gms | German Medical Science

Objective: Cranioplasty (CP) is a crucial procedure following decompressive craniectomy. Although CP is considered to be a “simple” neurosurgical procedure, it is performed extensively different depending on institutional conventions. So far, no reliable data on surgery-related complications after CP are available. Here, we report the first interim analysis of the German Cranial Reconstruction Registry (GCRR) with regard to early surgery-related complications.

Methods: All patients with a complete data set included in the GCRR, aprospective multicenter database, were analyzed. Early complications during the in-patient stay, medical history, mortality rates and neurological outcome at discharge, using the modified Rankin Scale (mRS), were evaluated. A multivariate logistic regression analysis was used to identify independent factors for the occurrence of complications.

Results: A total of 304 patients out of 351 registered in the GCRR were included. At least one complication during the initial hospital stay was declared in 75 patients (24.7%), a surgical revision was necessary in 27 patients (8.9%). A neurological deterioration was detected in 26 patients (8.5%), whereas an improvement was reported in 28 patients (9.2%). The surgery-related mortality rate was 1.3%. Most common complications were subdural or epidural hematoma (38 cases), seizures (13), new symptomatic hydrocephalus (7), CSF-fistula and wound dehiscence (each 5). Whether autologous bone or an artificial implant was used had no influence on the incidence of complications. The occurrence of any complication was associated with the duration of CP (132 vs 115 min; p=0.004), a worse preoperative mRS (3.4 vs. 2.7; p=0.001) and dural damage during surgery (46.6% in patients with complications vs. 32.9%; p=0.036). Additionally, a sinking-skin-flap-syndrome and an existing CSF-shunt were more frequently in patients with complications (24.0% vs 11.9%; p=0.011 and 16.0% vs. 8.0%; p=0.046). A multivariate regression analysis revealed duration of surgery (Odds ratio=OR=1.009; p=0.007), the preoperative mRS (OR=1.209; p=0.028) and a sinking-skin-flap (OR=2.287; p=0.029) as independent factors for early complications.

Conclusion: For the first time, we provide evidence-based multi-center results for this distinct patient group. Surgery related complications are common after CP and linked to specific pre-existing conditions and surgical skills. Upcoming analyses of the GCRR will further sharpen our understanding of this treatment.