OBJECTIVES/HYPOTHESIS: To compare cartilage palisades with fascia grafting in reconstruction of the eardrum after surgery for sinus or tensa retraction cholesteatoma in children, with respect to long-term postoperative eardrum retraction and perforation, cholesteatoma recurrence, and hearing acuity. METHODS: A total of 64 children underwent surgery for either sinus or tensa retraction cholesteatoma during the period 1995 to 2000 (mean age 9 years, range 5-15). The eardrum was reconstructed using cartilage palisades in 32 children (32 ears) and fascia or perichondrium in 32 children (33 ears). The patients were followed for at least one year postoperatively and re-evaluated 4 years after surgery, and again recently at a mean of 10 years. The main outcome measures were postoperative drum retraction and perforation, cholesteatoma recurrence, and hearing acuity (pure tone average, speech reception threshold, and pure tone air-bone gap). RESULTS: All but two patients in both groups attended the 10-year follow-up examination (94% attendance). The mean overall follow-up period was 119 months (115 months in the palisade and 125 months in the fascia group). Total number of retractions during follow-up and at the 10-year examination was six (19%) for the palisade group and 14 (42%) for the fascia group (P = .03; chi-square test). The accumulated numbers for a perforation were four (13%) for the palisade group and seven (21%) for the fascia group (difference not significant). Two residual cholesteatomas, which are not related to the graft material, occurred in the palisade group (6%), whereas both recurrencies, which may be related to the graft material, occurred in the fascia group (6%). The hearing acuity for children operated on for a sinus cholesteatoma and for children with type III tympanoplasties was significantly better when cartilage palisade grafting had been employed. CONCLUSIONS: The cartilage palisade grafting technique appears superior with respect to prevention of long-term eardrum retraction. The occurrence of cholesteatoma recurrency and eardrum perforation seem to be independent of grafting material, although these results may be due to type 2 error (low number of ears). In sinus cholesteatoma surgery and in type III tympanoplasty, the long-term hearing results appear better when grafting cartilage palisades.