Effects of proprioceptive neuromuscular facilitation technique on scapular dyskinesis in patients with subacute stroke

Dear Editor,

We read with great interest the article published by Rahman et al.^[1] in your esteemed journal. Investigating the “Effects of proprioceptive neuromuscular facilitation (PNF) on scapular dyskinesis in subacute stroke”, the authors found a few intriguing findings. The research concludes that scapular PNF significantly improves these patients’ shoulder discomfort and range of motion. Both the D1 flexion and D2 flexion groups demonstrated notable enhancements in range of motion, pain reduction, and decreased disability. Notably, D1 flexion showed superior improvement in shoulder rotation, whereas D2 flexion was more effective in reducing disability, although certain aspects require clarification.

There are several notable concerns raised about the study:

Clinical Trial Registration and CONSORT Flow Diagram: The study lacks a Clinical Trial Registration number or any link to assess the registry details and does not include a CONSORT flow diagram, both of which are essential.^[2]

Outcome Measure Standardization: During participant recruitment, it is recommended that a minimum value for the outcome measure be established to ensure consistency in evaluation and enhance study robustness.^[3]

Statistical Analysis: Using the Shapiro–Wilk test to assess data normality suggests that the descriptive statistics (mean and standard deviation) might not be appropriate. Median and range could be more suitable for not-normally distributed data. The Friedman test is used for the within-group analysis of the visual analog scale and shoulder range of motion, which is a non-parametric test. However, the data is presented in terms of mean and standard deviation. Ideally, it should be presented in terms of median and range. Furthermore, it is mentioned that the Shoulder Pain and Disability Index follows a normal distribution, so a parametric test should be used for within-group analysis, but instead, the Wilcoxon sign rank test is used, which is a non-parametric test, and data is also mentioned in terms of mean and standard deviation.^[4]

In summary, although the authors’ work is praiseworthy, attending to these specified concerns would notably improve the study’s precision and lucidity. Following established research protocols and including more specifics would aid readers and further scientific understanding in this area. We are optimistic that these recommendations will be considered, thus supporting the continuous advancement of this vital research.