Ophthalmic and Physiological Optics | 2021

Should a pooled analysis of FDA trials be considered representative for a population?

 
 

Abstract


To the Editor: We have read with interest the article by Wu et al., in which the authors have compared the risk of bestcorrected vision acuity (BCVA) loss associated with contact lens wear and laserassisted in situ keratomileusis (LASIK). The pooled analysis revealed that the risk of a ≥2 line BCVA loss following LASIK was equivalent to the risk of daily contact lens wear for 103 (95% CI: 103– 391) years. While it is useful to compare outcomes of LASIK to that of contact lenses, suggesting such a chasm between visual outcomes is inappropriate given the limitations of this study. To evaluate the total risk of BCVA loss following LASIK, the authors have identified randomised clinical trials published within the FDA database from 2003 to 2019. In our opinion, the pooled results of these investigations cannot be considered as representative for the population of patients undergoing LASIK. Several of these FDA studies have been conducted in patients with hyperopia; within the analyzed data, at least 36.7% of eyes (1,792 out of 4,882) were hyperopic. Importantly, the percentage of eyes with a loss of two or more BCVA lines is significantly higher in eyes treated for hyperopia than in those treated for myopia. In clinical practice, hyperopes might constitute only up to 20% of patients undergoing LASIK. In some particular groups e.g., in US Navy personnel undergoing LASIK, the proportion of hyperopes might be even lower; in the PROWL1 Study 10 out of 508 eyes (1.9%) undergoing LASIK were hyperopic, as were 14 out of 620 eyes (2.2%) in the PROWL2 Study. Moreover, microbial keratitis (MK) related to contact lens use is a more severe condition than corneal aberrations after hyperopic LASIK, while both might lead to a ≥2 line BCVA loss. It is also useful to note that the pooled studies used from the FDA database have less than 5,000 participants. Several significantly larger studies demonstrated a significantly lower risk of BCVA loss after LASIK. 7

Volume 41
Pages None
DOI 10.1111/opo.12884
Language English
Journal Ophthalmic and Physiological Optics

Full Text