Reducing the Pain Associated with Local Anaesthetic Infiltration for open Carpal Tunnel Decompression

INTRODUCTION

Open carpal tunnel decompression is commonly performed under local anaesthetic. Although the procedure itself is usually painless, the preoperative infiltration of the palm with lidocaine (with or without adrenaline) is painful. Three distinct elements of this pain have been identified (Vossinakis, 2001). The first, though not the most important, is the introduction of the hypodermic needle through the skin and this can be minimized by the use of a blue needle (23-G). The second element of pain is the tension that the volume of the local anaesthetic solution causes in the unyielding subcutaneous tissues of the palm. This can be significantly reduced by a slow rate of local anaesthetic administration (Scarfone et al., 1998).

Finally, the acidity of the anaesthetic solution causes a burning pain that most patients describe as severe (McKay et al., 1987; Parham and Pasieka, 1996). Buffering lidocaine with sodium bicarbonate has been reported to be very effective in reducing this pain in a variety of minor surgical procedures (Fitton et al., 1996; Friedman et al., 1997; Matsumoto et al., 1994), but has not been assessed in carpal tunnel decompression.

PATIENTS AND METHODS

This double-blinded, prospective and randomized trial was approved by the hospital’s Scientific Committee, for patients undergoing bilateral open carpal tunnel decompression under local anaesthetic. Using the data from the first eight patients, a sample size calculation indicated that 11 patients (22 hands) would be required to achieve 99% power, for a clinically significant difference of one point on a10-point visual analogue pain scale. Twenty-one consecutive patients (42 hands) participated in the study after giving informed consent. Fourteen were women and seven were men and their average age was 53 (range, 34–69) years. Patients with a known allergy to lidocaine, and pregnant or breastfeeding women, were excluded. All patients had their second carpal tunnel decompression operation within 6 to 12 weeks of the first one.

Patients were anaesthetized with an injection of 15 ml of 1% lidocaine with adrenaline 1:200,000 in one hand and a 15 ml solution of 1% lidocaine with adrenaline 1:200,000 which had been buffered with 8.4% sodium bicarbonate (5:1 ratio) in the other. The order of these two anaesthetics was randomized and neither the patient nor the operating surgeon knew which type of solution was injected in each hand. Randomization of the order of use of the two solutions was performed using sealed envelopes stating the type of solution for the first procedure for each patient. The solutions were prepared immediately preoperatively and administered by an assistant who withheld the information until completion of the statistical analysis. A 23-gauge needle was used in all patients and the solution was administered over 30 seconds.

Following each injection patients quantified their perceived pain by drawing a line on a 10-point visual analogue scale (VAS). Three other 10-point visual analogue scales separately quantified the three different phases of pain (needle sting, tension, burning).

The paired, two-tailed Student’s t-test was used for all the statistical analyses.

RESULTS

All patients reported some pain during local anaesthetic infiltration on both sides. However, the mean VAS for pain was significantly lower for the hands in which the buffered solution was used (paired t-test, P< 0.0000001). VAS scores for the three characters–elements of pain were also compared. The stinging pain from needle introduction was minimal in all patients and did not differ between the two types of local anaesthetic. The same was true for the pain due to tension during the infiltration, which was described as “moderate”. In contrast, the burning pain was significantly lower with the buffered solution (P< 0.0000001). Table 1 presents these results. Buffering was effective in reducing the pain in all patients (Fig 1) and this was due to the reduction in its burning element (Fig 2).

No differences were observed between the pain levels in men and women and there was no correlation between age and pain reduction by the buffered solution. No adverse effects were recorded for the use of buffered lidocaine. The depth of local anaesthesia was adequate in both hands of all patients. All patients verbally stated their preference for the buffered solution and quite a few complained that the other side was “much more painful”. The patients who had the buffered lidocaine for their first operation were particularly unhappy with the normal lidocaine.

DISCUSSION

It was not feasible to perform both operations simultaneously since most patients desire having full use of one hand during recovery. Specific statistical analysis of the effect of the order of use of the two solutions was not performed. However, patients who received the buffered solution first found the second operation more painful than expected and the opposite was true for those who received the normal anaesthetic first.

Although several patients verbally responded that there was “no pain” during needle insertion, none recorded a score of “0” on the VAS. This indicates that most patients consider the sting from a 23-G needle necessary, tolerable and minimally painful. All patients described the tension of the anaesthetic solution as mildly to moderately uncomfortable rather than painful. Undoubtedly, the rate of infiltration is critical for the tension pain and a slow rate significantly reduces pain during this phase (Scarfone et al., 1998).

In the control hand (normal local anaesthetic) patients considered the burning element of the pain as the most painful. Buffering was effective in reducing this burning pain to a level where the tension effect was considered more painful than the burning. Several previous studies have documented the effectiveness of buffering lidocaine with sodium bicarbonate in reducing the pain from local anaesthetic injections in a variety of settings, including hand surgery (Friedman et al., 1997; Fitton et al., 1996; Matsumoto et al., 1994). In addition, buffering was found to be more effective than reducing needle size for pain reduction (Palmon et al., 1998). This study confirms the effectiveness of this method in open carpal tunnel surgery.

A 10:1 ratio is recommended for lidocaine buffering, mainly due to concern about precipitation of the solution. Prior to the clinical trial we tested solutions of up to 3:1 and no precipitation was observed. Furthermore, by comparing 1 ml of 10:1 and 5:1 solutions given subcutaneously on healthy volunteers from our department, we concluded that the 5:1 solution was less painful. So this ratio was used in our clinical trial.

Further advantages of buffered local anaesthetics (lidocaine, bupivacaine) have been reported. Buffering reduces the latency time for local anaesthesia and enhances the intensity and spread of nerve blocks (Curatolo et al., 1998; Gandy, 1991). Moreover, it has significant antibacterial properties (Thompson et al., 1993). Buffered local anaesthetic solutions are stable in the short term, but preparation just before surgery is recommended (Fitton et al., 1996). In the present study no precipitation of the solution was observed.

We believe that buffering lidocaine with 8.4% sodium bicarbonate at a ratio of 5:1 is an inexpensive, simple and effective method of reducing pain levels (Masters, 1998) and requires no extra time, since the operating surgeon can prepare the solution just before administration.