Ischemic stroke versus hemorrhagic stroke – A comparative study on the outcome of stroke survivors attending a comprehensive stroke center of a tertiary care hospital

INTRODUCTION

Stroke, being a significant concern with 12.2 million cases annually in 2019, remains one of the leading causes of death and disability worldwide.^[1] The impact of stroke outcomes is not limited to physical disability alone. It also influences the social, psychological, and financial domains of stroke survivors and their caregivers. In India, outcome measures for stroke rehabilitation were found to use ordinal scales developed in the West, with marked socioeconomic and cultural differences.^[2] Being a significant contributor to disability-adjusted life year, a holistic approach is required to address how the quality of life of a stroke survivor is affected by the stroke.

Previous studies have shown that hemorrhagic stroke (HS) patients tend to present with more severe neurological deficits than those with acute ischemic stroke (AIS).^[3,4] Nevertheless, survivors of HS often demonstrate substantial functional gains following rehabilitation.^[5,6] Although mortality is initially higher in HS, long-term survival beyond 3 months appears to be similar between the two stroke types.^[7] Most comparative studies, however, have focused primarily on mortality and physical disability, with limited exploration of broader quality-of-life outcomes.^[2]

We use a culturally appropriate and comprehensive scale, the stroke impact assessment questionnaire (SIAQ), to compare outcomes in AIS and HS among Indian subjects.^[8] It covers eight domains of life (cognition, communication, sensory, motor, behavioral, emotional, social, and economic domains) that could be crippled by the burden of stroke. This approach provides a culturally sensitive lens to understand stroke recovery beyond survival, guiding more holistic rehabilitation and policy planning.

MATERIALS AND METHODS

RESULTS

We saw 130 stroke survivors in the stroke review clinic, out of which 112(86.2%) had an AIS and 18 (13.8%) had HS. The majority of the patients, 79 (60.7%), were male. The mean age was 60.37 (S.D. 11.28). The sociodemographic characteristics of the sample are summarized in Table 1. Age, category, sex, and educational status were statistically significant between AIS and HS patients on univariate analysis, but did not reach significance on logistic regression.

The most common cause of AIS was of undetermined etiology – 52 (40%), followed by 32 (24.6%) patients having large artery atherosclerosis, 20 (15.4%) having small vessel disease, 7 (5.4%) patients having cardio-embolic disease, and one patient having stroke of other determined etiology.

The mRS at discharge and follow-up were not normally distributed. The mRS score at discharge was higher in the HS group than the AIS group (P = 0.048). While at follow-up, there was no significant difference in mRS score in the two groups (P = 0.640).

There was no difference in SIAQ scores between HS (Median = 25.5 n = 18) and AIS (Median = 22.0, n = 112) survivors at follow-up (P = 0.177). A comparison of the eight domains of SIAQ across the two groups also showed no significant difference [Table 2 and Figure 1].

Close

Figure 1:

Comparison of stroke impact assessment questionnaire scores between hemorrhagic and ischemic stroke. Boxes indicate interquartile range, horizontal line shows the median, whiskers denote the range, and circles represent outliers. SIAQ: Stroke impact assessment questionnaire

Export to PPT

A one-way analysis of covariance (ANCOVA) was conducted to determine whether there was a statistical difference between AIS and HS on the SIAQ Score at follow-up, while controlling for follow-up duration. The test showed an insignificant effect of the type of stroke on SIAQ Score after controlling for follow-up duration ([F [1,127] = 2.969, P = 0.087]). There was a significant positive correlation (Spearman) between the motor domain of the SIAQ and the mRS score at follow-up (r [128] = 0.845, P < 0.001). A Spearman correlation between the stroke follow-up duration and the follow-up SIAQ score was negative. It was not significant for patients with <6 months of follow-up (r [128] = −0.088, P = 0.387) but was positive and statistically not significant for durations longer than 6 months (r [128] = 0.291, P = 0.112). A Spearman correlation between the stroke follow-up duration and the follow-up SIAQ motor domain was negative and statistically insignificant for durations <6 months. (r [128] = −0.065, P = 0.522). However, the correlation remained positive and insignificant for more than 6 months (r [128] = 0.284, P = 0.121).

DISCUSSION

The study utilized patient-reported outcomes, as measured by SIAQ, which incorporates physical disability and other domains relevant to stroke patients. When controlling for follow-up duration, a one-way ANCOVA revealed that the type of stroke did not significantly affect the SIAQ score. This indicates that after accounting for the follow-up duration, there was no significant difference in the stroke outcomes measured by the SIAQ between AIS and HS patients.

The study found no difference in long-term outcomes between HS and AIS patients attending a CSC of a tertiary care hospital. The sample population consisted of 130 acute stroke patients who came for review after discharge at diverse time intervals ranging from 1 month to 1 year after an acute stroke. The baseline characteristics, including the sociodemographic profile of stroke in this study, were similar to those found in a study analyzing stroke incidence, types, risk factors, and outcomes in the Trivandrum stroke registry.^[9]

The findings of our study show a higher incidence of stroke among men (60.7%). This finding is consistent with a systematic review of nine studies focused on incidence, prevalence, and case fatality of stroke in India.^[10] The protective effect of genetic and hormonal factors in females and differences in lifestyle choices may have contributed to this result. HS was more prevalent in patients under 60 (72.2%) years, and AIS was more prevalent in patients aged 60 and above (P = 0.006). The age distribution observed in the study is consistent with a study conducted in the Netherlands that identified the relative proportion of AIS to increase with age and the relative proportion of intracerebral hemorrhage to decrease with age.^[11] Advancing years may lead to increased comorbidities, risk factors, and alterations in brain vasculature, possibly contributing to increased AIS among the elderly.^[12] HS, although commonly associated with hypertension, various factors such as vascular abnormalities and coagulation disorders can contribute to its occurrence in the younger population.^[13]

Increased prevalence of Ischemic stroke in the pre-metric population could be attributed to the greater influence of conventional cardiovascular risk factors. A similar study conducted in New South Wales, Australia, demonstrated that low educational status was associated with an increased stroke risk in both men and women. This is attributed to the modifiable risk factors of stroke.^[14]

The association of age category, sex, and educational status with the type of stroke was insignificant in the regression model; the model explains only 9.9% of the variance in the type of stroke.

Our data showed that patients with HS had severe physical disability at discharge based on the mRS score compared to AIS (P = 0.048). This finding was consistent with a national stroke registry-based study by Andersen et al. in Denmark, which showed that HS had an initial high severity and high mortality in the first 3 months.^[7] However, outcome assessment, as measured by mRS and SIAQ at follow-up, showed no significant difference between the two groups. Further analysis of the cognitive, communication, sensory, motor, behavioral, emotional, social, and economic domains of SIAQ also did not reveal significant differences between AIS and HS patients. The higher early mortality in hemorrhagic stroke likely introduced survivorship bias, as only those with relatively better recovery attended follow-up. This may have contributed to the comparable outcomes observed between groups. A similar comparative study on rehabilitation outcomes by Perna and Temple also showed no significant group difference at admission or discharge.^[15] Some studies point toward a more significant gain in function for HS after rehabilitation.^[6,7]

Several factors can influence the outcome of stroke patients, including the severity of the stroke, age, educational status, and the type and duration of rehabilitation. These factors may have contributed to the convergence of scores, as observed in this study. Although socioeconomic status, assessed using the above poverty line/below poverty line classification, did not differ significantly between groups, financial capacity likely influences recovery through access to physiotherapy, medications, and rehabilitation support. This highlights the importance of considering economic and social factors when planning post-stroke care in low- and middle-income settings.

It was observed that motor, behavioral, and financial domains of SIAQ scored high on follow-up. Compared to mRS, which focuses mainly on ambulation, the motor domain of SIAQ is more sensitive in identifying motor disabilities, as it incorporates upper limb disabilities and culturally appropriate scenarios of physical disability. Hence, SIAQ helps clinicians to adopt more appropriate physical rehabilitation measures. The involvement of domains other than motor can be addressed only by SIAQ, since mRS deals with physical disability alone. The all-inclusive approach to outcome assessment provides more personalized care for stroke survivors. Patients with behavioral changes could be identified, and proper counseling and expert advice could be given. Studies show that stroke patients with depressive symptoms have a poorer health-related quality of life.^[10,16] In such patients, the rehabilitation plan should include measures to improve their social participation and assessment by a psychiatrist. Similarly, effective speech therapy could be given to patients with communication issues.

The economic burden was significant following a stroke. Consequent to loss of employment, many found that patients’ family income was insufficient and had to borrow or take out loans to cover the expenses of medicine and physiotherapy.

Correlation analysis revealed a significant and robust positive correlation between the SIAQ motor domain and the mRS score at follow-up. This finding suggests that motor impairments assessed by the SIAQ are related to functional disability measured by the mRS.

Stroke recovery, measured by the SIAQ, may vary based on the duration of follow-up. As proven by multiple studies conducted earlier, stroke patients tend to experience significant improvement in the initial stages of their recovery. A previous study conducted in Korea studying the 6-month functional recovery of stroke patients showed a statistically significant improvement in neurologic impairments (trunk control, motor function, sensory, and cognition) and recovery in functional impairments (activity of daily living and gait) over 6 months.^[17] However, the overall recovery process after a stroke follows a nonlinear, logarithmic pattern.^[18] This means the most substantial improvements occur early after the stroke, gradually slowing down. A previous study conducted by Meyer et al. demonstrated that, upon long-term follow-up, the condition of stroke patients deteriorated.^[19] This decline could be attributed to the fact that more intensive and improved care and rehabilitation are typically provided in the early stages of stroke recovery. Given our cross-sectional design and non-significant correlation between follow-up duration and SIAQ scores (r = 0.291, P = 0.112), we are unable to confirm this pattern of long-term decline definitively. However, these non-significant trends underscore the complexity of recovery beyond the acute phase, emphasizing the need for robust, prospective studies with standardized follow-up time points to characterize the dynamic long-term trajectory accurately. Finding new strategies is crucial for enhancing long-term outcomes.

This study has certain limitations. The main limitation is its small sample size. Although the number of HS patients was small compared to ischemic stroke, this reflects the real-world distribution observed in tertiary care centers. Despite a small sample size, ANCOVA and nonparametric tests were used to minimize bias from unequal groups. Acute reperfusion data (thrombolysis/thrombectomy) were unavailable for this study due to incomplete documentation. Departmental records indicate that ~40% of ischemic stroke patients at our center receive intravenous thrombolysis, compared to ~13% in the Indo-US Collaborative Stroke Project.^[20] These contextual figures suggest good access to acute care but cannot substitute for patient-level data, representing a potential source of treatment-related bias. Rehabilitation type, frequency, and intensity were not systematically documented and could not be included in our analysis. Given the substantial influence of rehabilitation on functional recovery, this represents an important confounder that could influence functional outcomes between stroke subtypes. Baseline stroke severity using the National Institutes of Health Stroke Scale (NIHSS) was inconsistently documented, and the mRS at discharge was used as a proxy. NIHSS is a more granular measure of acute neurological severity, and its absence limits the precision with which baseline stroke severity could be compared between AIS and HS groups. This may have influenced our interpretation of early severity differences. Information on comorbidities, including hypertension, diabetes, atrial fibrillation, and depression, was incomplete, which may have influenced outcomes. Future prospective studies incorporating these variables are warranted for a more comprehensive understanding of stroke recovery.

While no significant difference was found between hemorrhagic and ischemic strokes in SIAQ scores after controlling for follow-up duration, future studies should have a standardized follow-up duration and serial assessment of SIAQ score at definite intervals to have a clearer picture of functional changes over time.

The findings should therefore be interpreted with caution, and future multi-center studies with larger and more balanced cohorts are needed to validate these results.

CONCLUSION

This study, conducted at a CSC in Kerala, India, found differences in sociodemographic characteristics between patients with AIS and HS. However, no significant differences were observed in stroke outcomes or functional disabilities between the two groups, as measured by mRS and SIAQ scores. These results suggest further research is needed to identify additional factors influencing recovery. The SIAQ proved valuable for assessing stroke outcomes across multiple domains beyond motor function.