A prospective randomised controlled trial of diazepam (valium) vs emulsified diazepam (diazemuls) as a premedication...

Two hundred and forty four patients were randomised to receive either diazepam or emulsified diazepam (diazemuls) intravenously prior to routine upper gastrointestinal endoscopy. The groups were comparable with regard to age, sex, medical condition, vein size, and amount of drug administered. A quantitative assessment of symptoms at the injection site during the ensuing week was made by analysing questionnaires on pain and tenderness, which the patients graded each day, on a scale 0-3. Twenty-one out of 82 patients who received diazepam (25.6%) and 22 out of 84 who received diazemuls (26.2%) had local symptoms. Although the mean symptom scores over 7 days for the two groups were not different (5.95 +/- 5.19 and 7.27 +/- 6.30 respectively), more patients who received diazepam reported induration (P = 0.033). In 112 patients, signs of thrombosis in the injected veins were looked for 3-7 weeks later without knowledge of the preparation given. Thrombosis was present in 13 (25%) patients who had received diazepam and two (3.6%) who had received diazemuls (P less than 0.002). Only eight (53%) patients with thrombosed veins had symptoms. Thrombosis after intravenous injection of diazepam or diazemuls may be symptomless, and is significantly less likely following diazemuls.