Hospital mortality associated with stroke in southern iran.

BACKGROUND
Unlike the western hemisphere, information about stroke epidemiology in southern Iran is scarce. The aim of this study was to determine the main epidemiological characteristics of patients with stroke and its mortality rate in southern Iran.


METHODS
A retrospective, single-center, hospital-based longitudinal study was performed at Nemazee Hospital in Shiraz, Southern Iran. Patients with a diagnosis of hemorrhagic and ischemic strokes were identified based on the International Classification of Diseases, 9th and 10th editions, for the period between 2001 and 2010. Demographics including age, sex, area of residence, socioeconomic status, length of hospital stay, and discharge destinations were analyzed in association with mortality.


RESULTS
16351 patients with a mean age of 63.4 years (95% CI: 63.1, 63.6) were included in this analysis. Men were slightly predominant (53.6% vs. 46.4%). Forty-seven percent of the total sample was older than 65,17% were younger than 45, and 2.6% were children younger than 18. The mean hospital stay was 6.3 days (95% CI: 6.2, 6.4). Among all types of strokes, the overall hospital mortality was 20.5%. Multiple logistic regression revealed significantly higher in-hospital mortality in women and children (P<0.001) but not in patients with low socioeconomic status or from rural areas. During the study period, the mortality proportions increased from 17.8% to 22.2%.


CONCLUSION
In comparison to western countries, a larger proportion of our patients were young adults and the mortality rate was higher.


Introduction
There has been a significant decrease in stroke mortality rates in developed countries, but this success story has not been mirrored in developing countries. 1 Of 5.7 million stroke patients who died in 2005, 87% were from low and middle-income countries, where stroke is considered a major disabling health problem. 2,3 Iran is a middle-income country according to the World Bank classification. 4 Recent reports have shown that the prevalence of stroke in Iran is significantly higher than that in western countries; this is especially true for stroke in the young population. 5,6 These reports have emerged from northern and central provinces of Iran. In southern Iran, however, information on stroke epidemiology is limited.
Fars Province is located in southwestern Iran, and Shiraz is its provincial capital. According to a census in 2006, Fars Province had a population of 4.3 million, 60% of them residing in urban areas. 7 Nemazee Hospital is a tertiary center in Shiraz and admits patients from the entire Fars Province. Ethnic history of Iran abounds with successive waves of occupation and migration, with the largest ethnic group being the Persians. Mitochondrial DNA linage analysis has determined the main lineage to be western Eurasian. 8 In Iran, life expectancy is about 72 years for women and 69 years for men, which suggests an ageing population perhaps similar to those in developed countries. 9 Regarding health plans in Iran, about 90% of the Iranians are covered by at least one health insurance carrier. Several types of health organizations are available to provide health coverage and these include social security, medical services, armed forces, private insurances, and charities. The first three organizations cover mainly urban public and private sector employees, as well as members of the armed forces. In 2000, a rural health insurance system was implemented to provide health coverage to rural inhabitants. The main charity provider is "Imam Khomeini Charity Foundation", which covers individuals with low or no income that is reflective of a low socioeconomic status. 10 Similar to other regions of Iran, the population of Fars Province is covered by the same health insurance carriers, with those in the low socioeconomic status accounting for approximately 7%.
This study was performed to provide basic epidemiological data on stroke. Such information has been very scarce in our region. We sought to determine the main epidemiological characteristics of patients with stroke during the last decade in southern Iran and assess the mortality rate associated with all types of stroke in Fars Province.

Patients and Methods
All patients with any types of stroke (hemorrhagic or ischemic) were admitted to Nemazee Hospital, a major tertiary center affiliated with Shiraz University of Medical Sciences. We considered the International Classification of Diseases, 9th edition-Clinical Modification (ICD-9-CM) and ICD-10-CM codes as recorded in the hospital database. The final diagnosis was determined by a qualified neurologists and then coded by experienced medical record technicians. Over , and sequel of cerebrovascular disease (I69). The diagnosis of stroke in all patients was based on clinical findings with computed tomography or magnetic resonance imaging, and was confirmed by an experienced neurologist. Patients with epilepsy, brain tumors, cerebral infections, trauma or deficits due to metabolic causes, or incomplete records were excluded. The follow-up time was equal to the duration of hospital stay. Age, sex, area of residence, socioeconomic status, and length of hospital stay were sought for each patient in a specially-designed data matrix. Because there is a lack of a structured rehabilitation system in southern Iran and most patients are discharged regardless of their stroke severity, discharge destination was not assessed in this analysis. This study was conducted and approved by the Ethics Committee of Shiraz University of Medical Sciences (HP29-90). Since the information was gathered from hospital database and included subject identifiers, we requested and obtained and institutional review board waiver of informed consent.

Statistical Analysis
For univariate analysis, Student's t test or Chi-square test was used to compare the mean and proportions of the continuous and categorical variables. A multivariate logistic regression analysis was built for the outcome of hospital mortality with the following covariates: age groups, gender, area of residence, and socioeconomic status. To assess mortality with the basic demographics of age and gender, a stratified analysis by decades was performed. The trend of mortality over the first and last years in the study period was assessed using the chi-squared test. A probability value less than 0.05 was considered significant. Statistical package for social sciences (SPSS version 15.0) was used for all the statistical tests.

Cohort Demographics
Medical

Outcome of Hospital Mortality
A total of 3354 (20.5%) patients (95% CI: 20.2% to 20.8%) died during the same hospitalization.  figure 1. The mean hospital stay in the patients who died during the same hospitalization was longer than that of the surviving population (7.0 [95% CI: 6.7 to 7.25 days] vs. 6.1 [95% CI: 6.0

Discussion
Four important observations can be made from this analysis. First is the higher in-hospital mortality (20%) in comparison to developed countries. 1 Our result chimes in with the reported case fatality rate from any stroke in central Iran (24.6%). 6 Furthermore, mortality rates in central and southern Iran are higher than those reported from the nearby states. Thirty-day case fatality rate for stroke in Arab middle-eastern and North African countries, where socioeconomic characteristics of the population are generally similar to Iran, falls between 10% and 17.3%. 11 Several factors may have contributed to these results, including absence of health institution infrastructure such as specialized stroke units and underutilization of thrombolysis, both of which are known to positively influence outcomes in acute ischemic stroke. 12 Moreover, stroke awareness is lacking among most of the Iranian general population. 13 This can lead to the referral of stroke patients in late stages and increased mortality. Post-stroke care has been another issue which may  have influenced outcome. Surveys of Iranian stroke survivors suggested that the social, financial, and rehabilitative support for stroke was inadequate. 14 Unlike developed countries, nursing facilities are not available in Iran; consequently, most stroke survivors are discharged home. 6 The lack of organized rehabilitation care and the nonsystematic nature of family care can lead to lengthy recovery, probable readmissions, and perhaps higher mortality. 15 The second observation from this analysis is noted differences in epidemiological characteristics of the stroke population in Iran. Our results suggest that a higher proportion of stroke occurs in young adults and children (14% of all stroke cases occurred in those younger than 45). These rates are comparable to those reported in the nearby countries such as Qatar (18%) and Libya (19.1%), 16,17 but they are certainly higher than those reported in the western countries. 18 The in-hospital mortality rate for this group was 21.2%, which was higher than the 3.4% to 11.2% 30-day case fatality rate in Norwegian 19 and Italian 20 patients with young-adult stroke. This suggests that stroke afflicts a large number of patients in their reproductive years in Iran, with higher-than-expected mortality.
Thirdly, multiple logistic regression revealed significant higher in-hospital mortality in women and children but not in patients with low socioeconomic status or from rural areas. Similar to prior reports, we observed a slight male predominence in our sample; however, the mortality was higher in women compared to men. The high incidence of stroke mortality in women is probably due to longer life expectancy. 21 Poor prognosis of stroke in the pediatric age group can be explained by devastating underlying general causes which make the final outcome poor. 22 The last observation is the disturbing trend of a higher mortality rate over the study period (between 2001 and 2010). This stands in contrast to the recent trends reported from developed and a few developing countries. 23 The exact explanations are yet to be determined; nevertheless, contributing factors similar to those highlighted above may have played a role in this trend.
Some shortcomings in this study are worth mentioning. First, this study is a retrospective single-hospital experience and might as such not be reflective of national Iranian standards. Second, the relatively high in-hospital mortality rate should be interpreted with caution because patients with a worse prognosis may have been over-represented among the patients who were admitted to our tertiary referral center. Third, our cohort was identified based on the ICD-9 and ICD-10 coding systems; thus, coding error could not be eliminated. Fourth, stroke-specific characteristics such as stroke location, stroke severity scale, and 30-day mortality were not reported. Fifth, the specific causes of death were not determined according to the hospital database characteristics.

Conclusion
Our study reconfirmed that stroke is a crucial health problem in Iran. In comparison to western countries, a larger proportion of Iranian patients were young adults and the mortality rate was higher. Although Iran is considered a middle-income country, the allocation of resources to improve the health system may need to be revisited. There is an urgent need for Iranian hospitals to develop better measures to manage acute stroke patients. In a wider context, international organizations should propose guidelines to implement a specialized infrastructure for stroke care in developing countries; these guidelines may influence global outcomes associated with stroke.