Background/aims To determine if nasolacrimal massage or topical antibiotics are associated with higher rates of resolution compared with observation alone in a population-based cohort of infants with congenital nasolacrimal duct obstruction (CNLDO).

Methods The medical records of all children <5 years diagnosed with CNLDO while residing in Olmsted County, Minnesota from 1 January 1995 through 31 December 2004 were retrospectively reviewed for type of management and non-surgical resolution of tearing.

Results Among 1958 infants diagnosed and followed for CNLDO, 516 (26.4%) were merely observed, 506 (25.8%) were prescribed massage alone, 485 (24.8%) were prescribed at least one course of topical antibiotics, 397 (20.3%) were prescribed both topical antibiotics and massage, and 54 (2.8%) had no documented therapy. Non-surgical resolution, occurring in 1669 (85.2%) during a median follow-up of 3.1 months (range: 1 week–248 months), was 74.6% for the merely observed, 89.7% for those prescribed digital massage, 87.0% for those prescribed antibiotics and 90.7% for those treated with both. This comparison was significant in unadjusted (p<0.001) and multivariable comparisons (p<0.001).

Conclusion Prescribing topical antibiotics or digital massage for infants with CNLDO in this cohort, individually or in combination, was associated with a higher rate of spontaneous resolution than observation alone.