External validation of prognostic models predicting outcome after chronic subdural hematoma

Abstract

Background Several prognostic models for outcomes after chronic subdural hematoma (CSDH) treatment have been published in recent years. However, these models are not sufficiently validated for use in daily clinical practice. We aimed to assess the performance of existing prediction models for outcomes in patients diagnosed with CSDH. Methods We systematically searched relevant literature databases up to February 2021 to identify prognostic models for outcome prediction in patients diagnosed with CSDH. For the external validation of prognostic models, we used a retrospective database, containing data of 2384 patients from three Dutch regions. Prognostic models were included if they predicted either mortality, hematoma recurrence, functional outcome, or quality of life. Models were excluded when predictors were absent in our database or available for < 150 patients in our database. We assessed calibration, and discrimination (quantified by the concordance index C) of the included prognostic models in our retrospective database. Results We identified 1680 original publications of which 1656 were excluded based on title or abstract, mostly because they did not concern CSDH or did not define a prognostic model. Out of 18 identified models, three could be externally validated in our retrospective database: a model for 30-day mortality in 1656 patients, a model for 2 months, and another for 3-month hematoma recurrence both in 1733 patients. The models overestimated the proportion of patients with these outcomes by 11% (15% predicted vs. 4% observed), 1% (10% vs. 9%), and 2% (11% vs. 9%), respectively. Their discriminative ability was poor to modest (C of 0.70 [0.63–0.77]; 0.46 [0.35–0.56]; 0.59 [0.51–0.66], respectively). Conclusions None of the examined models showed good predictive performance for outcomes after CSDH treatment in our dataset. This study confirms the difficulty in predicting outcomes after CSDH and emphasizes the heterogeneity of CSDH patients. The importance of developing high-quality models by using unified predictors and relevant outcome measures and appropriate modeling strategies is warranted.

Synthetic minority over-sampling technique-enhanced machine learning models for predicting recurrence of postoperative chronic subdural hematoma

“…In recent years, several models predicting the recurrence of CSDH after surgery have been published. To evaluate the performance of these existing models in predicting postoperative recurrence in CSDH patients, Holl et al ( 13 ) utilized a retrospective database comprising data from 2,384 patients across three regions in the Netherlands. The study revealed that current predictive models perform poorly on the author’s dataset, failing to effectively forecast the recurrence of hematomas following CSDH treatment.…”

A Linear Regression Equation for Predicting the Residual Volume of Chronic Subdural Hematoma 1 Week After Surgery

“…[4][5][6][7] These models attempted to predict outcome measures of CSDH such as mortality, risk of recurrence, and functional outcome after surgical treatment. 2 Population differences and differences in the distribution of predictors between development and validation studies may affect model performance in the validation setting. 2 The factors that affect prognosis may vary greatly at different follow-up times.…”

“…2 Population differences and differences in the distribution of predictors between development and validation studies may affect model performance in the validation setting. 2 The factors that affect prognosis may vary greatly at different follow-up times. Due to drain removal, a repeat head computed tomography (CT) 1 week after the burr hole and drainage tended to show an increased residual cavity volume.…”

“…In an effort to evaluate the efficacy of published prognostic models for predicting patient outcome with unilateral CSDH, Holl et al conducted a retrospective analysis of a database from three regions in the Netherlands. 2,[4][5][6] However, none of the models demonstrated significant discriminative ability or calibration. [4][5][6] In particular, Alford and Jack's model dichotomized continuous predictors, which can lead to information loss and limited information applicability to future studies.…”

“…[4][5][6] In particular, Alford and Jack's model dichotomized continuous predictors, which can lead to information loss and limited information applicability to future studies. 2,[4][5][6] To avoid such shortcomings, this study opted to utilize linear regression analysis to establish the relationship between continuous and inverse variables and weeks of residual cavity after surgery, without grouping them together.…”