Multiple drugs in several classes are effective and well tolerated in patients with generalized anxiety disorder

Reference

Slee A, Nazareth I, Bondaronek P, Liu Y, Cheng Z, Freemantle N. Pharmacological treatments for generalised anxiety disorder: a systematic review and network meta-analysis. Lancet 2019;393(10173):768-777.

Synopsis

These authors searched more databases and registries than I knew existed to identify randomized trials that compared an active drug with a placebo or another active drug in treating patients with GAD. They excluded studies of patients with refractory GAD and any studies with tolerability run-in periods (YES! Thank you!). The researchers used the Cochrane Collaboration Risk of Bias Tool to assess each of the included studies. Ultimately, they included 89 studies of more than 25,000 patients and 22 different drugs from multiple drug classes. After extracting the data, the authors conducted a network meta-analysis to estimate the relative comparative effectiveness of each drug, as well as its tolerability. Most of the studies were double-blind and funded by industry. Studies from China (n = 16) were more likely to be have a higher risk of bias, and investigator-initiated studies were more likely to use an active comparator. Seven studies recruited patients older than 65 years. The median follow-up was 8 weeks and none were longer than 26 weeks. Overall, the most effective commercially available drugs, in order of effectiveness were bupropion, quetiapine, duloxetine, mirtazapine, hydroxyzine, sertraline, pregabalin, venlafaxine, escitalopram, fluoxetine, buspirone, "benzodiazepine," paroxetine, and citalopram. Several other agents did not statistically significantly decrease anxiety scores: imipramine, maprotiline, opipramol, tiagabine, vilazodone, and vortioxetine. However, several of the top performers were also associated with higher study discontinuation: quetiapine, paroxetine, benzodiazepines. The drugs with the best combination of effectiveness and tolerability were duloxetine, pregabalin, venlafaxine, and escitalopram. Finally, the magnitude of reduction in the best performers was only in the range of 2 to 3 points on a 56-point scale. Generally, 15% to 20% reductions are considered to be clinically meaningful, which would be 8 to 11 points in this case.