Evaluation of intrathecal phenol injection using a rat model

Introduction: Intrathecal phenol injections have been used in the treatment of both chronic benign and cancer pain for over forty years. We developed a model to allow further evaluation of its effects on visceral and somatic pain perception. Methods: The study was approved by the institutional animal care committee. Somatic pain thresholds of both hind paws were evaluated using a radiant heat device. Colorectal distension pain thresholds were measured using a dual balloon system as previously described '. A total of 30 male Sprague-Dawley rats weighing 300-350g were studied. Each animal had an 8cm V-l intrathecal catheter inserted in the manner previously described to lie at the second lumbar vertebra2. The animals were allowed to recover from implantation for one week. Animals with motor deficits after implantation were not used. Testing was performed for two days prior to injection. Each animal was then anesthetized with halothane 2%. Following initial dose finding studies, animals were divided into three groups: A received 50ul 5% phenol in glycerol (n=15), B received 50ul glycerol (n=4), C nothing (n=7). The injected animals were kept anesthetized and positioned in a 30 degree headup tilt to encourage dorso-caudal spread of the solution following injection for 30 minutes. Each animal was allowed to recover for one day, then tested daily. Statistical analysis was performed using repeated measures of variance and Fisher's post hoc test for time. Results: An initial group of four animals was used to determine the optimal volume of injection. Volumes of 25, 37.5, 50 and 75ul were injected in sequential animals. An optimum response was deemed to be one with substantial prolongation in hotplate latency without severe paralysis. A volume of 50ul was found to reliably prolong latencies without an unacceptably high incidence of paralysis. Lower volumes caused transient unpredictable responses. Subsequently all animals were injected with 50ul of 5% phenol in glycerol. 5/26 animals developed dense paralysis following injection and were excluded from further analysis. Responses (Mean +/-SEM) were compared between phenol and control groups, and are seen in Fig. 1 and 2. Most animals injected with phenol developed transient bilateral lower extremity weakness following injection. There was a significant short lived prolongation of thermal threshold latencies and colorectal distension threshold pressures. Discussion: Our study shows that in most animals the response to intrathecal phenol injection is transient, suggesting a prolonged local anesthetic type effect rather than a neurolytic effect. Few animals developed prolonged effect beyond the first few days. This is consistent with the reported clinical effects where many patients only derive transient benefit.