Complications of Anesthesia for Cesarean Delivery

Steven L. Bloom, MD, Catherine Y. Spong, MD, Steven J. Weiner, MS, Mark B. Landon, MD, Dwight J. Rouse, MD, Michael W. Varner, MD, Atef H. Moawad, MD, Steve N. Caritis, MD, Margaret Harper, MD, Ronald J. Wapner, MD, Yoram Sorokin, MD, Menachem Miodovnik, MD, Mary J. OSullivan, MD, Baha Sibai, MD, Oded Langer, MD, and Steven G. Gabbe, MD, for the National Institute of Child Health and Human Development MaternalFetal Medicine Units Network*
Objective: To quantify anesthesia-related complications associated with cesarean delivery in a well-described, prospectively ascertained cohort from multiple universitybased hospitals in the United States and to evaluate whether certain factors would identify women at increased risk for a failed regional anesthetic. Methods: A prospective observational study was conducted of women (n 37,142) with singleton gestations undergoing cesarean delivery in the centers forming the National Institute of Child Health and Human Development MaternalFetal Medicine Units Network. Detailed information was collected regarding choice of anesthesia and procedure-related complications, including failed regional anesthetic and maternal death. Potential risk factors for a failed regional anesthetic were analyzed. Results: Of the women studied, 34,615 (93%) received a regional anesthetic. Few (3.0%) regional procedures failed, and related maternal morbidity was rare. Increased maternal size, higher preoperative risk, rapid decision-to-incision
* For a list of other members of the National Institute of Child Health and Human Development MaternalFetal Medicine Units Network, see the Appendix. From the Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, Dallas, Texas. Supported by grants (HD34116, HD27915, HD27869, HD34208, HD34136, HD27917, HD21410, HD27860, HD34122, HD27905, HD21414, HD27861, HD34210, HD36801) from the National Institute of Child Health and Human Development. Presented at the 24th Annual Meeting of the Society for MaternalFetal Medicine, February 6, 2004, New Orleans, Louisiana. Corresponding author: Steven L. Bloom, MD, Department of Obstetrics and Gynecology, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-9032; e-mail: steven.bloom@utsouthwestern.edu. 2005 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/05

interval, and placement later in labor were all significantly related to an increased risk of a failed regional procedure. Of the general anesthetics, 38% were administered when the decision-to-incision interval was less than 15 minutes. Women deemed at the greatest preoperative risk (American Society of Anesthesiologists score 4) were approximately 7-fold more likely to receive a general anesthetic (odds ratio 6.9, 95% confidence interval 5.83 8.07). There was one maternal death, due to a failed intubation, in which the anesthetic procedure was directly implicated. Conclusion: Regional techniques have become the preferred method of anesthesia for cesarean delivery. Procedure-related complications are rare and attest to the safety of modern obstetric anesthesia for cesarean delivery in the United States.
(Obstet Gynecol 2005;106:2817)

Level of Evidence: II-2

n 2002, the rate of cesarean delivery in the United states was 26.1%, the highest ever recorded.1 According to the most recent National Hospital Discharge Survey, this rate corresponded to more than 1 million women being delivered by cesarean, making it the most prevalent major nondiagnostic operation performed in the United States in 2002.2 With contemporaneous concerns about the safety of vaginal birth after cesarean, the relationship between child birth and pelvic floor injury, and a burgeoning desire of women to be able to choose to deliver by elective cesarean, it is quite likely that the rate of cesarean birth will increase even further.35 As obstetricians weigh the risks and benefits of vaginal compared with cesarean delivery in such increasing numbers, one of the factors most often requiring consideration is the
OBSTETRICS & GYNECOLOGY 281

VOL. 106, NO. 2, AUGUST 2005

safety of anesthesia for cesarean delivery. Given the ascendancy of cesarean surgery, what is actually known about related anesthetic maternal mortality and morbidity? In 1987, the Centers for Disease Control and Prevention established an ongoing National Pregnancy Mortality Surveillance System to monitor maternal deaths at the national level and to conduct epidemiological studies of the deaths of pregnant women. When these vital statistics data were recently reviewed, the information regarding anesthesia-related morbidity and mortality was judged to be very limited.6 Reasons for this limitation included incomplete ascertainment of the number of deaths as well as incomplete information on anesthetic procedures and no information on anesthesia-related morbidity. Our purpose was to rigorously profile anesthesia-related maternal mortality and morbidity associated with cesarean delivery using a well-described cohort prospectively ascertained from multiple collaborating centers in the United States. In addition, we sought to evaluate whether certain risk factors would identify women at increased risk for a failed regional anesthetic.

MATERIALS AND METHODS

The Maternal-Fetal Medicine Units Network was established in 1986 by the National Institute of Child Health and Human Development (NICHD) to study clinical questions in obstetrics. Every 5 years university-based clinical centers compete to join the MaternalFetal Medicine Units Network, which at the time of this study comprised 13 institutions and a datacoordinating center. These centers conduct research under a cooperative agreement with one another and the NICHD. Each center and the data-coordinating center received institutional review board approval for this study. The cesarean registry was a prospective observational study conducted between 1999 and 2002 and designed to assess several specific contemporary issues related to cesarean delivery. For the first 2 years, data were collected on all women undergoing cesarean delivery or attempted vaginal birth after a prior cesarean at a participating center. During 2001 and 2002, data were collected only on repeat cesarean deliveries and attempted vaginal births after prior cesarean. That is, data on primary cesarean deliveries were not collected during the last 2 years. For the purposes of this analysis, only data collected during the first 2 years were included, thereby permitting complete ascertainment of all cesarean births performed within the MaternalFetal Medicine Units Network during a 2-year time span. Detailed information regarding medical and obstetric history, intrapartum course, postpartum com282 Bloom et al

plications diagnosed before hospital discharge, and infant outcome were abstracted directly from maternal and infant charts by specially trained and certified research nurses. For the purpose of this specific analysis, information was collected regarding choice of anesthetic as well as procedure-related complications. Other variables collected included body mass index (BMI), calculated using the formula weight (kg) divided by height (m2), and emergency cesarean delivery, defined as those operations with skin incision timed within 15 minutes of the first decision recorded to operate. American Society of Anesthesiologists (ASA) scores assigned by obstetric anesthesia providers during preoperative assessments of each patient were also recorded.7 Healthy parturients usually incur a minimum score of 1 or 2, using a 15 scale. As the score increases, the preoperative condition of the patient is considered more guarded. Specifically, a score of 1 denotes no increased risk, 2 implies mild-to-moderate preoperative risk, 3 equates to severe preoperative risk, and 4 represents the presence of a life-threatening illness with or without surgery. A score of 5 is reserved for the moribund patient who has little chance of survival but who is submitted to surgery as a last resort. Data from the 13 centers were transmitted weekly by telecommunications link to the data coordinating center, where they were edited for missing, out of range, and inconsistent values. Edit reports were then transmitted to each center for correction or clarification. Anesthetic procedures used for cesarean delivery were grouped into spinal (subarachnoid block), epidural, combined spinal-epidural, or general, based upon the first anesthetic administered. Women given a general anesthetic because the initial regional procedure was inadequate were categorized as failed regional anesthetics. Shown in Table 1 are the anesthetic complications ascertained in this study and their corresponding definitions. The charts of all maternal deaths were reviewed in detail to determine whether the death was related to the anesthetic procedure. The initial review was conducted by the principal investigator at each site, followed by a central review conducted by 3 members of the protocol subcommittee. The study sample size was calculated based on the primary goal of the cesarean registry, which was to study uterine rupture following trial of labor.3 For the purpose of that analysis, an estimated 12,000 women with prior cesareans undergoing a trial of labor were required. The study was terminated after 4 years when the sample size for the uterine rupture analysis was sufficient. Chi-square and Fisher exact tests were used to
OBSTETRICS & GYNECOLOGY

Anesthesia Complications for Cesarean Birth

Table 1. Definitions of Anesthesia-Related Complications

Complication
High spinal Failed intubation Failed regional anesthesia Spinal headache or blood patch Chemical meningitis Epidural hematoma Extradural abscess

Definition
Respiratory suppression requiring intubation Inability to place an endotracheal tube Epidural or spinal attempted but general anesthesia required Severe positional headache following delivery and/or instillation of blood into the epidural space for relief of headache Meningitis, not viral or bacterial, following regional anesthesia Collection of blood beneath epidural site causing back pain and neurological dysfunction An abscess outside the dura mater

compare the rates among women receiving different types of anesthesia. Means were compared with Student t tests. Odds ratios and 95% confidence intervals for newborn infant outcomes were calculated by using logistic regression models, with adjustment for cesarean indication, gestational age at delivery, and a binary indicator for emergency cesarean delivery. Confidence limits for rates of anesthetic complications are exact limits for the binomial proportions. Confidence limits for maternal mortality are based on a Poisson distribution. All P values reflect 2-tailed statistical analyses. Unless otherwise noted, all statistics are unadjusted for other factors.

Table 2. Demographic Characteristics

Characteristic
Maternal age (y) Mean ( SD) Range Race White Black Hispanic Other or unknown Nulliparous Education, years completed* Mean ( SD) Range Prenatal medical care

Overall Cohort (n 37,142)

27.8 (6.4) 1151 14,670 (39.5) 11,316 (30.5) 9,222 (24.8) 1,934 (5.2) 15,120 (40.9) 12.1 (2.8) 016 36,169 (97.4)

RESULTS
Of the 70,442 women entered into the cesarean registry, 47,112 delivered within the first 2 years. Of these, a total of 9,970 were excluded from this analysis, including 7,829 with prior cesareans who delivered vaginally, 2,073 with multiple fetuses, and 68 with incomplete anesthesia records. The remaining 37,142 women included 21,809 (59%) who underwent a primary cesarean delivery, 12,576 (34%) who underwent a repeat cesarean delivery, and 2,757 (7%) who had an unsuccessful attempt at a vaginal birth after a prior cesarean. Over 93% of the women studied received a regional anesthetic. Specifically, 14,797 (40%) women received a spinal as the first anesthetic attempted; 15,443 (42%) women received an epidural as the first anesthetic attempted; and 4,375 (12%) women received a combined spinal-epidural as the first anesthetic attempted. The remaining 2,527 (7%) women received a primary general anesthetic. Shown in Table 2 are the demographic characteristics for the 37,142 women analyzed in this study. The distribution of cesarean deliveries according to indication and type of anesthesia initially attempted is shown in Table 3. Subarachnoid block (spinal) was the most commonly used method in women undergoing a repeat cesarean, whereas epidural was the most commonly used method in women undergoing a primary cesarean delivery. General anesthesia was
VOL. 106, NO. 2, AUGUST 2005

SD, standard deviation. Data are expressed as n (%) unless otherwise indicated. * Years of education are truncated at 16.

commonly used when the need for delivery was most urgent. Indeed, 900 (38%) of the general anesthetics administered were in emergency cesareans in which the decision-to-incision interval was less than 15 minutes. Maternal preoperative status, as reflected by ASA score, was also related to choice of anesthesia. For example, women deemed at the greatest preoperative risk (ASA score 4 or more) were approximately 7-fold more likely to receive a general anesthetic (odds ratio 6.9, 95% confidence interval [CI] 5.83 8.07). Women with preeclampsia, however, were more likely to receive regional analgesia than a general anesthetic (90% versus 10%), as were women with eclampsia (70% versus 30%, overall P .001). Infant outcomes according to type of anesthesia for cesarean delivery are shown in Table 4. In this analysis, each anesthetic subgroup was compared with the remainder of the entire cohort. For example, the odds ratio of various outcomes for infants in the general anesthesia group were in relation to the infants in all other subgroups combined. Odds ratios were adjusted for indication for cesarean, gestational age at delivery, and emergency cesarean. Newborn infant condition at birth, as reflected in low Apgar
Anesthesia Complications for Cesarean Birth 283

Bloom et al

Table 3. Type of Initial Anesthetic Procedure According to Indication for Cesarean Delivery, Preoperative ASA Score, and Decision-to-Incision Time Less Than 15 Minutes
Spinal (n 14,797)
6,537 848 (13) 1,148 (18) 2,708 (41) 1,833 (28) 8,260 6,157 (75) 1,497 (18) 606 (7.3)

Characteristic
Primary cesarean Total Dystocia Fetal distress Malpresentation Other Repeat cesarean Total No labor Labor Failed VBAC Emergency cesarean* Decision-to-incision time 15 min ASA Score 1 or 2 3 4 5 Pregnancy hypertension None Preeclampsia Eclampsia

Epidural (n 15,443)
11,401 6,591 (58) 2,764 (24) 968 (8.5) 1,078 (9.5) 4,042 1,787 (44) 478 (12) 1,777 (44)

Combined Spinal-Epidural (n 4,375)

1,964 777 (40) 384 (20) 374 (19) 429 (22) 2,411 1,733 (72) 461 (19) 217 (9.0)

General (n 2,527)
1,907 51 (2.7) 908 (48) 276 (14) 672 (35) 620 295 (48) 168 (27) 157 (25)

241 (1.9) 12,754 (88) 1,646 (11) 33 (0.2) 0 13,028 (88) 1,748 (12) 21 (0.1)

922 (6.2) 12,802 (86) 2,093 (14) 64 (0.4) 0 12,775 (83) 2,634 (17) 33 (0.2)

100 (2.4) 3,696 (86) 599 (14) 7 (0.2) 0 3,734 (85) 631 (14) 10 (0.2)

900 (38) 1,648 (67) 710 (29) 85 (3.5) 0 1,947 (77) 552 (22) 27 (1.1)

ASA, American Society of Anesthesiologists; VBAC, vaginal birth after cesarean. Data are expressed as n (%). * Data are missing for 7% of patients. Includes women diagnosed either with preeclampsia, gestational hypertension, or hemolysis, elevated liver enzymes, low platelets (HELLP) syndrome.

Table 4. Odds Ratio of Selected Newborn Infant Outcomes According to Type of Anesthesia Compared With the Remainder of the Cohort
Type of Anesthesia Spinal (n 14,471)
0.52 (0.460.58) 0.52 (0.390.69) 0.49 (0.370.65) 1.13 (0.901.41)

Characteristic
Apgar score 3 at 1 min 3 at 5 min Umbilical artery blood pH 7.0* Neonatal death

Epidural (n 14,753)
0.89 (0.800.99) 0.57 (0.420.77) 0.84 (0.651.08) 0.71 (0.530.93)

Combined Spinal-Epidural (n 4,293)

0.70 (0.590.84) 1.01 (0.651.55) 1.08 (0.691.68) 0.96 (0.641.44)

General (n 2,485)
2.86 (2.523.25) 2.68 (2.033.53) 2.71 (2.063.56) 1.17 (0.901.52)

General for Failed Regional (n 1,043)

2.46 (2.032.97) 2.69 (1.764.10) 1.15 (0.721.82) 1.19 (0.672.09)

Excluded are 3 cases without birth data and 94 stillbirths. Data are expressed as odds ratio (95% confidence interval). Odds ratios are adjusted for cesarean delivery indication, birth gestational age, and emergency cesarean. * Data are missing for 57% of patients.

scores and umbilical artery blood pH, was related to the type of anesthesia for cesarean delivery. Although these measures were about twice as common with general anesthesia compared with regional techniques, there was no difference in the rate of neonatal death associated with general anesthesia. Of note, infant outcomes for those women who required general anesthesia for a failed regional procedure were
284 Bloom et al

not worse than for those women who received a primary general anesthetic. As shown in Table 5, maternal morbidity attributable to regional anesthesia was rare. The most common complication was a failed regional procedure requiring general anesthesia, which occurred in 1,053 (3.0%) of the 34,615 women who received a regional anesthetic. Epidural anesthetics failed signifOBSTETRICS & GYNECOLOGY

Anesthesia Complications for Cesarean Birth

Table 5. Anesthetic Complications According to the Method of Regional Anesthesia First Attempted
Spinal (n 14,797)
313 (2.1, 1.92.4) 73 (0.5, 0.40.6) 27 (0.2, 0.10.3) 9 (0.06, 0.030.12) 0 0 0

Complication
Failed regional Spinal headache Blood patch High spinal Extradural abscess Chemical meningitis Epidural hematoma

Epidural (n 15,443)
666 (4.3, 4.04.6) 41 (0.3, 0.20.4) 32 (0.2, 0.10.3) 11 (0.07, 0.040.13) 0 0 0

Combined Spinal-Epidural (n 4,375)

74 (1.7, 1.32.1) 21 (0.5, 0.30.7) 18 (0.4, 0.20.7) 3 (0.07, 0.010.20) 0 0 0

Data are expressed as n (%, exact 95% confidence interval).

icantly more than spinal or combined spinal-epidural procedures (4.3% versus 2.1% or 1.7%, P .001, respectively). The overall incidence of postdural puncture (spinal) headache was 0.4%, and 57% of these women were treated with a blood patch. Of the 62 women who developed a postdural puncture headache following an epidural or a combined spinalepidural procedure, 50 (81%) received a blood patch. Other more serious morbidities were exceedingly rare and did not differ based upon the type of regional anesthetic administered. Selected characteristics of women who suffered a failed regional anesthetic are shown in Table 6. Increased maternal BMI, higher preoperative ASA score, rapid decision-to-incision interval, and placement later in labor were all significantly related to an increased risk of a failed regional anesthetic. Of the 29 maternal deaths in the cohort, one was attributed directly to the anesthetic procedure. This death occurred in a woman with a skeletal dysplasia, which precluded regional anesthesia, and class F diabe-

tes. The patient suffered a cardiac arrest due to hypoxia following a failed attempt at awake intubation.

DISCUSSION
We performed a large study of anesthetic use and complications in more than 37,000 women undergoing cesarean delivery from 13 university-based hospitals in the United States. Our results indicate that regional techniques are clearly the most common method of anesthesia, accounting for 93% of the procedures performed. Few (3.0%) of the regional procedures failed, and these were more likely to be associated with increased maternal size, higher preoperative risk, rapid decision-to-incision interval, and placement later in labor. Maternal mortality and serious morbiditysuch as high spinal, meningitis, and central nervous system hematomawere rare, and this attests to the safety of modern obstetrical anesthesia for cesarean delivery. During the last 2 decades, there has been an increasing impetus to avoid general anesthesia in

Table 6. Characteristics of Women Who had a Failed Regional Anesthetic

Failed Regional Requiring General Anesthesia (n 1,053)
33.7 7.8 697 (71) 840 (83) 166 (16) 7 (0.7) 173 (17) 306 (40) 432 (57) 21 (2.8)

Characteristic
Body mass index at delivery (kg/m2) Mean ( SD) 29 ASA Category 1 or 2 3 4 Decision-to-incision time 15 min Cervical dilatation when regional placed (cm)* 03 49 10

Successful Regional (n 33,562)

33.0 7.2 21,715 (68) 28,412 (87) 4,172 (13) 97 (0.3) 1,090 (3.5) 10,143 (51) 9,392 (47) 510 (2.5)

P
.004 .06 .001

.001 .001

SD, standard deviation; ASA, American Society of Anesthesiologists. Data are expressed as n (%) unless otherwise indicated. * Data are missing for 40% of patients.

VOL. 106, NO. 2, AUGUST 2005

Bloom et al

Anesthesia Complications for Cesarean Birth

285

women undergoing cesarean delivery because of the increased incidence of failed endotracheal intubation in pregnant women compared with nongravid patients.8 This led to a shift toward regional anesthesia, such that by 1992, 89% of all cesareans performed in hospitals with at least 1,500 birthsas was the case for all of the centers that participated in the cesarean registrywere estimated to have been performed with regional techniques, compared with 62% in 1981.9 Our results suggest that this shift to regional anesthetics has continued. With such a significant change in practice, it was reassuring to find that the rate of anesthesia-related complications was low. Given the heightened concerns about endotracheal intubation in pregnant women, it is noteworthy that the one maternal mortality attributable to anesthesia in our study was a result of a failed intubation. The first national study of anesthesia-related maternal mortality in the United States was for the period 1979 1990.8 The 19851990 maternal case fatality rate for regional anesthesia in women undergoing cesarean delivery reported in this national study was 0.19 per 100,000, compared with 3.2 per 100,000 in women undergoing cesarean with a general anesthetic. These maternal death rates were based upon estimates using linked maternal-infant death certificates and extrapolations based upon the prevailing rates of cesarean delivery. It must be emphasized, however, that pregnancy-related deaths are underreported, and the true number of deaths related to pregnancy might increase 30 150% with improved ascertainment.10 The maternal mortality rates in our study attributable to regional and general anesthetics for cesarean delivery were 0 (95% CI 0 11) and 28 (95% CI 0.7156) per 100,000, respectively. It should be emphasized that these rates were calculated from a completely ascertained cohort and strengthened by a prospective review of patient records, both features not possible in the first national study.11 Of note, the rates measured in the earlier national study do fall within the 95% confidence intervals that we found. To put these mortality rates into some perspective, the rates of death in the United States due to accidental drowning and motor vehicle accidents in 1998 were 1.5 and 15.2 per 100,000, respectively.12 It is important to note that there are several potential limitations to extrapolating the complication rates observed in our study to the general population. First, we did not collect data on the qualifications of the individuals who actually administered the anesthetic. That is, the provider could have been a faculty anesthesiologist, a resident or fellow in training, or a certified nurse anesthetist. Second, the precise cause of failed regionals was not recorded. Third, because of the referral nature of the participating centers, it is
286 Bloom et al

possible that the women included in this study were at an increased risk of complication. Alternatively, it is possible that receiving care at well-equipped tertiary facilities favored a lower chance of an anestheticrelated complication. In spite of these limitations, we are of the view that the findings of our study provide the most accurate and comprehensive information currently available on the outcomes of anesthesia for cesarean delivery. This is buttressed by the fact that we were able to achieve complete ascertainment of all women delivering by cesarean during the study time period. Moreover, all charts were reviewed by a dedicated team of research nurses who received specialized training for this study, and all data collected were subjected to a highly regimented process designed to reduce error. Our purpose was not to determine the ideal anesthetic for cesarean delivery, but instead, to describe contemporary anesthetic practices and their attendant complications. We avoided making judgments about the effects of various anesthetic methods on maternal or infant outcomes. We thought this prudent because of the multiple interacting and confounding variables that enter into the decision to administer a specific anesthetic in a specific situation. Indeed, we noted remarkable integration of obstetric and anesthetic care practices in our review of the data. For example, general anesthetics were chosen for those cesareans in which the fetus was judged to need very rapid delivery. Similarly, preoperative maternal characteristics, as reflected in the ASA score, suggest that obstetric, pediatric, and anesthetic concerns are all considered and inevitably intertwined in the choice of anesthesia. Commenting in the 12th edition of Williams Obstetrics13 on a report from the Chicago Lying-In Hospital14 that described 48 anesthetic fatalities among 242 cesarean-related maternal deaths in 1952, Nicholson Eastman and Louis Hellman wrote that anesthesia indeed is becoming a greater hazard than the operation itself. Now, a half-century later, the story has dramatically changed. REFERENCES
1. Martin JA, Hamilton BE, Sutton PD, Ventura SJ, Menacker F, Munson ML. Births: final data for 2002. Natl Vital Stat Rep 2003;52(10):1113. 2. DeFrances CJ, Hall MJ. 2002 National Hospital Discharge Survey: advance data from vital and health statistics. No. 342. Hyattsville (MD): National Center for Health Statistics; 2004. 3. Landon MB, Hauth JC, Leveno KJ, Spong CY, Leindecker S, Varner MW, et al. Maternal and perinatal outcomes associated with a trial of labor after prior cesarean delivery. N Engl J Med 2004;351:25819. 4. Guise JM, Berlin M, McDonagh M, Osterweil P, Chan B, Helfand M. Safety of vaginal birth after cesarean: a systematic review. Obstet Gynecol 2004;103:420 9.

Anesthesia Complications for Cesarean Birth

OBSTETRICS & GYNECOLOGY

5. Minkoff H, Powderly KR, Chervenak F, McCullough LB. Ethical dimensions of elective primary cesarean delivery. Obstet Gynecol 2004;103:38792. 6. Hawkins JL. Anesthesia-related maternal mortality. Clin Obstet Gynecol 2003;46:679 87. 7. Barash PG, Cullen BF, Stoelting RK. Clinical anesthesia. 4th ed. Philadelphia (PA): Lippincott Williams & Wilkins; 2001. p. 474. 8. Hawkins JL, Koonin LM, Palmer SK, Gibbs CP. Anesthesiarelated deaths during obstetric delivery in the United States, 1979 1990. Anesthesiology 1997;86:277 84. 9. Hawkins JL, Gibbs CP, Orleans M, Martin-Salvaj G, Beaty B. Obstetric anesthesia work force survey, 1981 versus 1992. Anesthesiology 1997;87:135 43. 10. Chang J, Elam-Evans LD, Berg CJ, Herndon J, Flowers L, Seed KA, et al. Pregnancy-related mortality surveillanceUnited States, 19911999. MMWR CDC Surveill Summ 2003;52:(SS-2):1 8. 11. Chestnut DH. Anesthesia and maternal mortality. Anesthesiology 1997;86:273 6. 12. Brunner B. Time almanac 2003. Boston (MA): Family Education Network; 2002. p. 132 4. 13. Eastman NJ, Hellman LM. Williams obstetrics. 12th ed. New York (NY): Appleton-Century-Crofts; 1961. p. 1198. 14. Gordon CA. Cesarean section death. Am J Obstet Gynecol 1952;63:284 93.

2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

APPENDIX
Other members of the MaternalFetal Medicine Units Network include: 1. University of Texas Southwestern Medical Center:

K. J. Leveno, S. Sharma (Department of Anesthesiology), J. McCampbell, D. Bradford. The Ohio State University: J. Iams, F. Johnson, S. Meadows, H. Walker. University of Alabama at Birmingham: J. Hauth, A. Northen, S. Tate. University of Utah: M. Belfort, F. Porter, B. Oshiro, K. Anderson, A. Guzman. Thomas Jefferson University: A. Sciscione, M. DiVito, M. Talucci, M. Pollock. Wayne State University: M. Dombrowski, G. Norman, A. Millinder, C. Sudz, B. Steffy. University of Pittsburgh and Magee Womens Hospital: K. Lain, M. Cotroneo, D. Fischer, M. Luce. Wake Forest University: P. Meis, M. Swain, C. Moorefield, K. Lanier, L. Steele. University of Miami: G. Burkett, J. Gilles, J. Potter, F. Doyle, S. Chandler. University of Cincinnati: T. Siddiqi, H. How, N. Elder. University of Tennessee: W. Mabie, R. Ramsey. University of Chicago: J. Hibbard, P. Jones, M. RamosBrinson, M. Moran, D. Scott. University of Texas Health Science Center at San Antonio: D. Conway, S. Barker, M. Rodriguez. George Washington University Biostatistics Center: E. Thom, S. Leindecker, H. Juliussen-Stevenson, M. Fischer. National Institute of Child Health and Human Development: D. McNellis, K. Howell, S. Pagliaro.

VOL. 106, NO. 2, AUGUST 2005

Bloom et al

Anesthesia Complications for Cesarean Birth

287