Stroke Epidemiology and Outcome in Southern Sweden

Abstract: Background: Stroke is a common cause of mortality and disability worldwide and continuously updated, thorough information of itsoccurrence and outcome is needed. In recent years, multiple advances have been made in stroke prevention and treatment that mayhave changed the epidemiology and outcome of stroke.Aims: To investigate updated population-based stroke epidemiology as well as long-term survival, recurrence and functional outcomein southern Sweden.Methods: Paper I: A population-based cohort of 413 consecutive patients with first-ever stroke in 2015-2016 was assembled usingthe prospective hospital-based stroke studies Lund Stroke Register (LSR) and Riksstroke, as well as retrospective searches ofprimary care, outpatient clinic and autopsy sources. Age- and sex-standardized incidence rates were calculated and compared with aprior study from the same study area in 2001-2002.Paper II: A subset of the population-based cohort of 400 patients (only ischemic stroke and intracerebral hemorrhage) was comparedwith LSR and Riksstroke regarding case ascertainment and which patients were not included in the hospital-based stroke studies.Paper III: The 400 individuals in the population-based cohort were followed up after 3 years regarding survival and causes of deathby using the Swedish Causes of Death Register, and stroke recurrence by using medical record review. Index and recurrent strokeswere classified by pathogenetic mechanism using the TOAST classification. Stroke survival at 3 years was compared with priorstudies in our area from 1983-1985, 1993-1995, and 2001-2002.Paper IV: The population-based cohort was followed-up in-person or via telephone at 3-4 years regarding functional outcome (mRS),dependency in activities of daily living (ADL), and health-related quality of life (HRQoL). Data were also collected on potentiallycorrelated health problems after stroke such as fatigue, cognitive impairment and depression.Results: Paper I: The total population-based age- and sex-standardized rate of stroke incidence was 165 per 100 000 person yearsin our study area, a 33% decrease compared with a population-based study in our area from 2001-2002. However, stroke incidencerates did not change for hemorrhagic stroke or among those <65 years.Paper II: LSR detected 363 (91%) of cases while Riksstroke detected 328 (82%). Patients undetected by the hospital-based studieshad high case fatality (44% vs 9%; p=0.001), and those only detected in primary care (n=11) often lived in nursing homes (57%).Those not detected by Riksstroke had less severe stroke (median NIHSS 3 vs 5; p=0.013).Paper III: In total, 265 (66%) survived 3 years after first-ever stroke. Among individuals with ischemic stroke, cardio-aortic embolismas pathogenetic mechanism was associated with the lowest 3-year survival (51/91; 56%). Cardiovascular disease was the cause ofdeath in 59% of cases (79/135). Meanwhile, 8% (32/400) had a recurrent stroke within 3 years, and the pathogenetic mechanism ofischemic stroke changed between first-ever stroke and recurrence in 16/29 (55%) cases. Three-year survival improved between the1980s and the present study (56% vs 66%; p=0.002), but survival did not change significantly between the early 2000s and thepresent study (64% vs 66%; p=0.48. Paper IV: In all, 202 individuals were clinically followed-up after a median of 3.2 years, while 47(12%) stroke survivors were lost to follow-up. Among followed-up survivors, 147 (73%) had favorable functional outcome (mRS≤2)and 134 (69%) reported good-excellent HRQoL. Age, stroke severity, professional care pre-stroke and recurrent stroke (all p<0.001)were predictors of poor functional outcome. Among follow-up variables, fatigue (p=0.001), and stroke severity (p<0.001) wereassociated with dependency in ADL, and fatigue (p<0.001) was also associated with worse HRQoL.Conclusions: Stroke incidence and survival have improved over recent decades, however some subgroups of stroke have notimproved in the same manner. Thorough population-based epidemiological studies of stroke are important to avoid and to possibleselection bias in hospital-based stroke studies. Around 3 of 10 long-term stroke survivors have poor outcome, and fatigue may be asignificant contributor to post-stroke function and health.