The potential for untoward neurocognitive sequelae in the pediatric population associated with general anesthesia has garnered significant press in both the pediatric and anesthesia literature. In vivo animal studies have demonstrated histologic neurodegeneration with administration of anesthetics invigorating attention and concern.1–5 Jevtovic-Todorovic et al5 reported on administration of midazolam, nitrous oxide, and isoflurane to 7-day-old rats for the equivalent of 6 hours of anesthesia time. They observed cellular apoptosis in the multiple sites including hippocampus, anterior thalamic nuclei, mammillary bodies, and retrosplenial cortex—all areas associated with memory. Furthermore, specimens exposed to this anesthetic mixture exhibited deficits in spatial reference memory and slower learning capacity. Similarly, Brambrink et al1,3 reported widespread neuronal and oligodendrocyte apoptosis in 6-day-old rhesus macaque monkeys with 5 hours of exposure to isoflurane. Although the findings in these animal studies must be interpreted with caution and may not be extrapolated to humans, the effects should nonetheless be considered.
Research studies on humans linking anesthetic dosage or exposure are not definitive and are fraught with limitations but may also provide a tale of caution. The evidence lends suggestion that the risk for learning disabilities with repeated or prolonged anesthesia exposure (AE) may be dose dependent. In a retrospective matched cohort study of 5357 children between 1976 and 1982, Flick et al6 performed a multivariate analysis examining learning disability and need for an individualized education program. Three hundred fifty patients were exposed to combined nitrous oxide and halothane anesthesia (286 patients were exposed once and 64 exposed more than once). Patients exposed more than once had 36.6% cumulative incidence of learning disability compared with 23.6% exposed once and 21.3% for unexposed controls. Furthermore, multiple exposures to anesthesia was a risk factor for the need for an individualized education program. The criteria for admittance to a learning disability program is unclear, and single AE did not demonstrate any adverse outcome.7 Of note, halothane is no longer currently being utilized in routine pediatric anesthetic practices.
Dimaggio et al8 analyzed a cohort of 10,450 siblings born between 1999 and 2005 enrolled in the New York State Medicaid program. The determined hazard ratio for developmental or behavioral disorders associated with AE was 1.6 for the 304 children who underwent surgery before age 3. The ratio increased from 1.1 for 1 operation to 2.9 for 2 operations and 4.0 for 3 or more operations.
Wilder et al9 performed a retrospective birth cohort study of 5357 children, 593 of which had surgery before age 4 from 1976 to 1982. Children exposed 2 or 3 episodes of anesthesia had an increased risk for learning disabilities compared with those who only had a single episode (hazard ratios: 1.59, 2.60, and 1.00, respectively). In patients with multiple exposures, the incidence of learning disability was 35.1% versus 20% for controls. Similarly, Sprung et al10 identified an increased risk of attention-deficit hyperactivity disorder with repeated AE (10.7% with single exposure vs. 17.9% with 2 or more exposures). In a 2015 publication, Backeljauw and colleagues compared neurocognitive assessment test scores between children who had surgery under general anesthesia before age 4 to healthy, unexposed children. The children were matched by age, sex, handedness, and socioeconomic status, and the comparative analysis found the exposed children had lower language comprehension and performance IQ.11
Although the existing literature are primarily retrospective observational studies, they also do not definitively link anesthesia with behavioral changes or learning abilities. These findings have engendered concern for children who undergo repeated surgical procedures. Early-onset scoliosis (EOS) is a progressive spinal deformity that occurs in young children that often requires surgical intervention. Surgical treatment for EOS typically involves distraction-based spinal instrumentation, known as growing rod (GR) surgery, to allow continued growth of the spine and thorax while maintaining spinal alignment. GR treatment requires repeated surgical lengthenings of the instrumentation every 6 to 9 months, and it is not uncommon for patients to have at least 7 surgical lengthenings until reaching skeletal maturity and having a definitive spinal fusion.12
All patients enlisted to the GR treatment program will ultimately receive a minimum amount of anesthesia associated with the following: (1) diagnostic imaging; (2) initial placement of GR instrumentation; (3) repeated surgical lengthenings; and (4) definitive spinal fusion. In addition, due to the relatively high rate of complications, many GR patients may be subject to surgical procedures to resolve postoperative complications such as implant failure, surgical site infection, or other problems—all which lead to an increased cumulative exposure to anesthesia.13 In response growing body of evidence surrounding the potentially harmful effects of AE in young children, albeit inconclusive, we sought to quantify AE in our GR patients who are perhaps most susceptible to be exposed to multiple anesthesia episodes. To our knowledge this is the first study quantifying AE in GR treatment for EOS.
Forty-one patients who underwent initial GR surgery between November 1995 and February 2012 at a single institution were retrospectively reviewed. Sixteen of the 41 patients received “final” spinal fusion upon completion of their GR treatment and were included in the analysis. Demographics, duration of general endotracheal anesthesia, and sedation anesthesia related to spinal deformity care were obtained from the patients’ medical records.
AE from GR procedures was categorized into 2 categories: “standard care” and “associated care” procedures. AE related to standard care procedures included: (1) diagnostic imaging; (2) initial placement of GR instrumentation; (3) GR lengthening procedures; and (4) “final” spinal fusion and instrumentation. Associated care procedures included: (1) planned or unplanned revision surgeries; (2) wound-related surgeries to treat surgical site infection or other wound problems; and (3) adjunctive spine-related procedures such as neurosurgical procedures to address spinal cord or other central nervous system anomalies. It must be noted that a lengthening procedure could have been performed during a revision procedure; however, due to variability in the operative report dictations between surgeons it was difficult to discern whether a lengthening had been performed during a revision surgery. Therefore, whether or not a lengthening was performed during a revision, it was ultimately counted as a revision procedure.
Etiological diagnoses were classified based on the classification for EOS: congenital/structural, idiopathic, neuromuscular, and syndromic. The percentage of overall AE for standard care and associated care procedures was calculated and reported using descriptive statistics. Total anesthesia time (TAT) was defined as the time of first anesthetic administration until the time the patient was awakened and breathing independently. Mean TAT for different procedures was examined among etiologies using multiple analysis of variance tests. Spearman rank-order correlation was used to delineate correlation between TAT and age at initial GR surgery, severity of the spinal deformity, number of lengthening procedures, and number of complications. Final follow-up was defined as the most recent visit after the final fusion surgery. SPSS for Windows (version 21.0; Chicago, IL) was used for statistical analysis, and the α level was set at 0.05. Power analyses were performed using G* power (version 3.1; Düsseldorf, Germany).
Sixteen patients who underwent GR surgical treatment were included. Mean age at the time of initial GR surgery was 8 years (range, 3.9 to 14.4 y). Regarding the use of GR in the 14-year-old patient, this particular patient underwent an anterior spinal fusion of the lumbar spine at age 11. Approximately 2 years postoperatively, he experienced curve progression and deteriorating coronal balance. His mother was resistant to further spinal fusion surgery and preferred nonfusion techniques to address his worsening spinal alignment. It was ultimately decided to use GRs with the use of wedding band connectors. He subsequently underwent 2 lengthening procedures and at age 18 had a posterior spinal fusion. Of the 16 patients, 5 patients were syndromic, 8 patients were neuromuscular, and 3 patients were idiopathic. The mean preoperative major curve measured 70 degrees, which corrected to 37 degrees immediately after initial GR surgery and maintained at 41 degrees at the final follow-up. There were a total of 62 lengthening procedures for the entire cohort, and there was an average 5.2 lengthenings per patient (range, 0 to 11 lengthenings). There was a mean of 1.6 complications per patient (range, 0 to 7 complications).
The largest contributor of TAT among standard care procedures was definitive spinal fusion surgery [mean, 538 min (range, 334 to 920 min)] followed by index surgery [mean 356 min (range, 166 to 800 min)] (Table 1). Although the mean TAT for individual lengthenings was 85 minutes (range, 53 to 155 min), the mean cumulative TAT from all lengthenings per patient was 351 minutes (range, 75 to 763 min). Diagnostic imaging contributed the least amount of TAT—an average of 60 minutes per patient (range, 20 to 135 min). Standard care procedures constituted 55% of the total AE (Fig. 1).
The majority of TAT from associated care procedures was attributed to revision surgeries. The mean cumulative TAT for revision surgeries was 695 minutes per patient (range, 237 to 1517 min) for those who required at least 1 revision. The mean TAT for individual revision procedures was 181 minutes per patient (range, 24 to 573 min). Associated care procedures comprised 45% of the total AE.
With a sample size of 16 patients, we were able to detect a 60% correlation (0.6) with 80% power. There was a positive correlation between the preoperative major curve size and cumulative lengthening TAT (r2=0.624, P<0.05). There was also a positive correlation between preoperative major curve size and cumulative mean TAT from standard care procedures (r2=0.584, P>0.05). Age at index surgery and cumulative lengthening TAT demonstrated a negative correlation (r2=−0.654, P>0.05). Age at index surgery was also negatively correlated with cumulative mean TAT from standard care procedures (r2=−0.506, P>0.05). There was no correlation between age at index surgery or preoperative major curve and TAT from associated care procedures (individual procedures or cumulative mean).
This is the first study to quantify AE related to GR surgery. Much of the existing human data regarding the effects of anesthesia on young children are retrospective in nature and elicited from large-scale population review. For this reason and others, the data are controversial, must be interpreted carefully, and causality cannot be indisputably proven. In addition, modern anesthetic techniques and medications have improved and are now highly calibrated. Despite this, there is increasing awareness and concern among the pediatric medical community regarding AE in young patients. Case-controlled, prospective studies are underway to objectively evaluate neurodevelopmental outcomes with the use of inhaled versus neuraxial anesthesia.14
Our study was performed to gain a better understanding of the duration of anesthesia patients may expect from standard care and associated care procedures related to GR treatment. As expected, the majority of AE came from standard care GR procedures; however, associated care procedures added an additional 41% to the cumulative TAT. Therefore, avoidance of complications can reduce AE significantly. There were no significant differences between etiologies in AE for standard or associated care procedures.
The future of distraction-based surgical treatment for EOS will focus on less or noninvasive techniques with reduced need for repeated surgeries. Exciting data published by Akbarnia and colleagues showed promising early results of using magnetically controlled GRs in patients with EOS. Serial lengthenings are performed noninvasively in an outpatient setting without the use of anesthesia or other sedatives. It is anticipated that decreased number of surgical lengthening procedures will lead to a decreased AE throughout treatment.15
Spine surgeons, anesthesiologists, and pediatricians should be aware of potential neurotoxic risks with repeated AE and should recognize the limitations of the current literature. The uncertainty surrounding the human data at present merits further investigation and consideration for reduction in AE when possible if deemed safe. The US Food and Drug Administration and International Anesthesia Research Society have formed a partnership entitled SmartTots (Strategy for Mitigating Anesthesia-related neuroToxicity in Tots, http://www.smarttots.org)16 in hopes of identifying the effects of AE in children and defining safety parameters in anesthetic dosage and exposure.
Although this study was retrospective, had a relatively small sample size consisting mostly of neuromuscular EOS patients, we did not set out to evaluate neurocognitive effects of anesthesia on our patients, but instead to quantify AE during treatment and identify where efficiency can be introduced in our practice model. We utilized duration of AE as a surrogate for medication type and respective dosage. To our knowledge, a total dosage threshold has not yet been widely recognized as deleterious to neurocognitive function. Although we do believe medication type and administered dosages are significant and should ultimately be borne-out, the variability in anesthetic practices and difficulty in the interpretation of anesthesia records limited our ability to record these data retrospectively.
Furthermore, we included both sedation and “general” anesthesia in our “all-or-none” approach to data collection. Although we recognize these entities are disparate in terms of total dosage and types of medication administered, we felt both were important to define a global representation of AE in GR treatment.
Until a dosage threshold or clear demonstration of AE effect on neurocognition is proven, the significance of our data will remain in question. Despite this, we feel that the risk of anesthesia may be real and we will limit AE for our patients whenever possible. Similar to the recent efforts taken to minimize medical radiation exposure in young patients, efficiency measures should be implemented to reduce AE without adding morbidity or compromising the goals of GR treatment in patients with EOS.17
1. Brambrink A, Back SA, Riddle A, et al. Isoflurane-induced apoptosis of oligodendrocytes in the neonatal primate brain. Ann Neurol. 2012;72:525–535.
2. Zhu C, Gao J, Karlsson N, et al. Isoflurane anesthesia induced persistent, progressive memory impairment, caused a loss of neural stem cells, and reduced neurogenesis in young, but not adult, rodents. J Cereb Blood Flow Metab. 2010;30:1017–1030.
3. Brambrink A, Evers AS, Avidan MS, et al. Isoflurane-induced neuroapoptosis in the neonatal rhesus macaque brain. Anesthesiology. 2010;112:834–841.
4. McGowan F, Davis P. Anesthetic-related neurotoxicity in the developing infant: of mice, rats, monkeys and, possibly, humans. Anesth Analg. 2008;106:1599–1602.
5. Jevtovic-Todorovic V, Hartman RE, Izumi Y, et al. Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. J Neurosci. 2003;23:876–882.
6. Flick RP, Katusic SK, Colligan RC, et al. Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics. 2011;128:e1053–e1061.
7. Williams RK. The pediatrician and anesthesia neurotoxicity. Pediatrics. 2011;128:e1268–e1270.
8. Dimaggio C, Sun L, Li G. Early childhood exposure to anesthesia and risk of developmental and behavioral disorders in a sibling birth cohort. Anesth Analg. 2011;113:1143–1151.
9. 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.
10. Sprung J, Flick RP, Katusic SK, et al. Attention-deficit/hyperactivity disorder after early exposure to procedures requiring general anesthesia. Mayo Clin Proc. 2012;87:120–129.
11. Backeljauw B, Holland SK, Altaye M, et al. Cognition and brain structure following early childhood surgery with anesthesia. Pediatrics. 2015;136:e1–e12.
12. Akbarnia BA, Marks DS, Boachie-Adjei O, et al. Dual growing rod technique for the treatment of progressive early-onset scoliosis
: a multicenter study. Spine. 2005;30(suppl):S46–S57.
13. Bess S, Akbarnia BA, Thompson GH, et al. Complications of growing-rod treatment for early-onset scoliosis
: analysis of one hundred and forty patients. J Bone Joint Surg Am. 2010;92:2533–2543.
14. Rappaport B, Mellon RD, Simone A, et al. Defining safe use of anesthesia in children. N Engl J Med. 2011;364:1387–1390.
15. Akbarnia BA, Cheung K, Noordeen H, et al. Next generation of growth-sparing technique: preliminary clinical results of a magnetically controlled growing rod (MCGR) in 14 patients with early onset scoliosis. Spine. 2013;38:665–670.
16. Wyckoff AS. Consensus statement reflects mixed picture on anesthesia’s possible link to learning problems. AAP News. 2012;34:1.
17. Levy AR, Goldberg MS, Mayo NE, et al. Reducing the lifetime risk of cancer from spinal radiographs among people with adolescent idiopathic scoliosis. Spine. 1996;21:1540–1547; discussion 1548.