Page 36 - Clinical relevance of current materials for cranial implants

34
Chapter 2
Data extraction
The primary safety outcomes are defined as ‘infection’, ‘resorption’ and ‘removal’ rates.
From the selected articles the following parameters were extracted, if reported, on a pre-defined data extraction form: 1) study design; 2) number of patients; 3) number of cranioplasties; 4) cranioplasty material used; 5) sex; 6) age; 7) comorbidities; 8) smoking; 9) indication for cranioplasties; 10) previous cranioplasties; 11) location of the surgical intervention; 12) size of the defect in cm2; 13) use of antibiotics; 14) use of a drain; 15) involvement of the frontal sinus; 16) operation duration; 17) time interval between decompressive craniectomy and cranioplasty; 18) use of a mold, 3D printing or computer-aided design and computer-aided manufacturing; 19) all reported complications; 20) complication policy; 21) time interval between occurrence of complications and cranioplasty; 22) bacterial strain as cultured from the site of infection; 23) quality of life (QoL) measures; 24) esthetic outcome; 25) neurological functioning; 26) follow-up duration; 27) drop-outs. All data were extracted and verified by two authors (S.V. and T.M.) independently. Timing was recalculated to months according to the Gregorian calendar18.
Data analysis
Descriptive analysis of the included studies, outcome analysis, assessment of heterogeneity, and any subgroup analysis were performed using IBM SPSS v.24 (IBM Corp., Armonk, New York, USA).
If clinical heterogeneity was limited, a meta-analysis would be performed. Subgroup analysis was planned for each of the cranioplasty materials, and autologous versus alloplastic materials, regarding the primary outcomes. Differences in dichotomous outcomes are described as risk ratios (RRs), risk differences (RDs) and numbers needed to treat (NNTs), each with their 95% confidence intervals (95%CIs). Differences in continuous outcomes are reported as mean differences and their 95% CIs or medians with their interquartile ranges (IQRs). If the included studies are clinically heterogeneous, a range is provided of the outcomes for each of the cranioplasty materials.