Table 1.
Characteristics of the selected studies on cerebrovascular disease and frailty: study methods and main results.
| Country, study name, population type | N | Mean age ± SD | Women % | Cerebrovascular disease diagnosis/definition % prevalence | Frailty criteria and overall prevalence** | Prevalence % of cerebrovascular disease in each frailty group | Odds ratios, risks and other results | NOS | |
|---|---|---|---|---|---|---|---|---|---|
| Avila-Funes et al. (23) | France, AMImage study, Community (rural farmers) |
176 | 75 ± 5.2 | 40 | Self-reported history of physician diagnosed stroke. 4.4% | Fried criteria. Overall frail = 18.8% |
Robust = 4.0 Frail = 6.3 p = 0.630 |
White matter hyperintensities (mL) higher in frail (mean = 12.1, SD = 17.3) than in robust (mean = 4.8, SD = 9.6), p = 0.23 ADJ: age, sex, education, cardiovascular risk factors. Frail persons had lower Fractional Anisotropy and higher diffusity values in several white matter areas (corpus callosum, anterior limb of internal capsule, external capsule and posterior thalamic radiations). | 6 |
| Chen et al. (24) | Japan, Sasaguri Genkimon Study (SGS), Community | 1,565 | Range 65–93 | 60 | Self-reported history of having ever been diagnosed with stroke. 3.6% | Fried criteria. Overall frail = 9.5, prefrail = 43.9% | Robust = 2.6 Prefrail = 4.2 Frail = 6.0 p = 0.02 |
6 | |
| Calado et al. (25) | Brazil, FIBRA study (Study of Frailty in Elderly Brazilian Individuals), Community | 385 | 73.9 ± 6.5 | 64.7 | “Stroke.” Self-report (questionnaire) on any chronic diseases that had been recognized by a doctor during the past year. 2.1% | Fried criteria. Overall frail = 9.1, prefrail = 49.6% | Robust = 0 Prefrail = 2.6 Frail = 8.6 p = 0.02 |
6 | |
| de Albuquerque Sousa et al. (26) | Brazil, REDE FIBRA (Network of Studies on the Frailty of Elderly Brazilians), Community | 391 | 74.1 ± 6.6 Range 65–96 | 61.4 | Self-reported presence of stroke diagnosed in the last year. 1.8% | Fried criteria. Overall frail = 17.1%, prefrail = 60.1% | Robust = 1.1 Prefrail = 0.04 Frail = 7.5 p = 0.001 |
5 | |
| Espinoza et al. (27) | USA, San Antonio Longitudinal Study of Aging (SALSA), Community | 394 | Range 65–80 | 57.6 | Stroke was assessed according to self-report of physician-diagnosed disease. 10.7% | Fried criteria. Overall frail = 10.7% | Odds of frailty not significant in multi-adjusted models (data not shown) | 7 | |
| Lahousse et al. (28) | The Netherlands, Rotterdam study, Community | 2,833 | Median = 74 Range ≥ 55 | 55.9 | Stroke was “clinically validated.” Prevalence not reported. | Fried criteria. Overall frail = 6%, prefrail = 51% | Robust = 0.8 Prefrail = 0.9 Frail = 2.5 p = 0.001 |
5 | |
| Lee et al. (29) | Hong Kong, Community | 3,018 | 49.7 | Participants were asked whether they had ever been told by a physician that they had a stroke. Medical diagnoses were cross-checked in the computerized medical system database of the Hong Kong Hospital Authority. Diagnoses were counted as present if reported by the participant or recorded in the medical database. 5.2% in men, 3.5% in women | Fried frailty criteria. 2 year change in frailty status was assessed (e.g., robust worsening (from robust to prefrail or frail), prefrail worsening (from prefrail to frail). Overall baseline frail = 7.9%, prefrail = 50.6% | At 2 year follow-up, about half of prefrail persons remained prefrail, but 11.1% of men and 6.6% of women worsened into frailty, and a quarter recovered into the robust state. Among the frail at baseline, one-quarter remained frail. Change from prefrail to frail in men OR = 1.8 (0.8–3.8), women OR = 2.8 (1.01–7.8). Change from robust to prefrail/frail in men OR = 1.5 (0.7–3.2), women OR = 3.96 (1.4–10.5). In multivariate models stroke was associated with an improvement in men with baseline prefrailty OR = 0.4 (0.2–0.9) or frailty OR = 0.2 (0.1–0.9), and a worsening in prefrail status in women OR = 3.11 (1.05–9.18) and a change from robust to frail/prefrail in women OR = 3.5 (1.2–10.1). |
7 | ||
| Li et al. (30) | China, RulAS population-based survey | 1,757 | 75.3 (3.9) | 53.3 | Past medical history taken by physicians using a standard questionnaire. 7.3% | Fried criteria, frailty = 10.1% Plus a physical-cognitive frailty scale, frail = 19.4% |
Fried Robust = 6.2% Frail = 17.8% p = 0.001 Physical-cognitive frailty Robust = 5.9% Frail = 13.6% p = 0.001 |
7 | |
| Llibre Rodriguez et al. (16) | 8 countries, 10/66 Study, Community | 1,6886 | ≥65 | 62.4% | Stroke was self-reported, but confirmed by the interviewer as having characteristic symptoms lasting for more than 24 h. 6.7% | Modified Fried criteria (only four indicators measured) | Pooled estimates (10 sites in 8 countries) Stroke and frailty OR = 2.3 (2.1–2.6) Adj: age, sex, education |
7 | |
| Merchant et al. (31) | Singapore, HOPE (Healthy Older People Everyday), Community | 1,051 | 71.2 | 57.2 | Patients screened for chronic diseases, including stroke | 5-Item frailty scale (fatigue, resistance, ambulation, illness, loss of weight) (32) Overall frail = 6.2%, prefrail = 37% | Robust = 1.7 Prefrail = 6.4 Frail = 16.9 p < 0.001 |
5 | |
| Nadruz et al. (33) | USA, Atherosclerosis Risk in Communities Study, Community | 3,991 | 75.6 ± 5.0 | 59 | Previous Stroke. 2.7% | Fried criteria. Overall frail = 5.3% | Robust = 2 Frail = 7 p < 0.001 |
6 | |
| Ng et al. (18) | Singapore, SLAS—Singapore Longitudinal Aging Studies I and II, Community | 1,685 | 66.7 ± 7.76 | 64 | The self-report of a medical disorder diagnosed and treated by a physician(s) was recorded for 22 named diagnoses, including stroke. 32% | Fried criteria. Overall frail = 5%, prefrail = 42% | Robust = 1.6 Prefrail = 4.1 Frail = 12.1 p < 0.001 |
Significant correlates of prefrailty-frailty from binary logistic regression via backward stepwise variable selection: Stroke B = 0.76 OR = 2.1 (1.1–4.1), p = 0.23 | 5 |
| Seamon et al. (34) | USA, Medicare sample | 7,258 | 79.4 (8.4) | 56.7 | All patients hospitalized with a first-time acute ischemic stroke. 100% | Faurot Frailty Index | 39.1% of stroke patients were robust, 36.0% were prefrail, and 24.9% were frail. | 4 | |
| Serra-Prat et al. (17) | Spain, Community | 154 | 80.1 (3.5) | 47.5 | Information on comorbidities and medications was obtained from the electronic medical records held by the corresponding centers. All other information was obtained directly from the patient by trained healthcare professionals. Prevalence not reported. | Fried criteria. Overall frail = 53.7%, prefrail = 14.2% | Robust = 8.7 Prefrail = 7 Frail = 23.9 p = 0.003 |
Crude OR for frailty in stroke patients = 3.82 (1.7–8.58). Adjusted OR = 4.5 (1.35–14.97), p = 0.014.Adj: age, sex, education, anorexia, osteoarthritis, dyspepsia, number of medications, anemia, CRP, muscle mass, and creatinine. | 6 |
| Taylor-Rowan et al. (35) | UK, Patients consecutively admitted to acute stroke unit | 545 | 69 (14) | 46 | Physician diagnosed. 100% | 33 item frailty index. | 28% of stroke patients were frail and 51% were prefrail. | 4 | |
| Trevisan et al. (36) | Italy, Progetto Veneto Anziani, Community | 2,925 | 74.4 ± 7.3 Range ≥ 56 | 59.7 | Personal interview, medical interview and clinical examination including blood tests. Cardiovascular disease (CVD) was defined as atrial fibrillation; congestive heart failure; angina pectoris requiring a stent, angioplasty, or hospitalization; myocardial infarction; or stroke. Prevalence not reported. | Fried criteria. Overall frail = 6.6%, prefrail = 49.3% | Of the persons who were nonfrail at baseline 26.7% became prefrail at follow-up and 6.3% became frail. Progressing from prefrail to frail stroke OR = 1.96 (1.72–2.24), p < 0.001 | 8 | |
| Vaingankar et al. (19) | Singapore, Well-being of the Singapore Elderly study, Community | 2,102 | 69 Range ≥60 | Not specified. Field interviewers collected data on medical conditions. “Stroke” | Fried criteria. Overall frail = 40.1%, prefrail = 5.7% | Robust = 2.2 Prefrail = 7.4 Frail = 13.4 p < 0.001 |
Prefrailty OR = 2.6 (1.2–5.8), p = 0.018; Frailty OR not significant (data not shown) | 6 | |
| Winovich et al. (20) | USA, Cardiovascular Health Study, subsample of persons with incident ischemic stroke | 893 | 82 ± 6.4 | 61 | Incident stroke events identified through semi-annual phone calls, hospital discharge report review, and health insurance registries. | Fried criteria. | In patients with incident stroke, 27.8% were robust, 54.9% prefrail and 17.3% frail. No comparison between no stroke controls. | 5 |
ADJ, adjustment variables; ADL, activities of daily living; CMB, Cerebral Microbleeds; CRP, C-reactive protein; CT, Computed tomography; CVD, cardiovascular disease; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; HR, hazard ratio; MRI, magnetic resonance imaging; NOS, Newcastle Ottawa Scale risk of bias score; OR, odds ratio; VaD, vascular dementia.
*
All numbers are rounded to one decimal point.
**
In all studies using Fried criteria, the cutoff for frailty was 3 out of 5 symptoms, and prefrailty was 1–2 symptoms.