Secondary Logo

Journal Logo

Obstetric Anesthesia

Practice Patterns of Anesthesiologists Regarding Situations in Obstetric Anesthesia Where Clinical Management Is Controversial

Beilin, Yaakov MD; Bodian, Carol A. DrPH; Haddad, Elizabeth M. BS; Leibowitz, Andrew B. MD

Author Information


The obstetric anesthesiologist may be faced with situations in which clinical management is controversial. An example is the use of epidural anesthesia in a parturient with thrombocytopenia. Bromage [1] has recommended that an epidural anesthetic not be placed in a patient with a platelet count below 100,000 mm-3, but this has been challenged by others [2].

Although controversial areas of obstetric anesthesia are discussed in the literature [3], firm recommendations regarding the management of controversial situations are generally not offered, and it is not known how anesthesiologists actually practice when faced with such events.

The lack of published practice patterns makes it difficult to define the standard of care to guide the individual practitioner who is faced with a controversial issue. Also, we are not aware of any study that determines whether there are differences in practice patterns between those in academic and those in private practice, although one report implied that there may be such a difference [4].

Our study was designed to delineate current practice patterns of obstetric anesthesiologists in both academic and private practice in the United States when faced with controversial management situations, and to determine whether there is a difference in the practice patterns between these two groups.


This study was approved by our institutional review board. A 47-question survey was developed by members of the Division of Obstetric Anesthesia and The Department of Biomathematical Sciences. It was reviewed by members of the Research Committee of the Department of Anesthesiology at The Mount Sinai Hospital. Questions described commonly encountered clinical events in which management is controversial. The responding anesthesiologist was asked to answer the questions based on their individual practice and not based on any knowledge of the literature. The survey consisted mainly of multiple choice questions and was divided into two sections. Section one (seven questions) concerned individual practitioner and institutional characteristics. Section two (40 questions) covered the following areas: 1) preoperative laboratory testing for the healthy parturient; 2) preeclampsia and possible coagulopathies; 3) epidural catheter placement in women with "spinal problems"; and 4) use of epidural opioids and intravenous supplementation.

The survey was mailed in January 1995, to the chairpersons of all 153 anesthesiology programs listed in the 1993-1994 American Medical Association's Graduate Medical Education Directory [5] and to an equal number of directors of anesthesia departments in private practice hospitals. The private practice hospitals were selected from the 1993 American Hospital Association's Guide to the Health Care Field [6] and were matched by county with the academic centers. The only other criterion for inclusion was that at least 500 deliveries per year were performed at that institution. If more than one private practice hospital in the particular county met the inclusion criteria, then the selection was done randomly.

The department director was asked to give the survey to the director of obstetric anesthesia or, if there was no designated director of obstetric anesthesia, to the anesthesiologist who provided the most obstetric anesthesia care at that hospital. A cover letter describing the study and a stamped, self-addressed envelope were included for return of the survey. All surveys were designed to be completed anonymously in order to reduce the potential for self-report bias. Only the secretarial staff at the authors' institution were aware of institutions from which no response was received. After 2 mo, all nonresponders were sent a second mailing with the same survey, cover letter, and return envelope.

All data were entered in an Excel Registered Trademark database (Microsoft, Redmond, WA), and converted to a SAS file [7] for statistical analysis. Preliminary analysis of the data was done by performing univariate tests (chi squared tests or Student's t-tests as appropriate) to compare the responses of physicians in private practice versus academic practice. Apparent associations were explored further using Mantel-Haenszel tests to control for confounding variables. Data are presented as mean +/- SD. P < 0.05 was considered significant.


A total of 306 surveys were mailed; 196 anesthesiologists responded to the first mailing and an additional 10 responded to the second for a total of 206 responses (67% response rate). More of the academic (A) anesthesiologists (113/153, 74%) responded than those in private (P) practice (94/153, 61%). Three of the private practice surveys were discarded: one because the hospital reported less than 500 deliveries per year, and two because the respondents completed the survey incorrectly. Data analyses for the private practice anesthesiologists were based on the 91 completed surveys.

There was no detectable difference between the two groups with regard to average age (A = 43 +/- 10 yr, P = 43 +/- 8 yr), sex (A = 69% male, P = 76% male), or the average time since the respondents completed their residency (A = 11 +/- 10 yr, P = 11 +/- 8 yr). More of the academic respondents currently devote >50% of their time to obstetric anesthesia (A = 59%, P = 16%), practice at hospitals with more than 2500 deliveries per year (A = 51%, P = 36%, P < 0.05), and devoted more time during their residency to obstetric anesthesia Table 1.

Table 1
Table 1:
Demographic Data

Initial univariate analyses found the answers to 14 questions to be different between academic and private practitioners. However, further analyses revealed that only eight of these remained significant after accounting for the amount of clinical time currently devoted to obstetric anesthesia (>50% or <or=to50%).

The most frequently ordered laboratory tests prior to epidural placement in the healthy parturient is a complete blood count (A = 45%, P = 68%, P < 0.05) and platelet count (A = 17%, P = 31%, P < 0.05) Table 2. Most of the anesthesiologists (A = 66%, P = 55%) would place an epidural anesthetic in an otherwise healthy parturient who had a platelet count between 80,000 mm-3 and 100,000 mm-3 (Table 3, question 1). If a woman develops a coagulopathy after an epidural catheter has been placed, most (A = 74%, P = 67%) would remove the catheter only after the coagulation status has been documented as corrected (Table 3, questions 12 and 13).

Table 2
Table 2:
Preoperative Laboratory Testing
Table 3
Table 3:
Questions Related to Coagulation and Preeclampsia

Most anesthesiologists (A = 89%, P = 87%) would place an epidural anesthetic both in a woman taking 80 mg of aspirin ("baby" aspirin) per day, and in a woman taking 325 mg of aspirin ("adult" aspirin) per day (A = 67%, P = 66%). Among those who would not routinely place an epidural anesthetic in such women, most would if the bleeding time test were normal ("baby" aspirin, A = 89%, P = 87%; "adult" aspirin, A = 67%, P = 66%) (Table 3, questions 2 and 3).

As the severity of preeclampsia increases, the number of laboratory tests requested also increases. Most anesthesiologists requested a platelet count in mild (A = 93%, P = 93%) and severe (A = 98%, P = 100%) preeclampsia. Fewer anesthesiologists requested a prothrombin time (A = 23%, P = 39%), partial thromboplastin time (A = 21%, P = 39%), or fibrinogen (A = 5%, P = 18%, P < 0.05) in mild preeclampsia than in severe preeclampsia (prothrombin time, A = 73%, P = 90%, P < 0.05; partial thromboplastin time, A = 73%, P = 89%; or fibrinogen, A = 42%, P = 68%, P < 0.05) (Table 3, questions 4 and 5). The bleeding time test is frequently used (A = 48%, P = 76%, P < 0.05) to aid in the decision about placing an epidural anesthetic in a woman with preeclampsia and thrombocytopenia (Table 3, question 6). As the severity of preeclampsia increased there was a change in monitoring desired in both groups (A and P) from a central venous pressure catheter to a pulmonary artery catheter (Table 3, questions 8-11).

The majority of the responding anesthesiologists surveyed would place an epidural anesthetic in a woman with a Harrington rod (A = 86%, P = 81%), scoliosis (A = 96%, P = 93%), or a symptomatic (A = 71%, P = 71%) or an asymptomatic herniated lumbar disk (A = 91%, P = 93%).

Most respondents would not place an epidural anesthetic in a woman with a temperature of >or=to102 degrees F and increased white blood cell count (A = 34%, P = 25%), but would if the temperature decreased to 101 degrees F and antibiotics were given (A = 89%, P = 74%) (Table 4, question 1). Spinal anesthesia is commonly used for an emergent cesarean section (A = 88%, P = 80%), and epinephrine is commonly used as part of an epidural test dose (A = 61%, P = 66%) (Table 4, questions 2 and 3).

Table 4
Table 4:
Miscellaneous Issues

Ketamine is the most commonly used drug to supplement a regional anesthetic during cesarean section (A = 76%, P = 64%), followed by fentanyl (A = 65%, P = 44%, P < 0.05), and midazolam (A = 31%, P = 16%, P < 0.05) (Table 4, question 4). Most of the responding anesthesiologists add opiates to epidural infusions of local anesthetics during labor (A = 94%, P = 93%), and continue the epidural infusion during the second stage of labor (A = 90%, P = 95%) (Table 4, questions 5-7).


The purpose of this study was to assess practice patterns of obstetric anesthesiologists in both academic and private practice when faced with events where clinical management is controversial, and to determine whether there are differences in the practice patterns between these two groups. We want to emphasize that the results of the questionnaire are only descriptive and are not intended to predict outcome or to establish "standards of care."

There are several limitations to our survey that should be acknowledged. We are not certain whether the respondents were truly representative of their respective groups. This is especially true of the private practitioners, where there was only a 61% response rate as compared to a 74% response rate from the academic physicians. In addition, our study is limited to geographic areas that have academic institutions. Also, we cannot determine from the results of the survey whether time devoted to obstetric anesthesia during residency was as a resident or as a fellow.

All respondents were anonymous, and there is no way to verify that the responses truly represent the respondent's practice. It is possible that a negative response to a question where specialized equipment or skills are required (e.g., bleeding time test or thromboelastography) meant that the practitioner could not perform the test because the equipment was unavailable.

Overall, the responses the our questions were similar from academic and private practice physicians. This is in contrast to the results of a survey by Plumer and Rottman [4] of those who attended the 1993 meeting of The Society for Obstetric Anesthesia and Perinatology (Palm Springs, CA). The difference between our results and theirs could be related to the way in which the surveys were distributed. We surveyed one anesthesiologist at each academic institution and one from each of a matched set of nonacademic institutions selected by a random process. Plumer and Rottman [4] surveyed all attendees at an obstetric anesthesia meeting, including nurses, nurse anesthetists, obstetricians, and anesthesiologists, and selection bias may have played a greater role.

The amount of clinical time currently devoted to obstetric anesthesia mitigated some of the differences that we found between academic and private practitioners. Clinical time devoted to a specialty is an important factor in defining practice patterns of pediatric anesthesiologists [8], and should be considered in future studies of obstetric anesthesia practice.

A decrease in routine preoperative laboratory testing is important from both economic and patient standpoints [9]. The most frequently requested test, although only by a minority of anesthesiologists, was a complete blood count. Anemia of pregnancy and the risk of hemorrhage during labor makes the complete blood count a valuable routine test for the parturient although this would not be necessary if the same patient presented for a procedure (e.g., hysterectomy) that is associated with a low risk for bleeding.

Placing an epidural anesthetic when the platelet count was <100,000 mm (-3) is supported by a retrospective study [2]. We found that most anesthesiologists agree and will place an epidural anesthetic when the platelet count is between 80,000 and 100,000 mm-3Table 3.

Although there is one case report of a spontaneous epidural hematoma in a patient who was taking one adult aspirin per day [10], we are not aware of any reports of epidural hematoma in any patient taking aspirin who received an epidural anesthetic. Furthermore, a study found regional anesthesia to be safe in the parturient taking aspirin [11]. We found that most of the respondents would place an epidural anesthetic in a parturient taking either 80 mg or 325 mg of aspirin per day.

The results of our survey do not agree with the results of a survey of 47 British departments of anesthesia reported by Bushnell [12], who found that 77% of those surveyed requested a bleeding time prior to placing a regional anesthetic in a woman taking one "baby" aspirin per day. We find the difference between his results and ours to be surprising considering the medicolegal climate in which American physicians practice.

Interestingly, most of the responding anesthesiologists in our study who would not place a regional anesthetic in a woman taking either 80 mg or 325 mg of aspirin per day would do so if the bleeding time test was normal. The use of the bleeding time test is generally not recommended because it does not predict bleeding at other sites [13].

Preeclampsia may be associated with clotting abnormalities especially thrombocytopenia. We found that most anesthesiologists require that a platelet count be determined in women with both mild and severe preeclampsia. There was less agreement as to the necessity for determining other coagulation variables. This pattern of ordering laboratory tests is consistent with the results of Leduc et al. [14] who found, in the preeclamptic patient, that the prothrombin time, partial thromboplastin time, and fibrinogen concentration remained unchanged until the platelet count was <100,000 mm-3.

The majority of the respondents would request a bleeding time test if there were a doubt as to whether the platelet count is sufficient for placement of a regional anesthetic. This test was requested more frequently by those in private practice. Voulgaropoulos and Palmer [15] also found a significant number of anesthesiologists requesting a bleeding time test in women with preeclampsia.

Although there is no evidence that a normal tracing on a thromboelastogram can determine the safety of epidural anesthesia, some researchers are hoping that it can be used in this fashion [16]. We found that most of the anesthesiologists surveyed would not rely on the results of a thromboelastogram.

The most opportune time to remove an epidural catheter if the patient develops a coagulopathy after placement of the catheter is controversial [17]. We found that most anesthesiologists wait until the coagulation status is corrected before removing the catheter.

Our results are consistent with the literature [18] which supports placing a regional anesthetic in the woman with previous Harrington rod instrumentation and other spine pathologies (e.g., herniated disk), although 29% of anesthesiologists will not place an epidural anesthetic if the patient is symptomatic.

Bacteremia is a relative contraindication to regional anesthesia. Carp and Bailey [19] found an association between lumbar puncture and meningitis in bacteremic rats but prior administration of antibiotics eliminated the risk. Bader et al. [20], in a retrospective study, found no cases of meningitis in parturients who had received regional anesthesia while bacteremic with or without antibiotics. We found that most anesthesiologists prefer to delay regional anesthesia until antibiotics have been administered.

Marx et al. [21] found that Apgar scores were higher when a regional rather than a general anesthetic was used for a cesarean section when there was fetal distress. The current medicolegal climate may influence some anesthesiologists toward choosing general anesthesia for an emergent cesarean section because of the shorter onset time. Based on our results, however, this does not seem to be the case.

In the nonlaboring patient, epinephrine will cause a predictable increase in heart rate, but maternal heart rate varies greatly during labor, so that a tachycardiac response from epinephrine may be difficult to distinguish from the normal variations [22]. We found that approximately 60% of those surveyed use epinephrine in the test dose of an epidural local anesthetic to identify intravascular placement of a catheter.

Ketamine, which has been shown to be both safe and effective, and at small doses does not depress the neonate [23], is the intravenous drug most widely used by our respondents to supplement regional anesthesia during a cesarean section. Midazolam and fentanyl are used less commonly than ketamine, but they are used more by those in academic practice.

Some anesthesiologists are now using smaller concentrations of local anesthetics combined with opiates in the epidural infusion for analgesia during labor. Although the use of epidural opiates is safe for the fetus [24], some anesthesiologists will limit the total dose of opiates used during labor [25]. We found that almost all of the anesthesiologists who responded are using epidural infusions and more than 90% are adding opiates to the infusion.

In conclusion, our findings provide a sense of the practice patterns of anesthesiologists when faced with situations in obstetric anesthesia in which clinical management is controversial. There appears to be a uniform approach to these situations which for the most part is supported by the literature. The only exception may be the use of the bleeding time test to assess platelet function. We found an interaction between the variables 1) type of practice (academic versus private practice) and 2) current clinical time devoted to obstetric anesthesia (> or <or=to50%). Those in private practice are somewhat more cautious, requiring more laboratory tests for the healthy parturient and for the woman with preeclampsia or possible platelet abnormalities. Also, this group is more cautious when faced with the parturient who has "spine" problems or an increased temperature.

The authors express appreciation to James B. Eisenkraft, MD, Professor of Anesthesiology at The Mount Sinai Medical Center, New York, NY, for his critical review of this manuscript.


1. Bromage PR. Neurologic complications of regional anesthesia for obstetrics. In: Shnider SM, Levinson G, eds. Anesthesia for obstetrics. Baltimore: Williams & Wilkins, 1993;443-4.
2. Rasmus KT, Rottman RL, Kotelko DM, et al. Unrecognized thrombocytopenia and regional anesthesia in parturients: a retrospective review. Obstet Gynecol 1989;73:943-6.
3. Santos AC, Pedersen H. Current controversies in obstetric anesthesia. Anesth Analg 1994;78:753-60.
4. Plumer MH, Rottman R. How anesthesiologists practice obstetric anesthesia: responses of practicing obstetric anesthesiologists at the 1993 meeting of the Society for Obstetric Anesthesia and Perinatology. Reg Anesth 1996;21:49-60.
5. American Medical Association. Graduate medical education directory. Chicago: American Medical Association, 1991:166-75.
6. American Hospital Association. The 1991 American Hospital Association's guide to the health care field. Chicago: American Hospital Association, 1991.
7. SAS Institute Inc. SAS user's guide. Cary, NC: SAS Institute Inc., 1985.
8. Lunn JN. Implications of the national confidential inquiry into perioperative deaths for paediatric anaesthesia. Pediatr Anaesth 1992;2:69-71.
9. Roizen MF. The compelling rationale for less preoperative testing [editorial]. Can J Anaesth 1988;35:214-8.
10. Locke GE, Giorgio AJ, Biggers SL, et al. Acute spinal epidural hematoma secondary to aspirin-induced prolonged bleeding. Surg Neurol 1976;5:293-6.
11. De Swiet M, Redman CWG. Aspirin, extradural anaesthesia and the MRC collaborative low dose aspirin study in pregnancy (CLASP). Br J Anaesth 1969;69:109-10.
12. Bushnell TG. A survey of coagulation screening practices in preeclampsia and low-dose aspirin prophylaxis. Int J Obstet Anesth 1994;3:13-5.
13. Channing Rodgers RP, Levin J. A critical reappraisal of the bleeding time. Semin Thromb Hemost 1990;16:1-10.
14. Leduc L, Wheeler JM, Kirshon B, et al. Coagulation profile in severe preeclampsia. Obstet Gynecol 1992;79:14-8.
15. Voulgaropoulos DS, Palmer CM. Coagulation studies in the preeclamptic parturient: a survey. J Clin Anesth 1993;5:99-104.
16. Chadwick HS, Wall MH, Chandler W, Ross BK. Thromboelastography in mild and severe preeclampsia [abstract]. Anesthesiology 1993;79:A992.
17. Sprung J, Cheng EY, Patel S. When to remove an epidural catheter in a parturient with disseminated intravascular coagulation. Reg Anesth 1992;17:351-4.
18. Crosby ET, Halpern SH. Obstetric epidural anaesthesia in patients with Harrington rod instrumentation. Can J Anaesth 1989;36:693-6.
19. Carp H, Bailey S. The association between meningitis and dural puncture in bacteremic rats. Anesthesiology 1992;76:739-42.
20. Bader AM, Gilbertson L, Kirz L, Datta S. Regional anesthesia in women with chorioamnionitis. Reg Anesth 1992;17:84-6.
21. Marx GM, Luykx WM, Cohen S. Fetal-neonatal status following caesarean section for fetal distress. Br J Anaesth 1984;56:1009-13.
22. Leighton BL, Norris MC, Sosis M, et al. Limitations of epinephrine as a marker of intravascular injection in laboring women. Anesthesiology 1987;66:688-91.
23. Akamatsu TJ, Bonica JJ, Rehmet R, et al. Experiences with the use of ketamine for parturition. I. Primary anesthetic for vaginal delivery. Anesth Analg 1974;53:284-7.
24. Bader AM, Fragneto R, Terui K, et al. Maternal and neonatal fentanyl and bupivacaine concentrations after epidural infusion during labor. Anesth Analg 1995;81:829-32.
25. Vertommen J, Vandermeulen E, Van Aken H, et al. The effects of the addition of sufentanil to 0.125% bupivacaine on the quality of analgesia during labor and on the incidence of instrumental deliveries. Anesthesiology 1991;74:809-14.
© 1996 International Anesthesia Research Society