Answers are grounded in this paper and cite their sources.
We read with great interest the randomised, double-blind, placebo-controlled trial (RCT) testing topical dorzolamide for macular holes, proposed by Lee et al. [1]: they found no efficacy of the medical topical carbonic anhydrase inhibitor (CAI) for the closure of idiopathic full-thickness macular hole (FTMH) versus placebo. We congratulate the authors on conducting this first RCT following the hydration theory with a non-invasive approach, thus limiting the bias of previous non-RCT studies. However, there remain some important points we would like the authors to clarify. Regarding Table 1, Lee et al. reported that the mean FTMH diameter on presentation that closed was significantly smaller than that which did not close. They suggested that one of the limitations was to focus on small FTMH, which may include cases that have closed spontaneously. We do not fully agree with this according to the IVTS classification [2]: Lee et al. selected FTMH < 400 μm and defined them as “small”, but the exact definition should have been ≤ 250 μm. We believe that this detail could have changed their patient selection and possibly the final results: regardless of the medical treatment (placebo or topical dorzolamide), the six closed FTMH were true “small” FTMH (≤ 250 μm). Furthermore, among the non-closed 26 FTMH, 10 eyes were “medium” FTMH (> 250–400 μm): the diameter decreased in only three of them; others were enlarged at 8 weeks. The other 16 remaining non-closed FTMH were true “small” FTMH, with the diameter decreasing in only three, stabilising in one, and otherwise enlarging at 8 weeks. Thus, only 22/32 FTMH (69%) were truly “small” in this RCT, decreasing its calculated power. Furthermore, Garg et al. [3] reported that the time to spontaneous closure was longer for medium FTMH (7 ± 9.6 months) than for small FTMH (7 ± 9.6 months). We believe that the primary endpoint at 8 weeks was questionable; an evaluation later at 12 or 16 weeks for anatomical outcomes might have been interesting, especially for medium FTMH. For example, patient 10 decreased from 185 to 32 μm at 8 weeks; if the protocol was pursued in this case, medical closure might have been possible later. We did not understand why patient 26 underwent surgery (Table 1) despite medical closure. Regarding Table 2, we regret that no details were provided about epiretinal membrane (ERM) especially location, profile, 1 or more epicentre(s): the authors could have specified in Table 1 which patient was involved and whether ERMs were present in 6 medically closed FTMH, to better understand the underlying mechanisms of closure. Only complete posterior vitreous detachment was reported in Table 2; it would have been interesting to add stage according to the historical Gass classification [4], even if the ITVS classification seems to be more discriminating in predicting MH closure [3]. The authors could also have specified the presence of vitreous macular traction (VMT) or not to optimise the analysis of the potential effect of topical CAI: VMT was found to be inversely associated with successful medical FTMH closure in a recent retrospective series [5]. The duration of FTMH before treatment could also be reported, as it is known it is associated with anatomic and functional outcomes. Cystoid macular edema (CME)—which was not an inclusion criterion—present in almost all included patients may have potentialised the dorzolamide effect on aquaporin-4. CME presented in Table 2 (32 eyes/100%) did not have the same absolute numbers/percentage as the data in Table 1 or in the text (30 eyes/93.8%; end page 3-Section 3.1), although this would not change the general conclusion. To conclude on the efficacy or not of topical CAI, some points could be optimised. Topical CAI should be further retested in an RCT, ideally including true small FTMH with CME but no ERM. The authors declare no conflicts of interest. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.
No takes yet. Share an insight, caveat, or question.
Abihaidar et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: