of secondary brain injury. Artefactual and aberrant values of vital signs are not uncommon and can lead to unnecessary treatments or inadvertent withholding of otherwise necessary treatments. Kim and colleagues utilized a deep belief network machine learning algorithm to evaluate the electrocardiogram and arterial pressure waveforms for artefactual values of HR and blood pressure, respectively, in 99 patients with TBI.97 These artifacts could include aberrant signals on the electrocardiogram due to patient movement, or damping of the arterial pressure waveform. Hypertension was defined as a SBP>140 mm Hg or diastolic blood pressure >90 mm Hg, and hypotension as either a SBP<90 mm Hg or SBP<110 with the latter based on revised TBI management guidelines.98 Tachycardia and bradycardia were defined as HR>140 and <40 bpm, respectively. A deep belief network algorithm was effective at decreasing the incidence of hypotensive (SBP<90 mm Hg) and tachycardia episodes by 48% and 13% respectively, although the number of hypertensive or bradycardic episodes was not significantly reduced with this technique.
In patients with severe TBI, early nutritional support will meet the increased caloric needs following injury and is associated with improved outcome and decreased mortality.99 Early enteral nutrition is often supplied by nasogastric tube. In patients with prolonged inability to eat due to altered consciousness or dysphagia, percutaneous gastrostomy is often implemented to provide enteral nutrition, although the optimal timing of gastrostomy is not certain. Chaudhry and colleagues identified 96,625 patients who suffered TBI between 2011 and 2013 in the Nationwide Inpatient Sample, of whom 3343 received a percutaneous gastrostomy.100 Timing of gastrostomy placement was stratified into those who received it early (≤7 d after admission; n=877 [26%]), standard (8 to 14 d after admission; n=1300 [39%]), and late (>14 d after admission; n=1166 35%]). Patients with greater comorbidities and a more complicated postinjury course tended to receive later gastrostomy placement. Timing of gastrostomy placement correlated with hospital length of stay, with those having late placement having a longer hospital stay. However, highest mortality was observed in those with either early or late placement. The authors state that high comorbidities and high mortality was an expected finding in the late gastrostomy group as comorbidities and complications likely led to late placement and also to increased mortality risk. The authors attributed the high mortality with low comorbidity burden in the early placement group to less overt physiological derangement and gross complications leading to early transfer out of ICU and subsequent less vigilant surveillance for postinjury complications. The retrospective nature of this investigation leads to potential bias, and the effect of timing of percutaneous gastrostomy tube placement after TBI is probably best investigated prospectively.
Anemia is common in TBI patients and results from multiple etiologies including hemodilution, anemia of chronic illness, and TBI-associated coagulopathy.101,102 Clinicians must balance improving oxygen delivery to injured tissues by treating anemia with red cell transfusion against the risks associated with transfusion. Boutin and colleagues retrospectively reviewed hemoglobin concentration profile and transfusion practices in 215 patients admitted to a single institution following moderate or severe TBI.103 The greatest decrease in hemoglobin concentration occurred during the first 24 hours after injury. Sixty-six (31%) patients received a transfusion during their ICU stay with 80% of transfusions occurring within 3 days of admission; 49% of transfusions were given in response to a blood hemoglobin concentration of ≤7.5 g/dL. In-hospital mortality rate was higher among those who received a transfusion (41% vs. 24% in those that did not receive a transfusion; P=0.011), even after correction for confounding variables. Transfusion was independently associated with increased risk for neurological complications (relative risk [RR] 3.4, 95% CI=1.4-8.6; P<0.05) and with longer ICU and hospital lengths of stay. This association between transfusion and poor outcome is likely multifactorial and includes greater injury severity resulting in greater need for transfusion as well as increased risks associated with transfusion.
Traditionally, glucose has been considered the primary energy substrate for the brain. However, recent data suggests that ∼10% of the energy production of the healthy brain is derived from lactate and that this fraction can increase in certain disease states.104 To study the effect of hypertonic sodium lactate solution on outcome after TBI, Millet and colleagues assigned adult rats to 4 groups based on injury (sham vs. TBI) and treatment (hypertonic sodium lactate vs. isotonic sodium chloride).105 Thirty minutes after TBI or sham injury, rats received either 0.9% sodium chloride ([Na]=154 mEq/L) or 11.2% hypertonic sodium lactate ([Na]=1000 mEq/L) intravenously at 0.5 mL/kg/h for 3 hours. Compared with sham-treated animals, those that received TBI and saline had reduced cortical diffusion coefficients (determined by diffusion-weighted MRI) at 2 and 4 hours. This effect was attenuated in the hypertonic lactate group suggesting an attenuation of brain edema formation by this treatment. In rats subjected to TBI, compared with those that received saline rats that received hypertonic sodium lactate had improved oxygen utilization by mitochondria, higher brain tissue oxygen saturation and reduced mitochondrial structural changes. Unfortunately, the authors did not include a study group where animals received sodium chloride solution of similar osmolality to hypertonic sodium lactate; some of the reported effects may potentially be attributed to effects on reduction in brain edema improving brain microcirculation.
Because of concerns for potential exacerbation of intracranial hypertension, ketamine has often been avoided in patients with acute TBI. However, recent data support ketamine use not only as a treatment of intracranial hypertension in patients with TBI but also as a neuroprotectant.106,107 Following TBI, there is glial cell activation and proliferation as well neurogenesis in the hippocampus, an effect dependent on n-methyl-d-aspartate receptor activation.108,109 Peters and colleagues characterized the effect of ketamine exposure on hippocampal cell proliferation and functional outcome in a mouse model of TBI.110 Mice underwent either controlled cortical impact or sham injury and received either ketamine at a subsedative dose (30 mg/kg/d) or saline continuously via subcutaneous infusion for up to 7 days. Ketamine after TBI was associated with an increase in hippocampal cell proliferation at 3 days after injury. When new cells were stained for cell type, ketamine exposure after TBI inhibited both neurogenesis and production of astrocytes but enhanced the production of microglial cells, a surprising finding given the anti-inflammatory effects associated with ketamine.111 Ketamine had no significant effect on injury severity as measured by cortical cavitation, rates of apoptosis measured by cleaved caspase-3 staining, or on subclinical seizure activity measured by c-fos staining. Four weeks after injury, a cohort of mice underwent behavioral testing with the Morris water maze. Mice that sustained TBI initially had slower learning curves for the Morris water maze but there was no difference in performance by day 3. However, after learning the consistent location of the platform, the location was moved to a different quadrant to test neurogenesis-dependent hippocampal function,112 and mice that sustained TBI and did not receive ketamine had a delay in identifying the new location. This effect was attenuated in mice that sustained TBI but received ketamine. The authors attribute these findings to reduced glial scar formation or enhanced neuronal pruning by microglia.
Noninvasive Cerebral Monitoring
Accurate measurement of CBF can be performed with CT but requires patient transport to scanning suites. A noninvasive and reliable bedside technique to measure CBF is needed to guide management of critically ill patients. Currently, transcranial Doppler sonography (TCD) can be used to estimate CBF, but windows may not be accessible in all patients. Ultrasound-tagged near infrared spectroscopy (UTNIS) is a newer technology that may be used to noninvasively estimate regional CBF. Briefly, a sensor that emits both coherent monochromatic light and ultrasound waves is placed on the forehead. As the light is scattered by moving blood cells, it undergoes a shift in frequency due to the Doppler effect. However, the phase of light waves is shifted by the ultrasound waves allowing resolution of the depth that the photons have traveled; photons that have traveled to deeper tissue have a greater ultrasound-induced phase shift. Lipnick and colleagues compared changes in blood flow velocity in response to changes in carbon dioxide tension in healthy volunteers as measured with TCD and UTNIS.113 Middle cerebral artery blood flow velocity (measured by TCD) and CBF index (measured by UTNIS) were determined at eucapnia and during graded hypocapnia and hypercapnia in spontaneously breathing healthy volunteers induced by hyperventilation and inhalation of oxygen supplemented with carbon dioxide, respectively. With a mean nadir ETCO2 of 17±2 mm Hg, there was a decrease in middle cerebral artery blood flow velocity to 79%±22% of baseline (measured at eucapnia) but no change in CBF index as measured by UTNIS (101%±6% of baseline). With hypercapnia, peak ETCO2 was 59±3 mm Hg, and middle cerebral artery blood flow velocity increased by 153%±25% of baseline compared with an increase in middle CBF index as measured by UTNIS of 119%±11% compared with baseline. In addition, Caccioppola and colleagues showed that UTNIS indicated the presence of brain perfusion in those recently diagnosed with brain death.114 Collectively, these data suggest that UTNIS has limitations and requires additional refinement.
The gold standard technique to measure ICP is the ventriculostomy. Safer, accurate, and less invasive techniques to measure ICP are sought, and include TCD, optic nerve sheath diameter, and tympanic membrane displacement.115–117 Ganslandt et al118 report on the utility of a novel technique to noninvasively estimate ICP. The HS-1000 device (HeadSense Medical, Israel) produces an acoustic wave in one ear canal and the sound is modulated by cranial contents as it passes through the cranium and before detection in the opposite ear canal. The manufacturer claims that ICP can be estimated by changes in the acoustic wave properties. The authors recruited 14 patients who required ICP monitoring via ventriculostomy and simultaneously estimated ICP with the HS-1000 device collecting data for 30 to 60 minutes per patient at ∼4 measurements per minute and generating 2543 simultaneous data points. Data were obtained when the ventriculostomy was closed to drainage. Mean value of ICP measured via ventriculostomy was 10.0±6.1 mm Hg (range=0 to 26 mm Hg) and with the HS-1000 was 9.5±4.7 mm Hg (range=0 to 21 mm Hg). Of all ICP values measured with the HS-1000, 63% and 85% were within ±3 and ±5 mm Hg of those measured via ventriculostomy, respectively. The area under the curve for a receiver characteristic curve analysis was 0.895, indicating strong estimation of ICP by the HS-1000 device.
Intraoperative Electrophysiological Monitoring
Evoked potential (EP) monitoring is often used to assess tract integrity during a variety of surgical procedures, including spine and intracranial surgery and carotid endarterectomy. Nunes and colleagues recently reviewed these techniques, and the pharmacologic and physiological influences on monitored EP waveforms.119 EP monitoring requires establishment of a baseline waveform after induction of anesthesia, but before surgical manipulation, to which subsequent waveforms are compared to assess for changes. Kim and colleagues retrospectively reviewed the records of 345 adult patients who underwent motor evoked potential (MEP) recording during spine surgery to determine factors associated with limited ability to record an adequate baseline waveform.120 MEPs were performed with transcranial electrical stimulation with a train of 5 pulses each of 0.05 ms in duration with an interpulse interval of 1 to 2 ms and intensity of 250 to 500 V. Of the patients having cervical (n=171), thoracic (n=121), and lumbar (n=47) spine surgery, adequate baseline MEP tracings were not obtained in 36%, 21%, and 9%, respectively (P<0.001). For patients having cervical spine surgery, abnormal intramedullary T2-weighted signal and preoperative motor weakness were independently associated with inability to obtain baseline MEP recording. For those having thoracic surgery, male sex and preoperative lower extremity motor deficits were independently associated with inability to acquire adequate baseline MEP waveforms. Interestingly, on univariate or multivariate analysis no factor was associated with inability to acquire adequate baseline MEP waveforms in patients having lumbar spine surgery. The authors attribute the association with male sex in those having thoracic spine surgery to a higher prevalence of disk and ligamentous disease in men.121 However, limited study power may also be a factor as only 47 patients having lumbar surgery were included in this series.
The association between the presence of preoperative motor deficits and the acquisition of baseline MEP signals was also reported by Guo et al.122 Transcranial electric MEPs were prospectively attempted in 178 lower limbs of 89 patients having resection of thoracic intramedullary spinal cord tumors. All patients had motor strength testing before surgery graded on a scale of 5 (normal strength) to 0 (no motor function). Adequate baseline MEP waveforms were obtained in 100%, 90%, 25%, 13%, and 0% in those with motor strength scores of 5, 4, 3, 2, ≤1, respectively.
Chen and colleagues retrospectively reviewed the records of 631 adult patients who had somatosensory evoked potential (SSEP) monitoring during either intracranial or spine surgery for factors associated with the inability to obtain adequate baseline waveforms in the lower extremities.123 SSEP waveforms were generated with a signal intensity of up to 60 mA at a rate of 1.69 to 4.47 Hz with a square wave pulse width of 100 to 300 μs. Factors independently associated with failure to obtain adequate baseline lower extremity SSEP waveforms were increased patient height and weight, lower extremity edema, and preoperative lower extremity neurological examination abnormalities. Cranial versus spine surgery had no impact, but it was unclear if the authors assessed whether location of spine surgery (ie, cervical, thoracic, lumbar) impacted the ability to acquire adequate lower extremity baseline waveforms. Further, the authors did not stratify type of neurological deficit (ie, sensory, motor, reflex) when assessing the association between presence of preoperative neurological deficits and ability to acquire baseline SSEP waveforms.
Reddy and colleagues performed a meta-analysis that included both prospective and retrospective studies that assessed the utility of SSEP monitoring during carotid endarterectomy.124 Overall, 25 studies were identified that included 8307 patients. Fifty-four percent had symptomatic carotid stenosis before surgery and, of these, 13% experienced intraoperative SSEP changes during carotid occlusion. The incidence of stroke within 30 days of surgery in patients with SSEP changes was 10.7% compared with a rate of 0.97% in those without SSEP changes. The authors stratified signal changes by severity and response to shunting. Specifically, a signal change referred to a >50% decrease in amplitude or >10% increase in latency with a waveform still evident, whereas signal loss referred to complete loss of waveform. Transient changes were those that responded to shunting, whereas permanent changed persisted despite shunting. Transient changes, persistent changes, transient losses, and persistent losses were found in 86.2%, 3.2%, 7.4%, and 3.3%, respectively, of cases with SSEP changes. Thirty-day stroke rates were found in 4.2%, 47.6%, 21.7%, and 92.0% of patients who had transient changes, persistent changes, transient losses, and persistent losses, respectively. Overall, most (79%) strokes occurred either intraoperatively or during the first 24 hours after surgery, with the remaining (21%) occurring after 24 hours but within 30 days of surgery. Collectively SSEP monitoring had a sensitivity of 94% and specificity of 61% for predicting 30 days stroke rate after carotid endarterectomy.
Marino and colleagues compared intraoperative monitoring for cerebral ischemia during carotid endarterectomy with the combination of SSEP and MEP during general anesthesia (n=231) and neurological examination in those having an asleep-awake-asleep technique (n=100).125 Briefly, all patients received general anesthesia, tracheal intubation, and maintenance of general anesthesia with propofol and remifentanil. In the monitored group, general anesthetic doses were not changed during carotid occlusion, and monitoring for cerebral ischemia was accomplished with both SSEP and MEP monitoring. In the asleep-awake-asleep technique, propofol dose was significantly reduced while remifentanil was continued at 0.12 μg/kg/min with only minor adjustments, and cerebral ischemia monitoring was conducted via patient cooperation with a motor examination or the presence of agitation. Speech could not be assessed as the patients were intubated. Criteria for shunting were a reduction in amplitude of contralateral SSEP by >50% or a >25% decrease of baseline amplitude of contralateral MEP in the general anesthesia group, or development of motor deficits in the asleep-awake-asleep group after induced hypertension failed to resolve changes in both groups. Ischemic changes with carotid occlusion occurred in 25 of 231 (11%) and 15 of 100 (15%) of patients having general anesthesia and the asleep-awake-asleep technique, respectively, before blood pressure augmentation (P=0.28). After blood pressure augmentation, 13 of 231 (6%) and 12 of 100 (12%) still had persistent evidence of ischemia and required shunting in the general anesthesia and asleep-awake-asleep groups, respectively (P=0.02). Overall, surgical procedure duration was shorter in those having general anesthesia with EP monitoring. Unfortunately, the authors did not report changes in MEP and SSEP separately, only the fraction of patients with ischemic changes by either modality. Also, reported rates of shunting in the previous literature are lower in those having neurological assessment as a means to monitor for ischemia versus with surrogate means that are used in patients having general anesthesia.126
NEUROTOXICITY AND PERIOPERATIVE DISORDERS OF COGNITIVE FUNCTION
Exposure of animals to common anesthetic drugs at an early age is associated with widespread apoptosis in the brain.127 The interested reader is referred to a brief review by Jevtovic-Todorovic and Brambrink that summarizes some of the potential neuromodulatory effects of anesthetics that can account for neurotoxicity in the developing brain.128 The authors review anesthetic effects on brain microstructure, development, neuronal plasticity, and epigenetic modulations. Recent animal studies suggest that these deficits may persist into adulthood,129,130 and may even be passed on to subsequent progeny.131 Retrospective studies in humans exposed to general anesthesia at a young age suggest increased risk for developmental and behavioral disorders in those who received >1 general anesthetic.132–134 However, 2 recent well-designed trials in humans failed to show an association between early anesthetic exposure and subsequent neurodevelopmental deficits.135,136 In the Mayo Anesthesia Safety in Kids (MASK) trial, children born in Olmsted County Minnesota between 1994 and 2007 were retrospectively identified and stratified into 3 cohorts based on anesthesia exposure before the age of 3 years: (1) no anesthesia; (2) a single anesthetic exposure, and; (3) multiple anesthetic exposures.137 Study participants at age ranges of either 8 to 12 years or 15 to 20 years prospectively underwent neuropsychometric testing. A total of 997 subjects completed testing, consisting of 411, 380, and 206 in the unexposed, singly exposed, and multiply exposed groups, respectively. Samples were weighted in an effort to correct for differences in characteristics among groups including demographics, socioeconomic status, and medical comorbidities. There was no difference in the primary outcome measure, score on the Wechler Abbreviated Scale of Intelligence Full Scale, among groups. Children who received multiple anesthetic exposures scored lower on tests of processing speed and fine motor skills compared with those without an anesthetic exposure. Those with a single exposure scored similar to the unexposed group. One major limitation of the MASK trial, similar to other nonrandomized trials, is the issue of bias. Specifically, children who receive anesthesia, especially those who receive multiple anesthetics, are likely “different” from those who did not receive anesthesia. Although there were attempts to mathematically correct for these differences, correction is potentially incomplete.
As stated earlier, exposure of young animals to most sedative hypnotic drugs is associated with the development of widespread apoptosis in the brain. Unlike other drugs, dexmedetomidine has not been associated with apoptosis in the young brain,127 although its efficacy limits it from being used as a sole drug to maintain general anesthesia. Atluri and colleagues report on the utility of the neuroactive steroid (3β,5β,17β)-3-hydroxyandrostane-17-carbonitrile (HACN) to produce the state of general anesthesia without inducing apoptosis in the developing brain.138 The structure of HACN is depicted in Figure 1. All drugs were administered intraperitoneally in this study. The ED50 for loss of the righting reflex for HACN was determined in 7 day old rats and was found to be 39±4 mg/kg. The ED50 of ketamine and ketamine dissolved in the vehicle of HACN were determined to be 67±2 and 64±2 mg/kg, respectively. Ketamine at doses of 40 mg/kg administered as 6 doses every 2 hours to 7 days old mice leads to widespread apoptosis;139 an equipotent dose of HACN was calculated to be 10 mg/kg. Administration of 40 mg/kg ketamine and 40 mg/kg ketamine in HACN vehicle every 2 hours for 6 doses led to widespread increases in cleaved caspase-3, a biomarker for apoptosis, in multiple brain regions. Widespread increases in cleaved caspase-3 were not observed in 7 days old mice that received 10 mg/kg HACN every 2 hours for 6 doses. Mice exposed on day 7 of life to ketamine and ketamine in HACN vehicle showed significant impairment in performance in a radial arm maze on day 45 to 70 of life compared with unexposed animals whereas animals that received HACN performed similar on a radial arm maze to unexposed animals. The authors performed a series of pharmacodynamic studies showing that HACN antagonized T-type voltage-gated calcium channels in the dorsal root ganglion and thalamus resulting in decreased neuronal firing. HACN also acted presynaptically to decrease glutamate release in α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) but not n-methyl-d-aspartate glutaminergic synapses, and also decreased GABA release. Further study will be required to understand the anesthetic effects, pharmacology, and physiological effects of this novel drug.
Delirium is an acute condition characterized by fluctuations in mental status that can include alterations in consciousness and level of activity, inattention, and confusion. Delirium is common among older patients in the perioperative period and can increase hospital length of stay, long-term complications, and cost of care.140,141 Although some risk factors, such as older age and psychiatric history, are not modifiable, the search for modifiable risk factors continues in an effort to reduce risk for delirium in the periprocedural period. Weinstein and colleagues retrospectively identified 41,766 patients who underwent total hip or knee arthroplasty at a single institution between 2005 and 2014 to determine anesthetic factors associated with increased risk for delirium.142 Overall, 922 (2.2%) of patients developed delirium that was documented in the medical record, likely an underestimate of the true incidence. Patients who had neuraxial anesthesia had lower odds of developing delirium compared with those who received general anesthesia. Those who received ketamine, either intraoperatively or as an infusion postoperatively, or intraoperative opioids had increased odds of developing delirium. Benzodiazepines administered postoperatively increased odds for delirium, but not when administered intraoperatively. The authors hypothesize that the reasons for the lack of effect by intraoperative benzodiazepines may be related to doses being too small to modulate an effect and suggest future investigations might consider evaluating a dose effect.
Duan and colleagues performed a meta-analysis of prospective trials that assessed the effect of dexmedetomidine on delirium risk in the perioperative period.143 They identified 18 studies that comprised 3309 patients of whom 1616 received dexmedetomidine. Overall, perioperative dexmedetomidine decreased odds for delirium postoperatively (OR=0.35, 95% CI=0.24-0.51) and was effective in both cardiac (OR=0.41, 95% CI=0.26-0.63) and noncardiac surgery patients (OR=0.33, 95% CI=0.18-0.59). Dexmedetomidine was also effective at reducing delirium risk in those younger than 65 years old (OR=0.19, 95% CI=010-0.36) and those 65 years old and above (OR=0.44, 95% CI=0.30-0.65). Dexmedetomidine had no effect on ICU of hospital lengths of stay, or in-hospital mortality.
Lee and colleagues randomized 354 patients above 65 years old and having laparoscopic abdominal surgery to receive either: (1) dexmedetomidine as a 1 μg/kg bolus followed by an infusion at 0.2 to 0.7 μg/kg/h for the duration of surgery; (2) dexmedetomidine as a 1 μg/kg bolus only before the end of surgery, or; (3) normal saline.144 The authors did not provide parameters for dosing adjustment of the dexmedetomidine infusion in group 1. All patients received a standard general anesthetic with desflurane titrated to a BIS of 40 to 60 with analgesia provided with ketorolac and patient-controlled morphine with a basal continuous infusion. Screening for delirium was performed with the Confusion Assessment Method for up to 5 days after surgery. The incidences of delirium were 9.5%, 18.4%, and 24.8% in groups 1, 2, and 3, respectively, with rates of delirium only significantly lower in group 1 versus group 3 (P=0.03). Pain scores were also significantly lower 24 hours after surgery in group 1 compared with scores in the other groups (P<0.05). Dexmedetomidine decreased serum interleukin-6 concentrations at both 1 and 24 hours after surgery compared with those measured in the saline group, with the greatest reduction in group 1. These data suggest that a continuous infusion of intraoperative dexmedetomidine leads to greater reduction in risk for postoperative delirium compared with a single dose.
Haloperidol is often used to treat acute agitation. In a recent meta-analysis by Shen and colleagues, haloperidol at doses of ≥5 mg/d reduced the odds of delirium postoperatively without affecting hospital length of stay or mortality.145 Van den Boogaard and colleagues report on a multicenter randomized trial conducted in 21 ICUs in the Netherlands that assessed the utility of haloperidol at reducing risk of mortality and delirium.146 One thousand, seven hundred eighty-nine patients without delirium upon admission to ICU were randomized to receive either haloperidol 1 mg 3 times per day, haloperidol 2 mg 3 times per day, or saline placebo, all administered intravenously. Patients who were delirium free at 4 days received their assigned drug for 4 more days. There was no difference in the primary outcome measure, the number of days survived measured at 28 days. There was also no difference in the incidence of delirium, delirium free days, coma free days, duration of mechanical ventilation, and both ICU and hospital length of stay among survivors.
Daniels and colleagues retrospectively identified 449 patients who developed delirium during management in an ICU.147 Patients were stratified into 4 groups based on drugs used to treat delirium, those who received: (1) melatonin but not antipsychotic medications (n=39, 9%); (2) antipsychotic medications (aripiprazole, haloperidol, quetiapine; n=111, 25%) but not melatonin; (3) both melatonin and antipsychotics (n=71, 16%), or; (4) neither melatonin nor antipsychotics (n=228, 51%). Patients treated with either melatonin or antipsychotics had a greater severity of critical illness (based on Acute Physiology and Chronic Health Evaluation III scores) and were more likely to be exposed to invasive or noninvasive mechanical ventilation, sedative medications, and restraints. After correction for covariates, use of only antipsychotic medications was associated with a longer length of hospital stay (OR=0.65, 95% CI=0.47-0.91; P=0.01). Pharmacologic treatment of delirium was not associated with duration of delirium, length of ICU stay, or 28-day mortality. Unfortunately, the authors were unable to compare efficacy in those with hyperactive or hypoactive delirium due to the small sample size.
MacKenzie and colleagues report on a meta-analysis of randomized trials that assessed the utility of intraoperative processed electroencephalogram monitoring at reducing risk for postoperative delirium, excluding any studies that evaluated the effect on postoperative cognitive dysfunction.148 Data from 13 studies consisting of 2654 patients demonstrated that intraoperative processed electroencephalogram monitoring reduced the odds of delirium by 38% (OR=0.62, 95% CI=0.51-0.76; P<0.001). The authors suggested that the idea that processed electroencephalography results in lower anesthetic doses and, thus, lower rates of delirium, is a misconception based on current data. Instead, they suggest that, as preexisting cognitive dysfunction is a risk factor for delirium, depth of anesthesia may be a marker for the brain’s vulnerability.
Postoperative Cognitive Function
Older adults are at risk for transient impairment in cognitive function after surgery and anesthesia. However, the effect of surgery and anesthesia on long-term cognitive function in older adults is not well studied. Schulte et al149 utilized data from subjects enrolled in the Mayo Clinic Study on Aging150 to assess the impact of anesthesia and surgery exposure on long-term cognitive function. Briefly, the Mayo Clinic Study on Aging is a cohort of residents from Olmsted County, Minnesota without a prior diagnosis of dementia at the time of enrollment who were followed long-term, including cognitive assessments conducted every 15 months. Schulte and colleagues included 1819 subjects aged 70 to 91 years at initial assessment of whom 1564 had normal cognitive function and 255 had mild cognitive impairment. They found that any exposure to anesthesia and surgery resulted in a greater decline in long-term cognitive function compared with those without exposure despite correction for greater comorbidities in those who received anesthesia. The greatest declines in the exposed group were in memory and attention/executive functions, with no significant differences in language and visuospatial function. Recent work published by Bratzke and colleagues in a younger patient cohort (mean age of 964 patients was 54 y with SD or range not reported) also showed that anesthesia exposure resulted in a long-term decline in memory suggesting that long-term cognitive effects of anesthesia and surgery may not be limited to elderly patients.151
Zhang and colleagues identified adults aged 65 to 90 years undergoing either intra-abdominal or intrathoracic cancer surgery without prior significant cognitive impairment or neurological disease.152 Patients were randomized to receive general anesthesia maintained with either propofol infusion or sevoflurane titrated to a BIS of 40 to 60, with a standardized analgesic regimen. Three hundred seventy-nine patients underwent a battery of 7 neurospychometric tests before surgery and again 1 week after surgery. Demographics were well matched between groups but intrathoracic surgery was more common in the propofol (18%) compared with sevoflurane group (3%; P<0.001). To correct for learning effects, a cohort of 59 patients who did not undergo surgery received the same battery of tests twice, with a 1 week intervening interval. A cognitive deficit was defined as a corrected z-score of <−1.96 on at least 2 tests. The incidence of cognitive deficits at 1 week after surgery was greater in the sevoflurane (23%) compared with the propofol group (15%; P=0.038). The authors also reported lower daily pain scores in the propofol group on the first and second postoperative days. Unfortunately, it was not clear if differences in pain scores and surgery location (intra-abdominal vs. intrathoracic) were taken into consideration in the data analysis. Further, the authors did not explicitly state that regional anesthesia was not used to treat postoperative pain, leading one to question if use of regional anesthetics, especially thoracic epidural analgesia in those having intrathoracic surgery, may have accounted for difference in pain scores and could have contributed to differences in performance on cognitive tests.
Zorrilla-Vaca and colleagues performed a meta-analysis of randomized trials that assessed the utility of cerebral near infrared spectroscopy at reducing risk of postoperative cognitive dysfunction.153 They identified 15 studies that comprised 2056 patients of whom 1018 had cerebral near infrared spectroscopy monitoring. Interventions were generally performed in response to either an absolute regional cerebral oxygen saturation <55% to 60% or to a decrease to <75% of baseline saturation. Interventions included treatment of systemic hypotension, optimization of blood oxygen content and carbon dioxide tension, and increasing cardiac bypass pump flow in patients having cardiac surgery. Use of near infrared spectroscopy resulted, overall, in a reduction in risk for postoperative cognitive dysfunction (RR=0.60, 95% CI=0.40-0.89; P<0.001), but this was significant only in patients who were having cardiac surgery (RR=0.55, 95% CI=0.36-0.86; P=0.009) and not in those having noncardiac surgery (RR=0.79, 95% CI=0.61-1.02; P=0.07). Cerebral near infrared spectroscopy was not associated with a reduced risk for postoperative delirium (RR=0.90, 95% CI=0.77-1.10; P=0.05).
Neuroinflammation is thought to play a significant role in mediating postoperative cognitive dysfunction. In an adult mouse model, Hu and colleagues showed that anesthesia and surgery resulted in cognitive deficits, as measured by reduced freezing time on fear conditioning, an effect that was attenuated by blockade of the interleukin-6 receptor and after depletion of bone marrow monocytes, a secretor of interleukin-6.154 Systemic administration of interleukin-6 without surgery and anesthesia also resulted in reduced freezing time, and activation of microglia and infiltration of monocytes in the hippocampus. Hu and colleagues also showed that dexmedetomidine prevents cognitive decline by attenuating neuroinflammation, suggesting that dexmedetomidine may have a role in reducing risk for postoperative cognitive dysfunction.155
Orexin is a neuropeptide produced by neurons located in the perifornical area and lateral hypothalamus. These neurons project diffusely throughout the brain and regulate wakefulness. To determine if the orexin system plays a role in emergence time from general anesthesia, Ran and colleagues studied emergence from isoflurane anesthesia in middle-aged adult (age 4 to 6 mo) and elderly (age above 20 mo) rats.156 Following induction of anesthesia with intraperitoneal pentobarbital and fentanyl, rats inhaled 1.4% isoflurane for 30 minutes. Median emergence time, measured as the return of the righting reflex, was longer in the elderly group (1082 s [range=1010 to 1133 s]) versus the adult group (848 s [range=829 to 938 s]; P=0.0009). The authors report a similar density of expression of orexin in the perifornical areas and hypothalamus between groups. However, elderly rats had decreased expression of the orexin-1 but not of the orexin-2 receptor compared with adult rats. The authors did not report if orexin and orexin receptor determinations were made in unanesthetized or previously anesthetized rats. In separate groups of elderly rats, a viral vector was used to transfect the gene for a fluorescent protein with or without the gene for the orexin-1 receptors. Emergence time from general anesthesia was significantly shorter in the rats that were transfected with the receptor for orexin-1 (770 s [range=576 to 929 s]) versus the group that were transfected only with the gene for a fluorescent protein (1153 s [range=914 to 1200 s]; P=0.03). The authors did demonstrate widespread expression of the fluorescent protein and increased expression of orexin-1 receptor in those animals transfected with the gene for the receptor. The orexinergic system may represent a target to minimize risk for delayed emergence in humans, and possibly a role in mediating postoperative cognitive dysfunction in the elderly.
The Perioperative Neurotoxicity Working Group published a narrative review in the December 2018 issue of Anesthesia and Analgesia.157 The group focused on preoperative consent, preoperative assessment and both intraoperative and postoperative management strategies to minimize risk for delirium and postoperative cognitive dysfunction in older adults. The group recommended that all patients above 65 years old should be informed of the risk for postoperative cognitive deficits, and that screening for preoperative cognitive deficits should be considered. They suggested using various drugs with caution in older patients due concern for increased risk for postoperative cognitive deficits. These are summarized in Table 2. They also recommend monitoring age-adjusted minimum alveolar concentration of inhaled anesthetic drugs, optimizing cerebral perfusion, and using EEG-based anesthetic management for older adults.
A recent review article by Evered and Silbert summarizes the history and epidemiology of postoperative cognitive dysfunction, its characteristics, etiology, and possible means to reduce risk for its development.158 In a separate series of articles, Evered and members of the Nomenclature Consensus Working Group report on recommendations for nomenclature for periprocedural disorders of cognition.159–162 Briefly, they address preexisting cognitive disorders, delirium and postoperative cognitive dysfunction, describing specific qualifying terms and recommending that nomenclature be consistent with terms in the Diagnostic and Statistical Manual for Mental Disorders—Fifth Edition.
1. Rock AK, Opalak CF, Workman KG, et al. Safety outcomes following spine and cranial neurosurgery: evidence from the National Surgical Quality Improvement Program. J Neurosurg Anesthesiol. 2018;30:328–336.
2. Kobayashi K, Ando K, Nishida Y, et al. Epidemiological trends in spine surgery
over 10 years in a multicenter database. Eur Spine J. 2018;27:1698–1703.
3. Rajaee SS, Bae HW, Kanim LE, et al. Spinal fusion in the United States: analysis of trends from 1998 to 2008. Spine (Phila Pa 1976). 2012;37:67–76.
4. Cheney FW, Posner K, Caplan RA, et al. Standard of care and anesthesia liability. JAMA. 1989;261:1599–1603.
5. Kutteruf R, Wells D, Stephens L, et al. Injury and liability associated with spine surgery
. J Neurosurg Anesthesiol. 2018;30:156–162.
6. Brown DA, Himes BT, Major BT, et al. Cranial tumor surgical outcomes at a high-volume academic referral center. Mayo Clin Proc. 2018;93:16–24.
7. de Almeida CC, Boone MD, Laviv Y, et al. The utility of routine intensive care admission for patients undergoing intracranial neurosurgical procedures: a systematic review. Neurocrit Care. 2018;28:35–42.
8. Cinotti R, Bruder N, Srairi M, et al. Prediction score for postoperative neurologic complications after brain tumor craniotomy
: a multicenter observational study. Anesthesiology. 2018;129:1111–1120.
9. Asmussen S, Maybauer DM, Chen JD, et al. Effects of acupuncture in anesthesia for craniotomy
: a meta-analysis. J Neurosurg Anesthesiol. 2017;29:219–227.
10. Mamdami J, Andrzejowski JC, Pullman M. The effect of acupuncture at the yintang point on preoperative anxiety levels in neurosurgical patients: a randomized controlled trial. J Neurosurg Anesthesiol. 2017;29:84.
11. Bai WY, Yang YC, Teng XF, et al. Effects of transcutaneous electrical acupoint stimulation on the stress response during extubation after general anesthesia in elderly patients undergoing elective supratentorial craniotomy
: a prospective randomized controlled trial. J Neurosurg Anesthesiol. 2018;30:337–346.
12. Myles PS, Weitkamp B, Jones K, et al. Validity and reliability of a postoperative quality of recovery score: the QoR-40. Br J Anaesth. 2000;84:11–15.
13. Li LLM, Venkatraghavan L, Moga R, et al. Cost analysis of outpatient versus inpatient awake craniotomy
. J Neurosurg Anesthesiol. 2018;30:91.
14. Sheshadri V, Venkatraghavan L, Manninen P, et al. Anesthesia for same day discharge after craniotomy
: review of a single center experience. J Neurosurg Anesthesiol. 2018;30:299–304.
15. Dunn LK, Durieux ME, Elias WJ, et al. Innovations in functional neurosurgery and anesthetic implications. J Neurosurg Anesthesiol. 2018;30:18–25.
16. Jimenez-Ruiz F, Arnold B, Tatsui CE, et al. Perioperative and anesthetic considerations for neurosurgical laser interstitial thermal therapy ablations. J Neurosurg Anesthesiol. 2018;30:10–17.
17. Hester D, Hester D, Eastdown J, et al. Case series: anesthetic management of MR-guided laser interstitial thermal therapy (MRG-LITT) for epilepsy. J Neurosurg Anesthesiol. 2017;29:529.
18. Brockerville M, Unger Z, Rowland NC, et al. Airway management with a stereotactic headframe in situ—a Mannequin Study. J Neurosurg Anesthesiol. 2018;30:44–48.
19. Ciporen J, Gillham H, Noles M, et al. Crisis management simulation: establishing a dual neurosurgery and anesthesia training experience. J Neurosurg Anesthesiol. 2018;30:65–70.
20. Futier E, Constantin JM, Paugam-Burtz C, et al. A trial of intraoperative low-tidal-volume ventilation in abdominal surgery. N Engl J Med. 2013;369:428–437.
21. Severgnini P, Selmo G, Lanza C, et al. Protective mechanical ventilation during general anesthesia for open abdominal surgery improves postoperative pulmonary function. Anesthesiology. 2013;118:1307–1321.
22. Ruggieri F, Beretta L, Corno L, et al. Feasibility of protective ventilation during elective supratentorial neurosurgery: a randomized, crossover, clinical trial. J Neurosurg Anesthesiol. 2018;30:246–250.
23. Kim E, Kim HC, Lim YJ, et al. Comparison of intra-abdominal pressure among 3 prone positional apparatuses after changing from the supine to the prone position and applying positive end-expiratory pressure in healthy euvolemic patients: a prospective observational study. J Neurosurg Anesthesiol. 2017;29:14–20.
24. Babakhani B, Heroabadi A, Hosseinitabatabaei N, et al. Cerebral oxygenation under general anesthesia can be safely preserved in patients in prone position: a prospective observational study. J Neurosurg Anesthesiol. 2017;29:291–297.
25. Jo YY, Kim JY, Kwak YL, et al. The effect of pressure-controlled ventilation on pulmonary mechanics in the prone position during posterior lumbar spine surgery
: a comparison with volume-controlled ventilation. J Neurosurg Anesthesiol. 2012;24:14–18.
26. Robba C, Bragazzi NL, Bertuccio A, et al. Effects of prone position and positive end-expiratory pressure on noninvasive estimators of ICP: a pilot study. J Neurosurg Anesthesiol. 2017;29:243–250.
27. Yadav M, Reddy EP, Sharma A, et al. The effect of position on PaCO2
and PETCO2 in patients undergoing cervical spine surgery
in supine and prone position. J Neurosurg Anesthesiol. 2017;29:298–303.
28. Ni L, Fan Y, Bian J, et al. Effect of body mass on oxygenation and intra-abdominal pressure when using a jackson surgical table in the prone position during lumbar surgery. Spine (Phila Pa 1976). 2018;43:965–970.
29. Lamm WJ, Graham MM, Albert RK. Mechanism by which the prone position improves oxygenation in acute lung injury. Am J Respir Crit Care Med. 1994;150:184–193.
30. Pelosi P, Croci M, Calappi E, et al. The prone positioning during general anesthesia minimally affects respiratory mechanics while improving functional residual capacity and increasing oxygen tension. Anesth Analg. 1995;80:955–960.
31. Brower RG, Matthay MA, Morris A, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342:1301–1308.
32. Soh S, Shim JK, Ha Y, et al. Ventilation with high or low tidal volume with PEEP does not influence lung function after spinal surgery in prone position: a randomized controlled trial. J Neurosurg Anesthesiol. 2018;30:237–245.
33. Annoni F, Fontana V, Brimioulle S, et al. Early effects of enteral urea on intracranial pressure in patients with acute brain injury and hyponatremia. J Neurosurg Anesthesiol. 2017;29:400–405.
34. Dostalova V, Schreiberova J, Dostalova V Jr, et al. Effects of hypertonic saline and sodium lactate on cortical cerebral microcirculation and brain tissue oxygenation. J Neurosurg Anesthesiol. 2018;30:163–170.
35. Ali A, Tetik A, Sabanci PA, et al. Comparison of 3% hypertonic saline and 20% mannitol for reducing intracranial pressure in patients undergoing supratentorial brain tumor surgery: a randomized, double-blind clinical trial. J Neurosurg Anesthesiol. 2018;30:171–178.
36. Lillemae K, Laine AT, Schramko A, et al. Effect of albumin in combination with mannitol on whole-blood coagulation in vitro assessed by thromboelastometry. J Neurosurg Anesthesiol. 2018;30:265–272.
37. Myers J, Sharma D. Anesthetic management of awake craniotomies: a descriptive analysis of institutional practice over 10 years. J Neurosurg Anesthesiol. 2017;29:530.
38. Soehle M, Wolf CF, Priston MJ, et al. Propofol pharmacodynamics and bispectral index during key moments of awake craniotomy
. J Neurosurg Anesthesiol. 2018;30:32–38.
39. Marsh B, White M, Morton N, et al. Pharmacokinetic model driven infusion of propofol in children. Br J Anaesth. 1991;67:41–48.
40. Schnider TW, Minto CF, Gambus PL, et al. The influence of method of administration and covariates on the pharmacokinetics of propofol in adult volunteers. Anesthesiology. 1998;88:1170–1182.
41. Ilhan O, Koruk S, Serin G, et al. Dexmedetomidine in the supratentorial craniotomy
. Eurasian J Med. 2010;42:61–65.
42. Mitra S, Purohit S, Sharma M. Postoperative analgesia after wound infiltration with tramadol and dexmedetomidine as an adjuvant to ropivacaine for lumbar discectomies: a randomized-controlled clinical trial. J Neurosurg Anesthesiol. 2017;29:433–438.
43. Surve RM, Bansal S, Reddy M, et al. Use of dexmedetomidine along with local infiltration versus general anesthesia for burr hole and evacuation of chronic subdural hematoma (CSDH). J Neurosurg Anesthesiol. 2017;29:274–280.
44. Yun Y, Wang J, Tang RR, et al. Effects of an intraoperative dexmedetomidine bolus on the postoperative blood pressure and pain subsequent to craniotomy
for supratentorial tumors. J Neurosurg Anesthesiol. 2017;29:211–218.
45. Zhao LH, Shi ZH, Chen GQ, et al. Use of dexmedetomidine for prophylactic analgesia and sedation in patients with delayed extubation after craniotomy
: a randomized controlled trial. J Neurosurg Anesthesiol. 2017;29:132–139.
46. Liu Y, Liang F, Liu X, et al. Dexmedetomidine reduces perioperative opioid consumption and postoperative pain intensity in neurosurgery: a meta-analysis. J Neurosurg Anesthesiol. 2018;30:146–155.
47. Artime CA, Aijazi H, Zhang H, et al. Scheduled intravenous acetaminophen improves patient satisfaction with postcraniotomy pain management: a prospective, randomized, placebo-controlled, double-blind study. J Neurosurg Anesthesiol. 2018;30:231–236.
48. Stone S, Burbridge M, Jaffe R. Acetaminophen does not reduce postoeprative opiate consumption in patients undergoing craniotomy
for cerebral revascularization. J Neurosurg Anesthesiol. 2018;30:459.
49. Sexton MA, Abcejo AS, Pasternak JJ. Comparison of anesthetic management and outcomes in patients having either transnasal or transoral endoscopic odontoid process surgery. J Neurosurg Anesthesiol. 2018;30:179–183.
50. Mahrous RSS, Ahmed AMM. The Shikani Optical Stylet as an alternative to awake fiberoptic intubation in patients at risk of secondary cervical spine injury: a randomized controlled trial. J Neurosurg Anesthesiol. 2018;30:354–358.
51. Lee D, Armaghani S, Archer KR, et al. Preoperative opioid use as a predictor of adverse postoperative self-reported outcomes in patients undergoing spine surgery
. J Bone Joint Surg Am. 2014;96:e89.
52. Walid MS, Zaytseva NV. Prevalence of mood-altering and opioid medication use among spine surgery
candidates and relationship with hospital cost. J Clin Neurosci. 2010;17:597–600.
53. Lawrence JT, London N, Bohlman HH, et al. Preoperative narcotic use as a predictor of clinical outcome: results following anterior cervical arthrodesis. Spine (Phila Pa 1976). 2008;33:2074–2078.
54. Wick JB, Sivaganesan A, Chotai S, et al. Is there a preoperative morphine equianalgesic dose that predicts ability to achieve a clinically meaningful improvement following spine surgery
? Neurosurgery. 2018;83:245–251.
55. Monitto CL, Kost-Byerly S, White E, et al. The optimal dose of prophylactic intravenous naloxone in ameliorating opioid-induced side effects in children receiving intravenous patient-controlled analgesia morphine for moderate to severe pain: a dose finding study. Anesth Analg. 2011;113:834–842.
56. Movafegh A, Shoeibi G, Ansari M, et al. Naloxone infusion and post-hysterectomy morphine consumption: a double-blind, placebo-controlled study. Acta Anaesthesiol Scand. 2012;56:1241–1249.
57. Yang CP, Cherng CH, Wu CT, et al. Intrathecal ultra-low dose naloxone enhances the antinociceptive effect of morphine by enhancing the reuptake of excitatory amino acids from the synaptic cleft in the spinal cord of partial sciatic nerve-transected rats. Anesth Analg. 2011;113:1490–1500.
58. Firouzian A, Gholipour Baradari A, Alipour A, et al. Ultra-low-dose naloxone as an adjuvant to patient controlled analgesia (PCA) with morphine for postoperative pain relief following lumber discectomy: a double-blind, randomized, placebo-controlled trial. J Neurosurg Anesthesiol. 2018;30:26–31.
59. Mohamed SA, Abdel-Ghaffar HS. Effect of the addition of clonidine to locally administered bupivacaine on acute and chronic postmastectomy pain. J Clin Anesth. 2013;25:20–27.
60. Abdel Hay J, Kobaiter-Maarrawi S, Tabet P, et al. Bupivacaine field block with clonidine for postoperative pain control in posterior spine approaches: a randomized double-blind trial. Neurosurgery. 2018;82:790–798.
61. Pendi A, Field R, Farhan SD, et al. Perioperative ketamine for analgesia in spine surgery
: a meta-analysis of randomized controlled trials. Spine (Phila Pa 1976). 2018;43:E299–E307.
62. Loftus RW, Yeager MP, Clark JA, et al. Intraoperative ketamine reduces perioperative opiate consumption in opiate-dependent patients with chronic back pain undergoing back surgery. Anesthesiology. 2010;113:639–646.
63. Avidan MS, Maybrier HR, Abdallah AB, et al. Intraoperative ketamine for prevention of postoperative delirium
or pain after major surgery in older adults: an international, multicentre, double-blind, randomised clinical trial. Lancet. 2017;390:267–275.
64. Mashour GA, Ben Abdallah A, Pryor KO, et al. Intraoperative ketamine for prevention of depressive symptoms after major surgery in older adults: an international, multicentre, double-blind, randomised clinical trial. Br J Anaesth. 2018;121:1075–1083.
65. Han Y, Chen J, Zou D, et al. Efficacy of ketamine in the rapid treatment of major depressive disorder: a meta-analysis of randomized, double-blind, placebo-controlled studies. Neuropsychiatr Dis Treat. 2016;12:2859–2867.
66. Lee EE, Della Selva MP, Liu A, et al. Ketamine as a novel treatment for major depressive disorder and bipolar depression: a systematic review and quantitative meta-analysis. Gen Hosp Psychiatry. 2015;37:178–184.
67. Romeo B, Choucha W, Fossati P, et al. Meta-analysis of short- and mid-term efficacy of ketamine in unipolar and bipolar depression. Psychiatry Res. 2015;230:682–688.
68. Carspecken CW, Borisovskaya A, Lan ST, et al. Ketamine anesthesia does not improve depression scores in electroconvulsive therapy: a randomized clinical trial. J Neurosurg Anesthesiol. 2018;30:305–313.
69. Sriganesh K, Jadhav T, Venkataramaiah S, et al. Cerebral oxygen saturation during electroconvulsive therapy: a secondary analysis of a randomized crossover trial. J Neurosurg Anesthesiol. 2018;30:314–318.
70. Berkhemer OA, Fransen PS, Beumer D, et al. A randomized trial of intraarterial treatment for acute ischemic stroke
. N Engl J Med. 2015;372:11–20.
71. Nogueira RG, Jadhav AP, Haussen DC, et al. Thrombectomy 6 to 24 hours after stroke
with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11–21.
72. Albers GW, Marks MP, Kemp S, et al. Thrombectomy for stroke
at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378:708–718.
73. Brinjikji W, Pasternak J, Murad MH, et al. Anesthesia-related outcomes for endovascular stroke
revascularization: a systematic review and meta-analysis. Stroke
74. Campbell BCV, van Zwam WH, Goyal M, et al. Effect of general anaesthesia on functional outcome in patients with anterior circulation ischaemic stroke
having endovascular thrombectomy versus standard care: a meta-analysis of individual patient data. Lancet Neurol. 2018;17:47–53.
75. Lowhagen Henden P, Rentzos A, Karlsson JE, et al. General anesthesia versus conscious sedation for endovascular treatment of acute ischemic stroke
: the AnStroke Trial (Anesthesia During Stroke
76. Schonenberger S, Uhlmann L, Hacke W, et al. Effect of conscious sedation vs general anesthesia on early neurological improvement among patients with ischemic stroke
undergoing endovascular thrombectomy: a randomized clinical trial. JAMA. 2016;316:1986–1996.
77. Simonsen CZ, Yoo AJ, Sorensen LH, et al. Effect of general anesthesia and conscious sedation during endovascular therapy on infarct growth and clinical outcomes in acute ischemic stroke
: a randomized clinical trial. JAMA Neurol. 2018;75:470–477.
78. van de Graaf RA, Samuels N, Mulder M, et al. Conscious sedation or local anesthesia during endovascular treatment for acute ischemic stroke
. Neurology. 2018;91:e19–e25.
79. Athiraman U, Sultan-Qurraie A, Nair B, et al. Endovascular treatment of acute ischemic stroke
under general anesthesia: predictors of good outcome. J Neurosurg Anesthesiol. 2018;30:223–230.
80. Jagani M, Brinjikji W, Rabinstein AA, et al. Hemodynamics during anesthesia for intra-arterial therapy of acute ischemic stroke
. J Neurointerv Surg. 2016;8:883–888.
81. Rasmussen M, Espelund US, Juul N, et al. The influence of blood pressure management on neurological outcome in endovascular therapy for acute ischaemic stroke
. Br J Anaesth. 2018;120:1287–1294.
82. Bernhardt J, Zorowitz RD, Becker KJ, et al. Advances in stroke
83. Wang A, Abramowicz AE. Endovascular thrombectomy in acute ischemic stroke
: new treatment guide. Curr Opin Anaesthesiol. 2018;31:473–480.
84. Gritti P, Akeju O, Lorini FL, et al. A narrative review of adherence to subarachnoid hemorrhage guidelines. J Neurosurg Anesthesiol. 2018;30:203–216.
85. Dankbaar JW, Slooter AJ, Rinkel GJ, et al. Effect of different components of triple-H therapy on cerebral perfusion in patients with aneurysmal subarachnoid haemorrhage: a systematic review. Crit Care. 2010;14:R23.
86. Wang Y, Han R, Zuo Z. Dexmedetomidine post-treatment induces neuroprotection via activation of extracellular signal-regulated kinase in rats with subarachnoid haemorrhage. Br J Anaesth. 2016;116:384–392.
87. Okazaki T, Hifumi T, Kawakita K, et al. Association between dexmedetomidine use and neurological outcomes in aneurysmal subarachnoid hemorrhage patients: a retrospective observational study. J Crit Care. 2018;44:111–116.
88. Schebesch KM, Brundl E, Schodel P, et al. Differences in neuropeptide Y secretion between intracerebral hemorrhage and aneurysmal subarachnoid hemorrhage. J Neurosurg Anesthesiol. 2017;29:312–316.
89. Engstrom G, Smith JG, Persson M, et al. Red cell distribution width, haemoglobin A1c and incidence of diabetes mellitus. J Intern Med. 2014;276:174–183.
90. Zoller B, Melander O, Svensson P, et al. Red cell distribution width and risk for venous thromboembolism: a population-based cohort study. Thromb Res. 2014;133:334–339.
91. Kim J, Kim YD, Song TJ, et al. Red blood cell distribution width is associated with poor clinical outcome in acute cerebral infarction. Thromb Haemost. 2012;108:349–356.
92. Fontana V, Bond O, Spadaro S, et al. Red cell distribution width after subarachnoid hemorrhage. J Neurosurg Anesthesiol. 2018;30:319–327.
93. Engquist H, Lewen A, Howells T, et al. Hemodynamic disturbances in the early phase after subarachnoid hemorrhage: regional cerebral blood flow studied by bedside xenon-enhanced CT. J Neurosurg Anesthesiol. 2018;30:49–58.
94. Engquist H, Rostami E, Ronne-Engstrom E, et al. Effect of HHH-therapy on regional CBF after severe subarachnoid hemorrhage studied by bedside xenon-enhanced CT. Neurocrit Care. 2018;28:143–151.
95. Haegens NM, Gathier CS, Horn J, et al. Induced hypertension in preventing cerebral infarction in delayed cerebral ischemia after subarachnoid hemorrhage. Stroke
96. McCredie VA, Alali AS, Scales DC, et al. Impact of ICU structure and processes of care on outcomes after severe traumatic brain injury
: a multicenter cohort study. Crit Care Med. 2018;46:1139–1149.
97. Kim H, Lee SB, Son Y, et al. Hemodynamic instability and cardiovascular events after traumatic brain injury
predict outcome after artifact removal with deep belief network analysis. J Neurosurg Anesthesiol. 2018;30:347–353.
98. Carney N, Totten AM, O'Reilly C, et al. Guidelines for the management of severe traumatic brain injury
, fourth edition. Neurosurgery. 2017;80:6–15.
99. Wang X, Dong Y, Han X, et al. Nutritional support for patients sustaining traumatic brain injury
: a systematic review and meta-analysis of prospective studies. PLoS One. 2013;8:e58838.
100. Chaudhry R, Kukreja N, Tse A, et al. Trends and outcomes of early versus late percutaneous endoscopic gastrostomy placement in patients with traumatic brain injury
: nationwide population-based study. J Neurosurg Anesthesiol. 2018;30:251–257.
101. Lelubre C, Bouzat P, Crippa IA, et al. Anemia management after acute brain injury. Crit Care. 2016;20:152.
102. Martin G, Shah D, Elson N, et al. Relationship of coagulopathy and platelet dysfunction to transfusion needs after traumatic brain injury
. Neurocrit Care. 2018;28:330–337.
103. Boutin A, Moore L, Green RS, et al. Hemoglobin thresholds and red blood cell transfusion in adult patients with moderate or severe traumatic brain injuries: a retrospective cohort study. J Crit Care. 2018;45:133–139.
104. Riske L, Thomas RK, Baker GB, et al. Lactate in the brain: an update on its relevance to brain energy, neurons, glia and panic disorder. Ther Adv Psychopharmacol. 2017;7:85–89.
105. Millet A, Cuisinier A, Bouzat P, et al. Hypertonic sodium lactate reverses brain oxygenation and metabolism dysfunction after traumatic brain injury
. Br J Anaesth. 2018;120:1295–1303.
106. Bell JD. In vogue: ketamine for neuroprotection in acute neurologic injury. Anesth Analg. 2017;124:1237–1243.
107. Zeiler FA, Teitelbaum J, West M, et al. The ketamine effect on ICP in traumatic brain injury
. Neurocrit Care. 2014;21:163–173.
108. Cameron HA, McEwen BS, Gould E. Regulation of adult neurogenesis by excitatory input and NMDA receptor activation in the dentate gyrus. J Neurosci. 1995;15:4687–4692.
109. Morganti-Kossmann MC, Rancan M, Stahel PF, et al. Inflammatory response in acute traumatic brain injury
: a double-edged sword. Curr Opin Crit Care. 2002;8:101–105.
110. Peters AJ, Villasana LE, Schnell E. Ketamine alters hippocampal cell proliferation and improves learning in mice after traumatic brain injury
. Anesthesiology. 2018;129:278–295.
111. Loix S, De Kock M, Henin P. The anti-inflammatory effects of ketamine: state of the art. Acta Anaesthesiol Belg. 2011;62:47–58.
112. Garthe A, Kempermann G. An old test for new neurons: refining the Morris water maze to study the functional relevance of adult hippocampal neurogenesis. Front Neurosci. 2013;7:63.
113. Lipnick MS, Cahill EA, Feiner JR, et al. Comparison of transcranial Doppler and ultrasound-tagged near infrared spectroscopy for measuring relative changes in cerebral blood flow in human subjects. Anesth Analg. 2018;126:579–587.
114. Caccioppola A, Carbonara M, Macri M, et al. Ultrasound-tagged near-infrared spectroscopy does not disclose absent cerebral circulation in brain-dead adults. Br J Anaesth. 2018;121:588–594.
115. Gwer S, Sheward V, Birch A, et al. The tympanic membrane displacement analyser for monitoring intracranial pressure in children. Childs Nerv Syst. 2013;29:927–933.
116. Maissan IM, Dirven PJ, Haitsma IK, et al. Ultrasonographic measured optic nerve sheath diameter as an accurate and quick monitor for changes in intracranial pressure. J Neurosurg. 2015;123:743–747.
117. Wakerley BR, Kusuma Y, Yeo LL, et al. Usefulness of transcranial Doppler-derived cerebral hemodynamic parameters in the noninvasive assessment of intracranial pressure. J Neuroimaging. 2015;25:111–116.
118. Ganslandt O, Mourtzoukos S, Stadlbauer A, et al. Evaluation of a novel noninvasive ICP monitoring device in patients undergoing invasive ICP monitoring: preliminary results. J Neurosurg. 2018;128:1653–1660.
119. Nunes RR, Bersot CDA, Garritano JG. Intraoperative neurophysiological monitoring in neuroanesthesia
. Curr Opin Anaesthesiol. 2018;31:532–538.
120. Kim JS, Jang MJ, Hyun SJ, et al. Failure to generate baseline muscle motor evoked potentials during spine surgery
: risk factors and association with the postoperative outcomes. Clin Neurophysiol. 2018;129:2276–2283.
121. Choi BW, Song KJ, Chang H. Ossification of the posterior longitudinal ligament: a review of literature. Asian Spine J. 2011;5:267–276.
122. Guo L, Li Y, Han R, et al. The correlation between recordable MEPs and motor function during spinal surgery for resection of thoracic spinal cord tumor. J Neurosurg Anesthesiol. 2018;30:39–43.
123. Chen JH, Shilian P, Cheongsiatmoy J, et al. Factors associated with inadequate intraoperative baseline lower extremity somatosensory evoked potentials. J Clin Neurophysiol. 2018;35:426–430.
124. Reddy RP, Brahme IS, Karnati T, et al. Diagnostic value of somatosensory evoked potential changes during carotid endarterectomy for 30-day perioperative stroke
. Clin Neurophysiol. 2018;129:1819–1831.
125. Marino V, Aloj F, Vargas M, et al. Intraoperative neurological monitoring with evoked potentials during carotid endarterectomy versus cooperative patients under general anesthesia technique: a retrospective study. J Neurosurg Anesthesiol. 2018;30:258–264.
126. Aridi HD, Nejim B, Locham S, et al. Does anesthesia technique modify the risk of routine and selective shunting during carotid endarterectomy? J Vasc Surg. 2018;67:e189–e190.
127. Andropoulos DB. Effect of anesthesia on the developing brain: infant and fetus. Fetal Diagn Ther. 2018;43:1–11.
128. Jevtovic-Todorovic V, Brambrick A. General anesthesia and young brain: What is New? J Neurosurg Anesthesiol. 2018;30:217–222.
129. Raper J, De Biasio JC, Murphy KL, et al. Persistent alteration in behavioural reactivity to a mild social stressor in rhesus monkeys repeatedly exposed to sevoflurane in infancy. Br J Anaesth. 2018;120:761–767.
130. Satomoto M, Sun Z, Adachi YU, et al. Neonatal sevoflurane exposure induces adulthood fear-induced learning disability and decreases glutamatergic neurons in the basolateral amygdala. J Neurosurg Anesthesiol. 2018;30:59–64.
131. Ju LS, Yang JJ, Morey TE, et al. Role of epigenetic mechanisms in transmitting the effects of neonatal sevoflurane exposure to the next generation of male, but not female, rats. Br J Anaesth. 2018;121:406–416.
132. Ing C, Wall MM, DiMaggio CJ, et al. Latent class analysis of neurodevelopmental deficit after exposure to anesthesia in early childhood. J Neurosurg Anesthesiol. 2017;29:264–273.
133. DiMaggio C, Sun LS, Ing C, et al. Pediatric anesthesia and neurodevelopmental impairments: a Bayesian meta-analysis. J Neurosurg Anesthesiol. 2012;24:376–381.
134. Wilder RT, Flick RP, Sprung J, et al. Early exposure to anesthesia and learning disabilities in a population-based birth cohort. Anesthesiology. 2009;110:796–804.
135. Davidson AJ, Disma N, de Graaff JC, et al. Neurodevelopmental outcome at 2 years of age after general anaesthesia and awake-regional anaesthesia in infancy (GAS): an international multicentre, randomised controlled trial. Lancet. 2016;387:239–250.
136. Sun LS, Li G, Miller TL, et al. Association between a single general anesthesia exposure before age 36 months and neurocognitive outcomes in later childhood. JAMA. 2016;315:2312–2320.
137. Warner DO, Zaccariello MJ, Katusic SK, et al. Neuropsychological and behavioral outcomes after exposure of young children to procedures requiring general anesthesia: the Mayo Anesthesia Safety in Kids (MASK) Study. Anesthesiology. 2018;129:89–105.
138. Atluri N, Joksimovic SM, Oklopcic A, et al. A neurosteroid analogue with T-type calcium channel blocking properties is an effective hypnotic, but is not harmful to neonatal rat brain. Br J Anaesth. 2018;120:768–778.
139. Zou X, Patterson TA, Sadovova N, et al. Potential neurotoxicity of ketamine in the developing rat brain. Toxicol Sci. 2009;108:149–158.
140. Inouye SK, Westendorp RG, Saczynski JS. Delirium
in elderly people. Lancet. 2014;383:911–922.
141. Siddiqi N, House AO, Holmes JD. Occurrence and outcome of delirium
in medical in-patients: a systematic literature review. Age Ageing. 2006;35:350–364.
142. Weinstein SM, Poultsides L, Baaklini LR, et al. Postoperative delirium
in total knee and hip arthroplasty patients: a study of perioperative modifiable risk factors. Br J Anaesth. 2018;120:999–1008.
143. Duan X, Coburn M, Rossaint R, et al. Efficacy of perioperative dexmedetomidine on postoperative delirium
: systematic review and meta-analysis with trial sequential analysis of randomised controlled trials. Br J Anaesth. 2018;121:384–397.
144. Lee C, Lee CH, Lee G, et al. The effect of the timing and dose of dexmedetomidine on postoperative delirium
in elderly patients after laparoscopic major non-cardiac surgery: a double blind randomized controlled study. J Clin Anesth. 2018;47:27–32.
145. Shen YZ, Peng K, Zhang J, et al. Effects of haloperidol on delirium
in adult patients: a systematic review and meta-analysis. Med Princ Pract. 2018;27:250–259.
146. van den Boogaard M, Slooter AJC, Bruggemann RJM, et al. Effect of haloperidol on survival among critically ill adults with a high risk of delirium
: the REDUCE Randomized Clinical Trial. JAMA. 2018;319:680–690.
147. Daniels LM, Nelson SB, Frank RD, et al. Pharmacologic treatment of intensive care unit delirium
and the impact on duration of delirium
, length of intensive care unit stay, length of hospitalization, and 28-day mortality. Mayo Clin Proc. 2018;93:1739–1748.
148. MacKenzie KK, Britt-Spells AM, Sands LP, et al. Processed electroencephalogram monitoring and postoperative delirium
: a systematic review and meta-analysis. Anesthesiology. 2018;129:417–427.
149. Schulte PJ, Roberts RO, Knopman DS, et al. Association between exposure to anaesthesia and surgery and long-term cognitive trajectories in older adults: report from the Mayo Clinic Study of Aging. Br J Anaesth. 2018;121:398–405.
150. Roberts RO, Geda YE, Knopman DS, et al. The Mayo Clinic Study of Aging: design and sampling, participation, baseline measures and sample characteristics. Neuroepidemiology. 2008;30:58–69.
151. Bratzke LC, Koscik RL, Schenning KJ, et al. Cognitive decline in the middle-aged after surgery and anaesthesia: results from the Wisconsin Registry for Alzheimer’s Prevention cohort. Anaesthesia. 2018;73:549–555.
152. Zhang Y, Shan GJ, Zhang YX, et al. Propofol compared with sevoflurane general anaesthesia is associated with decreased delayed neurocognitive recovery in older adults. Br J Anaesth. 2018;121:595–604.
153. Zorrilla-Vaca A, Healy R, Grant MC, et al. Intraoperative cerebral oximetry-based management for optimizing perioperative outcomes: a meta-analysis of randomized controlled trials. Can J Anaesth. 2018;65:529–542.
154. Hu J, Feng X, Valdearcos M, et al. Interleukin-6 is both necessary and sufficient to produce perioperative neurocognitive disorder in mice. Br J Anaesth. 2018;120:537–545.
155. Hu J, Vacas S, Feng X, et al. Dexmedetomidine prevents cognitive decline by enhancing resolution of high mobility group Box 1 protein-induced inflammation through a vagomimetic action in mice. Anesthesiology. 2018;128:921–931.
156. Ran M, Wang Z, Yang H, et al. Orexin-1 receptor is involved in ageing-related delayed emergence from general anaesthesia in rats. Br J Anaesth. 2018;121:1097–1104.
157. Berger M, Schenning KJ, Brown CHt, et al. Best practices for postoperative brain health: recommendations from the Fifth International Perioperative Neurotoxicity Working Group. Anesth Analg. 2018;127:1406–1413.
158. Evered LA, Silbert BS. Postoperative cognitive dysfunction and noncardiac surgery. Anesth Analg. 2018;127:496–505.
159. Evered L, Silbert B, Knopman DS, et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Anesth Analg. 2018;127:1189–1195.
160. Evered L, Silbert B, Knopman DS, et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Br J Anaesth. 2018;121:1005–1012.
161. Evered L, Silbert B, Knopman DS, et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Anesthesiology. 2018;129:872–879.
162. Evered L, Silbert B, Knopman DS, et al. Recommendations for the nomenclature of cognitive change associated with anaesthesia and surgery-2018. Can J Anaesth. 2018;65:1248–1257.