Comment on: “The Comprehensive Incidence and Risk Factors of Fracture in Kidney Transplant Recipients: A Meta‐Analysis”

We read with great interest the article by Jia et al., “The comprehensive incidence and risk factors of fracture in kidney transplant recipients: A meta-analysis,” published in Nephrology [1]. The authors have addressed a clinically significant issue by exploring the incidence and risk factors of fractures in kidney transplant recipients (KTRs), which are a growing concern given the high morbidity and mortality associated with such events. However, upon review, we have identified several methodological limitations and opportunities for improvement that could enhance the robustness and clinical applicability of the study's findings. One notable limitation of the study is the high degree of heterogeneity (I2 = 100%) observed in the pooled estimates of fracture incidence and risk factors. While the authors performed subgroup analyses based on geographic regions and publication years, these efforts alone may not sufficiently explain the variability across studies. The authors could have addressed this issue by conducting meta-regression analyses, which allow for a more nuanced investigation of potential sources of heterogeneity [2]. For example, study-level covariates such as follow-up duration, study design (retrospective vs. prospective), population characteristics (age, gender, and pre-existing comorbidities), or variations in immunosuppressive regimens (particularly steroid use) might have significantly influenced the reported outcomes. Meta-regression would have provided greater clarity on how these factors contribute to heterogeneity, offering more tailored insights into fracture risks among specific subgroups of KTRs. Another area for improvement lies in the statistical reporting. While the authors presented confidence intervals (CIs) for pooled estimates, the inclusion of prediction intervals (PIs) would have further strengthened the interpretation of their findings. Unlike CIs, which reflect the precision of pooled estimates, PIs provide the range of effect sizes expected in future similar studies, accounting for between-study variability [3]. This additional layer of analysis would have enhanced the clinical relevance of the findings, particularly given the high heterogeneity in fracture incidence across regions and time periods. The quality assessment of included studies was performed using the Newcastle-Ottawa Scale (NOS), which is a widely used tool for evaluating non-randomised studies. However, the authors could have complemented the NOS assessment with the GRADE (Grading of Recommendations Assessment, Development, and Evaluation) framework to provide a more comprehensive evaluation of the overall quality and strength of evidence. GRADE allows for the assessment of evidence across key domains such as risk of bias, inconsistency, indirectness, imprecision, and publication bias [4]. This approach would also have provided clinicians and researchers with clearer guidance on the reliability and applicability of the pooled estimates. Additionally, while publication bias was assessed using Begg's test, this method alone may not be sufficiently sensitive, especially when dealing with smaller sample sizes or highly heterogeneous studies. The authors could have used complementary methods such as Egger's test and funnel plot analysis to provide a more robust evaluation of publication bias. Furthermore, the application of the trim-and-fill method to other risk factors, beyond age, could have strengthened the credibility of the reported associations [5]. This meta-analysis highlights an important clinical issue; however, addressing these methodological limitations would improve the validity, transparency, and practical application of the findings. We commend the authors for their efforts in advancing the understanding of fractures in KTRs and hope these suggestions will support future research in this field. S.K., R.M., R.S., and A.N. critically provided comments on methodological aspects. S.K., A.N., and R.S. have written and edited the draft. The authors have nothing to report. The authors declare no conflicts of interest. Data sharing is not applicable to this article as no new data were created or analyzed in this study.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.