Objectives: To examine the clinical effectiveness and cost-effectiveness of tonsillectomy/adeno-tonsillectomy in children aged 4-15 years with recurrent sore throats in comparison with standard non-surgical management.
Design: A pragmatic randomised controlled trial with economic analysis comparing surgical intervention with conventional medical treatment in children with recurrent sore throats (trial) and a parallel non-randomised cohort study (cohort study).
Setting: Five secondary care otolaryngology departments located in the north of England or west of Scotland.
Participants: 268 (trial: 131 allocated to surgical management; 137 allocated to medical management) and 461 (cohort study: 387 elected to have surgical management; 74 elected to have medical management) children aged between 4 and 15 years on their last birthday with recurrent sore throats. Participants were stratified by age (4-7 years, 8-11 years, 12-15 years).
Interventions: Treatment was tonsillectomy and adeno-tonsillectomy with adenoid curettage and tonsillectomy by dissection or bipolar diathermy according to surgical preference within 12 weeks of randomisation. The control was non-surgical conventional medical treatment only.
Main outcome measures: The primary clinical outcome was the reported number of episodes of sore throat in the 2 years after entry into the study. Secondary clinical outcomes included: the reported number of episodes of sore throat; number of sore throat-related GP consultations; reported number of symptom-free days; reported severity of sore throats; and surgical and anaesthetic morbidity. In addition to the measurement of these clinical outcomes, the impact of the treatment on costs and quality of life was assessed.
Results: Of the 1546 children assessed for eligibility, 817 were excluded (531 not meeting inclusion criteria, 286 refused) and 729 enrolled to the trial (268) or cohort study (461). The mean (standard deviation) episode of sore throats per month was in year 1 - cohort medical 0.59 (0.44), cohort surgical 0.71 (0.50), trial medical 0.64 (0.49), trial surgical 0.50 (0.43); and in year 2 - cohort medical 0.38 (0.34), cohort surgical 0.19 (0.36), trial medical 0.33 (0.43), trial surgical 0.13 (0.21). During both years of follow-up, children randomised to surgical management were less likely to record episodes of sore throat than those randomised to medical management; the incidence rate ratios in years 1 and 2 were 0.70 [95% confidence interval (CI) 0.61 to 0.80] and 0.54 (95% CI 0.42 to 0.70) respectively. The incremental cost-effectiveness ratio was estimated as 261 pounds per sore throat avoided (95% confidence interval 161 pounds to 586 pounds). Parents were willing to pay for the successful treatment of their child's recurrent sore throat (mean 8059 pounds). The estimated incremental cost per quality-adjusted life-year (QALY) ranged from 3129 pounds to 6904 pounds per QALY gained.
Conclusions: Children and parents exhibited strong preferences for the surgical management of recurrent sore throats. The health of all children with recurrent sore throat improves over time, but trial participants randomised to surgical management tended to experience better outcomes than those randomised to medical management. The limitations of the study due to poor response at follow-up support the continuing careful use of 'watchful waiting' and medical management in both primary and secondary care in line with current clinical guidelines until clear-cut evidence of clinical effectiveness and cost-effectiveness is available.
Trial registration: Current Controlled Trials ISRCTN47891548.