Current Issue Previous Issues Published Ahead-of-Print CME Subjects Timely Topics Translations Podcasts For Authors Journal Info
Skip Navigation LinksHome > August 2003 - Volume 97 - Issue 2 > Isobaric Versus Hypobaric Spinal Bupivacaine for Total Hip A...
Anesthesia & Analgesia:
doi: 10.1213/01.ANE.0000070228.13782.ED

Isobaric Versus Hypobaric Spinal Bupivacaine for Total Hip Arthroplasty in the Lateral Position

Faust, Alexandre MD*; Fournier, Roxane MD*; Van Gessel, Elisabeth MD*; Weber, Anne MD*; Hoffmeyer, Pierre MD†; Gamulin, Zdravko MD*

Free Access
Article Outline
Collapse Box

Author Information

*Division of Anesthesiology,

†Clinic of Orthopedic Surgery, University Hospital Geneva, Geneva, Switzerland

Support was provided solely from institutional sources.

March 20, 2003.

Address correspondence and reprint requests to Dr. Alexandre Faust, Division d’ Anesthésiologie, Hôpital Cantonal—Hôpitaux Universitaires de Genève, Rue Micheli-du-Crest 24, 1211 Genève 14, Switzerland. Address e-mail to

Collapse Box


Total hip arthroplasty (THA) is frequently performed under spinal anesthesia using either isobaric or hypobaric anesthetic solution. However, these two solutions have never been compared under similar surgical conditions. In the present study, we compared the anesthetic and hemodynamic effects of isobaric and hypobaric bupivacaine in 40 ASA physical status I–II patients undergoing THA in the lateral decubitus position under spinal anesthesia. With operative side up, patients randomly received, in a double-blinded manner, a spinal injection of 3.5 mL (17.5 mg) of plain bupivacaine mixed with either 1.5 mL of normal saline (isobaric group) or 1.5 mL of distilled water (hypobaric group). Sensory level and degree of motor block were evaluated on the nondependent and dependent sides until regression to L2 and total motor recovery. Hemodynamic changes during the first 45 min after spinal injection, and the time between spinal administration and first analgesic for a pain score >3 (on a 0–10 scale) were noted. Demographic characteristics of both groups were comparable. Upper sensory level and maximal degree of motor block were comparable between the operative and nonoperative sides in each group and between corresponding sides in both groups. Compared with the isobaric group, in the hypobaric group there was a prolonged time to sensory regression to L2 on the operative side (287 ± 51 versus 242 ± 36 min, P < 0.004) and a prolonged time to first analgesic (290 ± 46 versus 237 ± 39 min, P < 0.001). No difference in quality of motor block was noted at the end of surgery. Hemodynamic changes were comparable. We conclude that for THA in the lateral position, spinal hypobaric bupivacaine seems to be superior to isobaric in that it prolongs the sensory block on the operative side and delays the use of analgesics after surgery without further compromising hemodynamic stability.

In our institution, total hip arthroplasty (THA) in the lateral decubitus position is frequently performed using single-shot spinal anesthesia with 15–17.5 mg of plain bupivacaine 0.5% which provides a surgical anesthesia for 3–4 h. In such patients, hypobaric bupivacaine could in theory provide a more selective subarachnoid distribution of local anesthetic on the nondependent (operative) side. This can result in a more profound sensory and motor block of longer duration which could be advantageous in the case of unexpectedly prolonged surgery.

The use of hypobaric local anesthetic has been reported for single-shot injection (1,2) and continuous spinal anesthesia (3,4). However, possible advantages of hypobaric over isobaric solutions have not been tested in these specific surgical conditions. The aim of the present study was to compare the anesthetic and hemodynamic effects of isobaric (plain bupivacaine mixed with normal saline) and hypobaric bupivacaine (plain bupivacaine mixed with distilled water) solutions for THA performed with patients in the lateral decubitus position.

Back to Top | Article Outline

Patients and Methods

After approval from the Ethics Committee of our institution and written informed consent, 40 orthopedic patients, aged 40–75 yr, ASA physical status I or II, scheduled to undergo elective THA under single-shot spinal anesthesia were enrolled in the study. Exclusion criteria were coagulation disorders, local infection, and obvious spinal postural abnormalities (kyphosis) as well as inability to comprehend basic aspects of the study. Preoperative medication consisted of morphine 0.1 mg/kg subcutaneously administered 1 h before arrival in the operating room, to ameliorate discomfort of the dependent shoulder during prolonged lateral decubitus position. Standard noninvasive monitoring consisted of continuous electrocardiogram, peripheral pulse oxymetry, and automatic noninvasive blood pressure measurement on the nondependent arm. More invasive monitoring (i.e., central venous pressure, invasive arterial pressure, or an indwelling urinary catheter) was used only if required by the patient’s clinical condition. After placement of a peripheral IV catheter in the dependent forearm, preanesthetic hydration consisted of 10 mL/kg of a crystalloid solution. During the first hour after spinal injection, 5 mL/kg of the same solution was further infused. Thereafter, fluids were given on the basis of changes in arterial blood pressure and estimated blood loss, replaced with a crystalloid solution on a 3:1 mL basis. When available, autologous blood was given if the hematocrit decreased to <30% and homologous blood was only administered if the hematocrit decreased to <26%.

Using a sealed envelope system, the patients were randomly assigned to receive either hypobaric or isobaric bupivacaine solutions, which were prepared as follows. Isobaric bupivacaine: 3.5 mL (17.5 mg) of plain bupivacaine 0.5% (5-mL vial Carbostesine® 0.5%; Astra Zeneca, Grafenau, Zug, Switzerland) diluted with normal saline to a total of 5 mL; measured density at 37°C was 0.999406 g/mL. Hypobaric bupivacaine: 3.5 mL (17.5 mg) of plain bupivacaine 0.5% diluted with distilled water, to a total of 5 mL; measured density at 37°C was 0.997302 g/mL.

According to the randomization, an anesthesiologist not participating in patient care or data collection prepared 10 mL of distilled water or normal saline. The anesthesiologist in charge of the patient withdrew 3.5 mL of plain bupivacaine 0.5% from a 5-mL vial and added 1.5 mL of the prepared solution. The density measurements of study solutions were performed using an Anton Paar DMA 4500 densitometer (Anton Paar GmbH, Graz, Austria). For each solution, three measurements were performed and the mean value was considered.

With the operating table horizontal, the patients were placed in the lateral decubitus position with the operated hip up. Lumbar puncture was performed at the L2–3 interspace with a 25-gauge Whitacre needle. After observing free cerebrospinal fluid (CSF) reflux, the needle aperture was oriented upward and 5 mL of the study solution injected at a rate of approximately 0.5 mL/s. The patients remained in the lateral position until the end of surgery, at which time they were turned supine.

The following variables were measured throughout the study:

* Evolution of upper sensory block level on nondependent (operative) and dependent sides. A pinprick test (24-gauge needle) was performed on the midthoracic line every 5 min during the first 45 min after spinal injection, and then every 15 min until sensory regression to L2 (level of the surgical incision). Maximal upper sensory block level, its onset time, and time to regression to L2 on both sides were recorded.

* Evolution of degree of motor block using a modified Bromage scale (5) ranging from 0 to 4 (0 = able to move hip, knee, ankle and toes; 1 = unable to move hip, able to move knee, ankle, and toes; 2 = unable to move hip and knee, able to move ankle and toes; 3 = unable to move hip, knee, ankle, able to move toes; 4 = unable to move hip, knee, ankle, and toes) on both limbs, every 5 min during the first 45 min after spinal injection. In order not to interfere with the surgical procedure, motor block was not tested during the operation. At the end of surgery, degree of motor block was determined for both limbs and tested every 15 min until total motor recovery. Maximal degree of motor block, its onset time, and time to total motor recovery of both limbs were recorded.

* Mean arterial blood pressure (MAP) and heart rate (HR) were recorded every 2.5 min during the first 45 min after spinal injection, every 5 min during surgery, and every 15 min in the recovery room until the study termination (defined as sensory regression to L2 on both sides and/or total motor recovery of both limbs). Maximal decrease in MAP and HR from baseline value (determined with patients in the lateral decubitus position just before spinal injection) was recorded for the first 45 min after spinal injection. Ephedrine 5–10 mg IV was given if MAP decreased >20% from baseline value or if systolic pressure decreased to <90 mm Hg. Atropine 0.5 mg IV was given if HR decreased to <45 bpm.

* Duration of anesthesia was defined as time between spinal injection and the end of the surgery.

* Duration of surgical analgesia was defined as the time between spinal injection and the first analgesic requirement for a pain score at the operated site >3 on the visual analog scale ranging from 0 to 10.

All the above variables were determined during anesthesia by the anesthesiologist in charge of the patient, and in the recovery room by nurses who were trained by the investigators. Discomfort related to the lateral position during surgery was treated with fentanyl 1 μg/kg IV (maximal 2 doses) and anxiety with midazolam 1 mg IV.

Prospective power tests defined the sample size using sensory block level regression time to L2 of 157 ± 37 min using 3 mL of plain bupivacaine 0.5%(6). The sample size was computed to detect a 25% difference in favor of the hypobaric group, i.e., a longer duration of block with a power of 80% and a two-tailed significance level of 5% (β = 0.2; α = 0.05). A minimal sample of 14 patients for each group met these criteria. Results are expressed as mean ± sd or median (ranges) for discrete variables. Comparisons between groups or between both sides in the same group were performed using the Student’s t-test for unpaired or paired data, the Mann-Whitney U-test, and the χ2 test as required. A P value < 0.05 was considered statistically significant.

Back to Top | Article Outline


Twenty patients were allocated to each group. One isobaric spinal anesthetic failed. This patient was not considered for further analysis. Patient demographic and preanesthetic hemodynamic data were comparable between the two study groups (Table 1).

Table 1
Table 1
Image Tools

Median upper sensory levels with ranges in nondependent (operative) and dependent sides and onset times (mean ± sd) are illustrated in Figure 1. There was no difference between corresponding sides in the two groups or between operative and nonoperative sides in the same group.

Figure 1
Figure 1
Image Tools

Maximal degree of motor block achieved and onset times were also comparable between the two groups and in the same group between both sides. In the isobaric group, these data were 4 (4) and 16 ± 4 min for the nondependent and 4 (3–4) and 22 ± 9 min for the dependent side. In the hypobaric group, these data were 4 (4) and 13 ± 7 min for the nondependent and 4 (1–4) and 16 ± 10 min for the dependent side.

Duration of anesthesia, defined as the time between the spinal injection and the end of the surgery, was comparable between the two groups—170 ± 25 min for isobaric and 168 ± 23 min for hypobaric.

When comparing the sensory regression times to L2 between the nondependent and the dependent sides, patients in both groups showed significantly prolonged sensory regression to L2 on the nondependent (operative) side (242 ± 36 versus 219 ± 30 min, P < 0.005; and 287 ± 51 versus 233 ± 38 min, P < 0.0001, for isobaric and hypobaric group, respectively). When comparing both groups (Fig. 2), regression to L2 on the nondependent side was significantly prolonged in the hypobaric group (287 ± 51 versus 242 ± 36 min, P < 0.004); likewise, time to first analgesic requirement was significantly longer in the hypobaric group (290 ± 46 versus 237 ± 39 min, P < 0.001).

Figure 2
Figure 2
Image Tools

Because the degree of motor block was not tested during surgery, and because at the end of surgery complete motor recovery of one or both limbs was observed in some patients, only relevant data were available after surgery was completed in all patients, i.e., 225 min after spinal injection. At this time in the hypobaric group, 5 patients had a complete motor recovery on the nondependent side compared with 15 on the dependent side (P < 0.0001). In the isobaric group, the number of patients was not statistically different between the two sides (8 versus 11). Also, there was no statistical difference when corresponding sides between the two study groups were compared.

Hemodynamic changes, observed during the first 45 min after spinal injection, were comparable between the two groups. Maximal decrease in MAP was 32% ± 13% versus 31% ± 16% and for HR 14% ± 11% versus 14% ± 10% for the isobaric and hypobaric group, respectively. Ten patients in the isobaric and nine in the hypobaric group received ephedrine; one patient in the isobaric group received atropine.

Finally, two patients in the isobaric group received fentanyl for discomfort and four received midazolam for anxiety. In the hypobaric group, three patients received fentanyl and five midazolam.

Back to Top | Article Outline


The results of the present study demonstrate an advantage of hypobaric over isobaric spinal anesthesia in patients undergoing THA in the lateral decubitus position. Although both solutions provide satisfactory analgesia, hypobaric bupivacaine showed a significantly delayed sensory regression to L2 on the nondependent side of 45 minutes, thus postponing the need for first analgesic.

Although described as a possible anesthetic technique (7), there are only a few studies reporting the use of hypobaric spinal anesthesia (1–4,8). Before the present study, two other reports compared hypobaric and isobaric bupivacaine. Van Gessel et al. (9) reported fewer failures with isobaric versus hypobaric bupivacaine during continuous spinal anesthesia for hip surgery. However, both injection of local anesthetic and surgery were performed with patients supine. Kuusniemi et al. (10) tested small doses (6 mg) of hypobaric and plain bupivacaine in the lateral position for knee arthroscopy; 20 minutes after spinal injection, the patients were turned supine for surgery. A differential spread for both sensory levels and motor block between nondependent and dependent sides was demonstrated for each solution. However, no difference was found between the two solutions, when comparing the same sides.

The present study is the first comparing isobaric and hypobaric bupivacaine in patients undergoing surgery in the lateral position. During progression of spinal anesthesia, both solutions demonstrate qualities of isobaricity (no difference in upper sensory level and maximal degree of motor block between nondependent and dependent sides). The results of other studies investigating subarachnoid distribution of hypobaric local anesthetic in the lateral position suggest a dose-related effect in favor of the nondependent side. Kuusniemi et al. (10) reported a differential sensory and motor block with hypobaric solution (6 mg of bupivacaine in 3.4 mL). Van Gessel et al. (4) observed a differential spread only for motor block when using hypobaric solutions of tetracaine or bupivacaine (7.5 mg in 3 mL). Atchison et al. (1), studying the effects of speed of injection, documented a differential sensory spread between nondependent and dependent sides with 10 mg of hypobaric tetracaine in 5 mL when injected over 250 seconds with an electrically driven syringe pump through a Whitacre needle with the aperture oriented upward. When the same solution was injected rapidly over 10 seconds, approximating the usual clinical spinal injection speeds, no differential sensory block was found. Using an identical methodology, Horlocker et al. (2) did not show any difference in sensory levels between the nondependent and dependent sides in either slow or fast injection groups, using 15 mg of hypobaric bupivacaine in 5 mL. Consequently, the appearance of a differential block seems to be favored by using small dose hypobaric solution injected very slowly.

In our study, the absence of early clinical signs of preferential distribution in favor of the nondependent side in the hypobaric group can be explained essentially by the following mechanism. Unlike in hyperbaric solutions (11), there is a relatively small difference in density between the hypobaric bupivacaine solution used in the present study (0.997302 g/mL) and CSF using measurements previously made (12) in our institution (1.000529 ± 0.000107 g/mL). Given this slight difference in density and given the relatively large dose (17.5 mg) and volume (5 mL) of hypobaric bupivacaine, injected rapidly over 10 seconds, it is not surprising that a dense initial bilateral subarachnoid block developed initially. Thus, despite injecting the anesthetic solution through a directional Whitacre needle, there was no early evidence of a preferential spread.

However, qualities of hypobaricity appeared during regression of spinal anesthesia in both groups, but were clinically more relevant in the hypobaric group: 1) The regression time to L2 between nondependent and dependent sides was significantly different in the two groups; 2) Unlike in the isobaric group, significantly fewer patients in the hypobaric group had complete motor block recovery on the nondependent compared with the dependent side.

The appearance of this delayed asymmetrical block can be attributed to the differences in densities between anesthetic solutions and CSF, associated with prolonged lateral position of approximately 3 h in the present study. It is presumed that part of a local anesthetic injected intrathecally remains free in CSF for at least 60 minutes, because one hour after administration, variations in upper sensory level were found when changing patient position (13,14). We speculate that in the present study, three hours of lateral decubitus positioning allows more neural fixation on the nondependent roots of hypobaric than isobaric bupivacaine. These arguments could explain the more pronounced differential spread of hypobaric over isobaric bupivacaine observed during regression of the spinal anesthesia.

In the present study, some degree of differential block is documented for isobaric bupivacaine. Similar findings are reported by others for patients receiving plain bupivacaine and tested in the lateral position (10,15), questioning whether plain bupivacaine is isobaric or hypobaric (15). Greene (7) stated that the limit between hypobaric and isobaric local anesthetic solutions is a baricity of 0.9990, which is calculated by dividing density of local anesthetic with that of CSF. In 1954, Davis and King (16) stated that this limit is a density of local anesthetic lower than three standard deviations below mean human CSF density. Using more precise techniques of measurement of CSF density, Richardson and Wissler (17) determined the upper limits of hypobaricity as density of local anesthetic between 1.00016 to 1.00037 g/mL according to the variations of CSF density in a different subgroup of patients and considered the mixture plain bupivacaine-morphine with a density of 0.99941 as hypobaric (18). Recently, we reported the density of plain bupivacaine 0.5% of 0.999343 ± 0.000004 g/mL at 37°C (12). Compared with this value, the density of hypobaric bupivacaine investigated in the present study was less (0.997302 g/mL) and that of the isobaric solution more (0.999406 g/mL). Their baricities calculated with density of CSF of 1.000529 g/mL (12) were 0.996774 and 0.998876 for hypobaric and isobaric solutions, respectively. Thus, hypobaric bupivacaine seems to be hypobaric according to the definition of Greene, Davis, and Richardson, whereas isobaric bupivacaine is at the limit between hypobaric and isobaric for Greene but remains hypobaric for Davis and Richardson. It should be noted that these different limits of baricity are given arbitrarily and that, besides baricity, there are >20 demonstrated or theoretical factors influencing subarachnoid distribution of local anesthetics (7). Nevertheless, it is admitted that plain bupivacaine 0.5% administered at the usual volume of 3 mL in patients placed supine after injection behaves clinically as isobaric (7). However, the results of the present study suggest that local anesthetic solutions considered isobaric, with a density even more than that of plain bupivacaine but less than that of the CSF, can show some signs of hypobaricity in patients kept in prolonged lateral position.

In summary, for patients undergoing THA in the lateral position under spinal anesthesia, 17.5 mg of hypobaric bupivacaine, compared with the identical dose of isobaric bupivacaine, prolonged sensory regression to L2 and delayed the use of first analgesic, without further compromising systemic hemodynamics. We believe that 45 minutes longer duration of spinal block is clinically relevant and increases the reliability of hypobaric spinal anesthesia for this type of surgical procedure.

Back to Top | Article Outline


1. Atchison SR, Wedel DJ, Wilson PR. Effect of injection rate on level and duration of hypobaric spinal anesthesia. Anesth Analg 1989; 69: 496–500.

2. Horlocker TT, Wedel DJ, Wilson PR. Effect of injection rate on sensory level and duration of hypobaric bupivacaine spinal anesthesia for total hip arthroplasty. Anesth Analg 1994; 79: 773–7.

3. Kallos T, Smith TC. Continuous spinal anesthesia with hypobaric tetracaine for hip surgery in lateral decubitus. Anesth Analg 1972; 51: 766–73.

4. Van Gessel EF, Forster A, Gamulin Z. Surgical repair of hip fracture using continuous spinal anesthesia: comparison of hypobaric solutions of tetracaine and bupivacaine. Anesth Analg 1989; 68: 276–81.

5. Martin-Salvaj G, Van Gessel E, Forster A, et al. Influence of duration of lateral decubitus on the spread of hyperbaric tetracaine during spinal anesthesia: a prospective time-response study. Anesth Analg 1994; 79: 1107–12.

6. Racle JP, Jourdren L, Benkhadra A, et al. Effect of adding sodium bicarbonate to bupivacaine for spinal anesthesia in elderly patients. Anesth Analg 1988; 67: 570–3.

7. Greene NM. Distribution of local anesthetic solutions within the subarachnoid space. Anesth Analg 1985; 64: 715–30.

8. Maroof M, Khan RM, Siddique M. Hypobaric spinal anesthesia with bupivacaine (01%) gives selective sensory block for ano-rectal surgery. Can J Anaesth 1995; 42: 691–4.

9. Van Gessel EF, Forster A, Schweizer A, Gamulin Z. Comparison of hypobaric, hyperbaric, and isobaric solutions of bupivacaine during continuous spinal anesthesia. Anesth Analg 1991; 72: 779–84.

10. Kuusniemi KS, Pihlajamäki KK, Pitkänen MT, Korkeila JE. Low-dose bupivacaine: a comparison of hypobaric and near isobaric solutions for arthroscopy surgery of the knee. Anaesthesia 1999; 54: 540–5.

11. Lui ACP, Polis TZ, Cicutti NJ. Densities of cerebrospinal fluid and spinal anaesthetic solutions in surgical patients at body temperature. Can J Anaesth 1998; 45: 297–303.

12. Schiffer E, Van Gessel E, Fournier R, et al. Cerebrospinal fluid density influences extent of plain bupivacaine spinal anesthesia. Anesthesiology 2002; 96: 1325–30.

13. Povey HMR, Jacobsen J, Westergaard-Nielsen J. Subarachnoid analgesia with hyperbaric 05% bupivacaine: effect of a 60-min period of sitting. Acta Anaesthesiol Scand 1989; 33: 295–7.

14. Niemi L, Tuominen M, Pitkänen M, Rosenberg PH. Effect of late posture change on the level of spinal anaesthesia with plain bupivacaine. Br J Anaesth 1993; 71: 807–9.

15. Blomqvist H, Nilsson A. Is glucose-free bupivacaine isobaric or hypobaric? Reg Anesth 1989; 14: 195–8.

16. Davis H, King WR. Density of cerebrospinal fluid of human beings. Anesthesiology 1954; 15: 666–72.

17. Richardson MG, Wissler RN. Density of lumbar cerebrospinal fluid in pregnant and nonpregnant humans. Anesthesiology 1996; 85: 326–30.

18. Richardson MG, Collins HV, Wissler RN. Intrathecal hypobaric versus hyperbaric bupivacaine with morphine for cesarean section. Anesth Analg 1998; 87: 336–40.

© 2003 International Anesthesia Research Society


Become a Society Member