Secondary Logo

Journal Logo

Obstetric Anesthesiology: Research Reports

A Randomized Comparison of Low Doses of Hyperbaric Bupivacaine in Combined Spinal-Epidural Anesthesia for Cesarean Delivery

Section Editor(s): Wong, Cynthia A.Leo, Serene MMED; Sng, Ban Leong MMED, FANZCA; Lim, Yvonne MMED; Sia, Alex T. H. MMED

Author Information
doi: 10.1213/ANE.0b013e3181b72d35


Combined spinal-epidural (CSE) anesthesia is a common technique for cesarean delivery anesthesia.1 The spinal component allows for rapid onset of dense sensory and motor blockade. The indwelling epidural catheter confers flexibility to supplement the neuraxial block intraoperatively and to extend analgesia into the postoperative period.

Earlier studies reported that the CSE technique was associated with a higher level of sensory block and thus a reduced need for local anesthetic as compared with a single-shot spinal anesthesia.2–4 However, the most recent trial, conducted by Horstman et al.,5 has suggested that dose adjustment for the CSE compared with single-shot spinal anesthesia may not be necessary.

In this study, we aimed to compare the neuroblockade characteristics of 3 different intrathecal hyperbaric bupivacaine doses (7, 8, and 9 mg) when administered as part of a CSE technique in an attempt to elucidate the dose that allows quick and adequate sensory blockade for cesarean delivery while minimizing the side effects of hypotension, nausea, and vomiting. We hypothesized that there would be no difference in block height among 7, 8, and 9 mg of intrathecal hyperbaric bupivacaine used in CSE anesthesia for cesarean delivery.


With the approval of the hospital research ethics committee and written informed patient consent, we enrolled 60 ASA physical status I and II patients with singleton full-term pregnancies who presented for elective cesarean delivery into our study. We excluded parturients who were in labor, those in whom tubal ligation was planned in the same setting, those with allergy to the study drugs, contraindications to central neuraxial blockade, and obstetric complications, such as preeclampsia, multiple gestation, or placenta previa. Parturients who were at the extremes of height and weight (body mass index <20 or >35 kg/m2, height <145 cm or >180 cm) were also excluded.

Before the administration of neuraxial anesthesia, an 18-gauge IV cannula was inserted and standard monitors were applied to each patient. Baseline arterial blood pressure (BP) and heart rate measurements were recorded before the procedure and at 2.5-min intervals throughout the period of study from a noninvasive BP cuff on the right brachial artery with the patients lying supine with a 15° left lateral tilt. All procedures were performed by the principal investigator and a coinvestigator who were proficient in CSE anesthesia in the obstetric population and who each had more than 5 yr of anesthetic experience.

Patients were randomized into 3 groups after enrollment. Group assignment was concealed in opaque envelopes that had been sorted by computer-generated random allocation. Group 7 received hyperbaric bupivacaine 7 mg (1.4 mL of 0.5% bupivacaine with 8% glucose) (Marcain, AstraZeneca, Sodertalje, Sweden) and morphine 100 μg intrathecally. Group 8 received hyperbaric bupivacaine 8 mg (1.6 mL of 0.5% bupivacaine with 8% glucose) with morphine 100 μg, and Group 9 received hyperbaric bupivacaine 9 mg (1.8 mL of 0.5% bupivacaine with 8% glucose) with morphine 100 μg. Intrathecal drugs were drawn into a 3-mL syringe and administered over 10–15 s. All patients received a rapid infusion of warmed Voluven® (hydroxyethyl starch 130/0.4) 15 mL/kg (Fresenius Kabi, Bad Homburg, Germany), which was commenced at the start of intrathecal drug administration and completed within 10–15 min (coload).6–9

Patients were placed in the right lateral decubitus position for the CSE procedure. An 18-gauge Tuohy needle was inserted into the L3-4 interspace using loss of resistance to 2 mL of air for identification of the epidural space. The dura mater was punctured with a 27-gauge Whitacre spinal needle using a needle-through-needle technique (Espocan, B. Braun, Melsungen, Germany). After verifying free flow of cerebrospinal fluid (CSF), the previously prepared spinal injectate was administered with the spinal needle orifice facing cephalad. An epidural catheter was inserted 4 cm into the epidural space. Patients were immediately positioned supine with a 15° left lateral tilt. If unintentional dural puncture with the epidural needle occurred, the epidural catheter was threaded into the subarachnoid space and intrathecal drugs were administered via the catheter.

An anesthesiologist blinded to group allocation was assigned to evaluate the patient’s hemodynamic status and block profile at 2.5-min intervals for the first 30 min and at 5-min intervals thereafter. The level of dermatomal sensory block was tested bilaterally (defined by a complete loss of sensation to ice) after the induction of spinal anesthesia, in an ascending fashion starting from the T12 dermatome. Patient assessment during this period included BP, heart rate, respiratory rate and oxygen saturation, maximum level of sensory block, time needed to reach maximal sensory level, degree of lower limb motor blockade as determined by the modified Bromage score (0 = no impairment; 1 = unable to raise extended legs but able to move knees and ankles; 2 = unable to raise extended legs or to flex knees but able to move feet; 3 = unable to flex ankles, knees, or hips), and time taken to reach maximal Bromage score. In addition, the following data were recorded: duration of surgery and total dose of oxytocin administered postdelivery, neonatal Apgar scores at 1 and 5 min, incidence of hypotension (systolic BP <90 mm Hg or decrease >20% from baseline BP), nausea, vomiting, and shivering (defined as involuntary contraction or twitching of the muscles). Treatment of hypotension consisted of titrated IV boluses of phenylephrine 100 μg or ephedrine 5 mg (when maternal heart rate was <60 bpm) administered by an anesthesiologist blinded to group allocation. Technical problems during CSE anesthesia, such as venous puncture, paresthesia, inability to obtain CSF backflow, and inadvertent dural puncture, were also documented.

Surgery via a Pfannenstiel skin incision was allowed to commence once a bilateral T4 sensory block was demonstrated. If this were not achieved after 15 min, 5-mL aliquots of epidural solution comprising alkalinized 1.5% lidocaine with epinephrine 5 μg/mL was given for block supplementation, up to a maximum of 20 mL. Inadequate anesthesia was defined as a sensory block that failed to reach T4 bilaterally 15 min after intrathecal block or that required epidural supplementation before delivery of the baby. After delivery, patients were given IV oxytocin 10 U in a slow bolus over 10 min. Uterine tone was assessed by the surgeon, and patients were started on an oxytocin infusion (30 U oxytocin in 500 mL at 60–100 mL/h) if necessary. Uterine repair was performed using the surgeon’s preferred technique. Patients were asked to report any intraoperative pain or discomfort using a visual analog scale (VAS) of 0–100 mm. In the event that the VAS was reported as 40 mm or more, or when the sensory block level had receded to T6, boluses of epidural solution would be administered in 5-mL aliquots to supplement or extend the sensory block as required. If patients reported pain or discomfort but sensory block level was higher than T6 and VAS <40 mm, adjuvant systemic analgesics (such as IV fentanyl and inhaled nitrous oxide) or sedative drugs, such as IV midazolam, were administered at the discretion of the attending anesthesiologist blinded to group allocation.

After surgery, all patients were cared for in the recovery area. Hemodynamic and block profile monitoring was continued at 5-min intervals by qualified postanesthetic care unit nurses. Patients were deemed fit for discharge when sensory block level to cold had receded to T10 and the Bromage score had recovered to 2. Perioperative data collection ended upon discharge of the patients back to their wards. All patients were followed up the next day by a blinded observer for postprocedure complications, such as postdural puncture headache, backache, urinary retention, and neurological deficits.

Our primary outcome was the maximum cephalad sensory block height. A sample size of 20 patients per group was calculated to detect a 2-segment difference in maximum sensory block height, based on previous studies2,3 (Kruskal-Wallis test, assuming α = 0.05, β = 0.20). Secondary outcomes included the incidence of hypotension, time taken for regression of sensory block to T10, and recovery of modified Bromage score to 2.

Patient demographic and parametric data (baseline hemodynamic variables, duration of surgery, time to maximal sensory block, and time for block regression) were analyzed using the analysis of variance, with post hoc Bonferroni correction applied if necessary. The Kruskal-Wallis test was used to compare nonparametric data (maximal sensory block level) and post hoc Mann-Whitney U-test for pairwise comparison with Bonferroni adjustment if applicable. The χ2 test was used for categorical data, such as incidence of side effects (shivering, nausea, vomiting, and high block). A P value <0.05 was considered significant. All statistical analyses were performed using Statistical Package for Social Sciences Version 11.5™ software (SPSS, Chicago, IL).


Sixty patients were recruited into our study over a 7-mo period between May 2007 and December 2007, with 20 patients in each group. All 60 patients completed the study. Baseline demographic profiles and patient characteristics were similar in all 3 groups (Table 1).

Table 1
Table 1:
Patient Demographics

The 3 groups differed in the maximum sensory block height achieved (Fig. 1). Group 7 attained a median block height of T2 (interquartile range T2–T3); Group 8 attained a median block height of T2 (interquartile range T1–T2); and Group 9 attained a median block height of T1 (interquartile range C8–T2); P = 0.02. Pairwise comparison showed a difference between Groups 7 and 9 (P = 0.01). There was also a difference in the time taken to reach maximum sensory block height among the 3 groups (Table 2). However, time taken to reach T4, defined as the adequate block height for surgery, was similar in all 3 groups. No patient had inadequate anesthesia for surgery. All patients achieved a maximum Bromage score of 3, and the time taken to reach Bromage score of 3 was similar among groups (Table 2). None of the patients required additional boluses of epidural local anesthetic before delivery of the baby and none required conversion to general anesthesia. Five patients in Group 9 developed blockade > T1, compared with one patient in Groups 7 and 8 (P = 0.08). None of the patients reported any respiratory difficulties or exhibited hypoxemia (defined as Spo2 <92%).

Figure 1
Figure 1:
Figure 1.
Table 2
Table 2:
Block Onset Characteristics

Eight patients in Group 7 required intraoperative block supplementation via the indwelling epidural catheter, compared with 4 patients in Group 8 and 3 patients in Group 9 (P = 0.16). There was no difference in the duration of motor blockade (recovery to Bromage score 2) among groups. After excluding patients who were given intraoperative epidural lidocaine, recovery from sensory blockade to T10 dermatome and recovery from lower limb motor blockade to a Bromage score of 2 was longer in Group 9 compared with Groups 7 and 8 (Table 3). The number of patients requiring supplemental analgesics in all 3 groups was similar. Two patients required Entonox during surgical closure (one from Group 7 and one from Group 8). Eight patients required IV fentanyl postdelivery (4 from Group 7, 2 from Group 8, and 2 from Group 9).

Table 3
Table 3:
Block Recovery Characteristics

There were significant differences in the incidence of hypotension between Groups 7 and 9, although the median total dosages of phenylephrine and ephedrine administered for treatment of hypotension were similar among groups (Table 4). There were no differences in the incidences of nausea, vomiting, and shivering. No neonates had 1- or 5-min Apgar scores <9.

Table 4
Table 4:
Hemodynamic Profiles and Side Effects

Three patients (one from Group 7 and 2 from Group 8) had inadvertent venous puncture. One patient in Group 9 had an inadvertent dural puncture. Her spinal catheter was removed 24 h postdelivery, and she did not develop a postdural puncture headache during the first 2 postoperative days. All patients’ data were included in the analysis according to intention-to-treat principles.


The feasibility of using low-dose CSE for cesarean delivery has been explored in previous studies.10–13 Most of these studies have included short-acting opioids for block supplementation. However, the side effects of such opioids, namely pruritus, may sometimes be uncomfortable or disconcerting in an awake patient undergoing surgery under neuraxial anesthesia. We designed our study to compare block characteristics of varying low doses of hyperbaric bupivacaine without short-acting opioids in order to elucidate an optimal dose which provides rapid and effective anesthesia for cesarean delivery while minimizing side effects. We found that the 3 groups differed in the maximum sensory block height achieved. Although this reached statistical significance, we thought that its clinical significance was debatable, particularly because the cephalad extent of sensory blockade was T4 or higher in all patients, and the mean time to T4 was <5 min and similar in all 3 groups. We noticed that the time taken to reach maximum block height was progressively longer with larger doses. This could be due to a larger mass of drug taking a longer time to reach equilibrium in the CSF.

We also found a significant reduction in the incidence of hypotension when progressively smaller doses were used. This concurs with a randomized controlled trial conducted by Van de Velde et al.,14 which demonstrated that small-dose spinal anesthesia (hyperbaric bupivacaine 6.5 mg) compared with conventional-dose spinal anesthesia (hyperbaric bupivacaine 9.5 mg) in combination with sufentanil better preserves maternal hemodynamic stability with equally effective anesthesia that is of shorter duration.

The literature has suggested that colloids may be superior to crystalloids for fluid administration before neuraxial anesthesia.6–9 In addition, it is possible that fluid coloading may be superior to preloading.15 However, despite our technique of coloading with colloid, the overall incidence of hypotension was rather high. One possible explanation is that our speed of coloading was inadequate, because the mean time to completion of IV 15 mL/kg hydroxyethyl starch administered via pressurized infusion bags ranged from 12 to 13 min. Also, we only studied the occurrence of hypotension and not its severity or duration. The efficacy of intravascular volume expansion with colloid may have been demonstrated in the relatively small requirement for vasopressors in all 3 groups, suggesting that the degree of postblock hypotension was easily treated, perhaps by being of lesser severity or shorter duration.

We acknowledge that there are certain limitations in our study. For instance, we studied a homogeneous population in a single obstetric unit and this may influence the generalizability of our results to patients from other centers. We did not conduct a detailed analysis on the timeframe during surgery when our patients developed hypotension. Hypotension immediately after CSE anesthesia may be correlated with the degree of ensuing sympathetic blockade, but hypotension postdelivery can be influenced by factors such as rapidity of oxytocin infusion, surgical blood loss, or extension of sympathetic blockade with epidural local anesthetic administration. Our study was underpowered to detect a difference in the need for supplemental epidural anesthesia among groups. Another possible limitation lies in the difficulty of accurately withdrawing 7, 8, and 9 mg of hyperbaric bupivacaine into a 3-mL syringe. Lastly, we did not record the time period from administration of neuraxial blockade to first maternal request for supplementary analgesia. It is possible that low-dose CSE anesthesia, with its faster regression of sensory blockade, may lead to larger consumption of nonsteroidal antiinflammatory analgesics or additional opioids in the first 24 h after surgery.

We emphasize that the success of low-dose CSE techniques depends on the placement of a catheter into the epidural space that enables anesthesiologists to extend the central neuraxial block intraoperatively if surgical complications necessitate prolongation of surgery. Unfortunately, accuracy of catheter placement cannot be assessed before commencement of surgery. In the event of catheter misplacement, the faster regression of sensory block height associated with low-dose CSE anesthesia may necessitate conversion to general anesthesia if surgery is protracted. In addition, performing CSE instead of a single-shot spinal anesthesia exposes the patient to the risk of postdural puncture headache from unintentional dural puncture with a large-gauge Tuohy needle. Therefore, the CSE technique should ideally be performed by experienced anesthesiologists or under close supervision by trained anesthesiologists.

In conclusion, hyperbaric bupivacaine 7 mg with 100 μg morphine used in our study provided fast and effective induction of surgical anesthesia for uncomplicated cesarean delivery while reducing the incidence of maternal hypotension compared with 8 and 9 mg. However, the limited duration of surgical anesthesia necessitates the availability of a functioning epidural catheter to supplement/extend neuroblockade if necessary. The need for epidural supplementation may be reduced by using higher bupivacaine doses of 8 and 9 mg, at the expense of increasing the risk of maternal hypotension. Understanding the block characteristics of different doses thus allows better individualization of therapy, with adjustments being made according to the estimated duration of surgery or risk of maternal hypotension.


1. Rawal N, Holmstrom B, Crowhurst JA, Van Zundert A. The combined spinal-epidural technique. Anesthesiol Clin North America 2000;18:267–95
2. Goy RW, Sia AT. Sensorimotor anesthesia and hypotension after subarachnoid block: combined spinal-epidural versus single-shot spinal technique. Anesth Analg 2004;98:491–6
3. Ithnin F, Lim Y, Sia AT, Ocampo CE. Combined spinal epidural causes higher level of block than equivalent single-shot spinal anesthesia in elective cesarean patients. Anesth Analg 2006;102:577–80
4. Goy RW, Yoong CS, Sia AT, Koay CK, Shen L. The median effective dose of intrathecal hyperbaric bupivacaine is larger in the single-shot spinal as compared with the combined spinal-epidural technique. Anesth Analg 2005;100:1499–502
5. Horstman DJ, Riley ET, Carvalho B. A randomized trial of maximum cephalad sensory blockade with single-shot spinal compared with combined spinal-epidural techniques for cesarean delivery. Anesth Analg 2009;108:240–5
6. Siddik SM, Aouad MT, Kai GE, Sfeir MM, Baraka AS. Hydroxyethylstarch 10% is superior to Ringer’s solution for preloading before spinal anesthesia for Cesarean section. Can J Anaesth 2000;47:616–21
7. Dahlgren G, Granath F, Pregner K, Rosblad PG, Wessel H, Irestedt L. Colloid vs. crystalloid preloading to prevent maternal hypotension during spinal anesthesia for elective cesarean section. Acta Anaesthesiol Scand 2005;49:1200–6
8. Ko JS, Kim CS, Cho HS, Choi DH. A randomized trial of crystalloid versus colloid solution for prevention of hypotension during spinal or low-dose combined spinal-epidural anesthesia for elective cesarean delivery. Int J Obstet Anesth 2007;16:8–12
9. Cyna AM, Andrew M, Emmett RS, Middleton P, Simmons SW. Techniques for preventing hypotension during spinal anaesthesia for caesarean section. Cochrane Database Syst Rev 2006:CD002251
10. Choi DH, Ahn HJ, Kim JA. Combined low-dose spinal-epidural anesthesia versus single-shot spinal anesthesia for elective cesarean delivery. Int J Obstet Anesth 2006;15:13–7
11. Coppejans HC, Vercauteren MP. Low-dose combined spinal-epidural anesthesia for cesarean delivery: a comparison of three plain local anesthetics. Acta Anaesthesiol Belg 2006;57:39–43
12. Lim Y, Loo CC, Goh E. Ultra low dose combined spinal and epidural anesthesia for cesarean section. Int J Obstet Anesth 2004;13:198–200
13. Ben-David B, Miller G, Gavriel R, Gurevitch A. Low-dose bupivacaine-fentanyl spinal anesthesia for cesarean delivery. Reg Anesth Pain Med 2000;25:235–9
14. Van de Velde M, Van Schoubroeck D, Jani J, Teunkens A, Missant C, Deprest, J. Combined spinal-epidural anesthesia for cesarean delivery: dose-dependent effects of hyperbaric bupivacaine on maternal hemodynamics. Anesth Analg 2006;103:187–90
15. Dyer RA, Farina Z, Joubert IA, Du Toit P, Meyer M, Torr G, Wells K, James MFM. Crystalloid preload versus rapid crystalloid administration after induction of spinal anaesthesia (coload) for elective caesarean section. Anaesth Intensive Care 2004;32:351–7
© 2009 International Anesthesia Research Society