Douglas T Hutchinson1, Ajinkya A Rane2, Anthony Montanez3. 1. Department of Orthopaedic Surgery, University of Utah, Salt Lake City, UT. Electronic address: Douglas.Hutchinson@hsc.utah.edu. 2. Department of Orthopaedic Surgery, University of Utah, Salt Lake City, UT; Department of Orthopedics, Kaiser San Jose Medical Center, San Jose, CA. 3. The San Antonio Orthopaedic Group, San Antonio, TX.
Abstract
PURPOSE: Surgical release of pediatric trigger thumbs has been recommended as definitive treatment, although controversy exists over the natural history of pediatric trigger thumb. This study sought to evaluate the incidence of spontaneous resolution of pediatric trigger thumb and the factors that may influence resolution. METHODS: Pediatric patients were prospectively enrolled by a single surgeon from August 2009 to July 2015. All patients were initially treated with observation. They were followed annually and we collected pain scores (Parental visual analog scale), subjective dysfunction as perceived by parents, and physical examination information including the presence of flexion contracture of the thumb interphalangeal (IP) joint, thumb metacarpophalangeal joint laxity, and medial-lateral plane IP joint angular deformity. A competing risk framework was used to estimate the cumulative incidence at 5 years from the initial visit, and a subdistribution hazards model was used to compare patient characteristics with spontaneous resolution. Hazard ratios (HRs), 95% confidence intervals (95% CIs), and P values were reported. RESULTS: Seventy-eight patients (93 thumbs) with an average age of 20 months ± 1 year (mean ± SD) were enrolled at the first clinic visit and followed for 4.3 years (interquartile range, 3.1-5.5 years). At 5 years from the initial visit, 32% (95% CI, 20%-43%) of thumbs had resolved spontaneously, and 43% (95% CI, 30%-54%) had elected to proceed to surgery. Among those who had surgery, the median time to surgery was 4.1 years (interquartile range, 2.9-5.3 years). Bilateral thumb involvement increased the risk of surgery (subdistribution HR, 2.38; 95% CI, 1.23--4.6). Each degree increase in initial IP joint flexion decreased the occurrence of spontaneous resolution by 3% (subdistribution HR, 0.97; 95% CI, 0.94-0.99). Initial IP joint flexion 30° or less was associated with spontaneous resolution at 3 years (sensitivity, 0.73, 95% CI, 0.37-1.00; specificity, 0.70, 95% CI, 0.38-0.94; positive predictive value, 0.18, 95% CI, 0.13-0.41; negative predictive value, 0.76, 95% CI, 0.71-0.83; area under the curve, 0.78), whereas only 2.5% (95% CI, 0.4%-17%) of patients with an IP joint flexion greater than 30° resolved. CONCLUSIONS: A third of pediatric trigger thumbs resolved spontaneously, but most parents desired eventual surgical release. Patients with IP joint flexion contractures greater than 30° at baseline often lacked spontaneous resolution at 3 years and may be reasonable early surgical candidates. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic II.
PURPOSE: Surgical release of pediatric trigger thumbs has been recommended as definitive treatment, although controversy exists over the natural history of pediatric trigger thumb. This study sought to evaluate the incidence of spontaneous resolution of pediatric trigger thumb and the factors that may influence resolution. METHODS: Pediatric patients were prospectively enrolled by a single surgeon from August 2009 to July 2015. All patients were initially treated with observation. They were followed annually and we collected pain scores (Parental visual analog scale), subjective dysfunction as perceived by parents, and physical examination information including the presence of flexion contracture of the thumb interphalangeal (IP) joint, thumb metacarpophalangeal joint laxity, and medial-lateral plane IP joint angular deformity. A competing risk framework was used to estimate the cumulative incidence at 5 years from the initial visit, and a subdistribution hazards model was used to compare patient characteristics with spontaneous resolution. Hazard ratios (HRs), 95% confidence intervals (95% CIs), and P values were reported. RESULTS: Seventy-eight patients (93 thumbs) with an average age of 20 months ± 1 year (mean ± SD) were enrolled at the first clinic visit and followed for 4.3 years (interquartile range, 3.1-5.5 years). At 5 years from the initial visit, 32% (95% CI, 20%-43%) of thumbs had resolved spontaneously, and 43% (95% CI, 30%-54%) had elected to proceed to surgery. Among those who had surgery, the median time to surgery was 4.1 years (interquartile range, 2.9-5.3 years). Bilateral thumb involvement increased the risk of surgery (subdistribution HR, 2.38; 95% CI, 1.23--4.6). Each degree increase in initial IP joint flexion decreased the occurrence of spontaneous resolution by 3% (subdistribution HR, 0.97; 95% CI, 0.94-0.99). Initial IP joint flexion 30° or less was associated with spontaneous resolution at 3 years (sensitivity, 0.73, 95% CI, 0.37-1.00; specificity, 0.70, 95% CI, 0.38-0.94; positive predictive value, 0.18, 95% CI, 0.13-0.41; negative predictive value, 0.76, 95% CI, 0.71-0.83; area under the curve, 0.78), whereas only 2.5% (95% CI, 0.4%-17%) of patients with an IP joint flexion greater than 30° resolved. CONCLUSIONS: A third of pediatric trigger thumbs resolved spontaneously, but most parents desired eventual surgical release. Patients with IP joint flexion contractures greater than 30° at baseline often lacked spontaneous resolution at 3 years and may be reasonable early surgical candidates. TYPE OF STUDY/LEVEL OF EVIDENCE: Prognostic II.