Effects of skin cooling on intra-articular knee temperature in healthy subjects

Abstract

Purpose: Cooling is widely used to control pain associated with acute injury and surgery involving the knee. Cooling may affect pain by exerting effects on the inflammatory cascade, transduction channels in nociceptors, sodium channel activity in neurons, or may suppress pain through masking effects mediated by central mechanisms. CNTX-4975, a highly purified, synthetic trans-capsaicin, is under investigation in phase 3 trials to determine efficacy and safety when administered via intra-articular (IA) injection in patients with moderate to severe knee osteoarthritis pain. Cooling has been demonstrated to decrease pain associated with topical capsaicin without interfering with the therapeutic response. In other studies, it has been determined that cooling markedly decreases the initial pain associated with capsaicin injection. Here we report temperature variations of both the skin and the IA space in healthy volunteers using 2 different cooling techniques currently used in clinical practice Methods: Two knee cooling techniques were used in healthy volunteers: an ice pack placed on the patella of one knee, and a commercially available circumferential circulating ice water wrap (CCIWW) placed on the skin surrounding the other knee. To reduce bias from order effects, 3 subjects received the CCIWW on the left knee and ice pack on the right, and the other 2 subjects received ice pack cooling on the left knee and CCIWW on the right. Two temperature probes were used for each knee: an IA probe and a skin probe placed adjacent to the superior pole of the patella under the cooling apparatus. After baseline temperature measurements in the left knee, cooling was applied for 15 minutes, at which point subjects received an IA injection of 15 mL of 2% lidocaine without epinephrine (to simulate the procedures used with the CNTX-4975 IA injection procedure). IA CNTX-4975 was not administered. Cooling was restarted and continued up to a total of 120 minutes. The process was repeated for the right knee after a rest period (≤4 hours) following left knee measurements. Temperatures within the knee were measured at approximately 5-minute intervals from probe placement until removal. Probe removal was approximately 30 minutes following cessation of cooling. Subjects were asked to rate their pain using a numeric pain rating scale (0-10; 0=no pain; 10=worst imaginable pain) at various time points throughout the procedure. At subsequent visits, knees were inspected for any injection site issues, and adverse events were recorded. Results: Five healthy male subjects were enrolled; mean age, 30 years. All subjects completed the study for both knees. The Figure shows the mean temperature change at the skin surface and in the joint over time under the 2 cooling conditions. The mean baseline skin and IA temperatures were approximately 29°C and 33°C, respectively. The decrease in skin temperature was similar between techniques for the first 40 minutes, reaching a temperature of approximately 22°C. With the ice pack, the skin temperature plateaued after 50 to 60 minutes and had a temperature of 19°C at 140 minutes when cooling was removed. In contrast, with the CCIWW technique, the skin temperature continued to decrease to 14°C at 140 minutes. The ice pack did not have an appreciable effect on the IA temperature at any point. With the CCIWW technique, 40 minutes after the start of cooling, the mean IA knee temperature had decreased by approximately 2°C, and by 140 minutes had decreased by approximately 10°C to a temperature of 23.9°C (95% CI, 22.6-25.2). Even after the cooling was stopped, there was a delay before the IA temperature began to rise. Subjects reported no pain from either the skin or the joint. The procedures overall were well tolerated and there were no complications. Conclusions: The baseline knee IA temperature under ambient conditions was considerably below core temperature and approximately 33°C. The mean baseline skin temperature overlying the joint was approximately 29°C. Circumferential cooling provided by the CCIWW technique was superior to cooling with the ice packs, resulting in a 9°C decrease in skin temperature and an approximate 10°C decrease in joint temperature over 140 minutes. Although there was initially a lag, decreases in joint and skin temperature were temporally parallel. The thermal inertia was such that the IA temperature decreased even after cessation of cooling before it began to rise. Effective joint cooling appears to favor a method that provides circumferential cooling of the joint. Subsequent studies will determine how procedural pain from trans-capsaicin injection into an osteoarthritic knee relates to joint temperature.