Spinal anesthesia is a widely used anesthetic technique for orthopedic surgery in the elderly, even though it has never been proven to have lower morbidity and mortality than epidural or general anesthesia [1,2]. Despite this, spinal anesthesia is often preferred for its efficacy, rapidity, minimal effect on mental status, reduction of blood loss, and protection against thromboembolic complications [2-4]. However, in the elderly there is a high prevalence of medical problems and a reduction in physiologic compensatory mechanisms. Spinal anesthesia is associated with a risk of severe and prolonged hypotension. This is due to the rapid extension of the sympathetic block, hindering cardiovascular adaptation and causing significant morbidity and mortality [2,3,5].
Continuous spinal anesthesia (CSA), by enabling the reduction and fractionation of the induction dose through a catheter, theoretically reduces the hemodynamic consequences of spinal anesthesia, but results are controversial and depend on many factors, such as the type of local anesthetic, its density, the technique used, and the methodology [2,6-11]. Moreover, CSA has been implicated in causing specific complications, such as infection, headache, or cauda equina syndrome [2,12,13]. Therefore, the purpose of this study was to compare the hemodynamic consequences of CSA versus single-dose spinal anesthesia (SDSA) for hip fracture repair in the elderly. The incidence of complications associated with both techniques was also recorded.
Sixty patients, aged 70 yr or more, ASA physical status I-III, undergoing hip fracture surgery were studied prospectively during an 18-mo period (September 1992-April 1994) in the Emergency Department of the Bordeaux University Hospital. After approval by the local ethics committee and obtaining the patient's or family's consent, the patients were included in the study if they accepted regional anesthesia and did not have an absolute contraindication for spinal anesthesia or a severely altered mental status.
Patients were randomized to two groups: SDSA (n = 30) and CSA (n = 30) groups. All patients had hip fractures and underwent surgery within 24 h after admission. Premedication consisted of midazolam in titrated doses (0.5-3 mg). An infusion of lactated Ringer's solution (8 mL/kg) was administered before the anesthetic procedure over 15-20 min.
The SDSA group received a single dose of 0.5% hyperbaric bupivacaine, based on age and height (15 mg between 70 and 79 yr and/or over 170 cm height, 12.5 mg between 80 and 90 yr and/or between 150 and 170 cm, 10 mg more than 90 yr and/or under 150 cm). The local anesthetic was injected intrathecally over 30 s in the lateral position through a 22-gauge needle (Becton Dickinson, Franklin Lakes, NJ) without aspiration at the L3-4 interspace, and the patient was immediately placed in a 30 degrees head-up supine position for the initial 45 min of the study.
The CSA group received 0.5% hyperbaric bupivacaine, injected intrathecally over 30 s through a 20-gauge catheter (Perifix Registered Trademark; B. Braun, Melsungen, Germany) inserted 2-4 cm cephalad at the L3-4 interspace through a 18-gauge Tuohy needle, while the patient lay in the lateral position. An initial dose of 5 mg (1 mL) of bupivacaine was injected; 15 min later, if the level of sensory block was lower than T10, 2.5 mg (0.5 mL) were injected every 5 min until a T10 or higher level of block was achieved. The patient was placed in a 30 degrees head-up supine position after the first injection for the initial 45 min of the study. In this group, reinjections of a third of the total induction dose were performed when pain reappeared at the operative site.
In both groups, the median approach was used for puncture and bupivacaine was injected at room temperature (20 degrees C). All the patients had their arterial blood pressure (BP) and heart rate (HR) taken at rest before the procedure. They were continuously monitored by clinical observation, automatic noninvasive BP (Dynamap Trademark; Critikon, Creteil, France), HR (CM5 derivation), and pulse oximetry. All data were recorded every 5 min for 4 h, including in the recovery room. The sensory level was assessed by the pinprick method and motor block according to the modified Bromage scale (0 = no motor block; 1 = hip flexion with extended leg blocked; 2 = knee flexion blocked; 3 = complete motor block).
Surgery was performed with the patients in the lateral or semilateral position as required. During the surgical procedure, all patients received mask oxygen, lactated Ringer's solution (5 mL centered dot kg-1 centered dot h-1), and perioperative bleeding was immediately treated with either colloid infusion or packed red cells depending on hematocrit (maintaining a 30%-32% hematocrit level). Any decrease in mean arterial pressure (MAP) below 25% of the preoperative value was treated with a 3-mg ephedrine bolus, which was repeated until hypotension was corrected. Duration of sensory block was defined by the regression of the block below the T12 level. Catheters in the CSA group were removed after 12 h. All patients were visited during the first month postoperatively for follow-up interviews.
For statistical analysis, the t-test was used for comparison of normally distributed data and the Mann-Whitney U-test for nonparametric values. For comparison of frequencies the chi squared test was used. Results are expressed as mean +/- SEM unless indicated otherwise. A P value of less than 0.05 was considered significant.
Both groups were comparable with regard to age, weight, height, ASA physical status, sex ratio, and preoperative values of MAP and HR Table 1. Spinal anesthesia was successful in all patients. One CSA patient and two in the SDSA group had difficult puncture of the subarachnoid space, requiring more than one attempt. No patients required additional sedation or analgesia during surgery. Both groups were comparable with respect to operative time, perioperative bleeding, and colloid or blood transfusion Table 2.
The maximum sensory block level was not significantly different in the two groups, but the time to reach this level and the time to reach the T10 level were significantly longer in the CSA group Table 3. In both groups, bilaterally equal levels of anesthesia were obtained. Motor block was not significantly different in either group: no patient had a Grade 0 Bromage score, one patient had a Grade 1 in the CSA group, 5 and 10 patients had a Grade 2 in the SDSA group and the CSA group, respectively. In the CSA group, motor block was independent of the side. Incomplete motor block did not affect the surgical procedure or require additional sedation.
The variations in MAP were significantly greater in the SDSA group at any time from 5 min after the first injection to 4 h later Figure 1. The decrease in MAP in the SDSA group also occurred earlier, within the first 10-15 min, whereas the decrease in the CSA group occurred later, between 45 and 60 min. When reinjections were necessary in the CSA group after 45-60 min, they did not affect hemodynamic stability. The maximum decrease in MAP was significantly greater in the SDSA group (40.2% +/- 1.9% of the baseline value) than in the CSA group (19.9% +/- 1.6%, P < 0.0001). In the SDSA group, 80% of the patients experienced at least one episode of significant hypotension (decrease in MAP greater than 30% of preoperative value). In the CSA group, no patient experienced episodes of severe hypotension or bradycardia. The variations in HR were small and not significantly different in the two groups Figure 1.
The mean dose of ephedrine was significantly larger in the SDSA group than in the CSA group (19.4 +/- 3.3 mg vs 1.8 +/- 0.7 mg, P < 0.0001). All SDSA patients required ephedrine (3 mg to more than 50 mg) to maintain pressure; only 11 (37%) CSA patients required ephedrine and only 3 patients received more than 3 mg ephedrine Figure 2.
In the immediate postoperative period, two SDSA patients and three CSA patients experienced temporary confusion. No cardiovascular complications were observed in either group. Nausea and/or vomiting were observed in one patient in each group. No patient suffered from headaches in the following days, and no late complications related to spinal anesthesia were observed in either group. No patient in the series died during the 1-mo observation period. The length of hospitalization in the surgical unit was comparable in the SDSA and CSA groups (13.5 +/- 3.7 and 14.0 +/- 3.1 days, respectively).
Spinal anesthesia is a commonly used technique for lower limb surgery in the elderly. Since the elderly population is increasing, and since these patients often have concomitant medical problems and reduced physiologic adaptation capacities, spinal anesthesia can be an ideal anesthetic technique [3,4,14].
Although widely used, spinal anesthesia is associated with the risk of potential complications, the main one being hypotension. The hemodynamic consequences of spinal anesthesia related to the degree of sympathetic block might be of greater importance in the elderly, leading to possible serious postoperative consequences [9,15]. Therefore, the question arises as to the possibility of limiting hemodynamic consequences by reducing and titrating the local anesthetic dose, without causing other complications with the CSA technique.
It is well known that many factors influence the kinetics of local anesthetics in the intrathecal space . However, only some are of major importance such as the dose of local anesthetic related to height and age, and its specific gravity [16-19]. Reduction of the local anesthetic dose according to age and height seems logical. Although it is established that the dose influences the spread of spinal anesthesia , it did not affect the efficacy of spinal anesthesia in our study. However, a greater reduction of local anesthetic doses, if theoretically resulting in decreased hemodynamic consequences, may lead to a high rate of anesthesia failure (insufficient sensory level and duration). Moreover, the specific gravity of the anesthetic solution is highly important in the interpretation of the local anesthetic spread. Hyperbaric bupivacaine at 0.5% usually leads to a higher level of anesthesia than iso- or hypobaric bupivacaine for comparable doses. Spinal anesthesia with hyperbaric bupivacaine is more predictable, making it possible to control the level of block by the choice of the dose and position . However, using the CSA technique, Van Gessel et al.  reported a more controllable and favorable anesthetic with respect to hemodynamic consequences. The same author found greater hemodynamic consequences with larger doses in older patients with plain bupivacaine .
In our study, all the SDSA patients and approximately 40% of the CSA group received ephedrine, a much larger number than reported in previous studies [9,15]. This is mainly because we started to administer small doses of ephedrine when there was a decrease of 25% in baseline MAP, in order to avoid severe hypotension. However, none of the patients in the CSA group received more than 12 mg ephedrine. The slower progression and decreased hemodynamic consequences in the CSA group may be explained by a slower onset of segmental block in this group. This allows the cardiovascular system to adapt more easily than when the sympathetic block appears abruptly, such as with SDSA. This slow onset of block was found in previous studies using different protocols of CSA [9,15] whereas a rapid decrease in BP is usually reported with SDSA. There are substantial differences in the severity of hypotension, depending on various factors, including the dose of local anesthetic, age, physical status, and position of the patient [9,15,19,21].
Experimentally, injection through an intrathecal catheter appears to be a critical factor in influencing the distribution of local anesthetic in the subarachnoid space . In some studies, the use of intrathecal catheters has led to technical difficulties, with a significant rate of failure [23,24]. The use of larger catheters may reduce this rate and avoids slow injection through high resistance microcatheters which has been implicated as a possible cause of inadequate anesthesia and cauda equina syndrome. The quality of anesthesia was completely satisfactory in both groups of our study. None of our patients required additional sedation or analgesia, which might seriously compromise the elderly patient. No late complications directly related to the spinal anesthesia techniques were found.
In summary, by enabling reduction and fractionation of the local anesthetic, CSA using 0.5% hyperbaric bupivacaine was safe and effective for management of hip fracture in the elderly. CSA provided better hemodynamic stability during the perioperative periods and required lower doses of ephedrine than SDSA. As a result, CSA may be the technique of choice for lower limb surgery in the elderly, when hemodynamic stability is critical.
1. Djokovic JL, Hedley-Whyte J. Prediction of outcome of surgery and anesthesia in patients over 80. JAMA 1979;242:2301-6.
2. Sutter PA, Gamulin Z, Forster A. Comparison of continuous spinal and continuous epidural anesthesia for lower limb surgery in elderly patients: a retrospective study. Anaesthesia 1989;44:47-50.
3. Covert CR, Fox GS. Anaesthesia for hip surgery in the elderly. Can J Anaesth 1989;36:311-9.
4. Valentin N, Lomholt B, Jensen JS, et al. Spinal or general anaesthesia for surgery of the fractured hip? Br J Anaesth 1986;58:284-91.
5. Chan VWS, Chung F, Gomez M, et al. Anesthetic and hemodynamic effects of single bolus versus incremental titration of hyperbaric spinal lidocaine through microcatheter. Anesth Analg 1994;79:117-23.
6. Bonnet F, Marcandoro J, Minoz O, et al. Comparaison entre rachianesthesie conventionnelle et rachianesthesie continue utilisant la bupivacaine. Ann Fr Anesth Reanim 1990;9:280-4.
7. Palas TAR. Continuous spinal anesthesia versus single shot technique in the elderly [abstract]. Reg Anesth 1988;13:S9.
8. Pitkanen M, Rosenberg P, Silvanto M, Tuominen M. Haemodynamic changes during spinal anaesthesia with slow continuous infusion or single dose of plain bupivacaine. Acta Anaesthesiol Scand 1992;36:526-9.
9. Sabate A, Asbert R, Gracia T, et al. Regional anesthesia and elderly patients: continuous subarachnoid anesthesia versus single dose in peripheral vascular surgery. Reg Anesth 1994;19:79-84.
10. Schnider ThW, Mueller-Duysing S, Johr M, Gerber H. Incremental dosing versus single-dose spinal anesthesia and hemodynamic stability. Anesth Analg 1993;77:1174-8.
11. Van Gessel EF, Forster A, Gamulin Z. Surgical repair of hip fractures using continuous spinal anesthesia: comparison of hypobaric solutions of tetracaine and bupivacaine. Anesth Analg 1989;68:276-81.
12. Riggler ML, Drasner K, Krejcie TC, et al. Cauda equina syndrome after continuous spinal anesthesia. Anesth Analg 1991;72:275-81.
13. Rasmusen BS, Blom L, Hansen P, Mikkelsen SS. Postspinal headache in young and elderly patients: two randomised, double-blind studies that compare 20- and 25-gauge needles. Anaesthesia 1989;44:571-3.
14. Carpenter RL, Caplan RA, Brown DL, et al. Incidence and risk factors for side effects of spinal anesthesia. Anesthesiology 1992;76:906-16.
15. Klimscha W, Weinstabl C, Ilias W, et al. Continuous spinal anesthesia with microcatheter and low-dose bupivacaine decreases the hemodynamic effects of centroneuraxis blocks in elderly patients. Anesth Analg 1993;77:275-80.
16. Greene NM. Distribution of local anesthetic solutions within the subarachnoid space. Anesth Analg 1985;64:715-30.
17. Bannister J, McClure JH, Wildsmith JAW. Effect of glucose concentration on the intrathecal spread of 0.5% bupivacaine. Br J Anaesth 1990;64:232-4.
18. Veering BTh, Burm AGL, Van Kleef JW, et al. Spinal anesthesia with glucose-free bupivacaine: effects of age on neural blockage and pharmacokinetics. Anesth Analg 1987;66:965-70.
19. Shesky MC, Rocco AG, Bizzarri-Schmid M, et al. A dose response study of bupivacaine for spinal anesthesia. Anesth Analg 1983;63:931-5.
20. 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.
21. Pitkanen M, Haapaniemi L, Tuominen M, Rosenberg PH. Influence of age on spinal anaesthesia with isobaric 0.5% bupivacaine. Br J Anaesth 1984;56:279-84.
22. Riggler ML, Drasner K. Distribution of catheter-injected local anesthetic in a model of subarachnoid space. Anesthesiology 1992;75:684-92.
23. Hurley RJ. Continuous spinal anaesthesia. Int Anesthesiol Clin 1989;27:45-50.
24. Nagle CJ, McQuay J, Glynn CJ. 32-gauge spinal catheters through 26-gauge needles. Anaesthesia 1990;45:1052-4.