To the Editor We have read the study by Roizenblatt and collaborators 1 and were delighted to notice heightened and ongoing interest in microsurgical education, as well as in microsurgical skill improvement and assessment. The influence of caffeine on microsurgical skill has been a topic of debate for almost 30 years in the scientific literature, including being mentioned by Robert D. Acland in Acland's Practice Manual for Microvascular Surgery from the 1980s.The authors define their study as a "single-blind crosssectional study." 1(p819) A more suitable definition is an n-of-1 randomized trial, with each ophthalmology fellow acting as their own control participant. Because this was an n-of-1 trial, days 1 and 2 represent treatment blocks A and B. In this respect, it is unclear to us why the authors 1 did not choose to perform a systematic counterbalanced design, such as ABBA. 2 Evaluating caffeine on day 1 might introduce bias through familiarity with the task on day 2. This makes drawing conclusions from the β-blocker-associated improving technique difficult. Furthermore, repeated measurements in a time series are often autocorrelated, so perhaps fitting a linear autoregressive model would have given a better estimation of the true effect size, while not ignoring the inherent features of time-series data. 3 In microsurgical simulation studies, adjusting for baseline skill is a daunting task, as is establishing predictive validity. All participants 1 had less than 2 years' experience, but given the limited sample size, confounder adjustment would have been necessary, since it is likely that caffeine affects individuals with more skill less than beginners. 4 A more pressing question, however, is whether the difference measured between the 2 groups, 1 if unbiased, is important enough clinically to warrant using a β-blocker in a healthy physician.We are firm believers that tremor, while inevitable, can be trained using a dedicated microsurgical program supervised by experienced microsurgical teachers. 5 This holds true even for anastomoses of vessels smaller than 0.5 mm or toe-to-thumb transfer in rats. Controlling tremor is as much a skill as it is a mindset, and it requires patience and continuous training. Selfadministering drugs without a medical indication should warrant a large-scale study to prove that β-blockers ultimately improve patient outcomes through better surgical skill. While the authors were diligent and avoided making treatment recommendations, the trial could create an impulse for those willing to try β-blockers as a means to control tremor. One should be wary of recommending drugs for novice trainees to support better technique, especially when proper training programs are available.