Geographical variations in the incidence of stroke may be

Original Contributions Low Incidence of Stroke in Southern Italy A Population-Based Study Giovanni Manobianca, MD; Stefano Zoccolella, MD; Antonella Petruzzellis, MD; Annamaria Miccoli, MD; Giancarlo Logroscino, MD, PhD Background and Purpose Although a lower incidence of stroke has been observed in the Mediterranean area compared to other European countries, this is based on only a few studies. We sought to determine the incidence and 28-day case-fatality of stroke through a population-based stroke register in a rural area in Southern Italy, characterized by a stroke unit in the referral hospital. Methods We established a multisource prospective population-based register in a well defined geographic area of 38 735 inhabitants in Puglia, Southern Italy. We identified all subjects in the study area with a first-ever stroke between January 1, 2001 and December 31, 2002. Results We identified 127 first-ever strokes (77 males, 50 females) during the two-year study period. Hospitalization was 95%: 92 cases (72.4%) were cerebral infarction, 24 (18.9%) intracerebral hemorrhage, 3 (2.4%) subarachnoid hemorrhage, and 8 (6.3%) were unclassifiable strokes. The overall crude annual incidence was 1.6 per 1000 (95%CI: 1.4 to 1.9), 2.0 for males (95% CI:1.6 to 2.5), and 1.3 for females (95% CI:0.9 to 1.6). The incidence rates standardized to the 2001 European and world populations were respectively 1.5 (2.0 for males and 1.3 for females) and 0.8 (0.9 for males and 0.6 for females). Incidence rates progressively increased with age in both sexes, reaching their peak at 85 years or more (21.4/1000 overall, 35.0 for men and 13.4 for women). The 28-day case-fatality was 18.1%. Conclusions Our study supports previous findings of lower incidence of stroke in the Mediterranean area, whereas the case-fatality in our study was lower than in previous studies from Italy. Further studies are needed to determine the role of prompt referral and stroke units on prognosis in population-based setting. (Stroke. 2008;39:2923-2928.) Key Words: stroke incidence registry stroke unit population-based Italy Mediterranean area Geographical variations in the incidence of stroke may be related to differences in the prevalence of genetic and environmental risk factors in populations. 1,2 The Mediterranean populations are characterized by a specific diet, lifestyle, and other cultural factors that may contribute to differences in the incidence, pathology, and sex distribution of stroke. 2 Compared with Northern European populations, a trend toward lower incidence for stroke has been observed in the Mediterranean area by five population-based studies. 3 7 However, these lower estimates could be attributable to limitations in some of these studies, including retrospective collection of data 4 and absence of specialized care in the study area. 3,7 We established a prospective population-based stroke register to determine the incidence and 28-day case-fatality of stroke in a well-defined area in Puglia, Southern Italy, with a high level of a specialized care, including a stroke unit in the referral hospital. Materials and Methods In 2000, we established a registry based on the prospective, community-based collection of data on stroke incidence and outcome. Surveillance began on January 1, 2001 and was continued until December 31, 2002. The study was based on the standard ideal criteria for population-based stroke studies as proposed by Malmgren et al in 1987, 8 Bonita et al in 1995, 9 and updated by Sudlow and Warlow in 1996 and 1997. 10,11 Study Area and Population The study was conducted in a well-defined geographic district in Puglia, a region located in south-east Italy (4 086 613 residents according to the 2001 census, 19 357 km 2 ). 12 The district is 208.46 km 2 in area and includes two towns, Acquaviva delle Fonti and Casamassima. The study population on January 1, 2001 comprised 38 735 inhabitants (19 068 males, 19 667 females). The proportion of individuals aged 65 years or more was 15.7% (14.2% of men, 17.1% of women), which is comparable to that in the entire Italian population (18.6% overall, men 15.8%, and women 21.2%). 12 To be counted as a resident, each stroke case should have been resident in the study area for at least one year. Received March 14, 2008; accepted April 11, 2008. From the Department of Neurology (G.M., A.P., A.M.), Hospital F. Miulli, Acquaviva delle Fonti, Bari, Italy; Azienda Ospedaliero-Universitaria Ospedali Riuniti (S.Z.), Medical and Neurological Sciences, Clinic of Nervous System Diseases, University of Foggia, Italy; and the Department of Neurology and Psychiatry (G.L.), University of Bari, Italy. G.M. and S.Z. contributed equally to this study. Correspondence to Giancarlo Logroscino, MD, PhD, Department of Neurology and Psychiatry, University of Bari, Italy. E-mail lggn01ne@neurol.uniba.it 2008 American Heart Association, Inc. Stroke is available at http://stroke.ahajournals.org DOI: 10.1161/STROKEAHA.108.519421 2923

2924 Stroke November 2008 The study area is served by only one regional general hospital (Ospedale F. Miulli), which has a department of neurology and a semi-intensive stroke unit of 6 beds. Almost all patients with acute neurological diseases are referred to the department of neurology, and stroke patients admitted to the internal medicine or cardiology departments are routinely assigned to a neurologist. The population is also served by 32 general practitioners (GPs) and 2 first aid stations that were periodically contacted by the study staff. This study has been approved by the Ethical and Scientific Committee of Ospedale F. Miulli. Stroke Definition and Classification Stroke was defined according to the WHO criteria as rapidly developing clinical symptoms or signs of focal, and at times global, loss of cerebral function, with symptoms lasting more than 24 hours or leading to death, with no apparent cause other than that of vascular origin. 13,14 This definition excludes on clinical grounds cases of primary cerebral tumors, cerebral metastasis, subdural hematoma, postseizure palsy, and brain trauma. Stroke was classified as either first-ever (occurring for the first time in a patient s lifetime) or recurrent (defined though anamnestic recall, general physician interview, or hospital records). Only cases of first-ever stroke were included in the study; patients with transient ischemic attacks (TIA) or recurrent stroke were excluded. Strokes subtypes were identified on the basis of either computerized tomography (CT) or MRI (MR) performed within 28 days of symptom onset. Patients with first-ever strokes were categorized based on neuroimaging as cerebral infarction, intracerebral hemorrhage, subarachnoid hemorrhage, or as undetermined, when no neuroimaging tests were performed. Case Ascertainment Complete case ascertainment was ensured through multiple sources of information in the study area. Hospitalized patients included those admitted directly to the department of neurology, those admitted to the departments of medicine or cardiology, and patients who, for any reason, were identified by either the referring physician or the neurologist acting as consulting physician. Emergency room records were periodically searched for cases who were not admitted to Ospedale F. Miulli but were referred to other hospitals because of a lack of beds in the appropriate departments or because of death before admission. We also regularly examined the Hospital Discharge Diagnosis Data Bank, a regional computerized system that includes the diagnosis of each hospitalized patient in the study area in both public and private clinics. We looked for any patient with a discharge diagnosis including codes 430 to 438, according to the International Classification of Disease, 9th revision. The search for nonhospitalized stroke cases was performed in collaboration with the GPs in the study area. Each GP was informed of the aims and methods of the study and was provided with the study protocol and the research team s telephone numbers and addresses. The GPs were invited to promptly report any suspected case of nonhospitalized stroke to the research team. GPs were contacted on average every 3 months. Nursing and residential homes were contacted every month. To identify possible cases not caught by the other sources of information, the death certificates of all area residents were reviewed every month. Suspected cases identified by death certificates were reviewed with the referring physician, using the GP s medical records, and interviews with relatives. Statistical Methods Data were stored in a computerized database, in separate, anonymous files. Age- and sex-specific incidence rates of first-ever stroke were estimated per 1000 inhabitants. Rates were calculated as the number of cases divided by the person-year. Age- and sex-specific incidence rates were also adjusted by the direct method 15 to the 2001 Italian, 12 European, 16 and world 16 populations. To make results comparable with other population-based studies, data were presented in 10-year age bands. The 95% confidence intervals (CI) were estimated according to Poisson distribution. Univariate analyses employed either t test (for continuous variables) or 2 (for categorical variables). Significance was tested at the 5% level. Results During the 2-year surveillance period, 127 first-ever strokes were identified. Seventy-seven (60.6%) were males with a mean age of 72.5 years (range 13 to 93), and 50 (39.4%) were females with a mean age of 77.5 years (range 47 to 96). Of these 127 cases, 121 (95.2%) were hospitalized (2 of whom died in the emergency room), and 6 were treated at home and died without a neuroimaging test. One hundred twenty-five cases were prospectively included in the study, mostly within 24 hours of their event. Two cases hospitalized outside the study area were identified through examination of the Hospital Discharge Diagnosis data bank, and no patients were identified only by death certificates examination. Of these 127 strokes, 92 were classified as cerebral infarction (72.4%), 24 intracerebral hemorrhage (18.9%), 3 subarachnoid hemorrhages (2.4%), and 8 (6.3%) undetermined strokes. The 8 cases of undetermined stroke did not undergo a neuroimaging test, because they were already deceased (6 at home and 2 in the emergency room). At least one CT or MR was performed on the other 119 patients (93.7%). The overall crude incidence rate was 1.6 (95% CI: 1.3 to 1.9) per 1000, 2.0 (95% CI: 1.6 to 2) for males and 1.3 (95% CI: 0.9 to 1.6) for females. The stroke incidence rate adjusted to the 2001 Italian population was 1.8 overall (95% CI: 1.5 to 2.1), 2.0 (95% CI: 1.6 to 2.5) for males and 1.4 (95% CI: 1.1 to 1.9) for females. The incidence rate adjusted to the 2001 European population was 1.5 overall (95% CI: 1.3 to 1.9), 1.7 (95% CI: 1.3 to 2.0) for males and 1.3 (95% CI: 0.9 to 1.7) for females. Finally, the incidence rate adjusted to the 2001 world population was 0.8 overall (95% CI: 0.6 to 0.9), 0.9 (95% CI: 0.7 to 1.1) for males and 0.6 (95% CI: 0.4 to 0.7) for females. Stroke incidence progressively increased with age in both sexes, reaching a peak at 85 years or more (21.4/1000 overall, 35.0 for men and 13.4 for women; Table 1). The overall crude incidence rates per 1000 population for the different pathological subtypes of first-ever stroke were: cerebral infarction 1.2 (1.5 for men, 0.9 for women); intracerebral hemorrhage 0.3 (0.4 for men, 0.3 for women); subarachnoid hemorrhage 0.04 (0.05 for men, 0.02 for women); and undetermined stroke 0.1 (0.1 for both men and women). The incidence rates for cerebral infarction and intracerebral hemorrhage increased significantly with age; the highest incidence for both these pathological subtypes was among men aged 85 years or more (28/1000 for CI and 5.3/1000 for IH). The overall 28-day case-fatality rate was 18.1% (23/127 patients), 15.5% for males and 22% for females. The overall mean age at death was 83 years (81.5 for males, 84.8 for females). The highest case-fatality rate was among undetermined strokes (100%), compared to intracerebral hemorrhage (20.8%), cerebral infarction (10.8%), and subarachnoid hemorrhage (0%). Discussion In this population-based study, the incidence rates of firstever stroke were similar to those in other population-based

Manobianca et al Low Stroke Incidence in Southern Italy 2925 Table 1. Age- and Sex-Specific Incidence Rates of First-Ever Stroke per 1000 in Acquaviva and Casamassima Southern Italy All Males Females Age Group Population at Risk Cases Rate 95% CI Population at Risk Cases Rate 95% CI Population at Risk Cases Rate 95% CI 0 14 12 651 1 0.07 nc 6486 1 0.15 nc 6165 0 0 nc 15 24 10 261 0 0 nc 5248 0 0 nc 5013 0 0 nc 25 34 11 815 0 0 nc 5770 0 0 nc 6045 0 0 nc 35 44 12 474 1 0.08 nc 6191 1 0.16 nc 6283 0 0 nc 45 54 10 156 6 0.6 0.2 1.3 5038 5 1.0 0.3 2.2 5118 1 0.2 nc 55 64 7817 17 2.2 1.2 3.5 3838 12 3.1 1.7 5.4 3979 5 1.2 0.4 2.9 65 74 6585 23 3.5 2.2 5.2 3148 16 5.1 2.9 8.2 3437 7 2 0.8 4.2 75 84 4174 46 11 8 14.7 1775 22 12.4 7.7 18.7 2399 24 10 6.4 14.9 85 1537 33 21.4 14.8 30.1 570 20 35 21.4 54.2 967 13 13.4 7.1 23 TOTAL 77 470 127 1.6 1.4 1.9 38 064 77 2.0 1.6 2.5 39 406 50 1.3 0.9 1.6 1.8* 1.5 2.1* 2* 1.6 2.5* 1.4* 1.1 1.9* 1.5 1.3 1.8 1.7 1.3 2.0 1.3 0.9 1.7 0.76 0.62 0.90 0.94 0.70 1.17 0.57 0.41 0.73 *Adjusted to the Italian population; Adjusted to the European population; Adjusted to the world population. nc indicates not calculated. studies from Southern Italy 3 7 but lower than those from Central and Northern Italy 17 20 and other European populations 21 37 (Table 2). In this study the 28-day case-fatality was lower than in previous studies conducted in Italy (3 to 5, 17 to 20). Only 5 population based-studies on stroke incidence and pathology in the Mediterranean area have been published as full papers 3 7 : 4 were prospective 3,5 7 and 1 was retrospective. 4 Three of these studies have certain limitations: the incidence of subarachnoid hemorrhage and undermined stroke were not evaluated in the study from Foggia province 4 ; adjusted data on stroke incidence were available only for patients aged between 45 and 85 years in the study from Greece 6 ; and although the small sample size and isolation of the population in the Aeolian archipelago study can help ensure complete ascertainment, the study area has no neurologists or other specialized health care services. 3 Although the lower stroke incidence in Puglia could be related to the age structure of the study population, 1 the proportion of subjects in the age groups at high risk of stroke is very similar to both the overall Italian and European populations, as the comparison of crude and adjusted estimates of incidence rates clearly shows. 12 16 We found a strong relationship between stroke incidence and advancing age. This is consistent with the vast majority of epidemiological studies on stroke, 1 3,5,6,11,17 in which incidence progressively increases with age, particularly after 65 years, reaching a peak in individuals aged 85 years or more. This finding may have important implications for the burden of stroke in future years, because this is the fastest growing age group in industrialized countries. 38,39 Age is one of the most important predictors of unfavorable stroke outcomes, and the oldest average age of incident stroke will affect the statistics for stroke prognosis. 40,41 As observed in most of population-based studies including the Mediterranean area, 1,5,6 we found that incidence rates were higher for men than women. This difference was observed in all age groups and across the different pathological subtypes. Although this could be explained by underascertainment of stroke among women, this seems unlikely, as the difference in age-incidence rates between the two sexes was consistent across age groups, which would be unlikely if the referral process were different for men and women. A more likely explanation is that the difference in incidence between genders is real. Differences in certain lifestyle factors, such as smoking, could at least partially determine these results, and a lower frequency of smoking in women of the birth cohorts that are now at risk for stroke is usually observed in Southern Italy. The distribution and incidence of stroke subtypes observed in this study were similar to those observed by other studies, 1,5,6,11 with the exception of the higher incidence of intracerebral hemorrhage, which is similar to the results in the Greek study. 6 In our study, only 5% of stroke patients were not hospitalized, which is lower than reported in other studies from industrialized countries conducted in recent years. 5,17,19,30 Because of the widespread use of neuroimaging, a subtype diagnosis of stroke was reached in 94% of our cases. In this study the 28-day overall case-fatality of 18.1% was the lowest reported in Italy compared to 24% in Vibo Valentia, 25% in L Aquila, and 31% in Valle d Aosta, 5,17,20 and this good prognosis was consistently present both for cerebral infarction and hemorrhage. As recently demonstrated in the Dijon Stroke Registry, 32,42 the differences between current and previous stroke fatality rates found in studies conducted in the same geographical area could be related to better overall health in the population at large 42 as well as to improvements in the standard of care provided by stroke units. 43,44 Our epidemiological study presents several strengths. A stroke event was defined according to the established WHO criteria, 13,14 and only first-ever strokes were included in the

2926 Stroke November 2008 Table 2. Crude and Adjusted Stroke Incidence Rates Among European Populations Incidence for Strokes per 1000 Study (Reference) Study Period Crude Incidence for Strokes per 1000 Person-Years (95%CI) Person-Years Adjusted to the European Population (95%CI) Northern Europe Malmoe, Sweden (21) 1989 2.3 (na) na Oxfordshire, UK (31) 1981 2004 1.9 (1.7 2.1) na Innherred, Norway (24) 1994 1996 3.1 (2.8 3.4) 2.2 (na) Novosibirsk, Russia (27) 1982 1992 2 (na) 2.2 (na) Uzhgorod, west Ukraine (26) 1999 2000 2.8 (2.5 3.1) na Soderhamn, Sweden (25) 1987 1990 3.6 (3.3 4) 2.6 (2.4 2.9) Frederiksberg, Denmark (29) 1989 1990 3.1 (2.7 3.4) na South London, UK (36) 1995 1998 1.3 (1.2 1.5) 1.28 (1.2 1.35) Warzaw, Polland (37) 1991 1992 1.27 (1.1 1.5) 1.1 (1 1.3) Espo-Kauniainen, Finland (28) 1989 1991 2.2 (2.0 2.4) na Erlangen, Germany (30) 1994 1998 1.7 (1.6 1.9) 1.3 (1.2 1.4) Scottish Borders, UK (33) 1998 2000 2.8 (2.6 3.0) na Tblisi, Georgia (34) 2000 2003 1.7 (1.4 1.9) 1.5 (1.3 1.7) Orebro, Sweden (35) 1999 2000 3.1 (2.8 3.5) 2.5 (2.3 2.8) Tartu, Estonia (22) 2001 2003 2.2 (2 2.4) 1.9 (1.7 2.1) Southern Europe Dijon, France (32) 1985 2004 na 1 (0.9 1.1) Porto, Portugal (23) 1999 2000 2.8 (2.6 3) 1.8 (1.6 2) Umbria, Central Italy (18) 1986 1989 2.5 (2.3 2.8) 1.5 (1.4 1.8) Belluno, Northern Italy (19) 1992 1993 2.2 (1.7 2.3) 1.7 (na) Valle d Aosta, Northern Italy (20) 1989 2.2 (2 2.5) 2.1 (na) L Aquila, Central Italy (17) 1994 2.8 (2.6 2.9) 2.3 (na) Mediterranean area Aeolian Archipelago, Southern Italy (3) 1999 2002 1.5 (1.2 1.9) 1.5 (1.3 1.8) Vibo Valentia, Southern Italy (5) 1996 1.8 (1.6 2.0) 1.4 (1.2 1.5) Benghazi, Libya (7) 1984 0.6 (na) na Acquaviva-Casamassima, Southern Italy 2001 2002 1.6 (1.4 1.9) 1.5 (1.3 1.8) na indicates not available. analyses. The study was conducted in a well-defined geographic district in the Mediterranean area, using a prospective cohort design with a multisource population-based registry as the primary source of case ascertainment. Complete case ascertainment in our study is highly likely. First, the national health system in Italy is the only medical provider of stroke services, free of charge to all residents, regardless of age or socioeconomic status. Second, the study area contains only one hospital with a stroke unit, which is well known to the community. This prompts immediate and accurate detection with neuroimaging for almost all stroke cases that came to medical attention in the area. Compared to studies conducted in population-settings in several other European countries, the rate of hospitalization and use of neuroimaging for stroke in our study area is one of the highest reported. 1 Our study has limitations that are common to populationbased studies compared to clinical-based studies: the small sample size of the reference population, and the short period of observation (2 years). The low incidence found in our study may be attributable to protective factors related to lifestyle, such as diet. 2,10 The diet in the study area, especially in the age groups at high risk for stroke, includes olive oil, fruits, vegetables, and other fiber-containing foods that may be protective against stroke, particularly of the ischemic subtype. 45 50 It is possible that the Mediterranean diet, characterized by a high intake of antioxidants such as omega-3 fatty acids, oleic acid, polyunsaturated fatty acids, and vitamins, 45,47 50 at least partially explains the low incidence of stroke in this population, particularly of cerebral infarction, which is more closely associated with atherosclerosis. The presence of a permanent stroke surveillance system in the study area may help us better understand the temporal trends for stroke incidence and survivorship. If the best prognosis will be confirmed, the increased number of stroke survivors will be important for health care planning and for resource allocation to stroke services and long-term care in the region. Further studies are needed to determine the positive role of prompt referral and stroke units on the short-term prognosis

Manobianca et al Low Stroke Incidence in Southern Italy 2927 of incident strokes also in population-based setting similarly to what has been already well established in clinical setting. 43,44 Acknowledgments The authors thank Donald Halstead of Harvard School of Public Health for his critical comments on the manuscript. None. Disclosures References 1. Feigin VL, Lawes CM, Bennet DA, Anderson CS. Stroke epidemiology: a review of population-based studies of incidence, prevalence, and casefatality in the late 20th century: a non-communicable disease of increasing socioeconomic importance in ageing populations. Lancet Neurol. 2003;2:43 53. 2. Bejot Y, Benatru I, Rouaud O, Fromont A, Besancenot JP, Moreau T, Giroud M. Epidemiology of stroke in Europe: geographic and environmental differences. J Neurol Sci. 2007;262:85 88. 3. Musolino R, La Spina P, Serra S, Pastorino P, Calabro S, Savica R, Salemi G, Gallito G. First-ever stroke incidence and 30-day case fatality in the Sicilian Aeolian Archipelago, Italy. Stroke. 2005;36: 2738 2741. 4. Intiso D, Stampatore P, Zarrelli MM, Guerra GL, Arpaia G, Simone P, Tonali P, Beghi E. Incidence of first-ever ischemic and hemorrhagic stroke in a well-defined community of southern Italy, 1993 1995. Eur J Neurol. 2003;10:559 565. 5. Di Carlo A, Inzitari D, Galati F, Baldereschi M, Giunta V, Grillo G, Furchì A, Manno V, Naso F, Vecchio A, Consoli D. A prospective community based study (VISS). Methodology, incidence and case fatality at 28 days, 3 and 12 months. Cerebrovascular Dis. 2003;16: 410 417. 6. Vemmos KN, Bots ML, Tsibouris PK, Zis VP, Grobbee DE, Stranjalis GS, Stamatelopoulos S. Stroke incidence and case fatality in southern Greece. The Arcadia Stroke Registry. Stroke. 1999;30:363 370. 7. Ashok PP, Radhakrishnan K, Sridharan R, El-Mangoush MA. Incidence and pattern of cerebrovascular disease in Benghazy, Libya. J Neurol Neurosurg Psychiatry. 1986;49:519 523. 8. Malmgren R, Warlow C, Bamford J, Sandercock P. Geographical and secular trends in stroke incidence. Lancet. 1987;2:1196 1200. 9. Bonita R, Broad JB, Anderson NE, Beaglehole R. Approaches to the problems of measuring the incidence of stroke; the Auckland Stroke Study, 1991 1992. Int J Epidemiol. 1995;524:535 542. 10. Sudlow CL, Warlow CP. Comparing stroke incidence worldwide: what makes studies comparable? Stroke. 1996;27:550 558. 11. Sudlow CL, Warlow CP. Comparable studies of the incidence of stroke and its pathological types: results from an international collaboration. Stroke. 1997;28:491 499. 12. Statistica IId. demo.istat.it/pop1/start.html. Istituo Italiano di Statistica [Web Site]. Available at: demo.istat.it/pop1/start.html. 13. World Health Organization: manual of the international statistical classification of diseases, injuries and causes of death, 9th revision. Geneva, WHO, 1997. 14. Aho K, Harmsen P, Hatano S, Marquardsen J, Smirnov VE, Strasser T. Cerebrovascular disease in the community: results of WHO collaborative study. Bull World Health Organ. 1980;58:113 130. 15. Woodward M. Epidemiology: study design and data analysis. Boca Raton, FL: Chapman & Hall/CRC Press; 1999:159 167. 16. Population Division of the Department of Economic and Social Affairs of the United Nations Secretariat, World Population Prospects: The 2006 Revision and World Urbanization Prospects: The 2005 Revision, http:// esa.un.org/unpp. 17. Carolei A, Marini C, Di Napoli M, Di Gianfilippo C, Santalucia P, Baldassarre M, De Matteis G, Di Orio F. High stroke incidence in the prospective community-based L Aquila Registry (1994 1998). First year s results. Stroke. 1997;28:2005 2006. 18. Ricci S, Celani MG, La Rosa F, Vitali R, Duca E, Ferraguzzi R, Poalotti M, Seppoloni D, Caputo N, Chiurulla C, Scaroni R, Signorini E, Chiurulla C, Scaroni R, Signorini E, Sapiva C. A community-based study for stroke incidence in Umbria, Italy. J Neurol Neurosurg Psychiatry. 1991;54:695 698. 19. Lauria G, Gentile M, Fassetta G, Casetta I, Agnoli F, Andreotta G, Caneve G, Cavallaro A, Cielo R, Mongillo D, Mosca M, Olivieri P. Incidence and prognosis of stroke in the Belluno Province, Italy: first-year results of a community-based study. Stroke. 1995;26: 1787 1793. 20. D Alessandro G, Di Giovanni M, Roveyaz L, Iannizzi L, Compagnoni MP, Blanc S, Bottacchi E. Incidence and prognosis of stroke in the Valle d Aosta, Italy: first-year results of a community-based study. Stroke. 1992;23:1712 1725. 21. Jerntorp P, Berglund G. Stroke registry in Malmoe Sweden. Stroke. 1992;23:357 361. 22. Vibo R, Korv J, Roose M. The third stroke registry in Tartu, Estonia, from 2001 to 2003. Acta Neurol Scand. 2007;116:31 36. 23. Correia M, Silva MR, Matos I, Magalhaes R, Lopes JC, Ferro JM, Silva MC. Prospective community-based study of stroke in northern Portugal: incidence and case fatality in rural and urban populations. Stroke. 2004; 35:2048 2053. 24. Ellekjær H, Holmen J, Indredavik B, Terent A. Epidemiology of stroke in Innherred, Norway, 1994 to 1996 incidence and 30-day case-fatality rate. Stroke. 1997;28:2180 2184. 25. Terént A. Trends in stroke incidence and 10-year survival in Söderhamn, Sweden, 1975 2001. Stroke. 2003;34:1353 1358. 26. Mihalka L, Smolanka V, Bulecza B, Mulesa S, Bereczki D. A population-based study of stroke in West Ukraine. Incidence, Stroke services and 30-day case fatality. Stroke. 2001;32:2227 2231. 27. Feigin VL, Wiebers DO, Whisnant JP, O Fallon M. Stroke incidence and 30-day case-fatality rates in Novosibirsk, Russia, 1982 Through 1992. Stroke. 1995;26:924 929. 28. Numminen H, Kotila M, Waltimo O, Aho K, Kaste M. Declining incidence and mortality rates of stroke in Finland from 1972 to 1991 results of three population-based stroke registers. Stroke. 1996;27: 1487 1491. 29. Jorgensen HS, Plesner AM, Hubbe P, Larsen K. Marked increase of stroke incidence in men between 1972 and 1990 in Frederiksberg, Denmark. Stroke. 1992;23:1701 1704. 30. Kolominsky-Rabas PL, Sarti C, Heuschman PU, Graf C, Siemonsen S, Neundoerfer B, Katalinic A, Lang E, Gassmam KG, von Stockert PT. A prospective community-based of stroke in Germany. The Erlangen Stroke Project (ESPRO): incidence of case fatality at 1,3 and 12 months. Stroke. 1998;29:2501 2506. 31. Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM, et al. Change in stroke incidence, mortality, case-fatality, severity and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet. 2004;363:1925 1933. 32. Benatru I, Rouaud O, Durier J, Contegal F, Couvreur G, Bejot Y, Osseby GV, Ben Salem D, Ricolfi F, Moreau T, Giroud M. Stable stroke incidence rates but improved case-fatality in Dijon, France, from 1985 to 2004. Stroke. 2006;37:1674 1679. 33. Syme PD, Byrne AW, Chen R, Devenny R, Forbes JF. Community-based stroke incidence in a Scottish population: the Scottish Borders Stroke Study. Stroke. 2005;36:1837 1843. 34. Tsiskaridze A, Djibuti M, van Melle G, Lomidze G, Apridonidze S, Gauarashvili I, Piechowski-Józwiak B, Shakarishvili R, Bogousslavsky J. Stroke incidence and 30-day case-fatality in a suburb of Tbilisi: results of the first prospective population-based study in Georgia. Stroke. 2004;35: 2523 2528. 35. Appelros P, Nydevik I, Seiger A, Terént A. High incidence rates of stroke in Orebro, Sweden: Further support for regional incidence differences within Scandinavia. Cerebrovasc Dis. 2002;14:161 168. 36. Wolfe CD, Rudd AG, Howard R, Coshall C, Stewart J, Lawrence E, Hajat C, Hillen T. Incidence and case fatality rates of stroke subtypes in a multiethnic population: the South London Stroke Register. J Neurol Neurosurg Psychiatry. 2002;72:211 216. 37. Czlonkowska A, Ryglewicz D, Weissbein T, Baranska-Gieruszczak M, Hier DB. A prospective community-based study of stroke in Warsaw, Poland. Stroke. 1994;25:547 551. 38. The World Bank: world development report 1993, New York, Oxford University Press, 1993. 39. Murray CJL, Lopez AD. Evidence-based health policy. Lessons from the Global burden of disease study. Science. 1996;274:740 743. 40. Warburton E. Stroke management. Clin Evid. 2004;12:236 252.

2928 Stroke November 2008 41. Nakayama H, Jorgensen HS, Raaschou HO, Olsen TS. The influence of age on stroke outcome. The Copenhagen stroke study. Stroke. 1994;25: 808 813. 42. Bejot Y, Rouaud O, Durier J, Caillier M, Marie C, Freysz M, Yeguiayan JM, Chantegret A, Osseby G, Moreau T, Giroud M. Decrease in the stroke case fatality rates in a French population-based twenty-year study. A comparison between men and women. Cerebrovasc Dis. 2007;24: 439 444. 43. Rudd AG, Hoffman A, Irwin P, Lowe D, Pearson MG. Stroke unit care and outcome: results from the 2001 National Sentinel Audit of Stroke (England, Wales, and Northern Ireland). Stroke. 2005;36:103 106. 44. Stroke Unit Trialists Collaboration: Collaborative systematic review of the randomised trials of organised inpatient (stroke unit) care after stroke. BMJ. 1997;314:1151 1159. 45. Joshipura KJ, Ascherio A, Manson JE, Stampfer MJ, Rimm EB, Speizer FE, Hennekens CH, Spiegelman D, Willet WC. Fruit and vegetables intake in relation to risk of ischemic stroke. JAMA. 1999; 282:1233 1239. 46. Trichopoulou A, Lagiou P. Healthy traditional Mediterranean diet: an expression of culture, history, and lifestyle. Nutr Rev. 1997;55: 383 389. 47. Katsouyanni K, Skalkidis Y, Petridou E, Polychronopoulou- Trichopoulou A, Willet W, Trichopoulos D. Diet and peripheral arterial occlusive disease: the role of poly-, mono- and saturated fatty acids. Am J Epidemiol. 1991;133:24 31. 48. De Lorgeril M, Salen P, Martin JL, Mamelle N, Moujaud I, Touboul P, Delaye J. Effect of a mediterranian type of diet on the rate of cardiovascular complications in patients with coronary artery disease. Insights into the cardioprotective effect of certain nutriments. J Am Coll Cardiol. 1996;28:1103 1108. 49. He K, Song Y, Daviglus ML, Liu K, Van Horn L, Dyer AR, Goldbourt U, Greenland P. Fish consumption and incidence of stroke. A metaanaylisis of cohort studies. Stroke. 2004;35:1538 1542. 50. Bernal-Pacheco O, Romàn GC. Environmental vascular risk factors: New perspectives for stroke prevention. J Neurol Sci. 2007;262:60 70.