Stroke epidemiology and outcomes of stroke patients in Nepal: a systematic review and meta-analysis

Background With an increasing burden of stroke, it is essential to minimize the incidence of stroke and improve stroke care by emphasizing areas that bring out the maximum impact. The care situation remains unclear in the absence of a national stroke care registry and a lack of structured hospital-based data monitoring. We conducted this systematic review and meta-analysis to assess the status of stroke care in Nepal and identify areas that need dedicated improvement in stroke care. Methods A systematic literature review was conducted to identify all studies on stroke epidemiology or stroke care published between 2000 and 2020 in Nepal. Data analysis was done with Statistical Package for Social Sciences (SPSS) and Comprehensive Meta-analysis (CMA-3). Results We identified 2533 studies after database searching, and 55 were included in quantitative and narrative synthesis. All analyses were done in tertiary care settings in densely populated central parts of Nepal. Ischemic stroke was more frequent (70.87%) than hemorrhagic (26.79%), and the mean age of stroke patients was 62,9 years. Mortality occurred in 16.9% (13-21.7%), thrombolysis was performed in 2.39% of patients, and no studies described thrombectomy or stroke unit care. Conclusion The provision of stroke care in Nepal needs to catch up to international standards, and our systematic review demonstrated the need to improve access to quality stroke care. Dedicated studies on establishing stroke care units, prevention, rehabilitation, and studies on lower levels of care or remote regions are required. Supplementary Information The online version contains supplementary material available at 10.1186/s12883-023-03382-5.


Background
Stroke has become a critical global public health challenge requiring prompt and effective intervention.In particular,12.2 million new cases, 101 million prevalent cases, and 6.55 million stroke-related deaths were reported [1].The data show a remarkable increase in stroke incidence and mortality rates from 1990 to 2019, with a 70% rise in stroke incidence and a 43% rise in stroke-related deaths [1].
The situation of stroke in Asian countries is not different from the global scenario.The reported incidence of stroke in Asia ranges from 116 to 483 per 100,000 per year [2][3][4].Furthermore, evidence suggests that South Asians have a twofold higher risk of getting a stroke than Europeans due to the higher prevalence of dyslipidemia, diabetes mellitus, and central obesity [5,6].Nepal, a South Asian country with a population of 29 million, has reported a relatively high crude and age-standardized prevalence of stroke in the southwestern region in 2018, with rates of 2368 and 2967 per 100,000 population, respectively [7].However, this data only represents a specific region and may not be generalizable to the Nepalese context [2].
Improving stroke care demands reliable data on stroke epidemiology, risk factors, treatment, and outcomes.However, such data are not available for Nepal.Therefore, this systematic review aims to fill this knowledge gap by exploring stroke studies conducted in the Nepalese population regarding stroke epidemiology, risk factors, treatment, and outcomes.This review will help to identify the needs in stroke care in Nepal.

Methods
This systematic review was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement.The study protocol was registered in PROSPERO [8] prior to the conduct of the review.

Selection of studies Inclusion criteria
We included studies published in English between January 1, 2000, and January 1, 2021, reporting empirical data (quantitative, qualitative) obtained in Nepal.Studies were included if participants had a confirmed stroke diagnosis and were at least 18 years of age.(REF: Sacco L et al., Stroke, 2013) or TIA (REF: Easton et al., Stroke, 2009) and reported on incidence, epidemiology, risk factors, etiology, stroke outcome, or stroke treatment (e.g., diagnosis, acute and post-acute care, rehabilitation, financing of stroke care, complications of stroke).

Exclusion criteria
We excluded articles that could not be classified as empirical literature (e.g., commentaries, discussion papers, journalistic interviews, policy reports), reviews, studies on stroke mimics (e.g., migraine), and studies on mixed populations (e.g., South Asians) unless separate results for people with stroke in Nepal could be isolated.Studies reporting on adults < 18 years were excluded.

Search strategy
The study followed the "Cochrane Guidelines for Systematic Reviews of Health Promotion and Public Health Interventions" in designing the search strategy.PubMed, Ovid, Cochrane Library, Web of Science, and clinicaltrials.orgwere searched for English-language articles published between 2000 and 2020.Google and Google Scholar identified grey literature not indexed in academic databases was identified.The search terms and keywords related to stroke, knowledge, epidemiology, and treatment.The ' Appendix 1 Search strategy' contains the detailed search strategy.Additionally, the reference lists of included papers were screened.

Study selection
Study selection was performed by (1) independent screening of titles and abstracts (RP, CT), and (2) Independent screening of full texts of all hits judged suitable in the first step (RP, CT).Discrepant ratings were discussed and agreed upon in consensus meetings (AC, BPG, LT, and PJ).Specificity (proportion of suitable articles in all hits) and sensitivity (proportion of suitable articles in all correct positives) were calculated as quality criteria for the search strategy based on a predefined test set.Subsequently, studies on stroke awareness were excluded to keep the review topic specific to stroke care.

Data extraction, synthesis, and analysis
A data extraction form was designed, including author, year, study title, sample characteristics, stroke prevalence, incidence, etiology, risk factors, treatment (recanalization therapy, length of hospital stay), mortality, complications, outcome, and diagnostic findings.The included articles were extracted by AA, KB, SA, and PJ and checked by RP and CT.Given the significant heterogeneity of the included studies, a narrative data synthesis was performed.The heterogeneity of the studies was calculated using the I2 statistics; for I2 ≤ 50, a fixed effects model was used.For I2 > 50, the random effects model was used and represented using forest plots with CMA-3 for metaanalysis and SPSS 22 for descriptive analysis.The quality of the included studies was assessed using the Oxford Centre of Evidence-Based Medicine: Level of Evidence (March 2009) [9].

Study selection
A total of 2533 studies were identified, and 141 duplicates were removed.The title and abstracts of 1250 studies were screened, and 1100 studies were excluded.Full texts of 150 studies were assessed, and 95 studies were excluded for definite reasons.A total of 55 studies were included in this systematic review (Fig. 1).Specificity was 3.9%, and sensitivity was 100%.

Types of strokes
Fifty-Four studies reported on the types of strokes 64].Pooled data showed ischemic stroke in 70.87% and hemorrhagic in 26.79%.The transient ischemic attack was reported only in 0.66% of patients (Table 1).

Carotid doppler findings
Three studies reported on carotid Doppler findings in ischemic stroke patients [58][59][60], where 70.78% of 178 patients were found to have plaque and 18.5% had 50-99% occlusion of the carotid artery.

Disability/functional outcomes
From 13 studies reporting on the disability and functional outcome of stroke patients, ten studies used the modified Rankin scale (mRS) [13, 21, 35, 43-45, 50, 51, 56, 57], two studies used the Glasgow outcome scale [32,36] and one used WHO disability assessment schedule [27].Assessment time ranged from discharge to 6 months.Most patients had mRS 3 [21,50], and the mean average mRS ranged from 2.66 to 3.48 [50,51].Table 3 presents the disability and functional outcomes of 13 studies.

Cost
No studies reported on the cost of stroke care or cost-effectiveness.

Discussion
Our study is the first systematic literature review to describe the overall picture of stroke patients and stroke care in Nepal and to analyze which aspects of stroke care have been scientifically investigated, what is known from these research results, and where there is an unmet need in research.Despite a comprehensive search strategy, we identified only 55 studies conducted in Nepal within the last 20 years and analyzed stroke outcomes or aspects of care.The low quantity of studies weighs even more seriously because half of the studies are also of low quality, and there has yet to be a randomized controlled trial (RCT) on stroke care in Nepal.Therefore, the most urgent implication of our work is that more high-quality research is needed.
Most studies were conducted in densely populated areas in central Nepal, with better health infrastructures than in western regions.The studies were all done  [43] At discharge, mRS 4 = 12/12 At 3 months, mRS 2 = 7/12, mRS 3 = 5/12 Thapa L et al. 2016 [44] At discharge, mRS 0 = 4/9, mRS 3 = 2/9, mRS 4 = 1/9, mRS 5 = 2/9 Thapa A et al. 2018 [45] At discharge, mRS < 2 = 558/946 (58.9%)At the end of 3 years, mRS < 2 = 115/130 (88.5%)Shakya D et al. 2019 [27] WHODAS mean score = 46.0 ± 23.6 HRQOLISP mean score = 60.6 ± 13.6 in tertiary care and teaching facilities and hence may not represent the situation for stroke patients in remote areas or at lower levels of care but rather overestimate the level of stroke care in Nepal.However, as with other diseases, patients from rural communities are referred to tertiary care centers for treatment, and the study population can be said to comprise patients from rural parts of the country.Hence, we need dedicated studies to get an accurate picture of the rural parts of the country.It is even more alarming that even in this setting, no stroke units are described, and the rate of thrombolysis is below 2.5%, so we must assume a thrombolysis rate of less than 1% for Nepal.As IVT, EVT, and stroke unit care are the mainstay of acute therapy in ischemic stroke [65], our systematic review emphasizes the need for dedicated and organized stroke care to improve the country's overall picture of stroke care.The mean age of stroke presentation varied from 68.6 years in men to 72.9 years in women [66].The pooled results in our study showed a mean age of 62.4 years, which is younger than the global average; 63.1 in lowmiddle income country (LMIC) vs. 68.6 in high-income countries (HIC), which might be attributed to limited stroke care quality and accessibility [67].Our study showed more men than women suffering from stroke.A systematic review done in 19 countries by Appelros et al. showed stroke incidence to be 30% higher in men than women and 41% more prevalent in men than women [66].Further research is necessary to understand if this gender gap is caused by a reduced incidence of stroke in women or restricted access to care.
Our review highlighted a high prevalence of preventable risk factors in stroke patients, aligning with other studies' findings [1,[68][69][70].Policymakers should focus on preventing noncommunicable diseases through effective primordial, primary, and secondary prevention strategies [71] and adapt WHO strategies (e.g., Tobacco Control Convention) to meet the needs of Nepal.
The unavailability of stroke units is all the more detrimental because the long-term outcome of stroke can be significantly improved by preventing complications and recurrent stroke, which usually happens in a stroke unit.Pneumonia and UTI were common post-stroke infections (18.8 and 7% of patients) associated with more extended hospital stays.This data is similar to other studies showing UTI and pneumonia as the most common complications [72,73].As only four studies reported complications, our pooled data may not accurately portray the country's real scenario, and more focused studies on post-stroke complications are required.Stroke recurrence is common, especially in large artery atherosclerosis and cardioembolic stroke.While we found a high rate of carotid plaque, only 3% of reported patients had atrial fibrillation, which might be due to insufficient detection methods.Studies on the prevention of complications, secondary prevention, and long-term functional outcome are scarce in Nepal [74].
The economic burden of stroke for an LMIC like Nepal cannot be overstated.The use of thrombolysis treatment ($1390) is expensive, especially under consideration of the annual per capita gross domestic product (GDP) of the country ($1208.22)[75,76].In addition, the loss of active earnings by stroke patients' family members, costs of rehabilitation, and nursing care might even exceed the costs of acute treatment [77].Hence, cost-effectiveness studies will be crucial to evaluating the direct and indirect costs of stroke care.Further, LMICs like Nepal need to focus more on preventing stroke and imparting knowledge to stroke patients' families, which can assist significantly in minimizing overall stroke care costs.

Strengths and limitations
As the first study of its type, this is a milestone in stroke care in Nepal, especially in the absence of a populationbased or hospital-based national stroke registry.However, our review has some limitations.We only included the articles in English and could not retrieve some fulltext articles.We have also likely not identified all relevant articles published in non-indexed journals.Also, most of the studies included in our review were observational and were of low level of evidence as per Oxford grading.We, therefore, highlight the need to allocate more resources for research and access to publication in international journals for scientists from LMICs.

Conclusion
Without a national stroke registry, our systematic literature review will be highly relevant to Nepal's medical community and policymakers.We observed the demographics of stroke patients to be similar to those from other regions, but the provision of stroke care needs to catch up to international standards.Based on the available literature, we highly recommend conducting more high-quality research in Nepal, especially in rural settings outside Kathmandu.Our systematic review emphasizes the absence of structured stroke care in Nepal and the urgent need to improve access to quality stroke care.Hence, with the collaboration of the medical fraternity, local bodies, and government, we must establish stroke

Fig. 3 Fig. 2
Fig. 3 Number of studies reporting the outcome parameters

Fig. 5 Fig. 4
Fig. 5 Funnel plot for detection of publication bias in meta-analysis of mortality rate in stroke patients

Fig. 6
Fig. 6 Forest plot showing complications in stroke patients

Table 2
Summary of included studies Sample size indicates total sample size.The percentage of females and comorbidities are derived from the total sample size.

Table 3
Disability and functional outcomes of the studies