Cognitive Function in Adults Aging with Fabry Disease: A Case–Control Feasibility Study Using Telephone-Based Assessments

Abstract

We examined the feasibility of recruiting US adults ≥45 years old with Fabry disease (FD) for telephone assessments of cognitive functioning. A case–control design matched each FD participant on age, sex, race, and education to four participants from a population-based study. Fifty-four participants with FD age 46–72 years were matched to 216 controls. Standardized cognitive assessments, quality of life (QOL), and medical histories were obtained by phone, supplemented by objective indices of comorbidities. Normalized scores on six cognitive tasks were calculated. On the individual tasks, scores on list recall and semantic fluency were significantly lower among FD participants (p-values < 0.05), while scores on the other four tasks did not differ. After averaging each participant’s normalized scores to form a cognitive composite, we examined group differences in composite scores, before and after adjusting for multiple covariates using generalized estimating equations. The composite scores of FD cases were marginally lower than controls before covariate adjustments (p = 0.08). QOL and mental health variables substantially attenuated this finding (p = 0.75), highlighting the influence of these factors on cognition in FD. Additional adjustment for cardiovascular comorbidities, kidney function, and stroke had negligible impact, despite higher prevalence in the FD sample. Telephone-based cognitive assessment methods are feasible among adults with FD, affording access to a geographically dispersed sample. Although decrements in discrete cognitive domains were observed, the overall cognitive function of older adults with FD was equivalent to that of well-matched controls before and after accounting for multiple confounding variables.

Competing interests: None declared

Appendices

Synopsis

In a relatively large sample of older adults with Fabry disease matched to controls on age, sex, race, and education, we demonstrated the feasibility of telephone assessments of cognitive function, accounted for important psychosocial and disease comorbidities, and found no substantial differences between groups in overall cognitive function based on six tasks.

Compliance with Ethics Guidelines

Conflict of Interest

Virginia Wadley has received an investigator-initiated award from Genzyme: a Sanofi Company. David Warnock has received investigator-initiated awards from Genzyme: a Sanofi Company and Amgen, Inc.; he is a consultant for Genzyme: a Sanofi Company and Amgen, Inc.

Robert Hopkin has received research funding from Genzyme: a Sanofi Company.

Dawn Laney has received investigator-initiated awards from Genzyme: a Sanofi Company, Amicus Therapeutics, and Shire Corporation.

Virginia Clarke has received a coinvestigator award from Genzyme: a Sanofi Company. Katherine Sims has received research funding from Genzyme: a Sanofi Company, Synageva, and Biogen Idec.

Leslie McClure, Caroline Lassen-Greene, Manjula Kurella Tamura, and George Howard declare that they have no conflict of interest.

Informed Consent

All procedures followed were in accordance with the ethical standards of the responsible committees on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000. Informed consent was obtained from all patients for being included in the study.

Author Contributions

Virginia Wadley designed the study, obtained funding, collected data, contributed to statistical analyses and interpretation, and drafted the manuscript.

Leslie McClure contributed to study design, randomly selected matched controls from the REGARDS study, created and merged the datasets, analyzed and interpreted the data, and critically revised the manuscript.

David Warnock contributed to the study conception and design, recruited participants, contributed to data interpretation, and critically revised the manuscript.

Caroline Lassen-Greene collected data, created databases, contributed to data interpretation, and critically revised the manuscript.

Robert Hopkin recruited participants, contributed to data interpretation, and critically revised the manuscript.

Dawn Laney recruited participants, contributed to data interpretation, and critically revised the manuscript.

Virginia Clarke recruited participants, contributed to data interpretation, and critically revised the manuscript.

Manjula Kurella Tamura contributed to data interpretation and critically revised the manuscript.

George Howard designed the REGARDS study, contributed to the design of the present study, contributed to statistical analyses and interpretation, and critically revised the manuscript.

Katherine Sims contributed to the study design, recruited participants, contributed to data interpretation, and critically revised the manuscript.

Appendix: Detailed Methods

Demographics

Matching variables of age, sex, race (black or white), and education level (less than high school, high school, some college, college graduate) were self-reported. Income also was self-reported (<$20 K, $20 K–$34 K, $35 K–$75 K, ≥$75 K, Refused).

Cognitive Function

Cognitive function was assessed with six tasks drawn from the National Institute of Neurological Disorders and Stroke–Canadian Stroke Network (NINDS-CSN) 5-minute battery (Hachinski et al. 2006) and the Consortium to Establish a Registry for Alzheimer’s Disease (CERAD) battery (Morris et al. 1989). The NINDS-CSN 5-minute battery was recommended for use in studies calling for very brief assessments, epidemiological studies, and/or telephone administration (Hachinski et al. 2006).

Learning and Memory

Learning and memory were assessed with word list learning (WLL) and word list delayed recall (WLDR) from the CERAD battery and NINDS-CSN 5-item recall and 6-item orientation (Hachinski et al. 2006; Morris et al. 1989; Nasreddine et al. 2005). WLL consists of three learning trials of a list of ten words which are presented in fixed orders that vary across the three trials, followed by a free recall trial (WLDR) after a 5-minute delay (Welsh et al. 1995). WLL and WLDR were administered according to standard protocols with minor modifications for telephone administration (Addison-Brown et al. 2014). Instructions and word lists were administered via a recording so that all participants were exposed to the same narrator. For WLL, correct responses on the three trials are summed. Scores range from 0 to 30. For WLDR, participants recall as many of the ten words as possible. Scores range from 0 to 10. Scores for NINDS-CSN 5-item recall and 6-item orientation from the Montreal Cognitive Assessment (Nasreddine et al. 2005) range from 0–5 and 0–6, respectively.

Executive Function

Executive function was assessed with letter fluency (letter F) from the NINDS-CSN battery and semantic fluency (animals) from the CERAD battery (Hachinski et al. 2006; Morris et al. 1989). Letter fluency and semantic fluency tests prompt participants to name as many words as they can beginning with the letter “F” in one minute and, subsequently, to name as many animals as they can in one minute. Scores on each consist of the total number of valid responses produced by each participant in 60 s. With explicit verbal permission, the assessments were recorded in digital waveform audio files and then played back later for scoring following standard scoring protocols.

Quality of Life

Health-related QOL was measured using a modified interviewer script version of the Short Form-12 (SF-12) survey (Ware et al. 1996; Quality Metrics). The survey yields Physical and Mental Component Summary scores (PCS-12 and MCS-12), each with a mean of 50 and SD of 10, facilitating comparisons to general population norms.

Perceived Stress

Perceived stress was measured with the 4-item Perceived Stress Scale (PSS-4; Cohen et al. 1983). This questionnaire elicits perceptions of stress during the past month. Each response is assigned a value of 0–4. Scores range from 0 (no stress) to 16 (high stress).

Depressive Symptoms

The Center for Epidemiologic Studies Depression Scale – 4-item version (CES-D-4; Melchior et al. 1993) – was used to evaluate depressive symptoms. Each of the 4 items assesses emotional symptoms of depression; no somatic symptoms are included in the scale. Each response is assigned a value of 0–3. Total scores range from 0 to 12; a score ≥4 suggests a clinically significant level of psychological distress (Melchior et al. 1993).

Cardiovascular Comorbidities

Cardiovascular comorbidities of hypertension, diabetes, dyslipidemia, and heart disease were assessed, as well as kidney function and history of stroke. A combination of self-report and objective assessments was used. In the FD participants, objective measures were drawn from Fabry Disease Registry information or medical records; for REGARDS participants, they were measured during the baseline home visit. Hypertension was defined as systolic blood pressure ≥140 mmHg, diastolic pressure ≥90 mmHg, documented or self-reported physician diagnosis, or use of antihypertensive medications. Diabetes was defined as fasting glucose >126 mg/dL, non-fasting glucose >200 mg/dL, documented or self-reported physician diagnosis, or use of diabetes medications. Dyslipidemia was defined as total cholesterol ≥240 mg/dL, low-density lipoprotein cholesterol ≥160 mg/dL, or high-density lipoprotein cholesterol <40 mg/dL; documented or self-reported physician diagnosis; or use of lipid-lowering medications. Heart disease was defined as presence of atrial fibrillation, left ventricular hypertrophy (LVH), coronary artery disease (CAD), or heart failure. Atrial fibrillation and LVH were defined by ECG or documented diagnosis. CAD was defined by ECG or documented or self-reported physician diagnosis of myocardial infarction or coronary revascularization. Heart failure was defined by self-reported or documented physician diagnosis of orthopnea or paroxysmal nocturnal dyspnea.

Kidney function was a continuous variable derived from blood assays. In REGARDS, serum creatinine was measured with isotope dilution mass spectrometry, and estimated glomerular filtration rate (eGFR) was calibrated using the Modification of Diet in Renal Disease equation. In FD participants, local laboratory values for eGFR were used, representing variable creatinine calibration methods.

For FD participants, stroke was defined as documented or self-reported physician diagnosis. In REGARDS, baseline history of stroke was defined by self-reported physician diagnosis, and reported incident strokes occurring during follow-up but prior to the initial cognitive battery assessment were confirmed by retrieval of medical records, which were adjudicated by REGARDS study physicians using the World Health Organization and clinical stroke criteria. For cases in which death occurred with no medical records surrounding the event, death certificates were examined, and proxy interviews were conducted for detection and adjudication of stroke events.