IMPORTANCE: The optimal surgical management of cholesteatoma remains controversial. Within pediatric otolaryngology, one of the most vital points of contention is the selection of canal wall-up (CWU) versus canal wall-down (CWD) procedures. Pediatric cholesteatoma has high rates of recurrence (16%-54%). In adults, there is evidence that the selection of surgical techniques affects recurrence rates. This has not been shown in children.

OBJECTIVES: 1. To systematically review the literature on recurrent and residual cholesteatoma after CWU and CWD in children and perform a meta-analysis of the data. 2. To assess the rates of recurrent and residual cholesteatoma between CWU and CWD techniques in pediatric patients. 3. To assess hearing outcomes by evaluating postoperative differences in the air-bone gap (ABG) between CWU and CWD techniques.

DATA SOURCES: A systematic search of PubMed, Embase, Scopus, and Cochrane Collaboration was performed from inception to May 1st, 2020, to identify studies that compared CWU and CWD procedures for acquired cholesteatoma in children.

STUDY SELECTION: Search records were screened in duplicate by four reviewers. Inclusion criteria consisted of comparative randomized clinical trials and observational studies assessing outcomes of CWU and CWD techniques in the pediatric population. Studies involving patients with congenital cholesteatoma were excluded.

DATA EXTRACTION AND SYNTHESIS: Four reviewers working independently and in duplicate systematically reviewed and extracted study data. Dichotomous variables were analyzed as risk ratios (RR), while continuous variables were compared using weighted mean differences (MD). The risk of bias was assessed using the CLARITY Scale.

PRIMARY OUTCOMES AND MEASURES: The outcomes were recurrence, residual disease, air-bone gap (ABG), and air conductive (AC) thresholds.

RESULTS: After screening 1036 publications, 17 retrospective cohort studies were selected. 1333 children were included; the overall mean age was ten years (SD 7.9), and the overall mean follow-up time was 5.9 years (SD 6.6). CWU and CWD techniques were performed in 60% (796) and 40% (537) cases. We did not find differences in cholesteatoma recurrence (RR: 1.50, 95% CI 0.94; 2.40; n = 544; I2 0%; Tau [2]: 0.00), or rates of residual cholesteatoma (RR 1.51, 95% CI 0.96; 2.38, n = 506; I2: 0%; Tau [2]: 0.00) in patients who underwent CWU and CWD mastoidectomy. The mean air-bone gap was lower with CWU than CWD (mean difference: 7.60, 95% CI -10.65; -4.54; n = 242; I2: 71%; Tau [2]: 5.98).

CONCLUSION: and relevance: We show similar rates of recurrence and residual disease after either CWU or CWD tympanoplasty. Our results challenge the fundamental principle of CWD surgery as a standard technique, as there is no difference in rates of recurrence and residual disease in CWU and CWD. Moreover, audiometric results support CWU with improved hearing outcomes.

TRIAL REGISTRATION: PROSPERO identifier: CRD42020184029.