Secondary Logo

Journal Logo

Obstetric anaesthesia

Spinal anaesthesia with low-dose bupivacaine in marginally hyperbaric solutions for caesarean section

A randomised controlled trial

Tang, Wen-Xi; Li, Jian-Jun; Bu, Hui-Min; Fu, Zhi-Jian

Author Information
European Journal of Anaesthesiology (EJA): July 2015 - Volume 32 - Issue 7 - p 493-498
doi: 10.1097/EJA.0000000000000112
  • Free



Baricity is the density ratio between spinal anaesthetic solution (SAS) and cerebrospinal fluid (CSF). The baricity influences the intrathecal spread of local anaesthetics and ultimately the height of the block.1–3 During the third trimester, CSF density decreases, and pregnant women are more sensitive to intrathecal local anaesthetic.4–8 Changes in baricity and dose might be expected to exert more exaggerated effects here than in other cohorts. In support of this hypothesis, conventional hyperbaric anaesthesia with low dose local anaesthetic is associated with a lower incidence of complications such as hypotension, compared with conventional doses.9–12 Clinical studies of nonobstetric anaesthesia have shown that marginally hyperbaric bupivacaine solutions containing ≤0.8% glucose but still slightly denser than CSF, induce a more consistent but lower block, and a lower incidence of hypotension than the conventional hyperbaric solution containing 8% glucose in a lateral position.2,13–15 However, the spinal anaesthesia efficacy of marginally hyperbaric low-dose bupivacaine has not been studied during caesarean section.

The objective of the current study was to compare the efficacy of three different marginally hyperbaric glucose concentrations with low-dose (7.2 mg) bupivacaine with that of the conventional hyperbaric solution in women undergoing elective caesarean section with spinal anaesthesia. We hypothesised that marginally hyperbaric low-dose bupivacaine solutions would be associated with a reduced cephalad spread and a reduced incidence of side effects compared with conventional hyperbaric solution during caesarean section.

Materials and methods

Ethical approval for this randomised, controlled clinical trial (Ethical Committee No. 2011-02) was provided by the Ethics Committee of the Dongying People's Hospital Affiliated with Binzhou Medical University, Dongying, China on 1 February 2011. Informed written consent was obtained from all patients. Pregnant women who were scheduled for elective caesarean section and met the criteria for the American Society of Anaesthesiologists (ASA) physical status classes 1 to 2 were enrolled in this study. Women were excluded if they were younger than 20 years or older than 42 years; had dysfunctional coagulation; had pregnancy-induced hypertension; or had infection around the anaesthesia puncture site; and if the gestational age of the infant was less than 36 weeks. Data were collected from the operating room at the Dongying People's Hospital Affiliated with Binzhou Medical University between November 2011 and October 2012. A computer-generated random number sequence was used for group allocation. The allocation ratio was 1 : 1:1 : 1, and the blocking size was six. To ensure the study was double-blinded, the anaesthesiologist who administered the anaesthesia took no other part in the study, whilst another anaesthesiologist blind to the allocation collected intraoperative and postoperative data. The patients were unaware of the treatment allocation.

Women were randomised into four groups: group A received SAS with 8% glucose (density = 1.02759 ± 0.00003 g ml-1); group B received SAS with 0.8% glucose (1.00255 ± 0.00002 g ml-1); group C received SAS with 0.5% glucose (1.00122 ± 0.00002 g ml-1); and group D received SAS with 0.33% glucose (1.00084 ± 0.00002 g ml-1). Density was determined by using a digital density meter (DMA4500M; Anton Paar GmbH, Graz, Austria) at 37°C, and the data are presented as mean ± SD from three samples to within 0.00001 g ml-1. Fifteen minutes prior to administration of anaesthesia, the SAS was prepared by one nurse who was blind to the patient allocation, according to a planned recipe (Table 1) that included 0.75% bupivacaine (Shanghai Zhaohui Pharmaceutical Co. Ltd., Shanghai, China), 50% glucose, sufentanil and isotonic saline. The doses of bupivacaine and sufentanil, and the volume of SAS were kept constant at 0.45% bupivacaine–sufentanil in 1.6 ml (7.2 mg bupivacaine and 2 μg sufentanil), and only the density of the solution varied.

Table 1
Table 1:
Preparation and composition of spinal anaesthetic solutions and related characteristics

During the operation, the room temperature was kept at 24°C. The body temperature of each patient was maintained by an electric warming blanket, and all solutions used in the surgery were kept at 24°C. Routine monitoring was applied to all. To establish a baseline, blood pressure was measured three times before anaesthesia. Lactated Ringers’ solution (3 ml kg–1) was infused prior to anaesthesia. The combined spinal-epidural anaesthesia (CSEA) was performed with a 16-gauge Tuohy needle and a 26-gauge pencil point spinal needle (Zhejiang Sujia Medical Equipment Co., Jiaxing City, Zhejiang, China) in the lateral position at the intervertebral space between L3 and L4. Then, 1.6 ml SAS was injected intrathecally precisely 32 s after CSF release. Three centimetres of epidural catheter were inserted. The patient was immediately placed in a supine position with 15° left tilt. Facial oxygen at a rate of 7 l min–1 was given and lactated Ringers’ solution at 0.4 ml kg–1 min–1 was infused.

Completion of the spinal injection was considered as time zero. The sensory block was determined by pinprick along both sides of the midclavicular line at 3, 5, 10, 15, 20, 25, 30, 40 and 50 min postinjection. If there was a discrepancy between the left and right side, the average block level was used. Surgery began 10 min after the spinal injection. If the sensory block had not reached T8 before surgery, or there was moderate pain during the skin incision, spinal anaesthesia was considered to have failed. In that event, epidural lidocaine was given by 5 ml bolus to a total amount of 20 ml. General anaesthesia was performed if there was no improvement.

The efficacy of anaesthesia was evaluated using three criteria: lower limb motor block, muscle relaxation and pain during skin incision and abdominal exploration. The lower limb motor block was evaluated using the following scale: 0, able to straight leg raise; 1, free flexion of the knees; 2, no knee movement but mobile ankle joints; 3, no ankle movement but toes still active; and 4, no movement in the lower limb. Muscle relaxation was evaluated using a similar subjective scale and judged by the obstetrician as good (satisfactory), poor (unsatisfactory but surgery possible) and very poor ( further intervention needed to permit surgery). Pain during skin incision and abdominal exploration was graded as none, mild (tolerable pain), moderate (treatment required) or severe (intolerable pain). Blood pressure was measured every 2 min after administration of the SAS. A patient was considered hypotensive if the systolic blood pressure was lower than 90 mmHg or decreased by more than 30% of the basal value. To treat intraoperative hypotension, 6 mg ephedrine was injected followed by 3 mg every 2 min until the blood pressure was restored to normal. Instances of nausea, vomiting, shivering and pruritus were also recorded.

On the basis of the results of a pilot study, 30 patients in each group were needed to compare a one-segment difference (standard deviation 3) in the upper sensory level and a 30% difference in the incidence of hypotension at a significance level of P less than 0.05 with 80% power. Two software packages, SPSS 16.0 (SPSS Inc., Chicago, Illinois, USA) and SAS 9.1 (SAS Institute Inc., Cary, North Carolina, USA), were used for data analysis. Kruskal–Wallis one-way analysis was used to assess the difference in sensory block level amongst the groups, and the Cuzick trend post test was used to test the association between maximum sensory block level and the groups. χ2 test for trends was used to test the association between adverse effects of anaesthesia and the groups. In all cases, a P value less than 0.05 was considered statistically significant.


Thirty women were recruited to each group. Severe pain during the skin incision indicated failure of spinal anaesthesia in one patient from group A and one from group D, leaving 118 for analysis. In these two patients, the operation was performed without incident after supplementary epidural anaesthesia. Because epidural drugs might have interfered with the intrathecal spread of bupivacaine and subsequent blood pressure control, these two patients were included only in the evaluation of the anaesthesia,and otherwise excluded from the analyses.

There were no important differences in personal data between the groups (P > 0.05; Table 2). There were no significant differences in the time interval from spinal injection to positioning supine between the four groups (P = 0.429; Table 3).

Table 2
Table 2:
Clinical characteristics of enrolled patients
Table 3
Table 3:
Maximum cephalad sensory block level and time to motor block in lower limbs

The sensory block of spinal cord segments was compared between the groups at several time points after the start of anaesthesia (Fig. 1). With the exception of the assessment at 3 min, the sensory block level showed a significantly descending trend at all time points as the glucose concentration decreased from group A to group D (P < 0.001, Fig. 1).

Fig. 1
Fig. 1:
Sensory block over time following combined spinal-epidural anaesthesia. Changes in the dermatomal level of loss of pinprick sensation [median (Q L, Q U)] over time after CSEA. * P < 0.001 among groups.

The median height of the maximum block level was higher than T6 in all groups (Table 3). The maximum cephalad sensory block level and the incidence of a high sensory block level (higher than or equivalent to T2) gradually decreased from group A to group D (Table 3, P < 0.001). The coefficient of variation for the highest sensory block level decreased in line with the density of the intrathecal solution.

The score for lower limb motor block did not differ significantly among the four groups at any of the measured time points (P > 0.05). The groups also required a similar amount of time to achieve a motor block score of 2 (P = 0.162; Table 3). The lower extremity motor block achieved or exceeded a score of 2 in all patients within 20 min of anaesthesia.

Except for pruritus and vomiting, the incidence of other adverse effects of anaesthesia exhibited a significantly decreasing trend from group A to group D (P < 0.05, Table 4). Specifically, the incidences of hypotension were 48.3, 30.0, 13.3 and 10.3% in groups A, B, C and D, respectively (P = 0.0003, Table 4).

Table 4
Table 4:
Adverse reactions and incomplete anaesthesia

The efficacy of anaesthesia was similar in all four groups (Table 4), and there was no significant difference in the time from spinal injection to the completion of surgery among the four groups (P = 0.173). Pain on exploration of the abdomen was never worse than moderate but two patients from groups A and D experienced moderate skin incision pain (defined as anaesthesia failure, although they both achieved a sensory block level higher than T8). Poor muscle relaxation (defined as incomplete anaesthesia) was not reported in any group. The incidence of failed anaesthesia did not differ significantly among the four groups (3.33, 0, 0 and 3.33% for groups A, B, C and D, respectively, P > 0.05). The incidence of incomplete anaesthesia also did not differ significantly among the groups despite the two cases of failed anaesthesia (P > 0.05).


Many factors can affect the intrathecal spread of local anaesthetics. Because the baricity of the SAS is a controllable factor, it is common for anaesthetists to change the baricity of SAS by manipulating the concentration of glucose.2,3 SAS containing 0.8% glucose or less, which is known as a marginally hyperbaric solution, is slightly denser than CSF but much less dense than the conventional 8% glucose. Clinical studies in nonobstetric anaesthesia indicate that marginally hyperbaric solutions may induce a more constant and relatively lower block level and be associated with a lower incidence of hypotension than the conventional hyperbaric solution2,13–15 SAS with a higher density results in rapid spread and may allow a larger proportion of the injected dose of bupivacaine to reach the lowest point of the thoracic kyphosis,14 but a similar obstetric anaesthetic study did not reach the same conclusion. Connolly et al.16 reported that there was no difference in outcome with anaesthetic solutions containing glucose concentrations that varied ten-fold (8 and 0.8%, corresponding to 1.02081 and 1.00164 g ml–1, respectively, at 37°C) with respect to the anaesthesia onset time, the time between onset to the maximum sensory block and the highest level of sensory block. In our study, we also found that the differences in the highest level of sensory block and the incidence of hypotension between the conventional hyperbaric solution and the 0.8% glucose hyperbaric solution were minor, although we observed a significantly downward trend in the incidence of hypotension with decreasing density of SAS. However, 0.5 and 0.33% glucose solutions led to a significantly lower maximum cephalad sensory block level, a lower incidence of cervical block and a reduced incidence of hypotension compared with conventional hyperbaric solution. Richardson and Wissler8 demonstrated that the density of CSF was 1.00030 ± 0.00004 g ml–1 in pregnant women during the third trimester, which is significantly lower than that in other cohorts, and this offers an explanation as to why pregnant women might differ from other groups. Thus, SAS containing 0.8% glucose may act in a more hyperbaric manner even though it is only marginally hyperbaric in nonobstetric patients. A density greater than 1.00042 g ml–1 is considered to be a hyperbaric solution for pregnant patients. In the current study, we tested three marginally hyperbaric solutions (1.00255, 1.00122 and 1.00084 g ml–1 at 37°C) with a density slightly higher than 1.00042 g ml–1 but much lower than that of conventional hyperbaric solution (1.02759 ± 0.00003 g ml–1). Bannister et al.14 reported that the highest level of sensory block decreased along with declining glucose concentration (9, 0.83 and 0.33%), but the variability of the cephalad spread of bupivacaine was relatively large with 0.33% glucose, similar to that with the hypobaric solution. However, the variability was actually lowest with the 0.33% glucose SAS in our study. This again can be explained by the differences in the CSF density; SAS containing 0.33% glucose behaved as a hyperbaric solution in obstetric patients but a hypobaric solution in nonobstetric patients.

It is generally recognised that a low dose of local anaesthetic can significantly influence the intensity and duration of block whilst having little effect on the block level.12,17–19 High block is only one reason why the incidence of spinal-induced hypotension is much greater in obstetric anaesthesia than that in nonobstetric,20,21 and we did not expect a reduction in block height with a smaller dose. In our study, the marginally hyperbaric 0.5 and 0.33% glucose solutions achieved a significantly lower maximum sensory block level and thus a relatively lower incidence of hypotension than the conventional hyperbaric solution. Further studies are necessary to confirm and clarify this finding.

The use of high-dose bupivacaine may be another reason why previous studies have failed to observe the influence of density on block characteristics in obstetric anaesthesia. Both Connolly et al.16 and other previous authors used a dose of more than 8 mg bupivacaine.9,22 A concentration of local anaesthetics in the upper spinal canal that is ineffective in nonpregnant patients may be quite potent in pregnant women. Therefore using a conventional dose of local anaesthetic might obscure the influence of baricity on the intrathecal spread of anaesthetics. A study on low-dose hyperbaric bupivacaine solution versus plain bupivacaine solution during spinal anaesthesia for caesarean section showed that these two distinct solutions possess similar block characteristics in obstetric and nonobstetric anaesthesia, suggesting that the influence of baricity on the intrathecal spread of low-dose anaesthetics is similar in both obstetric and nonobstetric anaesthesia.23

The anaesthesia onset time is usually defined as the time from injection to a certain sensory block level.14,16 In the current study, in many cases in group D, the sensory block reached T6 at 3 min after injection (just after the change to supine), without a further rise. The anaesthesia onset time did not prove to be a helpful variable to compare amongst groups. In addition, comparison of the cephalad sensory block level at each time point indicated that the marginally hyperbaric solution spread more slowly than the conventional hyperbaric solution. This might allow time for compensation to occur for the slowly increasing sympathetic blockade, which might explain the lower incidence of hypotension in groups C and D.

Despite the report of incision pain in only one patient (classified as anaesthesia failure), the mean maximum sensory block level in group D by pinprick was T6, Because this provided adequate anaesthesia for surgery in the remainder we can concur with Russell that a pinprick level of T6 indicates a block sufficient for comfort during surgery.24

Low-dose bupivacaine in combination with opioid can improve the effect of spinal anaesthesia.25,26 The effect of adding sufentanil to our low-dose bupivacaine did not appear to be influenced by varying the density.

The incidence of shivering was significantly lower in the group that received the marginally hyperbaric solution containing 0.33% glucose than it was in the group that received the conventional hyperbaric solution. This might be related to the lower anaesthesia level in this group, because shivering is positively correlated with the number of spinal cord segments blocked.27

We used a complex recipe to prepare SAS with precise concentrations and densities of drugs that currently are widely used in clinical practice in China, but commercial prepackaged bupivacaine solution containing 8% glucose is not available to us. However, there are some limitations in our study arising from this. Because the densities quoted by a similar study were measured in different units and at different temperatures, we could only compare the results of our study with previous results produced using an SAS with a similar glucose concentration, even though the same glucose concentration cannot guarantee the same density for different formulations of SAS. This complicates comparability between different studies. We could not perform multiple comparisons between groups due to the limited number of samples. To reduce the operative error, we prepared the SAS using diluted solutions of different agents, but the cumbersome preparation of SAS may create difficulty in its clinical application. Also, the spinal anaesthesia injection was performed with patients in a lateral position, and this does not extrapolate to patients in a sitting position.

In conclusion, marginally hyperbaric low-dose bupivacaine solutions of different densities have similar spinal anaesthesia efficacy for caesarean section. However, marginally hyperbaric low-dose solutions with 0.5 and 0.33% glucose can significantly reduce the extension of the cephalad spread and the incidence of hypotension compared with conventional hyperbaric solutions with 8% glucose. However, the regression of anaesthesia and the postoperative analgesic effect of these study solutions require further study.

Acknowledgements relating to this article

Assistance with the study: We would like to thank the two senior engineers Shao-Jun Wang and Mei-Yun Zhang from the Petrochemical General Plant of Sinopec Shengli Oil Field Co., Ltd, for assistance with density measurement techniques and instruments. We also acknowledge our colleagues for their help and support in conducting this study.

Financial support and sponsorship: none.

Conflicts of interest: none.

Presentation: none.


1. Greene NM. Distribution of local anesthetic solutions within the subarachnoid space. Anesth Analg 1985; 64:715–730.
2. Hocking G, Wildsmith JA. Intrathecal drug spread. Br J Anaesth 2004; 93:568–578.
3. Lui AC, Munhall RJ, Winnie AP, Selander D. Baricity and the distribution of lidocaine in a spinal canal model. Can J Anaesth 1991; 38:522–526.
4. Fagraeus L, Urban BJ, Bromage PR. Spread of epidural analgesia in early pregnancy. Anesthesiology 1983; 58:184–187.
5. Datta S, Hurley RJ, Naulty JS, et al. Plasma and cerebrospinal fluid progesterone concentrations in pregnant and nonpregnant women. Anesth Analg 1986; 65:950–954.
6. Abouleish EI. Postpartum tubal ligation requires more bupivacaine for spinal anesthesia than does cesarean section. Anesth Analg 1986; 65:897–900.
7. Fassoulaki A, Gatzou V, Petropoulos G, Siafaka I. Spread of subarachnoid block, intraoperative local anaesthetic requirements and postoperative analgesic requirements in Caesarean section and total abdominal hysterectomy. Br J Anaesth 2004; 93:678–682.
8. Richardson MG, Wissler RN. Density of lumbar cerebrospinal fluid in pregnant and nonpregnant humans. Anesthesiology 1996; 85:326–330.
9. Van de Velde M, Van Schoubroeck D, Jani J, et al. Combined spinal-epidural anesthesia for cesarean delivery: dose-dependent effects of hyperbaric bupivacaine on maternal hemodynamics. Anesth Analg 2006; 103:187–190.table of contents.
10. Arzola C, Wieczorek PM. Efficacy of low-dose bupivacaine in spinal anaesthesia for Caesarean delivery: systematic review and meta-analysis. Br J Anaesth 2011; 107:308–318.
11. McNaught AF, Stocks GM. Epidural volume extension and low-dose sequential combined spinal-epidural blockade: two ways to reduce spinal dose requirement for caesarean section. Int J Obstet Anesth 2007; 16:346–353.
12. Teoh WH, Thomas E, Tan HM. Ultra-low dose combined spinal-epidural anesthesia with intrathecal bupivacaine 3.75 mg for cesarean delivery: a randomized controlled trial. Int J Obstet Anesth 2006; 15:273–278.
13. Sumi M, Sakura S, Sakaguchi Y, et al. Comparison of glucose 7.5% and 0.75% with or without phenylephrine for tetracaine spinal anaesthesia. Can J Anaesth 1996; 43:1138–1143.
14. Bannister J, McClure JH, Wildsmith JA. Effect of glucose concentration on the intrathecal spread of 0.5% bupivacaine. Br J Anaesth 1990; 64:232–234.
15. Sanderson P, Read J, Littlewood DG, et al. Interaction between baricity (glucose concentration) and other factors influencing intrathecal drug spread. Br J Anaesth 1994; 73:744–746.
16. Connolly C, McLeod GA, Wildsmith JA. Spinal anaesthesia for Caesarean section with bupivacaine 5 mg ml(-1) in glucose 8 or 80 mg ml(-1). Br J Anaesth 2001; 86:805–807.
17. Roofthooft E, Van de Velde M. Low-dose spinal anaesthesia for Caesarean section to prevent spinal-induced hypotension. Curr Opin Anaesthesiol 2008; 21:259–262.
18. Khaw KS, Ngan Kee WD, Wong EL, et al. Spinal ropivacaine for cesarean section: a dose-finding study. Anesthesiology 2001; 95:1346–1350.
19. Leo S, Sng BL, Lim Y, Sia AT. A randomized comparison of low doses of hyperbaric bupivacaine in combined spinal-epidural anesthesia for cesarean delivery. Anesth Analg 2009; 109:1600–1605.
20. Chinachoti T, Tritrakarn T. Prospective study of hypotension and bradycardia during spinal anesthesia with bupivacaine: incidence and risk factors, part two. J Med Assoc Thai 2007; 90:492–501.
21. Carpenter RL, Caplan RA, Brown DL, et al. Incidence and risk factors for side effects of spinal anesthesia. Anesthesiology 1992; 76:906–916.
22. Sarvela PJ, Halonen PM, Korttila KT. Comparison of 9 mg of intrathecal plain and hyperbaric bupivacaine both with fentanyl for cesarean delivery. Anesth Analg 1999; 89:1257–1262.
23. Vercauteren MP, Coppejans HC, Hoffmann VL, et al. Small-dose hyperbaric versus plain bupivacaine during spinal anesthesia for cesarean section. Anesth Analg 1998; 86:989–993.
24. Russell IF. A comparison of cold, pinprick and touch for assessing the level of spinal block at caesarean section. Int J Obstet Anesth 2004; 13:146–152.
25. Birnbach DJ, Soens MA. Hotly debated topics in obstetric anesthesiology 2008: a theory of relativity. Minerva Anestesiol 2008; 74:409–424.
26. Lee JH, Chung KH, Lee JY, et al. Comparison of fentanyl and sufentanil added to 0.5% hyperbaric bupivacaine for spinal anesthesia in patients undergoing cesarean section. Korean J Anesthesiol 2011; 60:103–108.
27. Leslie K, Sessler DI. Reduction in the shivering threshold is proportional to spinal block height. Anesthesiology 1996; 84:1327–1331.
© 2015 European Society of Anaesthesiology