Cerebrospinal fluid neurotransmitter changes during the perioperative period in patients undergoing total knee replacement: a randomized trial.

BACKGROUND: Total knee replacement (TKR) is of enormous benefit to patients with osteoarthritis of the knee; however, the acute postoperative pain can be severe and difficult to manage. The role of major spinal cord neurotransmitters in this acute postoperative period is not clear, although there are a few studies in humans. We performed the first prospective clinical study undertaken to delineate the changes in the spinal neurotransmitters after a surgery such as TKR. Furthermore, we also determined whether antihyperalgesic drugs at clinically acceptable doses modulate spinal neurotransmitter concentrations in patients during the perioperative period.

METHODS: All patients had a spinal needle placed in the lumbar region and cerebrospinal fluid (CSF) obtained for baseline measurement of the neurotransmitters. An intrathecal catheter was then placed for spinal anesthesia for standard TKR and for continuous spinal postoperative analgesia. The spinal catheter was also used postoperatively to sample CSF at 2, 4, 8, 12, 24, and 32 hours after catheter placement. CSF samples were assayed for norepinephrine, substance P, calcitonin gene-related peptide (CGRP), and glutamate concentrations. SF-36 (36-item Short Form Health Survey) was measured preoperatively. Numerical rating scale (NRS) pain scores and intrathecal analgesic consumption were recorded postsurgery at 4-hour intervals for 32 hours. We performed a randomized, placebo-controlled, double-blind trial with 3 drug groups (n = 16 per group): placebo; single-dose pregabalin (150 mg administered before surgery); and multidose pregabalin (150 mg administered presurgery and 12 and 24 hours later), to determine the effect of an antihyperalgesic drug such as pregabalin on spinal neurotransmitters.

RESULTS: Forty-eight patients were randomly assigned to the 3 perioperative treatment groups, and multiple CSF samples were successfully obtained from 44 patients. Before surgery, increased bodily pain (from preoperative SF-36 measure) was correlated with increased CSF norepinephrine concentration (P = 0.044). Compared with presurgery values, norepinephrine levels were lower in the placebo group at the 2- and 4-hour time points (P < 0.005) whereas in the single and multidose groups, the reduction (P < 0.001) continued until 12 and 24 hours, respectively. Substance P CSF levels had an early peak value (at 2 hours) in all 3 groups, and then returned to baseline. Compared with baseline value, the CGRP CSF levels only decreased at the 32-hour time point in the placebo group, but in both pregabalin groups, CGRP levels decreased over the 4- to 32-hour period. In the placebo group only, CSF glutamate decreased over 4 to 32 hours compared with presurgery values. However, there was no difference in the CSF neurotransmitter concentrations among the 3 treatment groups over the 32-hour sampling period. In the placebo group, the early NRS pain score area under the curve, AUC [0-12 hours], was positively correlated (R = 0.67, P = 0.0088) with the CSF norepinephrine concentration AUC [12-24 hours], but none of the other neurotransmitters was correlated with the NRS. None of the CSF neurotransmitter concentrations correlated with postoperative analgesic consumption.

CONCLUSION: In the perioperative period, the concentration changes of the 4 spinal neurotransmitters have a distinct time course. CSF substance P seems to increase very rapidly with surgical intervention, whereas the CSF norepinephrine concentration tends to decrease. At clinical doses, pregabalin does not seem to modulate these spinal neurotransmitter concentrations.