Pharmacological interventions for benzodiazepine discontinuation in chronic benzodiazepine users

摘要: Background Prolonged treatment with benzodiazepines is common practice despite clinical recommendations of short-term use. Benzodiazepines are used by approximately 4% the general population, increased prevalence in psychiatric populations and elderly. After long-term use it often difficult to discontinue due psychological physiological dependence. This review investigated if pharmacological interventions can facilitate benzodiazepine tapering. Objectives To assess benefits harms discontinuation chronic use. Search methods We searched following electronic databases up October 2017: Cochrane Drugs Alcohol Group's Specialised Register Trials, CENTRAL, PubMed, Embase, CINAHL, ISI Web Science. We also ClinicalTrials.gov, WHO ICTRP, ISRCTN registry, checked reference lists included studies for further references relevant randomised controlled trials. Selection criteria We trials comparing versus placebo or no intervention another adults who had been treated at least two months and/or fulfilled criteria dependence (any criteria). Data collection analysis We standard methodological procedures expected Cochrane. Main results We 38 (involving 2543 participants), but we could only extract data from 35 2295 participants. Many different were studied, single was assessed more than four trials. extracted on 18 comparisons. The risk bias high all one. Trial Sequential Analysis showed imprecision comparisons. For discontinuation, found a potential benefit valproate end (1 study, 27 participants; ratio (RR) 2.55, 95% confidence interval (CI) 1.08 6.03; very low-quality evidence) tricyclic antidepressants longest follow-up 47 RR 2.20, CI 1.27 3.82; evidence). We potentially positive effects withdrawal symptoms pregabalin 106 mean difference (MD) -3.10 points, -3.51 -2.69; evidence), captodiame 81 MD -1.00 -1.13 -0.87; paroxetine (2 studies, 99 -3.57 -5.34 -1.80; -19.78 -20.25 -19.31; flumazenil (3 58 standardised -0.95, -1.71 -0.19; intervention. However, effect did not persist until 54 -0.13 -4.03 3.77; evidence). The reduced anxiety intervention: carbamazepine 36 -6.00 -9.58 -2.42; -4.80 -5.28 -4.32; -5.70 -6.05 -5.35; -6.75 -9.64 -3.86; -1.30 -2.28 -0.32; evidence). Two treatments seemed reduce proportion participants that relapsed use: 0.31, 0.11 0.90; cyamemazine 124 0.33, 0.14 0.78; evidence). Alpidem decreased 25 0.41, 0.17 0.99; number needed treat an additional harmful outcome (NNTH) 2.3 occurrence syndrome 145 4.86, 1.12 21.14; NNTH 5.9 Likewise, magnesium aspartate discontinuing 144 0.80, 0.66 0.96; 5.8; evidence). Generally, adverse events insufficiently reported. Specifically, one discontinued severe panic reactions. Authors' conclusions Given low quality evidence reported outcomes, small identified limited each comparison, possible draw firm conclusions regarding users. Due poor reporting, be reliably across More required less systematic errors ('bias') random ('play chance') better full reporting patient-centred outcomes. Such ought conducted independently industry involvement.