Revisiting the Methodology and Implications of the Network Meta-Analysis on Dupuytren Disease Treatments: A Letter to the Editor

To the editor,

The recently published article by Nann et al, ¹ “Surgical Management of Dupuytren Disease: A Systematic Review and Network Meta-analyses,” drew our attention, and we commend the authors for performing an extensive network meta-analysis to discern the most effective treatment for Dupuytren disease. Nonetheless, we would like to highlight some concerns that would benefit from consideration.

First, one of the issues lies missing the inclusion of randomized controlled trial (RCT) of Kan et al. ² This study is similar to the included study by Selles et al, ³ comparing percutaneous aponeurotomy and lipofilling with limited fasciectomy. Another clinical trial by Zhou et al ⁴ compared collagenase clostridium histolyticum versus limited fasciectomy. Such an omission not only brings into question the thoroughness of the authors’ literature search but also potentially weakens the strength of evidence underpinning their findings. The lack of inclusion of this crucial study undermines the robustness of the data presented, limiting the comprehensive nature of the analysis.

For the second point of critique, we would like to know if the authors consider leveraging the GRADE (Grading of Recommendations, Assessment, Development, and Evaluations) framework to evaluate the quality of evidence across each treatment comparison. Although the Cochrane Risk of Bias tool and GRADE framework may seem analogous, they offer unique insights into different facets of study quality, and therefore, they are more complementary than substitutable. Predominantly, the Cochrane tool is employed to appraise the internal validity of RCTs, while the GRADE framework is versatile, enabling the evaluation of the aggregate strength of evidence across a spectrum of study types. ⁵ The utility of the GRADE approach extends beyond an appraisal of individual studies; it assesses the consistency, precision, directness, and likelihood of publication bias within the entire body of evidence. Thus, despite the authors’ application of the Cochrane tool, employing the GRADE framework could have fortified their analysis by providing a holistic, nuanced perspective on the overall reliability and quality of the evidence presented.

Lastly, the authors’ network meta-analysis could have benefited from additional elements that would enhance its quality such as the inclusion of ranking scores like the Surface Under the Cumulative RAnking curve (SUCRA). ⁶ These scores are instrumental in evaluating treatment effectiveness in network meta-analyses and aid readers in discerning the hierarchy of interventions based on efficacy. Moreover, the study could have benefited from a network plot, a key visual tool that illuminates connections between different treatments and their direct and indirect comparisons. Its presence illustrates full understanding of the interplay between treatments and the synthesis of the available evidence. ⁷

Despite the areas for improvement, the authors displayed a commendable level of methodological rigor by adhering to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and provided a high level of evidence study. Their results identified the long-term superiority of fasciectomy, marking an important contribution to our understanding of Dupuytren disease treatments. However, the need for further clinical trials remains, to corroborate and expand upon these findings by comparing diverse treatment modalities.