Statistics from Altmetric.com
Systematic review with meta-analysis.
MEDLINE (1966–2001), EMBASE (1980–2001), PsycINFO (1967–2001), Current Contents (1995–2001), plus hand searches of reference lists and reviews.
Study selection and analysis:
Eligible studies were randomised controlled trials (RCTs) comparing pindolol plus SSRIs with placebo plus SSRIs in people with depression. RCTs examining early response (between 10 days and 2 weeks) and late response (3–6 weeks) were included. Studies involving non-SSRI antidepressants and crossover RCTs were excluded. Sensitivity and heterogeneity analyses were conducted.
Depressive symptoms: efficacy assessed by the number of participants responding to treatment (defined as a decrease of ⩾50%, or similar criterion, on the Hamilton Depression Rating Scale (HDRS) or Montgomery-Åsberg Depression Rating Scale (MADRAS)). Tolerability: proportion of the total study population not completing the study. Safety: proportion of total participants experiencing side effects.
Nine RCTs met inclusion criteria (594 participants).
Early response: five RCTs met inclusion criteria. Pindolol plus SSRIs significantly improved depressive symptoms compared with placebo plus SSRIs, for the first two weeks of treatment (OR 2.8, 95% CI 1.4 to 5.7; NNT = 6, 95% CI 4 to 20).
Late response: seven RCTs met inclusion criteria. There were no significant differences between groups after 3–6 weeks (OR 1.4, 95% CI 0.8 to 2.7). However, the late response studies were heterogeneous.
Tolerability and safety: there were no significant differences in tolerability or adverse events between groups (pindolol plus SSRIs v placebo plus SSRIs; OR for tolerability, 1.3, 95% CI 0.8 to 2.3; OR for adverse events, 1.3, 95% CI 0.7 to 2.1).
During the first two weeks of treatment, the addition of pindolol to an SSRI appears to increase response; however there was no evidence of improved efficacy beyond this period.
Amid all the current arguments about the mechanism and efficacy of antidepressants, virtually everyone can agree that there is a delay—the antidepressant “latency”—before the specific antidepressant effect is clear. It is common for the antidepressant effect to be delayed. The formulation of a rapidly acting antidepressant would be a major landmark. Manufacturers’ claims of rapid onset for their particular drug should be regarded with caution, as clinical trials are not usually designed with time of onset in mind.
The problem is complicated because we do not know why latency occurs. Selective serotonin reuptake inhibitors (SSRI) selectively inhibit serotonin reuptake by the pre-synaptic neuron immediately; but there is commonly a lag of 3–4 weeks before they exert an antidepressant effect. There have been several suggested mechanisms, but as yet no unifying theory.
In the mid-1990s, Artigas et al1 and Blier et al2 suggested that continued activity of presynaptic 5HT1A autoreceptors was responsible for antidepressant latency, and that specific blockade of these receptors by the β-antagonist pindolol could reduce this latency—and so it proved in open label studies. The key was of course to test the effect in randomised controlled studies.
Results have been mixed. Roughly speaking, in European hands (including our own3), the addition of pindolol from the outset of antidepressant (not only with SSRI4) treatment does indeed seem to accelerate the antidepressant response in many patients, especially if they have not been treated for depression before (the severity of the depression seems to be less important). However, some US studies5 have been less successful, as have studies looking at treatment resistance. So the true place of pindolol augmentation remains debated, with the honours so far in favour of the efficacy of the combination. This meta-analysis is a helpful synthesis of knowledge up to 2002.
The clinical advice would be that pindolol augmentation is worth trying in new patients, as long as β blockers are not otherwise contraindicated, and may lead to dramatic and rapid clinical improvements. There seems to be no point in giving pindolol for more than 2–3 weeks.
For correspondence: Javier Ballesteros, Department of Neuroscience, Medical Psychology and Psychiatry, University of the Basque Country, Barrio Sarriena, Leioa, Spain;
Sources of funding: none specified.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.