Long-term functional outcomes and caregiver burden in survivors after decompressive hemicraniectomy for malignant hemispheric infarction: a real-world experience from South India

Abstract

Background

Malignant hemispheric infarction (MHI) is associated with high early mortality when managed conservatively. Decompressive hemicraniectomy (DHC) improves survival, but long-term functional outcomes and caregiver burden in low-resource settings remain poorly characterised, particularly in India. This study aimed to evaluate long-term outcomes and caregiver burden among MHI survivors undergoing DHC at a tertiary care centre in South India.

Methods

We conducted an ambispective study of adult patients who underwent DHC for MHI between 2018 and 2023 and had at least 6 months of follow-up. Functional outcome was assessed using the modified Rankin Scale (mRS) and caregiver burden was evaluated using the Zarit Burden Interview. Subgroup analyses were performed based on infarct lateralisation and timing of surgery. Logistic regression was used to identify predictors of a favourable functional outcome.

Results

66 patients (mean age 44±12.8 years; 80% male) were included, with a median follow-up of 2 years (IQR: 1–4). Overall mortality was 24.2% (n=16). Among the 50 survivors, 26 (52.0%) achieved a good functional outcome (mRS ≤3) and 13 (26.0%) achieved an excellent outcome (mRS ≤2). Infarct lateralisation did not significantly influence mortality or functional recovery. Although early surgery and higher admission Glasgow Coma Scale scores were associated with better outcomes in univariate analysis, these factors were not significant in multivariate models. Caregivers of patients with mRS >2 reported significantly higher burden, particularly in emotional, physical and social domains.

Conclusions

DHC for MHI results in meaningful long-term survival and functional recovery in selected patients, regardless of hemisphere involvement. However, residual disability imposes a substantial caregiver burden. Stroke care pathways in resource-limited settings should incorporate comprehensive rehabilitation and long-term caregiver support.

WHAT IS ALREADY KNOWN ON THIS TOPIC.

WHAT THIS STUDY ADDS

• This study demonstrates that DHC was associated with meaningful long-term survival and functional recovery in selected patients with MHI, irrespective of hemisphere involvement. It further highlights that long-term recovery in low- and middle-income countries remains influenced by limited access to structured rehabilitation, financial constraints and reliance on family-based caregiving after hospital discharge.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• These findings underscore the need for integrated poststroke rehabilitation pathways, improved access to affordable long-term neurorehabilitation, and structured caregiver-support systems across healthcare settings in low- and middle-income countries. The study also supports the need for future research focusing on long-term quality of life, caregiver burden and health-system strategies to optimise recovery after MHI.

Introduction

Stroke is a leading cause of death and long-term disability worldwide, accounting for over 12 million new cases and 6.5 million deaths annually, with a disproportionately high burden in low and middle-income countries (LMICs) such as India.¹ Malignant hemispheric infarction (MHI), a subtype of ischaemic stroke, is associated with high early mortality (within 2 weeks) of up to 80% when managed conservatively.^{2 3} The DECIMAL, DESTINY and HAMLET trials collectively showed that decompressive hemicraniectomy (DHC), when performed within 48 hours of symptom onset, reduces mortality and can improve functional outcomes.^4–6 While these trials laid the foundation for DHC to become the standard of care in MHI, most were conducted in high-income countries with robust poststroke rehabilitation infrastructure. Their generalisability to LMICs remains uncertain, as inequalities in acute stroke care, limited access to neurocritical care services and inadequate rehabilitation support could alter both survival and quality of life outcomes.⁷ Although previous reports from India have reported improved survival and functional outcomes in patients undergoing Decompressive Hemicraniectomy (DHC) for MHI, none have reported long-term outcomes or the caregiver burden related to postdischarge management of these patients.⁸ Recent observational studies in LMICs have reported comparable mortality reductions with DHC, but variable functional outcomes, with significant numbers of survivors left with high disability scores.^9–14 Further, questions remain regarding the effect of hemispheric lateralisation on functional recovery, with some clinicians hesitant to recommend DHC for dominant (left-sided) infarctions due to concerns over language deficits and quality of life. This study aimed to bridge these gaps by evaluating long-term functional outcomes and caregiver burden among patients who underwent DHC for MHI at a tertiary care centre in South India. We aimed to provide real-world evidence to inform clinical decision-making and policy development for stroke care in LMIC settings.

Methods

Study design and setting

This study was conducted on a historical cohort of patients who underwent DHC for MHI between January 2018 and January 2023 at a tertiary care teaching hospital in southern India.

Study setting

The hospital, situated in Vellore, serves as an advanced stroke care centre, offering comprehensive neurology, neurosurgery and neurocritical care services. Standard of care offered includes initiation of poststroke rehabilitation services to all patients before discharge, which is tailored as per individual requirements

Patient selection

This study included all patients aged 18 or older who underwent DHC for radiologically confirmed MHI. MHI was defined as a large territorial infarction involving more than 50% of the middle cerebral artery territory, often accompanied by early imaging signs of mass effect, such as midline shift or sulcal effacement. Survivors with a minimum of 6 months of follow-up after DHC were included in the study.

Patients were excluded if they met the following criteria:

• Pre-existing significant neurological disability (modified Rankin Scale (mRS) ≥3).

• Underlying intracranial malignancy, trauma or haemorrhagic venous infarcts.

• Incomplete medical records.

Management protocol

All patients diagnosed with MHI at admission were given a trial of medical management. Indications for DHC included (a) Glasgow Coma Scale (GCS) score of <15 at admission with evidence of MHI and uncal herniation, (b) GCS score of <15 with radiological signs of impending uncal herniation or severe midline shift, (c) Failed medical management as evidenced by clinical or radiological deterioration and (d) Conversion of the initial small volume of infarct to MHI on subsequent CT.

Data collection

Baseline demographic and clinical data were retrieved from hospital records, including age, gender, comorbidities (hypertension, diabetes, ischaemic heart disease), stroke lateralisation (right vs left hemisphere involvement), GCS at admission and National Institutes of Health (NIH) Stroke Scale (NIHSS) at admission; details of acute reperfusion therapy, including intravenous thrombolysis, endovascular thrombectomy or both; and timing of DHC categorised as <24 hours, 24–48 hours or >48 hours from stroke onset. Disability status was assessed at hospital discharge using the mRS. Patients were followed up at stroke clinic visits, where the functional outcomes were documented at each visit.

Outcome measures

The primary outcomes were all-cause mortality during follow-up, and functional status was assessed using the mRS at the last follow-up visit. The outcomes were categorised as excellent when the mRS was <2 and as good when the mRS was <3. The secondary outcomes included caregiver burden, assessed among survivors using the Zarit Burden Interview (ZBI) scale, a validated 22-item questionnaire that measures caregiver stress, which was administered by a stroke nurse and is interpreted as mild burden (0–21), moderate burden (21–40) or severe burden (>60).¹⁵ The mean responses to the 22 ZBI items were compared between the groups using independent sample t-tests. Subgroup analyses were conducted to evaluate differences in outcome and caregiver burden by hemispheric lateralisation (right vs left MHI), the association between decompression timing and outcome and the correlation between caregiver burden and patient disability.

Statistical analysis

The data were analysed using SPSS V.2. Descriptive statistics were used to summarise the patient characteristics. Continuous variables were reported as mean±SD or median (IQR), as appropriate, and categorical variables as frequencies and percentages. Comparisons between groups (right vs left hemisphere, good vs poor outcome) were conducted using the χ² test, Fisher’s exact test for categorical variables, independent t-test or Mann-Whitney U test for continuous variables. Predictors of a good functional outcome (mRS ≤3) were assessed using binary logistic regression, adjusting for potential confounders, including age, admission GCS score, revascularisation status and time to surgery. Statistical significance was set at p<0.05.

Missing data were minimal for baseline demographic and clinical variables obtained from institutional medical records. Analyses of functional outcomes were conducted using complete-case data. Caregiver burden assessment using the ZBI was available only for survivors and caregivers who consented to participate in follow-up. No imputation techniques were applied because the extent of missingness in the primary outcome variables was low, and the caregiver burden analysis was exploratory.

The study findings were presented at the Indian Academy of Neurology conference 2024; the abstract has been published in the conference abstract booklet.¹⁶

Results

Patient demographics and clinical characteristics

A total of 66 patients who underwent DHC for malignant MHI of the middle cerebral arterial territory between January 2018 and January 2023 were included (figure 1). The mean (SD) age was 44 (12.8), and the majority were male (n=53, 80%). Right hemispheric infarction was observed in 37 patients (56.9%), whereas left hemispheric involvement was noted in 29 patients (43.1%). Common comorbidities included hypertension in 29 patients (43%) and diabetes mellitus in 27 (40.9%). The mean GCS score at admission was 11.9±2.6, and 25 patients (37.9%) had a GCS score <12. The mean NIHSS score at presentation was 15.27±4.86. Regarding acute stroke interventions, intravenous thrombolysis was administered to 15 (23.1%) patients, endovascular thrombectomy to 5 (7.7%) and both modalities to 12 (18.5%) patients. DHC was performed within 24 hours in 43 patients (65%). The mean mRS score at discharge was 5 (1.6), indicating a severe disability (online supplemental table 1).

Figure 1. STROBE. Follow-up of malignant hemispheric infarction (MHI) patients post decompressive hemicraniectomy (DHC). STROBE, Strengthening the Reporting of Observational Studies in Epidemiology.

Functional outcomes at follow-up

The overall mortality rate was 24.2% (16 of 66 patients), with nine deaths occurring during the initial hospital stay and an additional seven during follow-up, within the first year. Among the 50 surviving patients, 13 patients achieved excellent functional outcomes, defined as an mRS score of ≤2. 26 survivors (52%) had good outcomes, with mRS scores of ≤3. These improvements in functional status over time are illustrated in figure 2, which depicts the mean mRS distribution at follow-up.

Figure 2. The disability as measured by the Modified Rankin scale at discharge and follow-up.

Comparison between right and left hemisphere infarctions

There were no statistically significant differences in age, sex distribution, revascularisation therapies or timing of surgery between the right and left hemispheres. There was no significant difference in mortality (p=0.74) or favourable outcome rates (mRS ≤3; p=0.17) between the hemispheres (online supplemental table 2).

Predictors of good functional outcomes

Univariate analysis suggested that a higher GCS score at admission and early decompressive surgery (within 24 hours of symptom onset) were associated with a greater likelihood of achieving good functional outcomes (mRS ≤3). Early surgery was performed in patients with a greater midline shift and/or a poor GCS. However, these associations did not reach statistical significance in the multivariate analysis. Among patients with an admission GCS greater than 12, 52% achieved favourable outcomes, compared with 26% among those with lower scores (p=0.14). Early surgery did not contribute to good outcomes in our overall cohort as an independent predictor (p=0.44).

Caregiver burden analysis by functional outcome

Caregiver burden was evaluated using the ZBI among survivors stratified by long-term functional outcomes. Caregiver burden was assessed in 29 patients who consented to participate in the interview. Statistically significant differences were observed in the four key domains of caregiver burden. First, interpersonal strain was significantly higher among caregivers of patients with an mRS score greater than 2, who more frequently reported that their relationships with other family members or friends were negatively affected (p=0.044). Second, caregivers of patients with more severe disabilities (mRS more than 4) commonly felt that they lacked sufficient personal time due to caregiving responsibilities, indicating a greater time burden (p=0.010). Third, caregivers reported deterioration in their health due to caregiving duties more frequently (p=0.020). Finally, emotional exhaustion was markedly higher in the poor-outcome group, with caregivers more likely to report feeling unable to continue caregiving for much longer (p=0.002), reflecting substantial caregiver burnout. Caregivers of patients with higher disability reported more embarrassment about the patient’s behaviour (p=0.068), greater anxiety regarding the patient’s future (p=0.078) and a stronger sense of inadequacy in their caregiving role (p=0.051). In contrast, no statistically significant differences were observed between the groups in questions related to anger, privacy, financial strain or perceived patient dependency. There was no significant difference in caregiver burden scores between those with right-hemispheric or left-hemispheric infarctions (online supplemental table 2).

Discussion

This study presents real-world, long-term follow-up data on functional outcomes and caregiver burden among patients who underwent DHC for MHI at a tertiary care centre in South India. Our results demonstrate that a substantial proportion of survivors achieved favourable functional outcomes, with 39.3% attaining an mRS score ≤3 and 19.6% achieving excellent outcomes (mRS ≤2) at a median follow-up of 2 years. These findings are encouraging and align with landmark clinical trials, such as DECIMAL, DESTINY and HAMLET, which have shown a significant reduction in mortality and acceptable levels of disability with DHC in selected patients.^{4 5 17 18}

Importantly, our study provides evidence that lesion lateralisation (right vs left hemisphere infarction) did not significantly affect mortality or functional recovery, reinforcing prior observations that functional outcomes are more dependent on global neurological recovery than hemisphere dominance.^19–22 These findings challenge clinical hesitation, sometimes associated with performing DHC in dominant hemisphere strokes, where language and cognitive deficits are feared, contributing to poor quality of life.

While several patient-level variables, including a worse admission GCS and early decompression (<24 hours), were associated with better outcomes in univariate analysis, these associations did not reach statistical significance in multivariate analysis. This may be due to the relatively small sample size and the multifactorial nature of stroke recovery, which extends beyond acute clinical severity to include factors such as access to rehabilitation, comorbidities and postdischarge care.

One of the most significant contributions of our study is the systematic assessment of caregiver burden using the ZBI. Caregivers of patients with worse functional outcomes (mRS>2) reported a significantly higher burden in domains related to interpersonal strain, time constraints, personal health deterioration and emotional exhaustion. These findings are consistent with the existing literature, which emphasises the substantial psychological and physical toll on informal caregivers, particularly in LMICs where professional support services are often limited.^23–25 Caregiver responses also revealed trends toward greater embarrassment, anxiety and perceived inadequacy, underscoring the need for comprehensive psychosocial interventions targeting caregivers. The high burden observed in our cohort highlights a critical gap in stroke recovery frameworks, which often prioritise patient rehabilitation while overlooking the pivotal role and well-being of caregivers.

From a public health perspective, these findings underscore the importance of developing integrated stroke care pathways that extend beyond the acute phase and include long-term support for both patients and their caregivers.^{26 27} Although our institution is a tertiary referral centre with advanced stroke and neurocritical care services, the broader healthcare environment reflects many of the challenges encountered in LMICs. A substantial proportion of patients are referred from rural and semiurban regions where access to organised rehabilitation, speech therapy, home-based nursing care and long-term psychosocial support remains limited. Adding to it, stroke-related healthcare expenditure is largely out-of-pocket, and caregiver support systems are often informal and family-dependent. These contextual factors likely influence long-term functional recovery and caregiver burden, and they are an important aspect of the real-world relevance of our findings.

The strengths of this study include the real-world context, long follow-up duration and combined analysis of clinical and caregiver-centred outcomes.

Limitations

This study has several limitations that merit consideration. First, this was a single-centre study conducted at a tertiary referral hospital, which may limit the generalisability of the findings to other settings, particularly centres with differing access to neurocritical care and rehabilitation services. Second, the sample size was relatively modest, limiting the statistical power of subgroup and multivariable analyses. Consequently, effect size estimates for several associations had wide CIs and were not reported. The findings should therefore be considered exploratory and hypothesis-generating rather than definitive. Third, although baseline demographic and clinical variables had minimal missing data, caregiver burden assessments were available for only a subset of survivors, potentially introducing selection bias. Analyses were performed using complete-case data, and no imputation methods were applied. Fourth, caregiver burden was assessed cross-sectionally during follow-up and may not fully capture temporal changes in caregiver stress over the course of recovery. Finally, while functional outcomes were assessed using the mRS, more detailed assessments of cognitive function, language deficits, psychosocial recovery and quality of life were not performed. Future multicentre studies with larger cohorts and comprehensive long-term outcome assessments are needed to better define predictors of recovery and caregiver burden following DHC for MHI.

Conclusions

DHC provides significant long-term survival and functional benefits in carefully selected patients with MHI regardless of the hemisphere involved. However, substantial residual disability in many survivors imposes a considerable burden on caregivers, particularly in emotional, social and physical domains. These findings suggest the need to expand poststroke care models to incorporate targeted interventions and community rehabilitation services, particularly in low-resource settings.

Data availability statement

Data are available on reasonable request.