A longitudinal study on quality of life along the spectrum of Alzheimer’s disease

Participants

In this longitudinal study, we included n = 1140 patients from the Amsterdam Dementia Cohort (ADC). These included n = 105 amyloid-positive SCD, n = 342 amyloid-negative SCD, n = 144 amyloid-positive MCI, n = 132 amyloid-negative MCI, and n = 417 amyloid-positive dementia patients. All patients presented with complaints at the memory clinic of the Alzheimer center Amsterdam had their baseline visit between 2009 and 2020. Inclusion criteria were (1) a baseline diagnosis of Alzheimer’s disease dementia (AD), mild cognitive impairment (MCI), or subjective cognitive decline (SCD), (2) availability of amyloid PET and/or CSF biomarkers, and (3) availability of EQ-5D or VAS data. The study was approved by the Medical Ethics Review Committee of the VU University Medical Center. All patients provided written informed consent for the use of their medical data for research propose.

All participants presented as patients at the memory clinic of the Alzheimer center Amsterdam, where they received a standardized dementia diagnostic work-up, which consisted of medical history, neurological, physical and neuropsychological evaluation, MRI, laboratory tests, and lumbar puncture [11, 12]. Subsequently, clinical diagnosis (i.e., SCD, MCI or AD dementia) was made in a multi-disciplinary meeting. Patients were diagnosed with AD dementia or MCI according to the National Institute on Aging-Alzheimer’s Association (NIA-AA) criteria [13, 14]. Patients were labeled SCD when they presented with cognitive complaints, had normal clinical and cognitive test results, and did not meet the criteria for MCI, dementia or other neurologic or psychiatric conditions [15]. Annual follow-up visit included clinical assessment and neuropsychological evaluation [11, 12].

Quality of life

During the standardized dementia diagnostic work-up and the follow-up visits between 2009 and 2018, patients were asked to complete the EQ-5D questionnaire based on the three-level version of the questionnaire (EQ-5D-3L) and/or VAS on paper [16]. In 2020, we started onlineADC, an online data collection of questionnaires related to patient-relevant outcomes (PROs), including EQ-5D five level version (EQ-5D-5L) and VAS [17]. We invited patients who had ever visited the memory clinic and their caregivers by email to complete the questionnaires in our online platform. A previous study showed that patient-reported outcome measures (PROMs) administrated on paper are comparable with questionnaires administrated on an electronic device [18].

Patients with at least one completed EQ-5D or VAS questionnaire were included in the present study. In total, we included n = 2170 EQ-5D questionnaires from 1140 persons (EQ-5D-3L/on paper: n = 1290, EQ-5D-5L/online: n = 880) and n = 2345 VAS questionnaires (n = 1465 on paper, n = 880 online). There were median (IQR) 2.0 (1.0–2.0) completed EQ-5D questionnaires per person and median (IQR) 2.0 (1.0–3.0) completed VAS questionnaires per person. The median (IQR) time between first recorded diagnosis at the memory clinic and completing the first questionnaire was 1.0 (0.0-2.0) years. The total median (IQR) follow-up time was 3.0 (2.0–3.0) years.

The EQ-5D was developed by the EuroQoL group as a standardized, non-disease-specific instrument for describing and valuing health states [19]. Patients were asked to rate their current health state in terms of five domains: mobility, self-care, usual activities, pain/discomfort and anxiety/depression. In the EQ-5D-3L version, each domain has three possible responses: no problems, some problems, or severe problems. The EQ-5D-5L has five possible responses: no problems, slight problems, moderate problems, severe problems, or unable to/extreme problems. The utility tariffs map each combination of responses on the EQ-5D to a score between 1 (perfect health) and 0 (death) and has negative values indicating a health state worse than death. The EQ-5D-5L responses were converted into an EQ-5D utilities using a Netherlands-based algorithm [20]. A “reverse crosswalk” value set was used to convert the EQ-5D-3L responses to utilities based on EQ-5D-5L values [21]. The visual analogue scale (VAS) included in the EQ-5D assesses the current health status, ranging from 0 (the worst health) to 100 (the best health).

Amyloid status

We used amyloid-PET and CSF Aβ42 (measured at first recorded diagnosis) to determine amyloid status. Patients were categorized as amyloid-positive if they had a positive amyloid-PET scan (n = 164) or abnormal CSF amyloid-ß1-42 (Aβ42) values (n = 502). Patients were categorized as amyloid-negative if they had a normal amyloid-PET scan (n = 142) or normal CSF Aβ42 values (n = 332). If both amyloid-PET and CSF values were available, we used the result of the amyloid-PET scan.

CSF was obtained by lumbar puncture, collected in polypropylene tubes (Sarstedt Nurnberg, Germany) and processed according to international guidelines [22]. Before 2018, amyloid beta (Aβ42), total tau (t-tau), and phosphorylated threonine 181 (p-tau) were measured using sandwich ELISA’s (Innotest, Fujirebio, Gent, Belgium) (n = 633) [23]. Amyloid beta values were drift-corrected [24]. After 2018, CSF was analyzed using Elecsys (n = 201). CSF concentrations were considered amyloid-positive if CSF Aβ42 drift-corrected ELISA < 813 or CSF Aβ42 Elecsys < 1000 pg/ml. Amyloid-PET scans made using 3-Tesla Ingenuity TF PET/MRI, Ingenuity TF PET/CT, and Gemini TF PET/CT scanners (Philips healthcare, the Netherlands) were visually rated by an experienced nuclear medicine physician according to manufacturer guidelines. In general, images were rated as positive when unilateral binding in one or more cortical brain regions (or striatum in case of 18F-florbetaben or 11C-Pittsburgh compound B) was observed and negative when predominantly white matter uptake was seen. Amyloid-PET scans were assessed together with a T1-weighted MRI or CT-scan to assist reading in the presence of atrophy. The amyloid-PET procedure using 18F-florbetaben (n = 73), 18F-Florbetapir (n = 98), 18F-flutemetamol (n = 50), or 11C-Pittsburgh compound B (PiB) (n = 84) have been described in detail elsewhere [25, 26].

Other variables

Follow-up time was measured in years from the first recorded diagnosis at the memory clinic to the date of EQ-5D and/or VAS was completed. The following variables were recorded during the first visit at the memory clinic: comorbidity was defined using Charlson Comorbidity Index (CCI), which was calculated based on medical history and medication use (CCI score ranges from 0 (low comorbidity) to 37 (high comorbidity)) [27]. Educational level was assessed using the Verhage scale, ranging from one (none or low educational level) to seven (high educational level: university degree) [28].Other variables we used were Mini-Mental State examination (MMSE), Rey-Auditory Verbal Learning Test (RAVLT) immediate and delayed recall, and Geriatric Depression Scale (GDS).

Statistical analysis

Statistical analyses were performed using STATA SE version 14.0 and the figures were created in R (version 4.0.3, R Development Core Team). Normally distributed continuous variables were represented as means with standard deviations (SD), non-normally distributed continuous variables as medians with interquartile ranges (IQR), and categorical variables as the number of cases with percentages. We analyzed group differences using t-tests and ANOVAs for normally distributed continuous variables, Mann-Whitney and Kruskal-Wallis tests for non-normally distributed continuous variables, and chi-squared tests for categorical variables.

First, we used linear mixed-effects models (LMM) with random intercepts to compare QoL trajectories based on both EQ-5D utilities and VAS scores between amyloid-positive and amyloid-negative patients in the SCD or MCI stage. SCD and MCI patients were analyzed separately. We included terms for amyloid status and the interaction between time and amyloid status as determinants in the models. As a result, the main effect of amyloid status represents the average difference between amyloid-positive and amyloid-negative patients at baseline and the interaction effect represents the average difference in QoL over time between amyloid-positive and amyloid-negative patients. Second, we used LMM models with random intercepts to compare QoL trajectories between amyloid-positive SCD, MCI, and dementia groups using interaction terms between follow-up time and diagnosis groups. In these models, the AD dementia group was used as the reference category. In a post hoc analysis, we used LMM to compare cognitive functioning (MMSE and RAVLT) over time between amyloid-positive and amyloid-negative SCD or MCI patients.

We adjusted LMM models for EQ-5D and VAS for two confounder sets: model 1 was adjusted for age and sex, and model 2 was additionally adjusted for CCI, education, EQ-5D version (3 or 5 level; EQ-5D only). We additionally adjusted for GDS in model 3 in the models comparing QoL between amyloid-positive and amyloid-negative SCD or MCI patients.

留言 (0)

沒有登入
gif