Venlafaxine ER for GAD in Adults Over Age 60
From MindLinc Wiki
Bottom Line: Venlafaxine ER is equally safe and well tolerated by and shows similar efficacy in young and older patients in the treatment of GAD.
Background: Although GAD is a common psychiatric disorder in older adults, it is often under diagnosed and under treated. Estimates of 6 month prevalence in older (age>=60) community dwelling patients has recently been estimated as high as 17.3% in the U.S. Treatment for GAD in older adults often consists of low dose bzd, despite concerns that these are associated with memory loss, psychomotor impairment, auto accidents and falls. There is also concern re: abuse or dependency with chronic use. Venlafaxine is an alternative with good tolerability profiles and no dependence, tolerance or amnesitic effects.
Clinical Question: Evaluate utility of Venlafaxine ERTM specifically in older patients with a diagnosis of GAD.
Methods: Authors conducted a series of post hoc analyses comparing efficacy and safety of Venlafaxine ER vs placebo in sub samples of older and younger patients using pooled data from 5 separate manufacturer-initiated studies. These were all prospective, double-blind, randomized, placebo-controlled clinical trials. There was no apparent screening for studies, just pre selection of these 5. Subjects were outpatients, 18 or older, with a diagnosis of GAD per DSM-IV and a HAM -A score of >=18 with at least moderate symptoms of anxious mood and tension (scores >=2 for items 1 &2). Exclusion criteria: Diagnosis of MDD w/in 6 months of tudy entry; depressive symptoms (per Rankin Scale) greater than anxiety symptoms (per Covi Scale); any diagnosis of primary psychosis or BPAD; previous use of venlafaxine ER; drug allergy, clinically significant medical illness that might compromise the study, MI w/in 6 months; substance dependence within 1 year; > lg caffeine per day or clinically significant abnormalities on physical exam or lab tests. Primary Outcomes: HAM -A total score (Response = 50%. in HAMA, Remission = <7 HAMA); psychic anxiety factor; anxiety subscale of Hospital Anxiety & Depression Scale and CGI Impression of Improvement (Response =1 or 2). Secondary Outcomes: HAMA somatic anxiety factor, HAD depression subscale, CGI severity, Covi Scale for Anxiety ad Raskin Scale for Depression. Primary efficacy evaluations were based on intention-to-treat analyses. Analyses of continuous measures such as efficacy rating scales were conducted using three way analysis of covariance with treatment, study, age group and treatment by age group interaction in the model and baseline as a covariate. The percentages of responders and remitters and incidence of study events and reasons for termination were analyzed using logistic regression with treatment, study, age group and treatment by age group interaction in the model.
Results: As for efficacy outcomes, owing to limited statistical power for detection of decrements in venlafaxine ER placebo differences between younger and older sub samples, the authors felt guided to perform evaluation of effects of the drug with a series of post hoc analyses within each of the two age-defined subgroups of patients (young < 60yo; older >= 60yo). During first 8 weeks, significant drug-placebo differences were noted in older patients in LOU on HAMA psychic anxiety factor and HAD anxiety subscale. In 24 week analyses, drug-placebo differences were noted in HAMA anxiety factor and CGI in both LOCF and observed-case analyses. However, in both time periods, the difference between placebo and drug impact on primary and secondary measures is small, and in general less than one-half the size of the placebo effect.
As for analyses of responses, tests at Week 8 found significant drug-placebo differences for older adult patients in the LOCF analyses of responders on the CGI-I and both responders and remitters on the HAM A. In the 24 week analyses of the older subgroup, there was a significantly higher rate of responders to venlafaxine ER than to placebo on both the HAMA and CGI-I scales. However, the CGI-I is not specific as an objective measure of patient's anxiety symptoms (the focus here) and the "n" for the 24 week results is quite low for the "Older" group.
A review of rate of treatment-emergent adverse events demonstrated a main effect for age, with inverse relationship. Among categories of adverse events leading to discontinuation, there were main effects of treatment in events related to digestive and metabolic systems, but no age effects or age by treatment interactions. Percent of participants discontinuing due to lack of efficacy was significantly different, favoring venlafaxine, only in the younger group.
Conclusions: Though the current does study shows that venlafaxine produced improvement (as measured across multiple scales) statistically greater than that seen with placebo in elderly (age >60), it does suffer from some weaknesses. In looking to combine multiple pat studies the authors did not screen all past studies comparing venlavaxine vs placebo, but arbitrarily chose 5 manufactures initiated studies. Exclusion criteria were vague (physical or laboratory abnormalities) or excluded many who may by that same criteria may have significant anxiety (MI w/in 6 months); and also created a selection bias in that those - had been tried on venlafaxine (and who by being in this study must have GAD and HAMA > 18) who would be known non-responders, were excluded. Moreover, though there is statistical significance to the difference of primary measures for drug and placebo, the meaning of such small differences in terms of improvement of functioning is vague. Perhaps more rigorous results could come from screening literature for all studies which by similarity of design would allow pooling of results, allowing for greater N, hence greater power. Moreover, allowing the medical exclusion criteria to be only a strict contraindication to the use of venlafaxine XR would provide for greater validity.
Comments:
