Shivering is a frequent complication in the postoperative period. The relative efficacy of interventions that are used for the treatment of postoperative shivering is not well understood. We performed a systematic search (MEDLINE, EMBASE, Cochrane Library, hand searching, all languages, to August, 2000) for full reports of randomized comparisons of any pharmacological antishivering intervention (active) with placebo (control) in the postoperative period. Dichotomous data on absence of further shivering after treatment and adverse effects were extracted from original reports. Relative risk (RR) and number-needed-to-treat (NNT) were calculated with 95% confidence interval (CI) using a fixed effect model. Data from 20 trials (944 adults received an active intervention, 413 were controls) were analyzed. Antishivering efficacy depended on the active regimen and the length of follow-up. Efficacy with meperidine 25 mg, clonidine 150 microg, ketanserin 10 mg, and doxapram 100 mg was reported in at least three trials; all were significantly more effective than control. After 1 min, the NNT of meperidine 25 mg for no further shivering compared with placebo was 2.7 (RR, 6.8; 95% CI, 2.5-18.5). After 5 min, the NNT of meperidine 25 mg was 1.3 (RR, 9.6; 95% CI, 5.7-16), the NNT of clonidine 150 microg was 1.3 (RR, 6.8; 95% CI, 3.3-14.2), the NNT of doxapram 100 mg was 1.7 (RR 4.0; 95% CI, 2.4-6.5), and the NNT of ketanserin 10 mg was 2.3 (RR 3.1; 95% CI, 1.9-5.1). After 10 min, the NNT of meperidine 25 mg was 1.5 (RR 4.0; 95% CI, 2.5-6.2). After 15 min, the NNT of ketanserin 10 mg was 3.3 (RR 1.5; 95% CI, 1.2-1.9). Long-term outcome data were lacking. There were not enough data for alfentanil, fentanyl, morphine, nalbuphine, lidocaine, magnesium, metamizol, methylphenidate, nefopam, pentazocine, and tramadol to draw meaningful conclusions. Reporting of adverse drug reactions was sparse. Fewer than two shivering patients need to be treated with meperidine 25 mg, clonidine 150 microg, or doxapram 100 mg for one to stop shivering within 5 min who would have continued to shiver had they all received a placebo.
Implications: Less than two shivering patients need to be treated with meperidine 25 mg, clonidine 150 microg, or doxapram 100 mg for one to stop shivering within 5 min who would have continued to shiver had they all received a placebo.