# A parsimonious nomogram for individualized prediction of 1-year functional outcome after STN-DBS in Parkinson’s disease: a single-center retrospective study

## TL;DR

This study developed a simple tool to predict 1-year functional outcomes after a specific brain surgery for Parkinson’s disease, using age, mental state, and post-surgery complications.

## Contribution

A new, simplified nomogram was developed for predicting 1-year functional outcomes after STN-DBS in Parkinson’s patients.

## Key findings

- Older age, lower MMSE scores, and postoperative electrolyte disorders were independent predictors of poor outcomes.
- The nomogram showed good discrimination (AUC 0.846) and calibration for predicting functional outcomes.
- Decision curve analysis confirmed the model’s clinical utility across a wide range of threshold probabilities.

## Abstract

In the context of Parkinson’s disease (PD), subthalamic nucleus deep brain stimulation (STN-DBS) has been shown to alleviate motor symptoms; nevertheless, functional outcomes at follow-up continue to be inconsistent. Patients who are at a higher risk of unsatisfactory functional recovery could be identified with the use of a personalized risk stratification tool, which would also provide information for perioperative treatment.

We retrospectively reviewed consecutive PD patients who underwent STN-DBS at the Department of Functional Neurosurgery, Liaocheng People’s Hospital, from January 1, 2015 to August 1, 2024, with 1-year follow-up. The outcome was defined using the medication-off Schwab and England Activities of Daily Living Scale (S&E) at 1 year: good outcome (S&E > 70) versus poor outcome (S&E ≤ 70). Candidate predictors were prespecified and collected from three domains: general clinical characteristics, perioperative indicators, and preoperative specialist assessments. Through the use of multivariable logistic regression, independent predictors of poor outcomes were found, and a nomogram was produced. Bootstrap resampling was utilized in order to carry out the internal validation process. The area under the receiver operating characteristic curve [AUC (C-index)] was used to quantify the discrimination of the model, the calibration was investigated using a calibration plot and the Hosmer–Lemeshow test, and the clinical utility was evaluated using decision curve analysis (DCA).

184 people were included in the study, with 109 having a positive outcome and 75 having a negative outcome. Out of the 195 patients that were eligible, 11 were lost to follow-up. Independently, worse outcome was linked with older age (odds ratio 1.08, 95% confidence interval 1.03–1.14), a lower score on the Mini-Mental State Examination (MMSE) (odds ratio 0.66, 95% confidence interval 0.56–0.79), and postoperative electrolyte disorder (odds ratio 2.97, 95% confidence interval 1.28–6.91). There was a low level of optimism on the internal validation (optimism-corrected C-index 0.841), despite the fact that the nomogram shown decent discrimination (AUC 0.846; bootstrap 95% confidence interval 0.781–0.905). In general, the calibration revealed that the projected risks and the observed hazards were in agreement. The results of the DCA indicated that the model had a positive net benefit across a wide range of threshold probabilities (0.20–0.99).

For the purpose of predicting the functional prognosis of a patient with Parkinson’s disease 1 year after receiving STN-DBS, we constructed a parsimonious nomogram that included age, MMSE, and postoperative electrolyte imbalance. In the event that our model is validated by an external source, it may be able to facilitate individualized perioperative risk assessment and collaborative decision-making.