Does preoperative imaging for scalp non‐melanocytic skin cancer accurately predict invasion of the cranial vault? A systematic review and meta‐analysis

Abstract

AbstractPurposeThis study aimed to determine the diagnostic accuracy of CT and MRI in the preoperative detection of bone involvement for non‐melanoma skin cancers (NMSCs) located on the scalp. This study further aimed to evaluate the predictive value of these imaging modalities in determining the need for craniectomy and to identify gaps in the existing literature.MethodsElectronic searches of the MEDLINE, Embase, Cochrane and Google Scholar databases were performed for English language studies of any type. Studies reporting detection or exclusion of histopathologically confirmed bone involvement through preoperative imaging were identified according to PRISMA guidelines. Studies reporting dural involvement, non‐scalp tumours, and lacking tumour type(s) or outcome data were excluded. Outcomes were preoperative imaging result and histopathologically confirmed bone invasion. Meta‐analysis was performed and sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated (excluding case report and MRI data due to insufficient quality and quantity respectively).ResultsFour studies with a total of 69 patients were included in the final review, of which two studies totalling 66 patients were included in the meta‐analysis. Preoperative CT had a sensitivity of 38%, specificity of 98%, PPV of 90% and NPV of 73%.ConclusionsThe available data suggests that a preoperative CT finding of calvarial involvement by a scalp NMSC is likely to be real, but the absence of such a finding is unreliable. Current evidence suggests that preoperative imaging cannot exclude the necessity for craniectomy and future research is needed, particularly on the role of MRI.