The combined spinal-epidural (CSE) technique resulted in prolonged sensory and motor recovery as compared with the single-shot spinal (SSS) technique, despite identical doses of local anesthetic (LA) (1). Arterial hypotension after institution of subarachnoid block with CSE could also increase the need for vasoconstrictor administration (1,2). A seemingly appropriate LA dose to provide adequate anesthesia for SSS may result in a relative overdose when used in CSE. To reduce the incidence of these morbidities, an LA dose decrement may be necessary whenever CSE is intended. The magnitude of the dose adjustment remains to be determined.
The up-down sequential allocation method (3,4), as described by Dixon and Massey (5), is increasingly used to obtain the median effective doses (MEDs) of drugs used in anesthesia. In this study, we attempted to derive the MEDs of intrathecal (IT) hyperbaric bupivacaine in both CSE and SSS. We hypothesized that, under otherwise similar conditions, the MED for IT hyperbaric bupivacaine when administered via CSE would be smaller compared with SSS.
After institutional ethics committee approval and written informed consent, 60 male patients (age, 20–50 yr; weight, 50–70 kg; height, 1.5–1.8 m; ASA physical status, I and II) scheduled for perineal and lower limb operations were recruited. Patients with a history of hypertension, diabetes mellitus, and contraindications to regional anesthesia were excluded.
Participants were allocated into two groups in a double-blind, randomized, prospective study design. Randomization was achieved by means of the opaque sealed envelope technique. Each envelope contained one of the two designations: SSS or CSE. To avoid interoperator variability, all the blocks were performed by the principal author. Electrocardiogram, heart rate, and oxygen saturation were monitored preblock and continuously throughout the study. Arterial blood pressure was recorded preblock and at 5-min intervals postblock for 60 min by using an automated noninvasive blood pressure device (Dinamap; Critikon, FL) over the patient's right upper arm. All patients received 500 mL of warmed lactated Ringer's solution over 15 min before neuraxial block. All the blocks were performed with the patient in the right lateral position.
In the SSS group (n = 30), the subarachnoid space was entered by using a 27-gauge Whitacre spinal needle (Becton Dickinson, Franklin Lakes, NJ) at the L3-4 interspace. After confirming free flow of cerebrospinal fluid (CSF), the designated dose of LA was administered. The solution was injected over 15 s in each instance with the orifice of the spinal needle in the cephalad direction. In the CSE group (n = 30), the epidural space was identified with a 17-gauge Tuohy needle (Espocan; B. Braun, Germany) at the L3-4 interspace by using the loss-of-resistance-to-air technique, whereby 4 mL of air was injected. Through the epidural needle, a long 27-gauge pencil-point spinal needle was introduced. Upon puncturing the dural, the spinal needle was securely docked, and the designated dose of LA was given. The solution was injected over 15 s with the orifice of the spinal needle in the cephalad direction.
The dose of LA received by a particular patient in either group was determined by the response of the previous patient in that group to a larger or smaller dose, by using an up-down sequential allocation technique. Bupivacaine 1.9 mL 0.5% (9.5 mg) in 8% glucose solution (Marcain; AstraZeneca, Sweden) was given to the first patient in Group SSS, and 1.4 mL (7.0 mg) was given in Group CSE. The dosing adjustment was 0.5 mg for each group.
After drug administration, the patients were immediately placed in the supine position. The level of sensory anesthesia was assessed by an independent observer, who was blinded to the allocation of regional technique, by using loss of sensation to pinprick at regular 5-min intervals. Our arbitrary anesthetic target was sensory anesthesia at the T6 dermatome 60 min after drug administration. Hence, two possible outcomes were considered:
- Success: sensory anesthesia at or above the T6 dermatome 60 min after the initiation of regional block. This outcome, defined as success, would result in a 0.5-mg dose decrement in the next patient.
- Failure: sensory anesthesia below the T6 dermatome 60 min after the initiation of regional block. This outcome, defined as failure, would direct a 0.5-mg dose increment in the next patient.
Demographic data were collected and are presented as mean (sd). Means (sd) were analyzed with unpaired Student's t-tests. The MEDs were calculated from the up-down sequences by using the method of Dixon and Massey (5). A 95% confidence interval (CI) of the ratio MED CSE versus MED SSS was estimated with a Taylor series (delta method). The sequences were also subjected to probit regression analyses as backup or sensitivity tests.
There were no statistical differences in demographics among patients from both groups (Table 1). In both groups, the MED of hyperbaric bupivacaine was derived from the less common of the two outcome variables, i.e., a successful outcome. There were 13 successful outcomesin both the SSS and the CSE groups (Table 2, Figs. 1 and 2).
The MED for Group CSE, determined with the formula described by Dixon and Massey (5), was 9.18 mg (95% CI, 8.89–9.47 mg). In contrast, Group SSS required a larger MED of 11.37 mg (95% CI, 10.88–11.86 mg) to achieve the same anesthetic goal. According to probit analysis, the MED of Group SSS (11.40 mg; 95% CI, 11.03–11.83 mg) was significantly different from the MED of Group CSE (9.19 mg; 95% CI, 8.81–9.60 mg) (P < 0.001) (Table 3). On the basis of an arithmetic ratio of the MED of hyperbaric bupivacaine for Group CSE versus Group SSS, a 19.3% (95% CI, 14.9%–23.6%) discrepancy in LA dose was demonstrated.
Our study demonstrated a difference in the MED of hyperbaric bupivacaine between CSE and SSS. In quantifying the magnitude of difference from the ratio of the MEDs, CSE required 19.3% (95% CI, 14.9%–23.6%) less LA than SSS to achieve a similar clinical target. Because other variables that could have influenced the spread of the block (i.e., the baricity of the IT injectate in relation to patients' position or the site and speed of injection) (6) were controlled, the difference in the MEDs was due to the inherent differences between CSE and SSS. In our previous study (1), we compared the extent of sensorimotor anesthesia by using identical doses of LA administered via SSS and CSE. CSE produced prolonged sensory and motor recovery, as well as more frequent hypotension, as compared with SSS. Although the actual mechanism that accounts for this is undetermined, the loss-of-resistance-to-air technique in CSE could potentially disrupt the equilibrium of forces and introduce air pockets within the epidural space during insertion. This could, in turn, result in a reduction of the lumbosacral CSF volume, which bears an inverse relation with the peak sensory level during subarachnoid block (7).
An important premise for the up-down sequential allocation technique is the proximity of the doses used in each group to the actual MED (5). Because we had previously established that CSE resulted in significantly prolonged sensory block, we did not use a common starting dose of IT bupivacaine for CSE and SSS to avoid introducing inaccuracies to our results: their MEDs were unlikely to be the same. The a priori 25% difference of the initial doses was selected because this was deemed to be clinically significant. We managed to obtain a similar proportion of successful blocks for both study groups (13 of 30) and a comparable number of successful and failed blocks (i.e., 13 and 17, respectively) within each group. Arguably, the MEDs determined by our study could have been numerically different if we had used different starting doses. In fact, errors could have been introduced if the starting doses were distant from the true MEDs. However, successful outcomes were achieved within four increments in SSS and five increments in CSE, thus demonstrating the proximity of our starting doses to the actual MEDs. On the basis of this finding and the findings of other prospective and retrospective observations (1,2), we could infer that to achieve a similar anesthetic target, the appropriate dose for subarachnoid block for CSE is significantly smaller than for SSS.
In this study, our target of sensory anesthesia—a T6 or higher block for at least one hour—was an arbitrary end-point. This target was chosen because sensory anesthesia at this level would be clinically adequate for our patients undergoing perineal and lower limb procedures. Besides, unless the block did not extend beyond the lumbosacral dermatomes, a failure of anesthetic outcome in the context of the study would not result in patient discomfort or interfere with minor perineal surgery. Indeed, none of the patients in our study required anesthetic supplementation or conversion to general anesthesia for their scheduled procedures.
It is likely that the derived ratio of 20% is dependent on the choice of LA and the level or duration of sensory blockade. Our results should not be extrapolated to other LAs or other dermatological end-points. Moreover, it is also premature to extrapolate our findings to different patient populations, such as pregnant women. Neither the use of the up-down sequential allocation technique to compare the MED of SSS with CSE nor the application of the results of our study would be appropriate for procedures that require a higher sensory block, e.g., cesarean delivery (8). In this situation, even though a failure in CSE could be addressed by supplementing with epidural anesthesia, a similar event in the SSS arm would be likely to predispose the subject to the need for conversion to general anesthesia and its attendant risks. Conversely, accepting a higher sympathosensory block as indicative of a successful block would potentially increase the occurrence of harmful adverse effects such as hypotension and respiratory difficulties.
The outcome of our study has some implications for practitioners of subarachnoid block. Even though the use of small-dose spinal anesthesia has gained much popularity in recent years, many of these conclusions were drawn from studies involving CSE (9,10). According to the results of our study, the use of such minidoses might not be suitable in SSS because of an increased possibility of block failure and the lack of an avenue to supplement neuraxial block in this event. Similarly, several studies that have attempted to determine the dose-response curves of LAs for subarachnoid block were also based on the CSE technique and, as such, may not be applicable to SSS (9–11). Conversely, practitioners of SSS who are contemplating a switch to CSE should be mindful that a seemingly appropriate dose in SSS may be excessive in CSE.
In conclusion, our study revealed that the MED of IT hyperbaric bupivacaine in CSE was 20% less than that in SSS. This magnitude of dose decrement may be warranted whenever CSE is intended in place of SSS.
1. 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.
2. Klasen J, Junger A, Hartmann B, et al. Differing incidences of relevant hypotension with combined spinal-epidural anesthesia and spinal anesthesia. Anesth Analg 2003;96:1491–5.
3. Columb MO, Lyons G. Determination of the minimum local analgesic concentrations of epidural bupivacaine and lidocaine in labor. Anesth Analg 1995;81:833–7.
4. Lacassie HJ, Columb MO. The relative motor blocking potencies of bupivacaine and levobupivacaine in labor. Anesth Analg 2003;97:1509–13.
5. Dixon WJ, Massey FJ. Sensitivity experiments. In: Introduction to statistical analysis. 3rd ed. New York: McGraw-Hill, 1983:377–93.
6. Greene NM. Distribution of local anesthetic solutions within the subarachnoid space. Anesth Analg 1985;64:715–30.
7. Carpenter RL, Hogan QH, Liu SS, et al. Lumbosacral cerebrospinal fluid volume is the primary determinant of sensory block extent and duration during spinal anaesthesia. Anesthesiology 1998;89:24–9.
8. Camann W. The level is TEE-FORE [letter]. Anesth Analg 1996;82:439.
9. Urmey WF, Stanton J, Peterson M, et al. Combined spinal epidural anesthesia for outpatient surgery: dose response characteristics of intrathecal isobaric lidocaine using a 27-gauge Whitacre spinal needle. Anesthesiology 1995;83:528–34.
10. Zayas VM, Liguori GA, Chisholm MF, et al. Dose response relationships for isobaric spinal mepivacaine using the combined spinal epidural technique. Anesth Analg 1999;89:1167–71.
© 2005 International Anesthesia Research Society
11. Pawlowski J, Sukhani R, Pappas AL, et al. The anesthetic and recovery profile of two doses (60 and 80 mg) of plain mepivacaine for ambulatory spinal anesthesia. Anesth Analg 2000;91:580–4.