Investigating Risk Factors and Predicting Complications in Deep Brain Stimulation Surgery with Machine Learning Algorithms.

BACKGROUND: Deep brain stimulation (DBS) surgery is an option for patients experiencing medically resistant neurologic symptoms. DBS complications are rare; finding significant predictors requires a large number of surgeries. Machine learning algorithms may be used to effectively predict these outcomes. The aims of this study were to 1) investigate preoperative clinical risk factors and 2) build machine learning models to predict adverse outcomes. METHODS: This multicenter registry collected clinical and demographic characteristics of patients undergoing DBS surgery (n = 501) and tabulated occurrence of complications. Logistic regression was used to evaluate risk factors. Supervised learning algorithms were trained and validated on 70% and 30%, respectively, of both oversampled and original registry data. Performance was evaluated using area under the receiver operating characteristics curve (AUC), sensitivity, specificity, and accuracy. RESULTS: Logistic regression showed that the risk of complication was related to the operating institution in which the surgery was performed (odds ratio [OR] = 0.44, confidence interval [CI] = 0.25-0.78), body mass index (OR = 0.94, CI = 0.89-0.99), and diabetes (OR = 2.33, CI = 1.18-4.60). Patients with diabetes were almost 3× more likely to return to the operating room (OR = 2.78, CI = 1.31-5.88). Patients with a history of smoking were 4× more likely to experience postoperative infection (OR = 4.20, CI = 1.21-14.61). Supervised learning algorithms demonstrated high discrimination performance when predicting any complication (AUC = 0.86), a complication within 12 months (AUC = 0.91), return to the operating room (AUC = 0.88), and infection (AUC = 0.97). Age, body mass index, procedure side, gender, and a diagnosis of Parkinson disease were influential features. CONCLUSIONS: Multiple significant complication risk factors were identified, and supervised learning algorithms effectively predicted adverse outcomes in DBS surgery.

Deep Brain Stimulation Generator Replacement in End-Stage Parkinson Disease.

BACKGROUND: The decision to replace deep brain stimulation (DBS) generators in end-stage Parkinson disease (PD) is based on the patients' clinical status and the risks of this surgery. The infection rates of initial DBS implantation surgery and generator replacement surgery are well established. But the risks of DBS generator replacement for the severely disabled end-stage PD patient have not been described. METHODS: A retrospective analysis of all DBS procedures (n = 446) for PD spanning 20 years in a single institution was performed. The focus was on generator replacement procedure performed in end-stage PD. Infections related to DBS surgery were evaluated. RESULTS: Perioperative infections (<90 days) after generator replacements occurred in 0.6% of 172 cases and 2.5% of 232 primary lead implantation. Delayed infections (>90 days) occurred in 2.7% of all cases. Generator replacement was performed in 11 end-stage PD patients. None of these patients developed perioperative or delayed infections, and none were readmitted for medical or surgical complications. CONCLUSIONS: DBS generator replacement surgery is low risk, even in patients who have end-stage PD.

Pre-operative smoking history increases risk of infection in deep brain stimulation surgery.

Although general risk of deep brain stimulation (DBS) therapy has been previously described, application of risk prediction at the individual patient level is still largely at the discretion of a treating physician or a multidisciplinary team. To explore associations between potentially modifiable patient characteristics and common adverse events following DBS surgery, we retrospectively reviewed consecutive adult patients who had undergone new DBS electrode placement surgeries at two high-volume tertiary referral centers between October 1997 and May 2018. Among 501 patients included in the analysis (mean age (SD), 64.6 (10.4) years), 165 (32.9%) were female, 67 (13.4%) had diabetes, 231 (46.1%) had hypertension, 25 (5.0%) were smokers, 27 (5.4%) developed an infection, 15 (3.0%) had intracranial or intraventricular hemorrhage, and 53 (10.6%) had an unplanned return to the operating room. Patients who developed a surgical site infection were more likely to report history of smoking before DBS surgery (16% vs 5%, p = 0.04). There was a trend for patients with hypertension to be at risk for intracranial hemorrhage (p = 0.11). In conclusion, this multicenter study demonstrated an association between preoperative smoking and increased risk of infection following new DBS implantation surgery. Counseling about this risk should be considered in preoperative evaluation of patients who are considering undergoing a DBS procedure.

Postoperative delirium in Parkinson's disease patients following deep brain stimulation surgery.

Deep brain stimulation (DBS) surgery is an effective treatment for patients with advanced Parkinson's disease. Delirium in hospitalized Parkinson's disease patients is common and often leads to prolonged hospital stays. This study reports on the incidence and etiology of postoperative delirium following DBS surgery. Patients (n=59) with advanced Parkinson's disease underwent bilateral (n=56) or unilateral (n=3) DBS electrode implant surgery, followed 1 week later with surgical placement of DBS generators. The development of delirium during either hospital stay was evaluated retrospectively from the hospital chart. Potential causes of delirium were evaluated, including history of delirium, opiate equivalents, medication administration delays and missed doses during hospitalization, and Parkinson's disease duration. Delirium following implantation of DBS electrodes was common (22% of patients). It was less commonly associated with generator placement (10%). A history of delirium, age, and disease duration were positive predictors of delirium. Opiate equivalent doses were negatively correlated with delirium. Missed Parkinson's medication doses (53% of patients) and delayed administration (81% of patients) were common, and had a slight relation with delirium. Delirium was not related to complexity of medication regimen or use of dementia medications. Despite the presence of delirium most patients still only required a single night in the hospital post-surgery (67%). Prolonged hospital stay was due not only to delirium but also severe off states and other medical issues. Recognition and expectant management of delirium is best accomplished in a multidisciplinary setting, including the patient's family and nursing, pharmacy and neurological surgery staff.