Apr 19 2007
All physicians involved in the treatment of PBS/IC understand that the placebo effect likely plays a major role in treatment, and is important when evaluating drug trials.
In an extremely important review article, Schagen van Leeuwen and colleagues from Amsterdam reviewed the placebo effect in the pharmacologic treatment of patients with lower urinary tract symptoms. Specifically they reviewed papers related to urinary incontinence, overactive bladder, and benign prostatic hyperplasia and compared placebo responses with those results reported for nonurologic disorders. Their literature review and their own conclusions highlight many issues that impact on PBS/IC trials, both those ongoing and those in the literature.
Placebo treatment of LUTS yields reductions in incontinence episodes ranging from 32% to 65%, whereas prostate or urge incontinence symptom scores are reduced by 9-34%. Genuine drugs decrease incontinence episodes by 45-77% and symptom scores by 22-45%. Placebo responses are much lower when objective changes in voided volume or peak flow rate are assessed.
A systematic review in 2001 found no significant placebo effect in randomized, controlled trials (RCT) with binary outcomes (nausea or no nausea) or continuous objective outcomes (serum cholesterol). A significant placebo effect was found in trials assessing continuous subjective outcomes (e.g. pain). The authors of this study report that more subjective outcomes used in LUTS treatment including self-reported number of incontinence episodes and symptom scores yield much higher placebo responses that objective measurements. Likewise, in the finasteride / terazosin / combination therapy trial for LUTS, the high placebo improvement in nocturia (23%) compared to 32% response on combination therapy, is consistent with the notion that a higher placebo response is present when assessing subjective outcomes, especially those that are continuous variables.
The authors discuss many putative placebo mechanisms that may be in play in RCTs. An undulating natural disease history may lead to improvement related simply to regression to the mean, as patients who are at their worst choose to enter a clinical trial and tend to improve to their mean symptom level. Placebo analgesia may play a role, and is mediated by endogenous opioid release to at least some extent, as it may be partially blocked by opioid antagonists. Placebo response may also be caused by activation of the reward circuitry, which involves the dopaminergic system. The anticipated clinical benefit induced by the placebo represents the reward, and dopaminergic neurons are activated. The importance of advertising is illustrated by the notion that a priori expectations about the effects of a drug may modify the patient's response to it.
Generally, disorders liable to subjective experience by the patient show much higher placebo responses than those involving objectively measurable parameters. Surgery outcomes are also prone to high placebo effects as is illustrated by the history of internal mammary ligation as the gold standard for alleviating angina until it was shown that sham surgery produced similar results. Placebo response tends to diminish as patients accrue prior experience with multiple treatments for their disorder.
This superb review points out many of the issues of placebo that impact BPS/IC trials and should be kept in mind by providers treating this difficult disorder.
van Leeuwen JHS, Castro R, Busse M, Bemelmans BLH
European Urology , 50:440-453, 2006
By Phillip M. Hanno, MD