Introduction and objectives:
Photoselective vaporization of the prostate (PVP) with the so-called 532nm greenlight laser is an accepted treatment modality for non-neurogenic lower urinary tract symptoms secondary to prostate enlargement. Conflicts of interest (COIs) and industrial sponsorship (IS) have been shown to have a significant impact on the favourability of study results. The aim of the current study was to evaluate outcomes of comparative studies on PVP as a function of COIs and IS.
Methods:
MEDLINE and EMBASE were systematically searched for records in English language. Comparative studies (randomized controlled trials [RCTs] and non-randomized comparative studies [NRCSs]), in which PVP was one treatment modality, were considered eligible. Sponsorship assessment distinguished between IS and non-industrial sponsoring. Two reviewers screened all abstracts and full-text articles independently. Disagreement was resolved either by discussion or by reference to a third independent reviewer. Favorability of outcome was evaluated on a binary scale (PVP-favourable versus PVP-unfavourable) by two independent board-certified urologists external to the remaining study team. Descriptive statistics were used for data analysis.
Results:
In total, 749 records were screened after manual deduplication. Of these, 286 articles were eligible for full-text screening. Full-text analysis identified 65 studies (25 RCTs [38%], 40 NRCSs [62%]). The majority of the studies mentioned the absence/presence of potential COIs (78%). In contrast, a sponsorship statement was only found in 29% of the investigations. Our analysis identified 56 conflicted authors. In 24 (37%) studies at least one COI was declared (range 1 to 18). Interestingly, searching the author lists of all included studies a second time for more occurrences of the 56 conflicted authors led to the identification 36 initially undeclared COIs in a total of 6 studies. Fifty-six (86%) and 9 (14%) of all included studies were rated as PVP-favourable and PVP-unfavourable, respectively. Among PVP-favourable studies, we could identify both a higher proportion of COI (39% vs. 32%) and IS (7% vs. 0%).
Conclusions:
Within the field of comparative PVP literature, COIs are not only highly prevalent but also more frequent in studies reporting PVP-favourable outcomes. IS was exclusively found in PVP-favourable studies. A majority of all RCTs and NRCSs on PVP mention the absence/presence of potential COIs. However, a sponsorship statement was identifiable in only about one third of all studies.
