Dear editor

We are interested in the article entitled “Clinical Outcomes of Exercise Rehabilitation for Degenerative Tibial Meniscal Tears: A Systematic Review and Meta-Analysis of Randomized Controlled Trials”.¹ They conducted a systematic review and meta-analysis to integrate the evidence of randomized control trials to estimate the efficacy of exercise rehabilitation on pain, knee function, and muscle strength in patients with Degenerative Meniscal Lesions of the Tibia. The insights presented are valuable, especially given the increasing prevalence of meniscal injuries in aging populations individuals. However, there are still some points for consideration and improvement during the writing process.

Firstly, conducting a comprehensive literature search is an essential prerequisite for ensuring the reliability of a meta-analysis. The authors¹ reported in the results section an initial screening of 622 studies with the exclusion of 40 duplicate studies. Figure 1 indicates a discrepancy in the reported number of studies identified through database searches. The total should be 758 (12+15+80+591+60), rather than the stated 622, highlighting a clear error in the data. Additionally, Figure 1 indicates that 582 studies remained after excluding 40 duplicate records. However, it is noteworthy that the authors identified 7 additional studies from other sources, which were not included in the subsequent screening process. The error in Figure 1 from Mao et al has been outlined with a red box in Figure 1. Therefore, further verification by the authors is required.

Figure 1 The PRISMA flowchart from the study conducted by Mao et al.1 We indicated the data with errors using a red box.

Secondly, some of the included studies were sourced from the same clinical centers with overlapping time periods, which could lead to data duplication and affect the credibility of the analysis. For example, Østerås et al published two correlated articles in 2014,^2,3 conducted within the same time period. We reviewed the two studies and found that the subject recruitment process, inclusion criteria, exclusion criteria, and intervention protocols were highly consistent across both studies. The latter² is a subset of the former,³ so they should not be simultaneously included in the forest plots presented in Figures 3 and 4.¹ Furthermore, there is a minor error that needs to be rectified: The indicators label for Østerås² in Table 2 should be I, II, III, IV instead of I, II, III.¹ The year label for Østerås et al³ in Table 1 should be 2014 instead of 2012.¹

Finally, the quality assessment seems flawed. Berg et al published two correlated articles in 2020 and 2022,^4,5 both from the same clinical center, and the same ethics committee number (ref-no 2009/230). Berg et al clearly show that both articles were based on the same randomized controlled clinical trial (ClinicalTrials.gov ID: NCT01002794), which employed a computer-generated randomization sequence and opaque envelopes to ensure randomization concealment. However, the inconsistencies in the data presented in Table 4 are concerning and require further clarification.

In conclusion, we thank Mao et al for their contributions to this field. We look forward to the authors taking these mentioned points into consideration to enhance the rigor and reliability of their study findings.