The Timing of Alloplastic Cranioplasty in the Setting of Previous Osteomyelitis

BACKGROUND:Management of cranial osteomyelitis is challenging and often includes débridement of infected bone and delayed alloplastic cranioplasty. However, the optimal interval between the removal of infected bone and definitive reconstruction remains controversial. The authors investigated the optimal time for definitive reconstruction and factors influencing cranioplasty reinfection. METHODS:A retrospective review of 111 alloplastic cranioplasties for osteomyelitis between 2002 and 2015 was performed. Patients were divided into four subgroups based on timing of reconstructiongroup 1, less than 3 months; group 2, 3 to 6 months; group 3, 6 to 12 months; and group 4, more than 12 months. Multivariate logistic regression was used to calculate the probability of cranioplasty reinfection based on risk factors. Median follow-up was 45.9 months (range, 12.4 to 136.9 months). RESULTS:The combined reinfection rate was 23.4 percent. The reinfection rate in group 1 was 39.6 percent; group 2, 12.5 percent; group 3, 8.0 percent; and group 4, 0.0 percent (p < 0.001). The mean interval between the infected bone removal and cranioplasty was shorter in patients with reinfection than in patients without reinfection (2.2 ± 3.9 months versus 6.1 ± 8.3 months; p < 0.001). The strongest independent predictors of reinfection were chemotherapy (OR, 10.1; 95 percent CI, 2.9 to 35.2), composite defect requiring scalp reconstruction at the time of cranioplasty (OR, 3.3; 95 percent CI, 1.2 to 8.9), and early reconstruction. Each month of delay in reconstruction reduced the reinfection rate by 10 percent (OR, 0.9 per each month of delay; 95 percent CI, 0.8 to 1.0). Cranioplasty material was not significant. CONCLUSIONS:Early alloplastic cranioplasty following osteomyelitis carries an unacceptably high risk of reinfection. This risk decreases by 10 percent with each month of delay. The authors’ regression model can be used to predict the probability of reinfection for all time periods. CLINICAL QUESTION/LEVEL OF EVIDENCE:Therapeutic, III.