Secondary Logo

Journal Logo

Spinal Versus Epidural Anesthesia for Cesarean Section: A Comparison of Time Efficiency, Costs, Charges, and Complications

Riley, Edward T. MD; Cohen, Sheila E. MB, ChB; Macario, Alex MD; Desai, Jayshree B. MD; Ratner, Emily F. MD

Obstetric Anesthesia

Spinal anesthesia recently has gained popularity for elective cesarean section.Our anesthesia service changed from epidural to spinal anesthesia for elective cesarean section in 1991. To evaluate the significance of this change in terms of time management, costs, charges, and complication rates, we retrospectively reviewed the charts of patients who had received epidural (n = 47) or spinal (n = 47) anesthesia for nonemergent cesarean section. Patients who received epidural anesthesia had significantly longer total operating room (OR) times than those who received spinal anesthesia (101 +/- 20 vs 83 +/- 16 min, [mean +/- SD] P < 0.001); this was caused by longer times spent in the OR until surgical incision (46 +/- 11 vs 29 +/- 6 min, P < 0.001). Length of time spent in the postanesthesia recovery unit was similar in both groups. Supplemental intraoperative intravenous (IV) analgesics and anxiolytics were required more often in the epidural group (38%) than in the spinal group (17%) (P < 0.05). Complications were noted in six patients with epidural anesthesia and none with spinal anesthesia (P < 0.05). Average per-patient charges were more for the epidural group than for the spinal group. Although direct cost differences between the groups were negligible, there were more substantial indirect costs differences. We conclude that spinal block may provide better and more cost effective anesthesia for uncomplicated, elective cesarean sections.

(Anesth Analg 1995;80:709-12)

Department of Anesthesia, Stanford University School of Medicine, Stanford, California.

Section Editor: Richard J. Palahniuk.

Accepted for publication November 30, 1994.

Address correspondence to Edward T. Riley, MD, Department of Anesthesia, Stanford University School of Medicine, Stanford, CA 94305.

The current emphasis on cost containment mandates evaluation of the economic implications of changes in anesthetic practice [1,2]. With the advent of pencil-point needles and a consequent decreased risk of postdural puncture headaches (PDPH) [3,4], there has been a resurgence in the use of spinal anesthesia in place of epidural anesthesia for cesarean section. In a survey performed at the 1993 meeting of the Society for Obstetric Anesthesia and Perinatology (SOAP), 52% of the conference attendees reported using spinal anesthesia as their technique of choice for elective cesarean delivery (data collected by an interactive computer system at the 1993 meeting of the SOAP and summarized by Michael H. Plumer in the 1993 SOAP program abstracts and membership directory). After changing from epidural to spinal anesthesia for elective cesarean section in 1991, our impression was that spinal anesthesia was quicker to perform and that patients were more comfortable intraoperatively. In contrast to the advantages perceived by the anesthesiologists, our nursing staff claimed that patients remained longer in the postanesthesia care unit (PACU) after spinal than after epidural block. The goal of this study was to compare spinal and epidural anesthesia for cesarean section with respect to time spent in the operating room (OR) and the PACU, complications, and requirements for additional analgesia. We also compared hospital costs and patient charges with the two techniques to determine whether time or resource utilization differences carried significant economic implications.

Back to Top | Article Outline


After approval by the Stanford Human Subjects Committee, we randomly selected and reviewed the charts of 100 patients who had received epidural (n = 50) or spinal (n = 50) anesthesia for nonemergent cesarean section between 1990 and 1992. As our shift from epidural anesthesia to spinal anesthesia occurred in 1991, we were able to study the period immediately before and after changing techniques. During this 2-yr period, anesthesia attending staff members were the same, and nursing protocols and criteria for discharge from the PACU (i.e., 1 h minimum stay, return of lower extremity motor strength, and stable vital signs) remained unchanged. Patients who already had an epidural catheter in place and obese patients (>or=to115 kg) were excluded from the study. Six of the selected 100 patients also were later excluded because their anesthetic management may have been nonstandard and not comparable with the other elective cases. Three of these patients received spinal anesthetics at night and were nonelective cases, and three patients received epidural anesthetics for pregnancy-induced hypertension (2 patients) or thalassemia (1 patient). All remaining cases were performed during normal working hours by the regular obstetric anesthesia faculty working with a second- or third-year anesthesia resident.

The following data were obtained from the anesthetic and PACU records, and progress notes: time elapsed from entering the OR until surgical incision (OR-cut time), total time spent in the OR, and total time spent in the PACU. We also documented anesthetic complications (failed blocks, intravascular injections, inadvertent dural punctures, PDPH), and the use of intra- and postoperative intravenous (IV) opioids or anxiolytics. OR charges, anesthesia professional fees, equipment and drug acquisition costs, and average hourly nursing salaries from 1992 were used to calculate hospital costs and patient charges. Data were analyzed using Student's t-test and chi squared analysis as appropriate. P < 0.05 was considered statistically significant. Values stated are mean +/- SD unless otherwise indicated.

Back to Top | Article Outline


Hyperbaric bupivacaine (0.75% in dextrose) was used in all spinal anesthetics. Lidocaine 2% with epinephrine was the primary local anesthetic in 43 epidural anesthetics, and bupivacaine 0.5% in one epidural anesthetic. Lidocaine was administered in two additional epidural blocks which failed and were converted to spinal anesthesia. One dural puncture with a Tuohy needle was immediately converted to a spinal anesthesia; hyperbaric bupivacaine was used in these cases which we included in the epidural group. All but one patient in the epidural group received 4-5 mg of epidural morphine; all patients in the spinal group received 0.2 mg of spinal morphine. Fentanyl was added to 39 of the spinal anesthetics (10 micro gram dose) and to 15 of the epidural anesthetics (50-100 micro gram dose).

The groups were similar with respect to maternal age, weight, neonatal weight, and Apgar scores Table 1. Significantly less time elapsed from entering the OR until surgical incision with spinal anesthesia as compared with epidural anesthesia Table 2. This resulted in a shorter total time spent in the OR with spinal anesthesia. PACU times were not statistically different between the groups Table 2.

Table 1

Table 1

Table 2

Table 2

Thirty-eight percent of patients in the epidural group received IV analgesic or anxiolytic drugs intraoperatively, compared with only 17% of the spinal group (P = 0.04). A similar percentage of patients in both groups required postoperative pain medication in the PACU (15% in the epidural group and 23% in the spinal group).

Six patients in the epidural group and none in the spinal group had complications significant enough to be noted on the anesthetic record or postoperative notes (P = 0.03). We did not consider transient hypotension or the use of ephedrine a "significant" complication. The complications in the epidural group included one dural puncture, one intrathecal catheter, one intravascular needle and catheter, and three inadequate blocks. Of the three inadequate blocks, one was converted to a spinal anesthetic, one had another epidural catheter placed, and one had the catheter adjusted. Two of these anesthesia administrations with complications had excessively long times from entering the OR until incision (78 and 84 min). However, excluding these times from the analysis did not affect the results.

Cost analyses revealed that patients would have been charged an average of $260 more for OR use and anesthesia professional fees with epidural as compared with spinal anesthesia, based on charges for the additional 17 min spent in the OR. Because our institution has a fixed fee for anesthesia materials and equipment, regardless of duration of the procedure, those charges were the same. There were no extra patient charges incurred by complications of epidural blocks, other than the additional time required to perform the blocks. There was no professional fee for epidural blood patches, and neither of the two patients receiving a patch required prolonged hospitalization. Differences in direct hospital costs based on purchase prices for drugs and materials used for spinal and epidural anesthesia, and nursing salaries for the additional OR time were relatively small Table 3.

Table 3

Table 3

Back to Top | Article Outline


The increasing role of managed care in the health services market provides stronger incentives to deliver effective and efficient medical care. In a retrospective manner we have compared two anesthetic techniques, analyzed time and resource utilization, and documented associated complications.

Our data support our initial impression that spinal anesthesia results in a shorter OR time than does epidural anesthesia. This finding is not unexpected, as an epidural block is inherently more time-consuming than a spinal block. To administer an epidural block, the anesthesiologist must progress more slowly with the epidural needle to avoid a dural puncture, the epidural catheter must be threaded and taped, a test dose must be given and the patient observed for 3-5 min to exclude IV or intrathecal placement, and the entire local anesthetic dose must be administered incrementally. Furthermore, the onset of epidural anesthesia is slower than that of spinal anesthesia. The time differences resulting from these factors is enough to account for the observed 17 min difference in OR time. Although critics may claim that anesthesia time is unusually prolonged in a teaching hospital setting as compared with a private practice setting, we content that safe administration of epidural anesthesia must, of necessity, take longer than spinal anesthesia. Moreover, our results agree with those of others who have reported 22 min [5] and 18 min [6] differences between the techniques.

Despite the fact that spinal anesthesia may result in a denser motor block than epidural anesthesia, it did not increase time spent in the PACU. Choice of local anesthetic, rather than technique, is probably a more important determinant of PACU discharge time. Shorter PACU times might have been achieved by using lidocaine in spinal anesthetics and chloroprocaine in epidural blocks. However, the former may provide an inadequate duration of analgesia and the latter antagonizes the analgesia obtained with epidural morphine.

A lesser requirement for supplemental intraoperative analgesia with spinal anesthesia suggests that these patients were more comfortable than those with epidural anesthetics. This may reflect the superiority of spinal anesthesia. It may also relate to patients having received intraspinal morphine at the beginning of the case rather than after delivery, as occurred with epidural anesthesia. In addition, more patients in the spinal group received fentanyl in conjunction with morphine, a practice which improves the quality of spinal anesthesia [7]. Our data lend support to the theory that patients can be made more comfortable with spinal than with epidural anesthesia [8].

The lower complication rate associated with spinal anesthesia was an unexpected finding. Perhaps because it is technically simpler to perform, there is less potential for procedural problems with spinal anesthesia. Of particular note is that significant PDPH developed only in patients receiving epidural anesthesia, reflecting the very low incidence of PDPH with pencilpoint spinal needles [3,4]. The PDPH rate with epidural anesthesia is probably higher than would be expected among nontrainee anesthesiologists, although even in experienced hands a PDPH rate of approximately 0.5% still may be anticipated [9,10].

The charges to patients who had spinal anesthesia were significantly less than those to patients who had epidural anesthesia. However, patient charges do not reflect the true costs of medical procedures and are meaningless in managed or capitated systems. Examining costs is a better way to compare economic advantages among techniques. In this study, the difference in direct costs between the techniques was relatively small (approximately $20). In contrast, the increased indirect costs associated with epidural anesthesia may be of greater consequence. The American Society of Anesthesiologists' Guidelines For Regional Anesthesia In Obstetrics [11] state that there be "a physician with privileges in obstetrics who is readily available" before starting an anesthetic for cesarean section. Therefore, the additional time taken to perform the epidural block not only occupies the anesthesiologist, but also the obstetrician, surgical assistant, and OR nurses. In addition, the OR is occupied for a longer period, which may be significant in a busy obstetrical service. Patient discomfort with PDPH and greater intraoperative discomfort also could be considered additional indirect costs of epidural anesthesia.

One important factor not addressed in this study is the higher incidence of hypotension associated with spinal anesthesia [12]. We did not include this as a complication because hypotension is usually successfully managed with volume loading, vasopressors, and left uterine displacement. When transient, it is without untoward effects on the fetus [12,13]. Although difficult to evaluate in a retrospective study of this nature, we detected no major differences in the condition of the newborn between the techniques.

In conclusion, we believe that spinal anesthesia may be a better choice for elective cesarean section than epidural anesthesia. It is faster to perform, patients are more comfortable, complication rates are lower, and it is more cost effective. However, if a functioning epidural catheter is in place for labor analgesia in a patient who needs a cesarean section, there is no reason not to use the epidural for surgical anesthesia. Epidural anesthesia also may be preferable in medical conditions such as pregnancy-induced hypertension or cardiac disease, where a slower onset of sympathetic block is desirable, or in cases in which prolonged anesthesia may be required.

Back to Top | Article Outline


1. Hudson RJ, Friesen RM. Health care "reform" and the costs of anaesthesia. Can J Anaesth 1993;40:1120-5.
2. Lanier WL, Warner MA, New frontiers in anesthesia research. Assessing the impact of practice patterns on outcome, health care delivery, and cost. Anesthesiology 1993;78:1001-4.
3. Cesarini M, Torrielli F, Lahaye F, et al. Sprotte needle for intrathecal anaesthesia for caesarean section: incidence of post spinal puncture headache. Anaesthesia 1990;45:656-8.
4. Ross AW, Greenhalgh C, McGlade DP, et al. The Sprotte needle and post dural puncture headache following caesarean section. Anaesth Intensive Care 1993;21:280-3.
5. Vegfors M, Cederholm I, Gupta A, et al. Spinal or epidural anesthesia for elective caesarean section? Int Obstet Anesth 1992;1:141-4.
6. Lucy SJ, Naugler MA. Spinal anaesthesia for caesarean section [letter]. Can J Anaesth 1991;38:940-1.
7. Ward ME, Kliffer AP, Gambling DR, et al. Effect of combining fentanyl with morphine/bupivacaine for elective C/S under spinal anesthesia. Anesthesiology 1993;79:A1023.
8. Huang JS, I YY, Tung CC, et al. Comparison between the effects of epidural and spinal anesthesia for elective cesarean section. Chung Hua I Hsueh Tsa Chih [Chin Med J] 1993;51:40-7.
9. Okell RW, Sprigge JS. Unintentional dural puncture. A survey of recognition and management. Anaesthesia 1987;42:1110-3.
10. Norris MC, Leighton BL, De Simone CA. Needle bevel direction and headache after inadvertent dural puncture. Anesthesiology 1989;70:729-31.
11. Guidelines for regional anesthesia in obstetrics. American Society of Anesthesiologists. Approved by House of Delegates on October 12, 1988, and last amended on October 30, 1991.
12. Caritis SN, Abouleish E, Edelstone DI, et al. Fetal acid-base state following spinal or epidural anesthesia for cesarean section. Obstet Gynecol 1980;56:610-5.
13. Corke BC, Datta D, Ostheimer GW, et al. Spinal anaesthesia for caesarean section. The influence of hypotension on neonatal outcome. Anaesthesia 1982;37:658-62.
© 1995 International Anesthesia Research Society