The Gerard W. Ostheimer lecture is given annually as a keynote address at the Society for Obstetric Anesthesia and Perinatology (SOAP) annual meeting. The Ostheimer lecture summarizes “what is new” in the literature published in the previous calendar year that is of relevance to practicing obstetric anesthesiologists. To select the most relevant new literature, the contents of leading anesthesia, obstetric, pediatric, and general medical journals published in 2016 were reviewed (see Supplemental Digital Content, Table 1, http://links.lww.com/AA/C561). Drawn from these sources, articles thought to be of greatest interest to the field were selected. The top 126 of these articles are briefly summarized in Supplemental Digital Content, Syllabus, http://links.lww.com/AA/C561. This article reviews 10 articles from this list that are of particular interest. These include (1) landmark randomized controlled clinical trials (RCTs) that are expected to have a significant effect on the practice of obstetrics and obstetric anesthesia, (2) studies focused on neuraxial anesthesia and postoperative pain control, and (3) studies (and a guideline) focused on systems-based interventions to improve obstetrical outcomes.
Three RCTs reviewed here focus on the prevention of obstetrical complications, including postcesarean infection and postpartum hemorrhage.
The first is a study by Tita et al1 that examines the addition of azithromycin to standard antibiotics used for surgical site infection after cesarean delivery. Prior studies have shown that wound infection complicates up to 12% of nonelective cesarean deliveries.1 Ureaplasma species have been commonly associated with these infections and are not covered by the standard antibiotics used for prophylaxis of surgical site infections. The authors therefore posed the hypothesis that the addition of azithromycin, which provides good coverage of ureaplasma, would reduce the risk of wound infection.1
The study was conducted at 14 centers across the United States. It enrolled 2013 women undergoing cesarean delivery during labor or after the rupture of membranes. All patients were scheduled to receive standard antibiotic prophylaxis and then were randomized to either azithromycin 500 mg intravenously or placebo. The primary outcome specified for the trial was a composite of endometritis, wound infection, or other infection.
The investigators found that the addition of azithromycin reduced the frequency of the primary outcome by about half, from 12.0% in the placebo arm to 6.1% in the azithromycin arm (relative risk [RR], 0.51; 95% confidence interval [CI], 0.38–0.68). There were also significant reductions in the endometritis (RR, 0.62; 95% CI, 0.42–0.92) and wound infection (RR, 0.35; 95% CI, 0.22–0.56) components of the composite end point. There were lower rates of many of the outcomes that would be expected to track with postpartum infection, including postpartum fever, unscheduled clinic visits, readmissions, and postpartum treatment with antibiotics.
While there was a marked reduction in the primary outcome associated with azithromycin, these findings need to be interpreted considering the limitations associated with the study. One important limitation is that nearly three-quarters of the patients enrolled in the trial were obese; it is therefore possible that the primary antibiotic (eg, cefazolin) was underdosed for the patients’ weight.2 Thus, it may be that any reduction in the infection rate is due to the addition of azithromycin in the setting of inadequate dosing of cefazolin rather than broadening of the coverage to include ureaplasma species. Unfortunately, the study does not report culture data for ureaplasma species that might inform this question.
Until additional data examine the impact of adding azithromycin to primary prophylactic antibiotics that are dosed in a weight-based fashion, labor and delivery units must decide based on imperfect data whether to add azithromycin to the regimens used for cesarean deliveries after labor or rupture of membranes. Disadvantages of broadening coverage include the possibility of using antibiotics unnecessarily in an era where we are acutely aware of the need for good antibiotic stewardship to prevent the emergence of resistant organisms, the potential adverse effects of azithromycin for the mother, and the less well-understood risk of potentially adversely impacting the infant microbiome, which we are just beginning to appreciate is affected by peridelivery antibiotic exposure.3
The second study was a single-center, randomized trial in which 1147 women undergoing cesarean delivery were randomized to chlorhexidine–alcohol or iodine–alcohol skin preparation before incision to test the hypothesis that chlorhexidine–alcohol is superior to iodine–alcohol for the prevention of surgical site infection after cesarean deliveries.4 RCTs in other settings have demonstrated chlorhexidine to be a superior antiseptic to iodine, but these findings may not apply to cesarean delivery where wound infections are due to both skin and vaginal pathogens.
The primary outcome was a composite of superficial and deep surgical site infection. The use of chlorhexidine was associated with a statistically significant reduction in the frequency of this outcome from 7.3% to 4.0% (P = .02) and a trend toward lower rates of each component of the primary outcome.
These findings are biologically plausible because chlorhexidine has a number of properties that can account for its superior efficacy compared to iodine, including stronger binding to the skin, higher antibacterial activity against relevant species, and longer residual effects.4 Potential limitations of the study are that it was conducted at a single center, and because of the differences in appearance of chlorhexidine and iodine, it was not possible to maintain blinding. That said, the large effect size observed in this well-done study suggests that a transition from the routine use of iodine to chlorhexidine is merited for cesarean deliveries, except in cases of emergent cesarean delivery when there may be inadequate time for the chlorhexidine–alcohol solution to dry.
The third RCT addressed whether misoprostol added to oxytocin as part of the active management of the third stage of labor would decrease the occurrence of postpartum hemorrhage (PPH) when compared with oxytocin alone.5 Misoprostol has been demonstrated to have uterotonic properties in studies comparing it to placebo.6 However, its role as a second-line uterotonic (when administered in addition to oxytocin) is less clear. A large RCT examining its use as a second-line uterotonic used to treat postpartum hemorrhage demonstrated no benefit.7
The study was conducted at 3 centers in France and included patients with vaginal deliveries at or near term. All patients in the trial received oxytocin in the third stage of labor and then were randomized to either misoprostol 400 μg orally or placebo with the primary outcome being PPH, defined as >500 mL of blood loss within 2 hours of birth. The trial was designed to enroll 3100 patients, a sample size that would provide statistical power to show a reduction in the occurrence of PPH from 7.5% to 5%.
At a planned interim analysis after 1721 patients were enrolled, the study was stopped prematurely due to futility, along with an unexpectedly high rate of adverse effects associated with the use of misoprostol. At the time of the interim analysis, the rate of PPH was indistinguishable in those treated with misoprostol and placebo (8.4% vs 8.3%, P = .98). However, the frequency of fever was markedly elevated in the misoprostol compared to the placebo arm (30.4% vs 6.3%, P < .001), as were the rates of shivering, nausea, vomiting, and diarrhea (all P < .05).
This study adds to the accumulating data suggesting a lack of effectiveness of adjunctive misoprostol. Based on RCTs comparing it to placebo, misoprostol has uterotonic properties when used alone; however, it does not appear to provide an additive uterotonic effect when given in addition to oxytocin (either prophylactically or after PPH occurs). This finding, coupled with the high rates of adverse side effects associated with its use, suggests that misoprostol likely has a limited role in the prevention/treatment of PPH in high-resource settings where oxytocin is available. In low-resource settings where the lack of refrigeration and/or ability to inject medications precludes the use of oxytocin, misoprostol retains an important role as a first-line uterotonic.
NEURAXIAL ANESTHESIA AND POSTOPERATIVE PAIN CONTROL
Four studies evaluate approaches to optimizing neuraxial anesthesia and postoperative pain control. These studies examined the optimal dose of intrathecal morphine for elective cesarean delivery, the use of transversus abdominis plane blocks after cesarean, and the question of the “untested” catheter with combined spinal–epidurals (CSEs).
The first study aggregated RCTs that compared the duration and intensity of analgesia and side effects in low- versus high-dose intrathecal morphine administered for elective cesarean.8 Prior survey work had shown that there is wide variation in what providers use, with typical doses ranging from 50 to 250 µg.9 The authors defined low dose as 50–100 µg and high dose as >100–250 µg. They identified 11 studies, including 480 patients who met the inclusion criteria.
The mean difference (MD) in the primary outcome, the time to supplemental analgesia, was modestly increased in those receiving high- versus low-dose intrathecal morphine (4.49 hours; 95% CI, 1.85–7.13). However, there was no significant difference in pain scores at 12 (MD, 2.54; 95% CI, 2.55–7.63) or 24 (MD, 1.00; 95% CI, 2.50–4.50) hours or in supplemental opioid consumption across the first 24 hours after delivery (MD, 1.31 µg; 95% CI, 5.90–8.53 mg). Assessing side effects, low-dose intrathecal morphine was associated with a markedly lower risk of pruritus (odds ratio [OR], 0.34; 95% CI, 0.20–0.59) and vomiting (OR, 0.38; 95% CI, 0.19–0.75) and a trend toward a lower frequency of using antiemetics (OR, 0.69; 95% CI, 0.32–1.45). Given that low-dose intrathecal morphine seems to provide similar intensity and only modestly shorter duration of analgesia, coupled with a much better side effect profile, these data favor the use of low doses of intrathecal morphine for most parturients.
There may, however, be some patients who are at heightened risk for postoperative pain who would potentially benefit from higher than usual doses of intrathecal morphine to achieve effective pain control. This question was evaluated by Booth et al10 in an RCT. These investigators previously defined a 3-question survey that predicts pain with movement on the first postoperative day. The survey assesses patients’ anxiety level, anticipated pain, and anticipated analgesic needs.
The study randomized 74 patients with predicted pain at or above the 80th percentile to either (1) 300 μg of intrathecal morphine, along with 1 g of acetaminophen every 6 hours for 24 hours; or (2) the usual dose of intrathecal morphine used at their institution, 150 μg, and no acetaminophen. All patients received morphine for intravenous patient-controlled analgesia and scheduled ibuprofen.
In this high-risk group, the intensive analgesic protocol reduced postoperative pain, as measured by multiple primary and secondary outcomes. For the primary outcome, pain with movement at 24 hours measured on a 100-mm visual analog scale, the mean score in the low-dose morphine group was 46 (standard deviation [SD], 25), and the high-dose morphine plus acetaminophen was 31 (SD, 17; P = .009). The investigators found a statistically significant reduction in all secondary measures of pain in the first 24 hours (eg, pain at rest and with movement), except for the measure of maximal pain score across the first 24 hours. They did not observe a difference in the frequency of persistent pain or postpartum depression, although the study was underpowered to detect clinically meaningful differences in these outcomes. There was a trend toward a higher rate of pruritus associated with the higher dose of intrathecal morphine (70% vs 50%, P = .11), but little difference was found in the rate of nausea (60% vs 53%, P = .60).
This study suggests that some patients may benefit from higher doses of intrathecal morphine to achieve optimal analgesia. However, the intervention included a combination of a higher dose of intrathecal morphine and scheduled acetaminophen, and the improvements in analgesia can plausibly be attributable to either of these interventions. Future studies will need to disentangle the relative contribution of each of these components.
The next study is a meta-analysis that sought to clarify the clinical effectiveness of transverse abdominis plane (TAP) blocks after cesarean delivery.11 TAP blocks are peripheral nerve blocks that anesthetize the nerves of the anterior abdominal wall. The investigators divided published studies into 3 groups: (1) studies comparing TAP versus sham/placebo blocks in the absence of intrathecal morphine, (2) studies comparing TAP blocks directly to intrathecal morphine, and (3) studies comparing TAP versus sham/placebo blocks in the absence of intrathecal morphine.
Nine trials compared TAP versus sham/placebo blocks in the absence of intrathecal morphine. The aggregated results from these trials showed a significant reduction in pain scores at rest at 6 hours (MD, −1.43 on an 11-point scale; 95% CI, 2.82 to −0.04) and with movement at 24 hours (MD, −1.54; 95% CI, 3.02 to −0.05) and a trend toward lower pain scores with movement at 6 hours (MD, −1.65; 95% CI, −.35 to 0.22) and at rest at 24 hours (MD, −0.63; 95% CI, 1.38–0.11). Two trials compared TAP blocks to intrathecal morphine. At all time points, both at rest and with movement, intrathecal morphine provided superior analgesia to TAP blocks. Finally, 4 trials compared TAP versus sham/placebo blocks in patients who received intrathecal morphine. These studies demonstrated slightly lower pain scores at 6 hours (at rest: MD, −0.50; 95% CI, 0.92 to −0.08; with movement: MD, −0.89; 95% CI, 1.47 to −0.31), but by 24 hours, there was no difference between the 2 groups.
These data provide important insight into the potential clinical use of TAP blocks. They suggest that neuraxial morphine should be the preferred modality over TAPs for the control of postcesarean pain. They further suggest that TAP blocks are of limited use when given in addition to neuraxial morphine. However, they do point to some benefit when neuraxial opioids cannot be administered due to either a contraindication or because the patient has a general anesthetic. Future work is needed to define whether there is benefit to TAP blocks in patients who are resistant to the effects of neuraxial opioids because they are opioid tolerant or taking a partial agonist such as buprenorphine or if done with long-acting local anesthetics.
The next study addresses the question of the “untested epidural catheter” when placed as part of a CSE technique.12 CSEs offer certain advantages over traditional epidurals, including faster onset. However, historically, there has been concern about using CSEs in patients at high risk for cesarean delivery because the spinal component of a CSE may potentially mask a poorly functioning epidural catheter.
In this single-center, retrospective cohort study, anesthetic, obstetric, and quality assurance records were reviewed for 1440 CSEs and 955 traditional epidurals. The primary outcome was the rate of catheter failure; secondary outcomes assessed included the proportion of failed catheters, time to failure, and frequency of failure at cesarean delivery.
They found that the rate of failure was significantly lower for CSEs than traditionally placed epidurals (hazard ratio, 0.58; 95% CI, 0.44–0.77). The proportion of catheters that failed overall and at cesarean was significantly lower in the CSE group (6.6% vs 11.6%, P = .001), and the median time to recognize failure was also shorter (41 minutes, interquartile range, 7–267 vs 75 minutes; interquartile range, 15–324; P = .048). These findings debunk the concern regarding the untested catheter and suggest that the choice to use a CSE or traditional epidural technique should be guided by other considerations.
SYSTEMS-BASED INTERVENTIONS AND OBSTETRICAL OUTCOMES
The next 2 studies focus on systems-based initiatives that can improve obstetrical outcomes. The first focuses on an early warning trigger tool and the second on the impact of operating room temperature. Also discussed is a new consensus bundle from the National Partnership for Maternal Safety.
During the past 20 years, there has been a rapid rise in maternal morbidity in the United States. Early warning systems are promoted as a systems-based intervention to facilitate early recognition and timely treatment for women at risk for clinical deterioration.13 Shields et al14 published the first prospective observational design to demonstrate that maternal morbidity can be reduced using a Maternal Early Warning Trigger (MEWT) tool tied to clinical protocols. In this prospective observational cohort, 6 California hospitals served as pilot sites for the introduction of MEWT tools, and 23 provided data as control sites. In pilot hospitals, the MEWT system was structured to both identify abnormal vital sign patterns and activate 1 of 4 management pathways based on the specific pattern of vital signs for hemorrhage, sepsis, hypertensive disorders, or cardiopulmonary dysfunction. The primary outcomes were the Centers for Disease Control and Prevention severe maternal morbidity measure and a composite morbidity measure that also included markers of hemorrhage and intensive care unit admission. The investigators compared the rates of these and other secondary outcomes in a 24-month baseline to 13-month post-MEWT introduction study period in both the pilot and control sites.
The trigger tool divided abnormal vital signs into severe and nonsevere range values. A patient would activate the trigger if she had 1 severe range value or ≥2 nonsevere range values sustained for >20 minutes. After the introduction of the tool into the 6 pilot hospitals, there were 12,611 deliveries in the follow-up period, of which 260 (2%) activated the trigger tool. The tool had a sensitivity of 97% for detecting patients who eventually were admitted to the intensive care unit. Interestingly, sepsis was the most commonly activated pathway (71%), followed by the hypertension (15%), hemorrhage (8%), and cardiopulmonary (3%) pathways.
After the introduction of the MEWT system, the frequency of Centers for Disease Control-defined severe maternal morbidity and the composite morbidity measures significantly decreased in the pilot hospitals (2.0%–1.6%, P < .01 and 5.9%–5.1%, P < .01, respectively). Point estimates for all other measures of maternal morbidity, except for sepsis, also decreased. In contrast, in the nonpilot, control hospitals, all measures of maternal morbidity were either stable or increased during the same interval, suggesting that the declines in the pilot sites were not driven by secular trends and were attributable to the impact of the introduction of the MEWT and associated management pathways.
This study is important because it provides what is probably the strongest evidence to date for the efficacy of MEWT in reducing maternal morbidity. The low frequency of triggering, while maintaining a high degree of sensitivity for detecting high-risk patients, suggests that the tool can be deployed on labor and delivery units without causing alarm fatigue or adversely impacting physician or nursing workflow.
The next study is a randomized trial examining the impact of ambient operating room temperature on neonatal and maternal hypothermia.15 Prior studies have shown that hypothermia is common for both the mother and infant after cesarean delivery. Hypothermia has been demonstrated to be particularly morbid for newborn infants and has been associated with hypoglycemia, metabolic acidemia, intraventricular hemorrhage, and respiratory distress. As such, the World Health Organization recommends maintaining delivery room temperatures between 25°C and 28°C (77°F and 82°F),15 far higher than temperatures typically used in obstetric operating rooms. The authors sought to determine whether increasing the operating room temperature at the time of cesarean delivery, and within a range that would be tolerated by providers, would impact the frequency of neonatal and maternal hypothermia.
The study was a single-center, cluster RCT in which operating room temperatures were adjusted weekly to either 20°C (67°F), which is the institutional standard of the study center, or 23°C (73°F). In total, 809 patients participated in the study, 146 of whom were preterm. The primary study outcome was defined as neonatal hypothermia on admission to the nursery defined as <36.5°C.
Higher operating room temperature was associated with a reduction in the frequency of neonatal hypothermia from 50% to 35% (P < .001). The frequency of neonatal moderate/severe hypothermia, defined as <36°C, was also reduced, from 19% to 5% (P < .001). Measures of neonatal morbidity that might be related to hypothermia were not reduced with the intervention, but the authors note they were greatly underpowered to detect meaningful differences in these outcomes. The authors note, as an example, if the outcome of interest associated with avoiding hypothermia was the prevention of mechanical ventilation in infants >34 weeks’ gestation, then a sample size of 25,000 patients would be required.
Maternal hypothermia on admission to the postanesthesia care unit, which was a secondary outcome, decreased from 77% to 69% in association with the higher operating room temperature (P = .008). When members of the surgical team were surveyed, 56% reported discomfort during the periods of higher temperature. However, 93% reported that the increase would be acceptable if neonatal outcomes were improved. While the study did not demonstrate an impact of the intervention on measures of neonatal or maternal morbidity, given the strong association between hypothermia and adverse outcomes, the data support making the default temperature in the operating room on labor and delivery at least 23°C (73°F).
The National Partnership for Maternal Safety published a patient safety bundle on venous thromboembolism (VTE) in 2016.16 The National Partnership is a collaborative effort by 19 core organizations (including the American Society of Anesthesiology and the SOAP) to describe practices and protocols that should be implemented at all hospitals that provide obstetrical care. The National Partnership identified VTE as 1 of 3 key obstetrical complications amenable to improvement, along with hemorrhage and hypertension, which were also the subject of recent bundles.
The VTE bundle encourages that hospitals adopt a standardized approach to prophylaxis. It promotes more aggressive use of pharmacological prophylaxis than previous guidelines in the United States, and it is closer in content to the guidelines from the Royal College of Obstetricians and Gynecologist (the introduction of which is thought to have been instrumental in reducing maternal mortality from VTE in the United Kingdom). Key recommendations include the following: (1) pharmacological prophylaxis with low–molecular weight heparin or twice daily unfractionated heparin in all patients admitted to the hospital in the antepartum period for >72 hours; (2) either pharmacological prophylaxis for all patients after cesarean delivery with an opt-out or adoption of a risk factor–based approach; and (3) pharmacological prophylaxis for patients at high risk for VTE after vaginal delivery. The net result of these new recommendations is that many more obstetrical patients are likely to be anticoagulated, which obviously has important implications for the provision of neuraxial anesthesia. Recently, the SOAP issued a consensus statement on the anesthetic management of parturients receiving pharmacological VTE prophylaxis, which is a useful resource for practitioners navigating these issues.17
Each of the studies presented here provides new and important data that should guide our clinical practice. To that end, a proposed “to-do” list of actionable items based on the literature reviewed is provided (Table). Although the selection of low-dose intrathecal morphine is a straightforward practice change that could be accomplished by anesthesia providers alone, all other recommendations will require multidisciplinary and intraprofessional collaboration. Implementing these practices as part of a multidisciplinary quality improvement effort in collaboration with nurses and obstetricians can help ensure that labor and delivery units are providing up-to-date, evidenced-based care for all patients.
Name: Brian T. Bateman, MD, MSc.
Contribution: This author wrote the manuscript.
This manuscript was handled by: Jill M. Mhyre, MD.
1. Tita AT, Szychowski JM, Boggess K, et al.; C/SOAP Trial Consortium. Adjunctive azithromycin prophylaxis for cesarean delivery. N Engl J Med. 2016;375:1231–1241.
2. Weinstein RA, Boyer KM. Antibiotic prophylaxis for cesarean delivery: when broader is better. N Engl J Med. 2016;375:1284–1286.
3. Azad MB, Konya T, Persaud RR, et al.; CHILD Study Investigators. Impact of maternal intrapartum antibiotics, method of birth and breastfeeding on gut microbiota during the first year of life: a prospective cohort study. BJOG. 2016;123:983–993.
4. Tuuli MG, Liu J, Stout MJ, et al. A randomized trial comparing skin antiseptic agents at cesarean delivery. N Engl J Med. 2016;374:647–655.
5. Quibel T, Ghout I, Goffinet F, et al.; Groupe de Recherche en Obstétrique et Gynécologie (GROG). Active management of the third stage of labor with a combination of oxytocin and misoprostol to prevent postpartum hemorrhage: a randomized controlled trial. Obstet Gynecol. 2016;128:805–811.
6. Tuncalp O, Hofmeyr GJ, Gulmezoglu AM. Prostaglandins for preventing postpartum haemorrhage. Cochrane Database Syst Rev. 2012;8:CD000494.
7. Widmer M, Blum J, Hofmeyr GJ, et al. Misoprostol as an adjunct to standard uterotonics for treatment of post-partum haemorrhage: a multicentre, double-blind randomised trial. Lancet. 2010;375:1808–1813.
8. Sultan P, Halpern SH, Pushpanathan E, Patel S, Carvalho B. The effect of intrathecal morphine dose on outcomes after elective cesarean delivery: a meta-analysis. Anesth Analg. 2016;123:154–164.
9. Aiono-Le Tagaloa L, Butwick AJ, Carvalho B. A survey of perioperative and postoperative anesthetic practices for cesarean delivery. Anesthesiol Res Pract. 2009;2009:510642.
10. Booth JL, Harris LC, Eisenach JC, Pan PH. A randomized controlled trial comparing two multimodal analgesic techniques in patients predicted to have severe pain after cesarean delivery. Anesth Analg. 2016;122:1114–1119.
11. Champaneria R, Shah L, Wilson MJ, Daniels JP. Clinical effectiveness of transversus abdominis plane (TAP) blocks for pain relief after caesarean section: a meta-analysis. Int J Obstet Anesth. 2016;28:45–60.
12. Booth JM, Pan JC, Ross VH, Russell GB, Harris LC, Pan PH. Combined spinal epidural technique for labor analgesia does not delay recognition of epidural catheter failures: a single-center retrospective cohort survival analysis. Anesthesiology. 2016;125:516–524.
13. Mhyre JM, D’Oria R, Hameed AB, et al. The maternal early warning criteria: a proposal from the national partnership for maternal safety. Obstet Gynecol. 2014;124:782–786.
14. Shields LE, Wiesner S, Klein C, Pelletreau B, Hedriana HL. Use of Maternal Early Warning Trigger tool reduces maternal morbidity. Am J Obstet Gynecol. 2016;214:527.e1–527.e6.
15. Duryea EL, Nelson DB, Wyckoff MH, et al. The impact of ambient operating room temperature on neonatal and maternal hypothermia and associated morbidities: a randomized controlled trial. Am J Obstet Gynecol. 2016;214:505.e1–505.e7.
16. D’Alton ME, Friedman AM, Smiley RM, et al. National partnership for maternal safety: consensus bundle on venous thromboembolism. Anesth Analg. 2016;123:942–949.
17. Leffert L, Butwick A, Carvalho B, et al.; Members of the SOAP VTE Taskforce. The Society for Obstetric Anesthesia and Perinatology Consensus Statement on the anesthetic management of pregnant and postpartum women receiving thromboprophylaxis or higher dose anticoagulants. Anesth Analg. 2018;126:928–944.