The Influence of Conflicts of Interest on Outcomes in the Lumbar Disc Arthroplasty Literature

Presented at SMISS Annual Forum 2018
By Dil Patel BS
With B. Woods , G. Schroeder , D. Kaye , C. Kepler , V. Hsu , G. Lee , M. Kurd , J. Rihn , G. Anderson , A. Hilibrand , A. Vaccaro , K. Radcliff ,

Disclosures: Dil Patel BS None B. Woods None, G. Schroeder None, D. Kaye None, C. Kepler None, V. Hsu None, G. Lee None, M. Kurd None, J. Rihn None, G. Anderson None, A. Hilibrand None, A. Vaccaro None, K. Radcliff None,

Introduction:

Lumbar disc arthroplasty (LDA) is an alternative to arthrodesis that affords symptom relief while preserving segmental motion. Investigators may have financial relationships with LDA device companies, which may play an important role on study outcomes.

Aims/Objectives:

This study aims to determine the association between study outcomes and conflicts of interest (COI) in the LDA literature.

Methods:

A systematic review was performed using the PubMed database to identify articles reporting clinical outcomes of LDA. Financial COI disclosed in the articles were recorded and confirmed through Open Payments reporting and Pro-Publica databases. Study outcomes were graded as favorable, unfavorable, or equivocal. Pearson chi-squared analysis determined the association between COI and study outcomes. Favorable outcomes were tested for an association with study characteristics using Poisson regression.

Results:

57 articles were included in this analysis, 30 (52.6%) had a financial COI, while 27 (47.4%) did not. 90% (n=27) of the conflicted studies disclosed their COI. Studies with United States authors were more likely to be conflicted (p=0.019). A majority of studies reported favorable outcomes for LDA (n=39, 68.4%). Conflicted studies were more likely to report favorable outcomes than non-conflicted studies (RR=1.6, p=0.020). Articles with COI related to consultant fees (RR=1.6, p=0.003), research funding (RR=1.6, p=0.002), and stock ownership (RR=1.5, p<0.001) were more likely to report favorable outcomes.

Conclusions:

A majority of articles report favorable outcomes for LDA, with conflicted studies more likely to report favorable outcomes than non-conflicted studies. Interestingly, United States authors are more likely to be conflicted compared to non-United States authors. In addition, financial factors that increase the likelihood of favorable outcomes by conflicted authors include consultant fees, research funding, and stock ownership. This study highlights the importance for COI reporting. As such, orthopedic surgeons should critically evaluate study outcomes with regard to potential conflicts before recommending LDA to their patients.