Outcome of Post-Stroke Rehabilitation in Nonagenarians

1. Introduction

Stroke incidence increases markedly with advancing age, with approximately half of all strokes occurring in individuals over 75 years and nearly one-third in those over 85 years [1,2,3]. Stroke remains a leading cause of long-term disability and reduced quality of life among older adults [4].

Age has consistently been identified as a predictor of functional outcome, mortality, length of hospitalization, and institutionalization after stroke [5,6,7,8]. However, it remains unclear to what extent poor outcomes are attributable to age per se versus comorbidities, baseline functional status, and medical complications.

Importantly, many stroke rehabilitation trials underrepresent the oldest old. A review of studies included in the Cochrane Database demonstrated that the mean age of participants was nearly a decade younger than the typical stroke population encountered in clinical practice. Furthermore, many trials excluded patients with cognitive impairment, aphasia, or recurrent strokes—characteristics common among very elderly individuals.

Some studies report reduced rehabilitation efficacy among the oldest old [6,8,10,11,12], whereas others suggest that chronological age alone does not independently predict poor rehabilitation outcome [13,14].

A large community-based cohort study in Denmark (191 over 85 years) reported that more than 58% of the very elderly (85 years old and above) were discharged to nursing homes or died during hospital stay post stroke [15]. In a multicenter prospective cohort study of over 300 patients of at least 75 years of age with a first stroke, age was both significantly related to low Functional Independence Measure (FIM) score upon discharge and independently and inversely related to rehabilitation efficacy (Montebello Rehabilitation Factor Score) [16].

Given demographic aging and the growing proportion of nonagenarians (individuals 90-99 years) worldwide, understanding rehabilitation outcomes in this age group is of substantial clinical relevance. The present study aimed to compare functional recovery and discharge destination in patients aged ≥90 years with those aged 65–89 years.

2. Methods

2.1. Study Design and Ethics

This retrospective study was approved by the institutional ethics committee of Kaplan Medical Center, Rehovot, Israel, with the protocol number: 0204-19-KMC.

2.2. Population

Medical records of all post-stroke patients admitted to Herzfeld Geriatric Rehabilitation Hospital between January 1, 2015, and December 31, 2019, were reviewed.

Inclusion criteria: Age ≥65 years, diagnosis of ischemic or hemorrhagic stroke confirmed by CT or MRI, admission to inpatient rehabilitation.

Exclusion criteria: length of stay ≤7 days or in-hospital death.

Patients were categorized into two groups: 65–89 years and ≥90 years.

2.3. Outcome Measures

The primary outcome was discharge destination (home vs. long-term care facility).

Secondary outcomes included: FIM score at admission, FIM score at discharge, FIM gain (ΔFIM), length of stay and rehabilitation complications.

2.4. Statistical Analysis

Categorical and Nominal variables were reported by prevalence and percentages.

Categorical and Nominal variables were analyzed by Pearson's chi-square (χ²) for the relationship between two categorical variables.

Data are presented as means/median ± standard deviations or percentages.

Continuous variables between the various study groups were tested for normality by the Shapiro-Wilk test and when abnormal distribution was found, non-parametric tests were performed. The Mann-Whitney test was performed to compare two groups.

P value <0.05 was considered statistically significant. Data were analyzed using SPSS25.

The statistical analysis included: categorical variables with two values, Fisher's exact test; when there were more than two values, the Pearson's chi-squared test was used. For continuous and normally distributed variables, the t-test was used. The data are presented as mean and standard deviation. When the variables are continuous but not normally distributed, the Mann-Whitney test was used, and the values are presented as a median and the quarterly range. Normal distribution was examined by the Shapiro-Wilk test and the Kolmogorov-Smirnov test and by examining the scattering index of the distributions and normal Q-Q plots. The statistical analysis was performed in R software (R x64 4.0.2)

3. Results

A total of 906 patients were screened; 30 were excluded (9 deaths, 21 short stays), leaving 876 patients for analysis. Seventy-three patients (8.3%) were aged ≥90 years. Data were collected in a computer-based registry that included subjects aged 65 years and older with a diagnosis of ischemic or hemorrhagic stroke accident, confirmed by neuroimaging (computed tomography, CT, or Magnetic Resonance, MR) and classified according to Bamford criteria [18].

Baseline characteristics are presented in Table 1.

Nonagenarians had:

Lower prevalence of diabetes and hyperlipidemia

Higher prevalence of dementia

Similar rates of hypertension, coronary artery disease, heart failure, and chronic kidney disease

Stroke subtype and neurological deficits did not differ significantly between groups.

Rehabilitation Outcomes and Complications (Table 2)

Length of stay did not differ significantly (41 vs. 40 days).

Infections (pneumonia, urinary tract infections) were more common in nonagenarians. Urinary catheter use at discharge and constipation were also more frequent in the ≥90 group.

Admission FIM, discharge FIM, and ΔFIM were all significantly higher in the younger group. However, discharge to home was comparable between groups (78.2% vs. 76.7%).

Predictors of Discharge Home (Table 3)

In the 65–89 group, higher admission FIM predicted discharge home. Pneumonia, aphasia, atrial fibrillation, constipation, and urinary catheter at discharge were negatively associated with home discharge.

In the ≥90 group, no independent predictor of discharge destination reached statistical significance.

4. Discussion

The present study demonstrates that patients aged 90 years and older derive meaningful benefit from post-stroke rehabilitation, achieving functional improvement and discharge outcomes comparable to those of younger elderly patients. Although nonagenarians were admitted with lower baseline functional status and achieved smaller absolute gains in FIM scores, the proportion discharged home did not differ significantly between age groups. These findings challenge the common perception that very advanced age alone predicts poor rehabilitation outcomes.

Population-based data from North America and Europe demonstrate a steep age-related increase in long-term care residency. In the United States, approximately 1% of individuals aged 65–74 reside in nursing homes, increasing to 3% at ages 75–84 and approximately 8% among those aged 85 years and older [19]. Similarly, in England and Wales, 10.8% of individuals aged ≥85 years live in care homes [20]. Canadian census data report even higher proportions among the oldest old, with nearly 30% of those aged ≥85 residing in special care facilities [21]. These epidemiological trends underscore the substantial baseline risk of institutionalization in very elderly populations and highlight the clinical relevance of rehabilitation strategies that facilitate discharge to the community.

Against this background, the comparable home-discharge rates observed in our cohort (76.7% in ≥90 years vs. 78.2% in younger elderly patients) are particularly noteworthy. Despite lower admission FIM scores and higher rates of dementia, infections, urinary catheter use, and constipation, nonagenarians were not more likely to require long-term institutional placement. This finding suggests that chronological age alone should not be considered a limiting factor when allocating rehabilitation resources.

In both groups—nonagenarians and younger elderly patients—the median number of prescribed medications exceeded five, consistent with polypharmacy, and no significant difference was observed between the groups. This finding is noteworthy given the distinct morbidity profiles: nonagenarians demonstrated a higher prevalence of dementia, a lower prevalence of diabetes, and lower Functional Independence Measure (FIM) scores.

The high prevalence of polypharmacy aligns with the well-established association between multimorbidity, functional impairment, and increased healthcare utilization in geriatric populations [22,23,24]. Importantly, polypharmacy substantially elevates the risk of clinically significant drug–drug interactions (DDIs), a concern that is particularly relevant in older adults. Age-related alterations in pharmacokinetics and pharmacodynamics, diminished renal and hepatic reserve, and increased frailty further amplify susceptibility to adverse drug events [25,26,27,28]. Several medication classes that are more frequently prescribed in older adults, including antipsychotics [29], opioids [30] and monoclonal antibody therapies[31], carry considerable interaction potential and safety concerns. Previous studies have demonstrated that the risk of adverse drug reactions increases exponentially with the number of medications prescribed [32], with DDIs representing a major and potentially preventable cause of hospitalization in older adults [33].

The absence of significant differences in both polypharmacy prevalence and discharge destination between the two age groups—despite their differing morbidity profiles—is intriguing. This observation warrants further investigation in prospective, multicenter studies to better elucidate the underlying mechanisms and clinical implications.

Consistent with previous reports, in our work, younger elderly participants demonstrated higher FIM scores at admission and discharge and greater absolute functional gains [18,34]. However, the length of stay was nearly identical between groups (41 vs. 40 days), indicating comparable rehabilitation efficiency. The absence of prolonged hospitalization in the oldest group suggests that rehabilitation intensity and duration need not differ solely with age.

In the present work, multivariable analyses revealed that in the 65–89-year group, higher admission FIM predicted discharge home, whereas pneumonia, aphasia, atrial fibrillation, constipation during rehabilitation, and urinary catheter at discharge were negatively associated with community discharge. These findings are in line with prior literature identifying medical complications and reduced baseline function as principal determinants of institutionalization [17,18,34]. Interestingly, no independent prognostic factor was identified in the ≥90-year group. This may reflect pre-existing adaptation to functional limitations in very elderly individuals and their families, including established caregiving arrangements and environmental adjustments before the index stroke.

In our study, discharge destination was selected as the primary outcome measure because it represents a clinically meaningful endpoint with substantial implications for quality of life, caregiver burden, and healthcare expenditure. Given the high baseline prevalence of institutional residence among individuals aged ≥85 in Western countries, the ability to return nonagenarian stroke survivors to the community carries both humanitarian and economic significance.

This study has several limitations. Its retrospective, single-center design limits causal inference and may reduce generalizability to other healthcare settings. Important factors such as frailty, caregiver support, and pre-stroke functional trajectory were not systematically assessed and may have influenced discharge outcomes. The relatively small number of nonagenarians (n=73) limits statistical power, particularly in multivariable analyses, and may explain the absence of independent predictors in this subgroup. In addition, discharge destination, although clinically meaningful, does not capture long-term functional sustainability, quality of life, or post-discharge mortality. Finally, while medication burden was reported, a detailed assessment of potentially inappropriate medications and drug–drug interactions was not performed. Prospective studies incorporating comprehensive geriatric assessment and long-term follow-up are warranted.

Our findings support an inclusive approach to rehabilitation referral. Advanced age alone should not be used to restrict access to post-acute rehabilitation services. Even in a very advanced age, structured inpatient rehabilitation can facilitate functional improvement and successful reintegration into the community.