Dear Editor,We have read the article entitled, "Phacoemulsification, visco-goniosynechialysis, and goniotomy in patients with primary angle-closure glaucoma: A comparative study" by Shokoohi-Rad et al. 1 We would like to congratulate the authors on their research and provide our own constructive insights.We have some concerns regarding the recruitment process. Although the patients were randomised, it is not clear how the patients were initially selected. Also, a large proportion of patients were excluded. In the "flowchart of patient selection", one-hundred-and-ten patients were assessed for eligibility and forty of these were excluded. Of these, twelve were excluded for "other reasons" unknown to the reader. This raises concerns over selection bias. Furthermore, there are seven patients who were lost to follow-up who were not included in the final analysis. An intention to treat analysis would have reduced attrition bias. 2 The authors claim that this study shows that phacoemulsification combined with visco-goniosynechialysis and goniotomy is superior to phacoemulsification with visco-goniosynechialysis only in lowering intra-ocular pressure (IOP). However, on review of the numbers, it appears that there is significant overlap of the standard deviations of the two groups. The statistically significant result should be questioned. Furthermore, despite having no power calculation, we can assume this is an under powered study with a sample size of just sixtythree patients split across the two groups. Also, there are six outliers within the visco-goniosynechialysis and goniotomy group, three with a marked IOP rise and one with very low IOP. It would be helpful to elaborate on those cases.Finally, we arrive at the issue of external validity. The patients in this trial were recruited from a single centre in Iran. The severity of glaucoma based on Hodapp or similar classifications of participants is not clear. We know they had an average cup to disc ratio of 0.65 and 0.68 in the goniotomy versus non-goniotomy group respectively. However, visual field test results are not disclosed. Can we assume that that the patient population is Iranian with moderate glaucoma?To conclude, before we can confidently use goniotomy, which has risks of hyphaema and hypotony as demonstrated in this study, we should acknowledge the pit-falls of this trial and adapt our clinical practice cautiously. A third arm undergoing phacoemulsification alone would have also been insightful.