Related Experiment Video
Updated: Aug 14, 2026
A Bedside, Single Burr Hole Approach to Multimodality Monitoring in Severe Brain Injury
Published on: March 26, 2019
Derivation and internal validation of a bedside risk-stratification score for complications after cranioplasty
Prashanta Chihnara1, Suchanda Bhattacharjee1, K Ramnadha Reddy1
1Department of Neurosurgery, Nizam's Institute of Medical Sciences (NIMS), Panjagutta, Hyderabad 500 082, Telangana, India.
Background:
Cranioplasty (CP) following decompressive craniectomy (DC) carries substantial morbidity, but evidence-based risk-stratification tools for individual-patient counselling are lacking, particularly in South Asian populations. We aimed to derive and internally validate a clinical risk-stratification score for postoperative complications following CP in a large prospective Indian cohort.
Methods:
Single-centre prospective observational cohort of 486 adults who underwent elective CP after DC (January 2020 - December 2023). The primary outcome was any complication within 12 months. A pre-specified multivariable binary logistic regression model estimated adjusted odds ratios (aOR); internal validation used 1,000-iteration bootstrapping with optimism correction. A point-based risk score was derived from the risk-increasing model coefficients. Discrimination was assessed by the c-statistic and calibration by the Hosmer-Lemeshow test and calibration plot.
Results:
Of 724 patients screened, 486 met eligibility criteria. Mean age was 32.8 ± 11.4 years; 76.5 % were male. The 12-month complication rate was 33.3 % (162/486). On multivariable analysis, female sex (aOR 2.51, 95 % CI 1.57-4.00), small calvarial defect (aOR 1.71, 1.08-2.70), bone-flap-plus-mesh implant (aOR 3.60, 1.89-6.84), and mesh-only implant (aOR 5.06, 1.87-13.71) were independent predictors, whereas polymethylmethacrylate was protective (aOR 0.12, 0.02-0.92). The model showed acceptable discrimination (apparent c-statistic 0.720; optimism-corrected 0.693) and good calibration (Hosmer-Lemeshow χ2 = 2.63, p = 0.96). Integer scores stratified patients into low- (n = 144; 18.1 %), intermediate- (n = 234; 32.9 %), and high-risk (n = 108; 54.6 %) bands (χ2 = 37.2, p < 0.001).
Conclusions:
In the largest prospective Indian CP series to date, an integer-scored bedside risk-stratification index derived from sex, defect size, and implant material discriminates patients into clinically meaningful risk bands with good calibration. External validation in independent multicentre cohorts is the necessary next step.
