Preservation of bone flap after craniotomy infection.

INTRODUCTION: The estimated incidence of craniotomy infection is 5%, ranging from 1-11% depending on the presence of certain risk factors, such as, prior radiation therapy, repeated surgery, CSF leak, duration of surgery over 4h, interventions involving nasal sinuses and emergency surgeries. The standard treatment for infected craniotomies is bone flap discarding and delayed cranioplasty. Adequate cosmetic results, unprotected brain and disfiguring deformity until cranioplasty are controversial features following bone removal. We present a limited series of five patients with craniotomy infection, that were successfully treated with wound debridement, in situ bone sterilization, reposition of the bone flap and antibiotic irrigation through a wash-in and wash-out draining system, all in the same surgical procedure. All infections cleared and every patient saved his/her bone flap.

PATIENTS AND METHODS: We retrospectively reviewed the records of 5 patients with craniotomy infection that presented with wound swelling, purulent discharge and fever. The operative technique consisted on three manoeuvres: wound debridement, bone flap sterilization (either autoclaved or soaked in a sterilizing solution), and insertion of subgaleal/epidural drains for non-continuous antibiotic irrigation (vancomycin 50mg in 20cc of saline every 12h alternating with cephotaxime 100mg in 20cc of saline every 12h). Also, patients received equal systemic endovenous antibiotherapy and oral antibiotics after discharge, until complete resolution of infection and wound healing.

RESULTS: Patients in the series (2 women and 3 men) ranged in age from 36 to 77. No patient had received prior radiation therapy and only one had undergone surgery involving nasal sinuses. The initial operations correspond to craniotomies performed for two intracranial tumours (meningiomas), one arteriovenous malformation and two decompressive craniotomies (haemorrhagic contusions and acute subdural haematoma). The duration of surgeries ranged from 1h30' to 5h30', only two operations extending over 4 hours. The interval between the initial surgery and the reintervention ranged from 11 to 227 days. Staphyloccocus spp were cultured in all patients. For bone sterilization povidone scrubbing was used in all patients, autoclave in two and soaking the flap in a sterilizing solution in three. All patients cleared infection and achieved complete wound healing in 2-3 weeks after the re-operation. Follow up ranged from 4 to 18 months. One patient died as a consequence of sepsis in the context of pneumonia some weeks after wound healing.

DISCUSSION: Recent multivariate analyses have demonstrated that the presence of a CSF leak and the performance of repeated operations are the most important independent risk factors for craniotomy infection, with associated odds ratios for infection as high as 145 and 7, respectively. Regular antibiotic administration at anaesthesia induction seems to decrease the rate of craniotomy infection by half, both in the entire population and in low-risk subsets. Organisms involved in craniotomy infections are common pathogens usually contaminating neurosurgical procedures or normal skin flora germs. Auguste and McDermott have recently presented a case series of 12 patients in which successful salvage procedures for infected craniotomy bone flaps were performed using a continuous wash-in, wash-out indwelling antibiotic irrigation system, that needed close observation of the neurological status since obstruction of the outflow system could precipitate brain herniation. The method we present is as effective as theirs and avoids such complication since only small quantities of antibiotic solutions (20 cc) are instilled during each dose administration.