Cognitive Impairment is Common and Unrecognized in Vascular Surgery Patients

Background: Patients' capacities to understand and act upon healthcare information is crucial to decision-making and high-quality care. Cognitive impairment (CI) has been associated with adverse outcomes across a range of diseases and surgeries. Despite the importance of CI, there is little to no information on its prevalence and severity in vascular surgery patients in the United States. We therefore conducted a prospective observational study to better characterize the prevalence and severity of CI in a contemporary vascular surgery practice.

Methods: We enrolled 111 outpatients attending a vascular surgery clinic using pragmatic consecutive sampling. Patients were excluded if they had a previous diagnosis of blindness, deafness, or dementia. Subjects completed a demographic survey and the Montreal Cognitive Assessment (MoCA), which was administered by a trained proctor. Chart review was used to assess comorbidities. The MoCA is a validated tool consisting of tasks such as clock drawing for assessing CI. It has a lower educational bias and higher sensitivity for detecting mild impairment compared to other examinations. The MoCA is scored from 0-30 based on an objective grading system. Scores between 0-9, 10-17, 18-25, and 26-30 indicate severe, moderate, mild, and no CI, respectively. Statistical analysis, including multivariable modeling, was performed using SAS (SAS Institute, Cary, NC).

Results: Of 163 patients, our analysis included 111 consecutive vascular patients who completed the MoCA. The average age of the entire cohort was 64.1 years, and 58.6% were male. The majority of the patients in the study were White (80.1%). The mean MoCA score of the entire cohort was 22.6 (mild CI). Of all subjects, 77% had CI: 68% with mild and 9% with moderate CI. Hypertension (P = 0.024), congestive heart failure (CHF) (P = 0.028), fewer years of education (P = 0.032), and Medicaid enrollment (P = 0.046) all had significant univariate associations with CI. There was no statistically significant difference between age (P = 0.11) or the primary vascular diagnosis disease for which the patient sought treatment and CI (P = 0.49). Multivariable models demonstrated that only CHF (odds ratio 3.8, P = 0.046) was statistically significantly associated with risk of CI.

Conclusions: In this first-time prospective study of the entire spectrum of vascular patients in the United States, we found that nearly 4 of every 5 vascular surgery patients have undiagnosed CI. Furthermore, we found that having CHF was associated with a higher likelihood of CI. Given the implications on consent, decision-making, and postoperative care, future work should focus on enrollment of a larger cohort along with an examination of the impact of CI on mortality, length of stay, and other outcomes.