Association of Postoperative Delirium and Parkinson Disease After Common United States Surgical Procedures

Abstract

IntroductionTo determine the association of Parkinson disease (PD) and postoperative delirium following common surgical procedures. MethodsWe performed a retrospective database analysis of the National Inpatient Sample. We used a matched sample of patients with and without PD who underwent any of ten common surgical procedures in the US, 2005-2014. Primary outcome measure was postoperative delirium for patients with and without PD. Secondary measures included disposition, length of stay, and hospital costs. ResultsThere were 3,235,866 patients receiving any of the ten most common operative procedures, 2005-2014. There were 35,743 patients with and without PD matched based on age, sex, elective admission status, Charlson Comorbidity index, and presence of dementia. Median age was 77 y (interquartile range 72-82), median Charlson Comorbidity index was 1 (standard deviation 0-2), 46.6% were female, and 46.8% were admitted electively. The three most common operative procedures were hip arthroplasty (28.5%), knee arthroplasty (16.1%), and percutaneous coronary angioplasty (14.9%). Postoperative delirium was present in 1519 patients with PD compared to 828 matched patients without PD (4.2% versus 2.3%; P < 0.001). The adjusted odds ratio of postoperative delirium for PD compared to the matched cohort without PD was 1.88 (95% confidence interval 1.73-2.05). Those undergoing spinal fusion (adjusted odds ratio 2.99, 95% confidence interval 2.06-4.38) had the greatest odds of delirium. For patients with PD, adjusted length of stay, adjusted hospital costs, and adjusted odds of postacute care facility discharge were greater compared to the matched cohort without PD. ConclusionsPatients with PD are more likely to develop postoperative delirium and have a more complicated postoperative course with longer length of stay and greater hospitalization costs.