Why This Article Matters
Summary
This randomized clinical trial evaluated the discontinuation phase of the Low-Concentration Atropine for Myopia Progression (LAMP) study over an eight-year period. The study included 246 children (mean age 13.47 years) who had completed five years of the initial trial and were receiving atropine treatment. Participants were randomized 1:1 into two groups: a taper group, which received 0.05% atropine for six months followed by 0.025% for six months, and a stop group, which received 0.05% for a full 12 months.
Both groups then discontinued all treatment for a two-year monitoring period. Over the three-year discontinuation evaluation (years 6 through 8), the taper group demonstrated significantly slower spherical equivalent progression (SEP) compared to the stop group (-0.54 D vs -0.78 D, P=.02) and less axial length (AL) elongation (0.33 mm vs 0.44 mm, P=.01). Furthermore, 65.1% of the taper group achieved a “good response” to discontinuation (defined as SEP progression ≥ -0.5 D over the final two years) compared to 42.6% in the stop group (P=.003). It was also noted that younger children and those with higher baseline myopia derived the most benefit from the tapering approach.
Commentary
Limitations
While the statistical models (GEE and multiple imputation) are structurally appropriate, the robustness of the findings is compromised by a high attrition rate. Only 180 of the 246 randomized participants (73.2%) completed the full 8-year follow-up. A 27% dropout rate introduces significant potential for nonrandom loss to follow-up. If participants with more aggressive myopia progression were more likely to drop out, the “missing at random” assumption underlying the authors' multiple imputation sensitivity analysis would be violated, potentially nullifying the modest observed treatment effect. Moreover, the actual observed treatment effect (0.24 D) fell short of the 0.40 D estimate used in the a priori sample size calculation, meaning the secondary outcomes and subgroup analyses may be underpowered.
Clinical relevance
The data supports tapering atropine concentration before full cessation to mitigate rebound receptor hypersensitivity, especially for younger, highly myopic patients. However, the absolute clinical relevance of a 0.24 D and 0.11 mm difference spread over three years is marginal and must be weighed against the prolonged burden of continuing daily drop administration.
Comparison with existing evidence
These findings contrast with recent comparative study analyses (such as the Changes in Axial Length and Refraction After Gradual Versus Prompt Cessation of Atropine 0.01% Treatment for Myopia Management by Erdinest et al.), which found no significant difference between tapering and stopping. The LAMP authors hypothesize this discrepancy exists because tapering may only be effective when stepping down from higher concentrations (like 0.05%) rather than from 0.01%.
Unanswered questions
Future research must establish the true Minimum Clinically Important Difference (MCID) for rebound myopia to determine if a 0.25 D preservation justifies a year-long taper. Additionally, studies should explore whether tapering frequency (e.g., every other day) is as effective as tapering concentration.
Key Take-Home Messages
- Tapering 0.05% atropine to 0.025% over a year prior to cessation results in statistically less myopia progression than abrupt discontinuation.
- The protective effect of tapering against rebound progression is most pronounced in younger children and those with more severe baseline myopia.
- While statistically significant, the absolute difference in progression between tapering and stopping is clinically modest (∼0.25 D over 3 years), necessitating individualized patient discussions regarding treatment burden.
- Older children with less severe myopia may tolerate abrupt cessation with a lower risk of significant rebound.
Conflict of Interest Statement
The author declares no conflict of interest related to this article.

Strengths
The study strengths include a prolonged eight-year longitudinal follow-up and a strictly randomized design for the discontinuation phase. The statistical methodology used Generalized Estimating Equations (GEEs) with robust standard errors to account for intereye correlations and repeated measures, while adjusting for critical covariates like age and baseline spherical equivalent.