Investigación Científica y Tecnológica

Fredy Ariza Cadena*, Luis Fernando González Arboleda*, Jaime Humberto Jordán*, Jorge Humberto Mejía Mantilla*, Carmen Sofía Cárdenas*, Rafael Ordóñez*, Paulo José Llinás**

* Depto de Anestesiología, Reanimación y Clínica del Dolor. fredyariza@hotmail.com

** Departamento de Ortopedia. Fundación Valle de Lili. Depto de Anestesiología, centro de ortopedia y fracturas, Cali, Colombia. Correspondencia a Fredy Ariza Cadena. fredyariza@hotmail.com

Recibido: agosto 2 de 2010. Enviado para modificaciones: agosto 18 de 2010. Aceptado: agosto 31 de 2010.

Spinal anesthesia for outpatient procedures of the lower limbs may have several advantages over general anesthesia in terms of the use of resources, patient satisfaction and intra-operative and immediate post-operative analgesia (1). Recent publications have shown that low doses of local anesthetic associated with lateralized blocks are quick to create optimum conditions at the time of discharge, with lower use of post-operative opioids (2,3). Low-dose spinal racemic bupivacaine for arthroscopic procedures of the knee has been shown to have an adequate clinical profile, although little is known about the behavior of its levogiric isomere that has been considered as an alternative with lower cardiovascular effects, similar potency and comparable clinical performance (4-6).

Levobupivacaine (LBP) is a simple enantiomere of bupivacaine hydrochloride chemically described as (S)-1-butyl-2-pipeydulform-2’,6’-xylidide hydrochloride. Its pharmacodynamics in subaracnoidal administration allows for a greater sensory block as compared to racemic bupivacaine when isobaric solutions are used (7). Paradoxically, results from studies comparing hyperbaric solutions suggest longer spinal block resolution times when equal concentrations of equipotent doses are used (0.5 %, 8 mg) (8).

There are hardly no comparative studies regarding the effects of decreasing doses of LBP in terms of the duration of the motor block and the quality of the intra-operative analgesia, and it is not clear whether there is a true clinical difference in terms of the advantages of the spinal use of low doses of this s-enantiomere. The objective of this comparative randomized clinical trial was to assess the clinical profile and motor block probability through time using three different doses of 0.75 % hyperbaric LBP (HLBP) (7,5, 9,37 and 11,25 mg) with a short (5 minutes) unilateral spinal technique, recreating the situation of many operating rooms where the block is performed in the OR and where the effective management of the induction time is critical during the highly-demanding periods of outpatient unilateral arthroscopic procedures in the knee.

A prospective randomized double-blind study of parallel groups on a 1:1:1 ratio was designed with the approval of the IRB. The study included, prior informed consent, ASA I-II patients between 18 and 60 years of age scheduled for unilateral outpatient knee arthroscopy in a level IV university hospital and in a private orthopedic surgery clinic. Patients with uncompensated cardiac or pulmonary disorders who had used anti-platelet aggregation or anticoagulant medications during the week prior to the procedures, and patients scheduled for additional non-arthroscopic, bilateral procedures, or complex ligament reconstructions were excluded.

Using non-invasive standard monitoring, all patients received one bolus dose of 5 ml/kg of Ringer’s lactate and were then positioned in lateral decubitus with the limb to be operated in a dependent position, trying to maintain the table in a level as neutral as possible. After one bolus dose of midazolam IV (0.05 mg/kg), the lumbar puncture was performed through a medial approach into the L2-L3 or L3-L4 inter-vertebral spaces using a 27-gauge spinal Whitacre needle. At that point, patients were randomly assigned to receive 7.5 mg (Levo-7.5 group, n = 59), 9.37 mg (Levo-9.37 group, n = 61) or 11.25 mg (Levo-11.25 group, n = 60) of 0.75 % HLBP. Randomization was based on a dynamic balanced block, using a close-envelope allocation system.

After the expression of free cerebral-spinal fluid, the needle was directed towards the dependent limb. The dose mandated by the randomization envelope was placed in a 3 ml syringe and administered through the spinal needle at a rate of 0.1 ml/sec, with no additional aspiration maneuvers at the end of the injection. Each allocation and administration of the drug was conducted by any one of the anesthesiologists participating in the study who was available, as long as he/she was not directly responsible for the intra-operative management of the patient. All patients were kept in the same position for a period of five minutes before placing them back in a supine position for the start of the arthroscopic procedure.

An independent observer, blinded to the allocated dose, was responsible for assessing the motor and sensory blocks only on the operated side, immediately after returning the patient to the supine position and once the dressings had been placed on the surgical site upon completion of the procedure. The motor block was assessed using the modified Bromage scale (1 = unable to move feet or knees; 2 = able to move only the feet; 3 = starts to move the knees; 4 = detectable hip weakness in supine position [complete knee flexion]; 5 = no detectable hip weakness in supine position; 6 = able to bend knees partially while standing).

The sensory block was assessed on the basis of patient tolerance to transcutaneous electrical stimulation (TES) measured at four common surgical sites, namely, lateral aspect of the ankle (S1), medial aspect of the knee (L3), inguinal region (T12) and umbilicus (T10), using a peripheral nerve stimulator (Ministim® PNS model MS-IV, Organon, Dublin, Ireland), correctly calibrated, and 50-Hz tetanus during five seconds, initially at 10 mA, up to a maximum of 60 mA. The top limit was selected on the basis of prior studies that had shown that TES at 60 mA was equivalent to the intensity of the stimulus caused by a surgical incision. (9,10)

At the start of the procedure, the anesthesiologist in charge of the case, who had been instructed previously on the methodology, assessed and recorded the quality of the intra-operative analgesia as sufficient, partially sufficient or insufficient. In the event the block was found to be partially insufficient, the administration of a 50 μg bolus of fentanyl and 1-2 mg of midazolam was considered, followed by a five-minute waiting time before continuing. If after that time period the analgesia was inadequate to continue with the procedure, induction was performed and the patient was maintained under general anesthesia. All patients received an intra-articular infiltration of 20 ml of 0.25 % bupivacaine at the end of the procedure. In no case was a tourniquet used.

Relevant hypotension (fall > 30 % from baseline systolic pressure) was treated using an IV bolus dose of ethylefrine 1 mg (Effortil; Boehringer Ingelheim, Colombia) until adequate blood pressure was obtained; the presence of bradicardia (< 50 beats/min) was treated with IV boluses of atropine 0.5 mg. In the event of evidence of pulse saturations below 92 %, oxygen by nasal cannula was administered at a rate of 3 Lt/min. All adverse events during the intra-operative period were recorded in the intra-operative data collection form. At the end of the procedure, all patients received one dose of dipirone (metamizol) 2.5 g and/or sodium diclofenac 75 mg IV.

Regardless of the intra-operative course, all patients were followed by a nurse blinded to the randomization group and who had received training regarding the protocol, from the moment they arrived at the PACU and then every ten minutes until both the motor and the sensory blocks were considered to have fully resolved. During this period, the presence of adverse events and the need for analgesics other than those included in the protocol were recorded. Discharge criteria included stable vital signs, the ability to tolerate liquids by mouth and the absence of pain or nausea. All patients were instructed to come to the emergency room in the event of significant headache, back pain, neurological alterations in the lower limbs, or any other atypical symptom.

Throughout the research, the PNS was periodically calibrated in order to ensure correct assessment of the sensory block. An interim analysis was planned in order to determine study safety and viability halfway into the recruitment process. An inter-observer agreement test (Kappa index) was also performed during this period where two blinded observers assessed 13 (8 %) of the total number of subjects simultaneously.

The estimated sample size for this study was based on the mean and standard deviation for motor block resolution previously reported. (2,3) It was estimated that 165 patients would be required in order to detect an OR difference in motor block of 0.5 at 200 minutes, using an α error of 0.05 and a power value of 0.8. However, assuming approximately 10 % missing or lost information, the total sample was approximated to 180 patients (60 per group) using the Schoenfeld formula analyzing time to failure, based on the assumption that all patients would eventually present the event.

The analysis was conducted by an independent statistician using the Stata 11.0 software (StataCorp, LP). Continuous variables were statistically described, and data distribution as well as central and scatter trends were estimated. Considering that the assumption was that the motor block distribution and the discharge time would skew the data to the right, departing from the normalcy and homogeneity criteria implicit in the variance analysis, it was decided to perform an alternative logarithmic transformation in an attempt at normalizing the distributions.

In view of the above, the Bonferroni and Wilcoxon methods were used for comparison between the groups in the case of continuous variables (if the normalcy criteria were not met). Categorical variables were analyzed using the Xi2 test, estimated on the basis of the log likelihood. Probability curves of time to motor block resolution were built using overall survival curves and stratification by treatment group. Once the 95 % CI were calculated, statistical significance was estimated using the log-rank test. The possibility of adjusting ratios by covariates was determined using proportional risk multivariate models (Cox model), in the event the survival curves (Kaplan Meier) did not report significant differences between the groups. No subgroup analysis was considered.

Between June 2006 and December 2009, 183 consecutive patients were randomized until the sample size was completed. Of the 183, three had to be excluded from the analysis because they did not receive the treatment allocated (n = 2), and because of incomplete recovery-room data (n = 1) (Fig. 1). After randomization number 102, a new center was added, prior verification by CIEM and the Fundación Valle del Lili Clinical Research Institute (CRI). This center recruited patients simultaneously until the end of the study.

The remaining 180 patients were included in the analysis (Levo-7.5 group, n = 59; Levo-9.37 group, n = 61 and Levo-11.25 group, n = 60), and no differences were found in the demographic variables as a result of the analysis between the groups (table 1). There were no episodes of hemodynamic instability requiring additional volume expansion maneuvers or the use of vasopressors in any of the groups. Dual assessments showed a significant degree of inter-observer agreement (Cohen’s K = 0.775; p < 0,001) for consecutive motor and sensory block assessments.

When analyzing the patients in whom the block was insufficient to perform the surgical procedure, no significant differences were found (2/59 vs. 3/61 vs. 2/60 for the Levo-7.5, 8.25 and 11.25 groups, respectively; p = 0,303). Thirty-three patients (56 %) in the Levo-7.5 group, 27 (44 %) in the Levo-9.37 group and 21 (35 %) in the Levo-11.25 group were able to walk to the stretcher on their own at the completion of the surgical procedure (p = 0.022) (table 2).

Times to achieve complete motor block resolution were significantly different in the comparison between the groups (Levo-7.5 group, 145 ± 49 min; Levo-9.37 group, 156 ± 65 min; Levo-11.25 group, 170 ± 70.5 min; p = 0,006), with no differences in times to achieve complete sensory block resolution (table 2). Length of stay in the PACU was also significantly shorter for the Levo-7.5 group (155 ± 45 vs. 178 ± 70 and 184 ± 72; p = 0,004). The possibility of presenting motor block 200 minutes after the intra-thecal injection was significantly greater in the Levo-7.5 group [0,95 (95 % CI 0.84-0.98) vs. 0.80 (0.67-0.88), vs. 0.73 (0.59-0.82)], when analyzing the resolution through time (figure 2 and table 3. OR: 1.84 [95 % CI 1.28-2.64], log-rank test; p = 0,001).

None of the patients presented adverse events in the PACU or required strategies other than those established in the protocol for post-operative pain management. There were no cases of post-puncture headache or other complications requiring a visit to the emergency room or additional management.

This prospective, randomized, double-blind study compares three different doses of 0.75 % hyperbaric levo-bupivacaine in patients undergoing unilateral arthroscopic procedures of the knee and shows that a dose as low as 7.5 mg offers the best results in terms of length of stay in the recovery room and duration of the motor block, without affecting the rates of adequate blocks for these types of procedures.

A recent systematic review by Nair et al. concluded, after analyzing 17 randomized clinical trials, that 0.5 % hyperbaric bupivacaine at a dose as low as 4-5 mg, was sufficient to provide adequate intra-operative conditions in outpatient knee arthroscopy (11). With only a few studies comparing 0.5 % HLBP with the racemic mix and with ropivacaine, intra-operative efficacy and resolution of the motor block had not been assessed in a randomized design with three decreasing doses of 0.75 % HLBP.

The ex ante analysis of the main outcomes using a method with no covariate adjustment like the one described by Kaplan and Meier found an initial significant difference when comparing the group receiving the lower dose with the other groups (log-rank test). For this reason, no subsequent adjustments were required and there was no need to implement other analytical methodologies for this study. Unlike multiple reports in patients scheduled for lower limb surgery, our study showed block failure rates of 3 % - 4 % regardless of the total injected mass of the drug, including the work by Casati et al. who reported 100% effectiveness rates when they compared hyperbaric BPV (0.5 %; 8 mg), HLBP (0.5 %; 8 mg) and hyperbaric ropivacaine (0.5 %; 12 mg) (12). This may be explained by the substantial differences in the ability to overcome important factors like the technical difficulties during the injection of the local anesthetic, despite the standardization of the injection technique.

In this respect, a study by Brun-Buisson et al. in patients undergoing orthopedic surgery in a university hospital, reported spinal block failure rates of up to 9.6 % using BVP, with significant reductions when continuous spinal techniques (4.8 %) were used (13). This finding was consistent with a prior study by the same author, where a maximum reduction down to 3 % was achieved (14). This leads us to suggest that the failure rates in our study are within the expected range for the usual clinical practice, and to underscore the fact that that they were not significantly different when comparing the groups.

Regarding length of stay in recovery and complete resolution of the motor block, we found a significant reduction using lower doses of HLBP. This may be important in services where longer stays in the recovery room are a critical issue for outpatient flows. Spinal lidocaine was for many years the drug of choice for short outpatient procedures because of fast block resolution but, unfortunately, strong reports of persistent neurological symptoms flashed a warning regarding this route of administration (15,16).

Cappelleri et al. compared 7.5 and 5 mg of 0.5 % HLBP with 7.5 mg of 0.75 % hyperbaric ropivacaine and determined that HLBP 5 mg was as effective as, and comparable to, ropivacaine 7.5 mg in terms of motor block resolution, but with a higher potency (1:1.5 equipotency ratio). However, the authors had to perform the dilutions manually in order to obtain the hyperbaric mixtures (17). Commercial presentations of 0.75 % HLBP are available in many countries and that was the reason why we used those concentrations in our study. These observations would lead us to think that it is possible to reduce the HLBP dose still more since, ultimately, the total mass of the local anesthetic is more important than its concentration when trying to predict block intensity through the spinal approach.

After multiple discussions around the methodology, it was decided not to use opiods across the board in this study because of the issue of masking of events directly associated with the dose of HLBP such as post-operative pain and the sensitivity to TES, and because they would increase the incidence of adverse events such as itching and post-operative nausea and vomiting. The systematic review by Neil et al. showed that the concomitant use of fentanyl with low-dose hyperbaric local anesthetics administered by the spinal route had a smaller impact on discharge times and resulted in a significant increase of nausea and itching.

Although we did not find significant difference in the incidence of adverse events such as an increased use of analgesics, POPNV, itching or hemodynamic instability, our study failed to measure the incidence of other important events such as post-operative urinary retention. Voelckel reported recently that there was no significant variation in the incidence of this event with the use of the unilateral block, despite the use of low dose 0.5 % hyperbaric BPV, and we assume that it must not be very different for the case of HLBP (18). Acute urinary retention is a very frequent but underestimated event in patients receiving spinal anesthesia; for this reason, we believe more clinical studies with sufficient power are required in order to analyze this specific issue that we consider a common cause of delays in the discharge of patients undergoing spinal anesthesia for outpatient procedures.

In conclusion, our study shows that 7.5 mg of 0.75 % HLBP is an effective dose to ensure adequate spinal block conditions for unilateral outpatient arthroscopic procedures of the knee when a short unilateral spinal technique is used. Additionally, these lower doses of HLBP are associated with a higher probability of early motor block resolution and shorter length of stay, without a negative effect on post-operative analgesia or the occurrence of adverse events.

To Marisol Badiel and Jhon H. Loaiza from the Fundación Valle del Lili Clinical Research Center, for their constant contributions and positive critiques. Also to doctors Eddy Ríos and Oscar Ramírez, for their patience and dedication.

1. Jankowski CJ, Hebl JR, Stuart MJ, Rock MG, Pagnano MW, Beighley CM, et al. A comparison of psoas compartment block and spinal and general anesthesia for outpatient knee arthroscopy. Anesth Analg. 2003;97(4):1003-9.

2. Fanelli G, Borghi B, Casati A, Bertini L, Montebugnoli M, Torri G. Unilateral bupivacaine spinal anesthesia for outpatient knee arthroscopy. Italian Study Group on Unilateral Spinal Anesthesia. Can J Anaesth. 2000;47(8):746-51.

3. Borghi B, Stagni F, Bugamelli S, Paini MB, Nepoti ML, Montebugnoli M, et al. Unilateral spinal block for outpatient knee arthroscopy: a dose-finding study. J Clin Anesth. 2003;15(5):351-6.

4. Kokki H, Ylönen P, Heikkinen M, Reinikainen M. Levobupivacaine for pediatric spinal anesthesia. Anesth Analg. 2004;98(1):64-7.

5. Lee YY, Muchhal K, Chan CK. Levobupivacaine versus racemic bupivacaine in spinal anaesthesia for urological surgery. Anaesth Intensive Care. 2003;31(6):637-41.

6. Alley EA, Kopacz DJ, McDonald SB, Liu SS. Hyperbaric spinal levobupivacaine: a comparison to racemic bupivacaine in volunteers. Anesth Analg. 2002;94(1):188-93.

7. Lovo C, Montagna A, D’Angelo L, Volpetti S, Redivo M, Crupi D, et al. La levobupivacaina ipobarica nell’anestesia subaracnoidea in regime di day-surgery: confronto con la bupivacaina. Minerva Anestesiol. 2004;70(6):540-1.

8. Moizo E, Marchetti C, Bergonzi PC, Putzu M, Iemi K, De Luca M, et al. Unilateral spinal anestesia for inguinal hernia repair: a prospective, randomized, double-blind comparison of bupivacaine, levobupivacaine, or ropivacaine. Minerva Anestesiol. 2004;70(6):542-3.

9. Zbinden AM, Maggiorii M, Petersen-Felix, Lauber R, Thomson DA, Minder CE. Anesthetic depth defined using multiple noxious stimuli during isoflurane/oxygen anesthesia: I. Motor reactions. Anesthesiology. 1994;80(2):253-60.

10. McDonald SB, Liu SS, Kopacz DJ, Stephenson CA. Hyperbaric spinal ropivacaine. A comparison to bupivacaine in volunteers. Anesthesiology. 1999;90(4):971-7.

11. Nair GS, Abrishami A, Lermitte J, Chung F. Systematic review of spinal anaesthesia using bupivacaine for ambulatory knee arthroscopy. Br J Anaesth. 2009;102(3):307-15.

12. Casati A, Moizo E, Marchetti C, Vinciguerra F. A prospective, randomized, double-blind comparison of unilateral spinal anesthesia with hyperbaric bupivacaine, ropivacaine, or levobupivacaine for inguinal herniorrhaphy. Anesth Analg. 2004;99(5):1387-92.

13. Brun-Buisson V, Bonnet F, Liu N, Delaunay L, Saada M, Porte F, et al. Analysis of failures of spinal anesthesia as a function of practice development in a university hospital. Ann Fr Anesth Reanim. 1991;10(6):539-42.

14. Brun-Buisson V, Bonnet F, Boico O, Saada M. Failure of spinal anesthesia. Evaluation of the practice at a university hospital. Ann Fr Anesth Reanim. 1988;7(5):383-6.

15. Schneider MC, Hampl KF, Kaufmann M. Transient neurologic toxicity after subarachnoid anesthesia with hyperbaric 5% lidocaine. Anesth Analg. 1994;79(3):610.

16. Ackerman WE 3rd. Transient neurologic toxicity after subarachnoid anesthesia with hyperbaric 5% lidocaine. Anesth Analg. 1993;77(6):1306.

17. Cappelleri G, Aldegheri G, Danelli G, Marchetti C, Nuzzi M, Iannandrea G, et al. Spinal anesthesia with hyperbaric levobupivacaine and ropivacaine for outpatient knee arthroscopy: a prospective, randomized, double-blind study. Anesth Analg. 2005;101(1):77-82.

18. Voelckel WG, Kirchmair L, Rehder P, Garoscio I, Krappinger D, Luger TJ. Unilateral anesthesia does not affect the incidence of urinary retention after low-dose spinal anesthesia for knee surgery. Anesth Analg. 2009;109(3):986-7.