The prevalence and impact of undiagnosed cognitive impairment in older vascular surgical patients

Judith S L Partridge,Jugdeep K Dhesi,Jason D Cross,Jessica W Lo,Peter R Taylor,Rachel Bell,Finbarr C Martin,Danielle Harari

DOI: https://doi.org/10.1016/j.jvs.2014.04.041

Abstract:Objective: The objectives of this observational cohort study were to investigate the prevalence of undiagnosed cognitive impairment in older patients presenting for vascular surgery, to examine its association with adverse postoperative outcomes, and to test the feasibility of a preoperative cognitive assessment tool. Methods: Patients aged 60 years or older were recruited by consent on admission to the vascular surgical ward of an inner-city teaching hospital with a large tertiary referral practice for proposed elective or emergency aortic or lower limb arterial intervention. Cognition was assessed preoperatively by the Montreal Cognitive Assessment (MoCA), and a score below 24/30 indicated cognitive impairment or dementia. The mean length of time taken to complete the assessment was recorded. Baseline characteristics (medical multimorbidity, frailty, and laboratory tests), hospital length of stay (LOS), and postoperative complications were documented. Results: Preoperative MoCA was completed in 114 patients with a mean age of 76.3 years (standard deviation, 7.36 years); 67.5% were men, and 55.3% of procedures were elective. The MoCA was completed in 100% of patients and was quick and acceptable to patients in this setting. Cognitive impairment or dementia was found in 68% of patients (77 of 114) and was previously unrecognized in 88.3% of patients (68 of 77). Therefore, 60.5% of patients (68 of 114) aged 60 years or older presenting for vascular surgery had previously undiagnosed cognitive impairment. MoCA <24 was univariately associated with pre-existing frailty (Edmonton Frail Scale [EFS] score ≥6.5) and longer LOS (≥12 days). In logistic regression modeling, MoCA <24 was strongly independently associated with frailty EFS score ≥6.5 (odds ratio, 12.55; P < .001). By use of the area under the receiver operating characteristic curve (AUC), MoCA <24 was predictive of longer LOS of ≥12 days (AUC, 0.621; P = .049). The strength of predictive power increased with the addition of frailty (EFS score ≥6.5) to the models (AUC, 0.695; P = .002). Conclusions: The prevalence of cognitive impairment among older patients presenting for vascular surgery is high and frequently undiagnosed before admission. It is feasible to use the MoCA to identify cognitive impairment in this high-risk surgical group preoperatively. The combined assessment of frailty and cognition is predictive of adverse postoperative outcomes and longer LOS.