Prevalence and correlates of cognitive impairment in kidney transplant recipients

Adult kidney transplant recipients followed at the University of Kansas Kidney Transplant Clinic were included if they were i) English speaking; ii) without a history of acute stroke, concussion or traumatic brain injury within 2 months of testing and iii) with no acute illness at the time of participation. Patients were excluded if they i) had hearing or visual impairment; ii) were unable to read, write, speak or understand English; iii) had uncontrolled psychosis or seizure disorder; iv) were currently using antipsychotics or anti-epileptics v) received dual organ transplantation or vi) had a known history of cognitive impairment or dementia. Cognition was assessed during the clinic visit for post-transplant care. Eligible patients who attended the transplant clinic between May 2015 and June 2016 were approached. To minimize the acute effect of high dose steroids, surgery, and possible post-operative complications, cognitive assessments were performed only after the patient was at least 1 month post-transplant. As part of a standardized protocol, all patients were maintained on a mycophenolic acid compound and a calcineurin inhibitor with or without prednisone.

Cognition was assessed with the Montreal Cognitive Assessment (MoCA), a brief cognitive screening instrument (Additional file 1) [17]. The MoCA is a validated, clinic-based tool that samples from various domains of cognition and is sensitive in detecting mild cognitive impairment in several diseases, including Alzheimer’s disease and vascular dementia [18, 19]. We chose MoCA over the more commonly used mini-mental state exam (MMSE) for detecting cognitive impairment, as MoCA has more focus on executive function, a domain more commonly affected in kidney disease [20]. Moreover, MoCA has been successfully used in kidney disease, with a good correlation with detailed neuropsychological testing and better results than MMSE [21]. MoCA consists of a single page test with a maximum score of 30. The MoCA takes less than 10 min to complete and assesses seven domains of cognition: visuospatial/executive, naming, memory (delayed recall), attention, language, abstraction and orientation. The original English version 7.1 was used (http://​www.​mocatest.​org/​paper-tests/​moca-test-full/​). Based on published validation data on patients from a memory clinic, we defined cognitive impairment as MoCA score of less than 26 [17]. The MoCA was administered by medical assistants who underwent an hour of training that included detailed review of the online instructions along with practice sessions on mock patients. These medical assistants performed the test on the transplant recipients in a private examination room during the patients routine clinic visit to the transplant center.

Demographic and clinical variables that can potentially influence cognition such as age, gender, ethnicity, level of education, body mass index (BMI), blood pressure, history of smoking, diabetes, coronary artery disease, atrial fibrillation, transient ischemic attack (TIA) or stroke, cause for ESRD, serum hemoglobin, estimated glomerular filtration rate (eGFR), time on dialysis prior to transplant, and time after transplant were included in the analysis. Body mass index (BMI) and blood pressure were collected by medical assistants immediately before administration of MoCA. Demographic information and medical history were obtained through review of the patient’s electronic medical record. Diabetes was defined as current or past use of oral hypoglycemic agents or insulin. Coronary artery disease was defined as history of myocardial infarction, angina pectoris, angioplasty or coronary artery bypass surgery. Hypertension was defined by past or current use of antihypertensive agents. However, hypertension was not included in the final analysis as all patients were on one or more antihypertensive agent at some point in time. The eGFR calculated by the Modification of Diet in Renal Disease equation closest to the day of MOCA assessment (the same day in most cases) was collected.

The data collection was initiated as a quality improvement project and was approved by the Institutional Review Board. Data were collected on paper forms and managed in standard case report forms in REDCap (Research Electronic Data Capture) a web-based, electronic data capture tool hosted on a secure, password protected, HIPAA compliant server [22].

We compared baseline covariate values between patients with and without cognitive impairment using the t test for continuous variables (or rank sum test when t test assumptions where violated as indicated by residual analysis) and the Pearson’s chi square test for categorical variables (or Fisher’s exact test when expected cell counts were insufficient for Pearson’s chi square test). To ensure accuracy of data collection (MoCA scores as well as data collected from patients medical records), two authors reexamined all the data entered for 50 random patients. We calculated the mean and the standard deviation for continuous variables, and the frequencies and the proportions for categorical variables. We further analyzed the prevalence in patients categorized by five-year age groups. With the MoCA score being a continuous variable, we performed a multivariable linear regression analysis to study the association of these variables with cognitive performance. In model 1 of the multivariable regression analysis, we included all variables of interest that can affect cognition (age, gender, race, level of education, history of diabetes, coronary artery disease, smoking, stroke, and atrial fibrillation, BMI, serum hemoglobin, estimated glomerular filtration rate, systolic and diastolic blood pressure, time of dialysis, time since kidney transplant, and end stage renal disease secondary to diabetes). We used the adjusted R-square to determine model 2, in which we included selected variables from model 1 on the basis of their predictive relevance. The adjusted R-square criterion assesses the improvement in the amount of variation described relative to the number of parameters in the model. Residual analyses were conducted to assess the normality and constant variance assumptions of the multivariable linear regression models. We also assessed the adjusted associations of the variables selected for Model 2 with cognitive impairment defined as a MoCA score < 26. All tests were two-tailed. For all analyses, P < 0.05 was considered statistically significant. All analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, 2002–2012).

A total of 297 patients were approached for the MoCA test. Out of these, 22 patients were excluded from the analysis due to known history of mental retardation (five), stroke causing known cognitive deficits (two), hearing impairment (three), visual impairment (five), or inability to speak fluent English (seven). Eight patients refused to complete the assessment and two started the assessment, but did not complete it. After excluding the above 32 patients, 265 kidney transplant recipients completed the MoCA. Of these, 226 (>85%) had complete data and were included in the analyses. Examination of data for 50 random patients (total of 1250 data points) by two authors revealed high accuracy of data with a total of seven errors. As shown in Table 1, the mean age of the participants was 54.0 (SD 13.4) years. Participants were predominantly white, male, and with an education level of college or above. About half of the study participants were obese defined by a BMI ≥ 30 (50.9%). The mean eGFR was 52 ± 21 ml/min/1.73 m². The average time on dialysis prior to transplant was 2.3 years (SD 2.1) and the average time since transplant was 3.4 years (SD 4.1). Only 3% of the participants had a history of TIA or stroke and 21% had a history of coronary artery disease.

Fifty eight percent reached criteria for cognitive impairment. Table 1 compares the patients with and without cognitive impairment. The mean age of patients with cognitive impairment was higher than patients without cognitive impairment (p ≤ 0.01), and a lower proportion of female patients had cognitive impairment (p = 0.02). Race, level of education, BMI, systolic and diastolic blood pressure, eGFR, time since transplantation, time on dialysis prior to transplantation, and a history of smoking, coronary artery disease, atrial fibrillation, diabetes, or the cause of ESRD were not different among those with and without cognitive impairment.

In the bivariate analysis, higher MOCA scores were associated with lower age (p < 0.001), female gender (p < 0.001), lower serum hemoglobin levels (p = 0.015), and higher level of education (p = 0.003). There was no association of the MOCA score with race, BMI, blood pressure, eGFR, time since kidney transplantation, time on dialysis prior to transplantation, history of smoking, stroke, diabetes, coronary artery disease, atrial fibrillation, and ESRD secondary to diabetes. Multivariable linear regression demonstrated that higher MOCA score was associated with younger age, female gender, a higher level of education, and diabetic status (Table 2). Model 1 includes all variables of interest entered in the analysis. Model 2 includes variables selected on the basis of their predictive relevance based on the adjusted R-square (Additional file 2).

When considering the dichotomous outcome of cognitive impairment older age was associated with higher risk (Table 3). With the higher prevalence of cognitive impairment in older transplant recipients, we also analyzed the prevalence of cognitive impairment by five-year age groups (Fig. 1). These results demonstrate an increased prevalence of cognitive impairment with older age, but notably also a high prevalence of cognitive impairment even in individuals less than 50 years old.