Diagnostic performance of digital markings on head computed tomography for evaluating intracranial hypertension in neonates and infants
摘要
This study aimed to investigate the diagnostic performance of digital markings appearing on head computed tomography (CT): CT digital markings to evaluate intracranial hypertension (ICH) in neonates and infants.
Materials and methodsWe retrospectively selected 270 consecutive pediatric patients (163 males and 107 females) aged 0–23 (8.0 ± 7.7) months, who underwent head CT at our institution from April 2009 to April 2023. Two board-certified observers evaluated the presence of CT digital markings, which are defined as focal depressions of inner table of the cranial bone, on CT images with reconstructed slice thickness ranging from 0.5 to 1 mm at the bone window setting. Generated volume rendering (VR) and ray-sum images were used as references. A board-certified pediatrician and ophthalmologist reviewed patients’ medical records and assessed the presence of ICH. Chi-square test was used to evaluate the relationship between CT digital markings and ICH in all patients, patients aged < 1 year, and 1 year old patients. A p value < 0.05 was considered as statistically significant. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were also calculated, according to each comparison.
ResultsA total of 47 patients had ICH, and 128 patients demonstrated digital markings. There were no significant correlations between CT digital markings and ICH in all patients, patients aged < 1 year, or patients aged 1 year (p = 0.63, 0.38, and > 0.99, respectively). Sensitivity, specificity, PPV, and NPV were 0.51, 0.54, 0.19, and 0.84 for all patients; 0.51, 0.56, 0.24, and 0.81 for patients aged < 1 year old; and 0.50, 0.48, 0.08, and 0.92 for patients aged 1 year old, respectively.
ConclusionDigital markings on head CT observed in neonates and infants are not associated with ICH and thus not clinically useful for the diagnosis of ICH.