#216 Anxiously Awaiting Evidence: Pregabalin in generalized anxiety disorder
CLINICAL QUESTION
Is pregabalin effective for generalized anxiety disorder (GAD)?
BOTTOM LINE
Evidence for pregabalin in GAD is inconsistent and at high-risk of bias (industry-written, short-term, poorly described methods, high drop-outs, and run-in periods that overinflate benefit). If real, an additional one in 6-8 people may respond to pregabalin compared to placebo at 4-8 weeks. However, the change in anxiety scales was not clinically meaningfully different than placebo for the average patient.
Reading Tools for Practice Article can earn you MainPro+ Credits
Join NowAlready a CFPCLearn Member? Log in
EVIDENCE
- All Randomized Controlled Trials (RCTs) written by industry.
- One systematic review, four RCTs:1
- Versus placebo (one RCT, n=271):
- Response: 59% versus 44% placebo, not statistically different.
- Changes in anxiety scale: ~3-4 points out of 56, not always statistically significant, likely not clinically meaningful.
- Versus benzodiazepines:
- Response (one RCT, n=454): 300 mg statistically better than alprazolam (61% versus 43%) but higher doses no difference.1,2
- Change in anxiety scale (one RCT, n=271): No difference.
- Overall adverse effects: 67% placebo, 73% pregabalin 50 mg, 89% pregabalin 200 mg, 91% lorazepam.1
- Versus placebo (one RCT, n=271):
- Four other RCTs (273-374 patients each):
- Change in anxiety scale: ~3 points out of 56. Statistically different, not clinically meaningful.3-6
- Response rates 50-60% versus 27-46% placebo. Statistically different in ¾ studies.3,5,6 Number Needed to Treat (NNT)=6-8.
- Trend to higher response rates with lorazepam (61% versus 46%).6
- Other systematic reviews provided standard mean differences (clinically uninterpretable).7,8
- RCT versus sertraline:9 No difference anxiety scale or adverse effects.
- As adjunct: RCT of 356 patients.10
- If inadequate response to antidepressant, randomized to pregabalin 150-600 mg/day or placebo. At eight weeks:
- Mean change in anxiety scale: 1.2 (statistically, but not clinically different).
- Response (anxiety scale): 48% versus 35%, NNT=8.
- Response (global improvement scale) or remission: No difference.
- Stopped due to adverse effects: 4% versus 2%, Number Needed to Harm=47.
- If inadequate response to antidepressant, randomized to pregabalin 150-600 mg/day or placebo. At eight weeks:
- Limitations: <80% completed study;3-6 short-term (4-8 weeks);1-6 selective reporting;6 quality markers inadequately described;3-6 run-in which can overinflate benefit.2,4-6
- Several studies show lower anxiety scores within one week, but usually not a clinically meaningful difference.3,4,6
- Weight gain at one year (all indications): 17% gained >7% of their body weight and mean gain=2.2kg.11
- Canadian guidelines recommend pregabalin or antidepressants first-line or as adjunct.12
Latest Tools for Practice
#364 Facing the Evidence in Acne, Part II: Oral Antibiotics
How effective are oral antibiotics in treating acne of at least mild-moderate severity?
#363 Making a difference in indifference? Medications for apathy in dementia
In patients with dementia, how safe and effective are stimulants, antidepressants, and antipsychotics for treating apathy?
#362 Facing the Evidence in Acne, Part I: Oral contraceptives and spironolactone in females
How effective are combined oral contraceptives (COC) and spironolactone for treating acne of at least mild-moderate severity in females?
