A randomized trial comparing the cost-effectiveness of 2 approaches for treating unilateral nasolacrimal duct obstruction

Abstract

Objective: To compare the cost-effectiveness of 2 approaches for treating unilateral nasolacrimal duct obstruction (NLDO).

Methods: One hundred sixty-three infants aged 6 to less than 10 months with unilateral NLDO were randomly assigned to receive immediate office-based nasolacrimal duct probing (n = 82) or 6 months of observation/nonsurgical management (n = 81) followed by probing in a facility for persistent symptoms.

Main outcome measures: Treatment success was defined as the absence of clinical signs of NLDO (epiphora, increased tear lake, mucous discharge) on masked examination at age 18 months. Cost of treatment between randomization and age 18 months included costs for all surgical procedures and medications.

Results: In the observation/deferred facility-probing group, NLDO resolved within 6 months without surgery in 44 of the 67 patients (66%; 95% CI, 54% to 76%) who completed the 6-month visit. Twenty-two (27%) of the 81 patients in the observation/deferred facility-probing group underwent surgery, 4 of whom were operated on within the initial 6 months. At age 18 months, 69 of 75 patients (92%) in the immediate office-probing group were treatment successes, compared with 58 of 71 observation/deferred facility-probing group patients (82%) (10% difference in success; 95% CI, -1% to 21%). The mean cost of treatment was $562 in the immediate office-probing group compared with $701 in the observation/deferred facility-probing group (difference, -$139; 95% CI, -$377 to $94). The immediate office-probing group experienced 3.0 fewer months of symptoms (95% CI, -1.8 to -4.0).

Conclusions: The immediate office-probing approach is likely more cost-effective than observation followed by deferred facility probing if needed. Adoption of the immediate office-probing approach would result in probing in approximately two-thirds of infants whose obstruction would have resolved within 6 months of nonsurgical management, but would largely avoid the need for probing under general anesthesia.

Application to clinical practice: Although unilateral NLDO often resolves without surgery, immediate office probing is an effective and potentially cost-saving treatment option.

Trial registration: clinicaltrials.gov Identifier: NCT00780741.

Figures

Figure 1a. Data for Immediate Office Probing Group Patients (N=82)a

Prob of path = Probability of following specified pathway. Avg mos of sx = Average months of symptoms. Avg cost = Average cost of treatment between randomization and 18 months of age. aNote that Ns decrease over time due to 7 patients being lost to follow-up. bAll initial surgeries were office nasolacrimal duct probings. cAll seven reoperations occurred in a facility (ASC or HOPD)—one was a probing and six were intubations.

Figure 1b. Data for Observation/Deferred Facility Probing Group Patients (N=81)a

Prob of path = Probability of following specified pathway. Avg mos of sx = Average months of symptoms. Avg cost = Average cost of treatment between randomization and 18 months of age. aNote that Ns decrease over time due to 10 patients being lost to follow up. bOf the four surgeries performed within the first 6 months after randomization, 2 were office probing, 1 was facility intubation, and one was facility balloon catheter dilation. cOne patient appeared resolved at the 6 month visit but underwent intubation before 18 months of age. dOne deferred surgery was an office probing and one was balloon catheter dilation. eThe reoperation was a balloon catheter dilation. fFive patients in the observation/deferred facility probing group had clinical signs of NLDO at the 6-month visit but did not receive surgery (symptoms resolved before the scheduled surgery or the surgeon and/or parents felt surgery was not warranted).

Figure 2a. Cost-Effectiveness for Treatment Success at 18 Months of Age

The scatter plot indicates the extent of uncertainty in the observed cost effectiveness estimate. Each point represents the incremental cost effectiveness derived using one of the 10,000 bootstrap samples. The white ‘X’ indicates the observed incremental cost effectiveness for the treatment success outcome (the immediate office probing approach is 10% more successful and costs $139 less than the observation/deferred facility approach).

Figure 2b. Cost-Effectiveness for Months of NLDO Symptoms between Randomization and 18 Months of Age

The scatter plot indicates the extent of uncertainty in the observed cost effectiveness estimate. Each point represents the incremental cost effectiveness derived using one of the 10,000 bootstrap samples. The white ‘X’ indicates the observed incremental cost effectiveness for the months of symptoms outcome (the immediate office probing approach has 3.0 fewer months of symptoms and costs $139 less than the observation/deferred facility approach).