- Email: [email protected]
Epidemiologic Trends in Hospitalized Ischemic Stroke from 2002 to 2010: Results from a Large Italian Population-Based Study
Epidemiologic Trends in Hospitalized Ischemic Stroke from 2002 to 2010: Results from a Large Italian Population-Based Study Paola Santalucia, MD,*† Marta Baviera, MSc,‡ Laura Cortesi, MSc,‡ Mauro Tettamanti, MSc,x Irene Marzona, MSc,‡ Alessandro Nobili, MD,* Emma Riva, MD,x Ida Fortino, MSc,jj Angela Bortolotti, MSc,jj Luca Merlino, MD,jj and Maria Carla Roncaglioni, MSc‡
Background: To describe the incidence of ischemic stroke, short-term mortality, recurrences, and prescription patterns. Methods: Data from administrative health databases of the Lombardy Region from 2002 to 2010 (about 4 million people) were analyzed for stroke incidence and recurrence, mortality, and drug prescriptions after an ischemic stroke. Results: A total of 43,352 patients with a first hospital admission for ischemic stroke were identified. During 8 years, stroke incidence decreased from 3.2 of 1000 to 2.4 of 1000 (P , .001) in people aged 65-74 years, from 7.1 of 1000 to 5.3 of 1000 (P , .001) at ages 75-84 years and from 11.9 of 1000 to 9.4 of 1000 (P , .001) at age 85 years or older. Stroke recurrences dropped by 30% (from 10.0% to 7.0%, P , .001) and 30-day mortality rate also decreased. Prescription trends showed linear increase in antiplatelets and lipid-lowering drugs, respectively, from 60.2% to 65.0% (P , .001) and from 19.1% to 34.6% (P , .001), whereas antihypertensive prescriptions did not change appreciably. Anticoagulant prescription increased in patients with atrial fibrillation, from 64.8% to 72.1% in the 65-74 years age group, (P 5 .004) and from 40.2% to 53.7% in the 75-84 years age group (P , .001); less than 20% of the 85 years or older age group were treated with anticoagulants (P , .0001). Conclusions: Stroke incidence, recurrence, and 30-day mortality decreased from 2002 to 2010 concomitant with an increase in prescriptions of secondary stroke prevention drugs. Key Words: Stroke incidence— recurrences—mortality—secondary stroke prevention—administrative database. Ó 2015 by National Stroke Association
Stroke is a major public health problem worldwide, causing about 6 million deaths each year.1 Stroke is also considered the main cause of disability-adjusted life years with a total annual cost of 27 billion Euros.2,3 The past 2 decades showed a worldwide stroke incidence, and
mortality rates decrease in high-income countries, whereas it has been observed an increase in middle-tolow income countries.4,5 Age and aging are expected to be the most significant population factors affecting the epidemiologic
From the *Laboratory for Quality Assessment of Geriatric Therapies and Services, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; †Scientific Direction and Emergency Medicine Department, Fondazione IRCCS Ca Granda Ospedale Maggiore Policlinico, Milan; ‡Laboratory of General Practice Research, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; xLaboratory of Geriatric Neuropsychiatry, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; and jjRegional Health Ministry, Milan, Italy.
Received January 8, 2015; revision received March 10, 2015; accepted May 6, 2015. There are no conflicts of interest. Address correspondence to Paola Santalucia, MD, IRCCS–Istituto di Ricerche Farmacologiche Mario Negri, Via Giuseppe La Masa 19, Milan 20156, Italy. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.05.008
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2015: pp 1-7
1
P. SANTALUCIA ET AL.
2
facts and figures of stroke. In Europe, Italy is one of the countries with the highest percentage of people aged older than 65 years, amounting to about 20% of the whole population. By 2020, assuming stable incidence rates, 195,000 new stroke cases per year are expected because of age.3 Annual stroke incidence rates, standardized to the Italian population, range from 175 of 100,000 to 360 of 100,000 in men and from 130 of 100,000 to 273 of 100,000 in women; 30-day mortality for all strokes ranges from 18.1% to 33.0% and 1-year mortality rates from 37.9% to 40.2%.6 This epidemic and economic burden has, over the years, fostered joint action by scientific societies and public institutions to promote health policies and facilitate risk factor control and stroke prevention campaigns.7-9 Secondary stroke prevention includes conventional lifestyle and pharmacologic approaches as well as more specific interventions such as anticoagulation for prevention of cardioembolic stroke due to atrial fibrillation.10,11 The aim of this study was to describe the incidence of ischemic stroke and short-term mortality, including prescription trends in relation to recurrences, in a population of more than 4 million people yearly from 2002 to 2010. Population-based administrative database was used as a source of reliable data for estimating stroke epidemiology and prescription patterns.
Method Data Source We used the linkable administrative health databases of Lombardy, which include a population registry with the demographic data of all residents and detailed information on drug prescriptions and hospital admissions. Data were available from 2000 to 2010 for more than 4 million inhabitants per year. Lombardy is the most heavily populated Italian region, comprising urban, industrial, and rural areas. Health care in Italy is publicly funded for all residents, irrespective of social class or employment, and everyone is assigned a personal identification number kept in the National Civil Registration System. Citizens are cared for by general practitioners as part of the National Health System. The pharmacy prescription database contains the medication name and anatomic therapeutic chemical classification code, quantity, and dispensation date. The hospital discharge database records the date of hospital admission, date of discharge or death, diagnosis, and procedures done. All these databases are linked anonymously with the demographic data using unique encrypted patient codes, in accordance with Italian laws for the treatment of confidential data. Approval from an ethics committee is not required to analyze encrypted administrative data.
Study Population Patients with ischemic stroke were identified as new cases (incidence cases) if they were discharged with a diagnosis of an acute cerebrovascular event (primary diagnosis International Classification of Disease, Ninth Revision [ICD-9] code 434.xx, 436) between 1 January, 2002, and 31 December, 2009, with no previous stroke diagnosis in the 2 years before entering the cohort.
Stroke Incidence, Recurrence, and Mortality Stroke incidence figures, according to age, were calculated for 2002-2009. Hospital admissions for stroke recurrences, according to discharge code of acute cerebrovascular disease, were identified in the year after the first event. Thirty-day and mortality rates were also recorded.
Pharmacologic Treatments Prescription rates and trends for antihypertensive drugs (C02, C03, C07, C08, and C09), lipid-lowering drugs (C10), antiplatelets (B01AC), and anticoagulant drugs (B01AA03 and B01AA07) were taken into account. Data on prescriptions of at least one of the previously mentioned classes of drugs during the year before stroke were considered for the description of the patients characteristics at baseline. Prescriptions for secondary prevention were considered in the year after the first hospital admission for stroke.
Statistical Analysis Variables were classified as follows. Patients were grouped by age (0-64, 65-74, 75-84, and $85 years) taking the largest group as reference (65-74 years). Males were the reference group for sex. Total numbers of drugs except antihypertensives, antiplatelet drugs, anticoagulants, and lipid-lowering drugs were grouped in classes with similar frequencies; the reference group was the group that took fewer than 2 concomitant drugs (as a proxy for the healthiest subjects). Patients were considered censored at death (except for analyses in which death was the outcome), emigration, or admission to a nursing home. Atrial fibrillation (AF, ICD-9 code 42731, 42732), myocardial infarction (ICD-9 code 410.xx), and heart failure (ICD-9 code 428.x, 7855.x) referred to hospital admission for these causes in the 2 years preceding the stroke event. Diabetes was defined if the patient was treated with antidiabetic drugs (Anatomic Therapeutic Chemical classification code A10), hospitalized with diagnostic code of diabetes (ICD-9 code 250.xx), or recorded in the exemption registry for diabetes. The crude annual stroke incidence was calculated from the number of patients discharged with a stroke diagnosis and the personyears contributed by the residents, per 1000 person-years, by age group. Stroke recurrences were calculated on patients discharged alive after a stroke. Crude mortality rate at 30 days was the number of patients who died as a
STROKE TRENDS FROM 2002 TO 2010
3
Table 1. Characteristics of patients with ischemic stroke according to sex Total no. 43,352 Mean age 6 SD, y 75.2 6 12.5 Age classes (y) 0-64 7298 65-74 10,350 75-84 16,067 $85 9637 Treatments Antihypertensives* 30,953 Antiplatelet drugs* 18,014 Anticoagulants* 3136 Lipid-lowering drugs* 7214 Number of other treatmentsy 0-1 11,718 2-3 11,086 4-5 8281 6-7 5267 81 7000 Atrial fibrillation 6539 Diabetes 9451 Myocardial infarction 1104 Heart failure 3308
% 100
Females
%
Males
%
22,089 78.0 6 12.2
50.9
21,263 72.3 6 12.1
49.0
P value (F versus M) ,.0001
16.8 23.9 37.1 22.2
2485 4061 8751 6792
11.2 18.4 39.6 30.7
4813 6289 7316 2845
22.6 29.6 34.4 13.4
,.0001
71.4 41.5 7.2 16.6
16,680 8967 1671 3247
75.5 40.6 7.6 14.7
14,273 9047 1465 3967
67.1 42.5 6.9 18.7
,.0001 ,.0001 ,.0067 ,.0001
27.0 25.6 19.1 12.1 16.1 15.1 21.8 2.5 7.6
5178 5556 4394 2960 4001 4090 4464 531 1928
23.4 25.1 19.9 13.4 18.1 18.5 20.2 2.4 8.7
6540 5530 3887 2307 2999 2449 4987 573 1380
30.8 26.0 18.3 10.8 14.1 11.5 23.4 2.7 6.5
,.0001
,.0001 ,.0001 .0546 ,.0001
Abbreviations: F, female; M, male; SD, standard deviation. *In the year before entering the study cohort. yExcluding antihypertensives, antiplatelet drugs, anticoagulants, and lipid-lowering drugs.
percentage of the total number with stroke. Prescription rates referred to the year after the stroke event. Characteristics of males and females were compared using the chisquare test. The relationship between calendar year (treated as continuous variable) and stroke incidence, stroke recurrences, 30-day mortality, and each drug treatment were analyzed using multivariable logistic regression models adjusting for sex, age class, and number of concomitant medications. We also repeated the analyses using dummy year variables to explore possible nonlinearities in the trends. Since most subjects were present multiple times during the years of observation, a clustered sandwich estimator was used to correct for nonindependence among data. The results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical analyses were done using STATA 12.0 (StataCorp LP, College Station, TX) and JMP 11.1 (SAS Institute Inc, Cary, NC).
men (78.0 6 12.2 versus 72.3 6 12.1, P , .001) and consistently more in the 851 age group (30.7% versus 13.4%, P , .001). Among major stroke risk factors, AF at the time of first cerebrovascular event was present in 15.1% of the study cohort, more frequent in women than men, diabetes in 21,8%, more frequent in men than women, history of myocardial infarction in 2.5% equally represented between sexes, and heart failure in 7.6% of the population, more frequent in women than men. During the year before the index event, 71.4% of the subjects were taking antihypertensive drugs, 41.5% were on antiplatelets, 7.2% on anticoagulant therapies, and 16.6% subjects were on lipid-lowering therapy. Figure 1 shows the stroke incidence between 2002 and 2009, by age the overall incidence decreased from 1.4 of
Results Out of a population of more than 4 million per year, 43,352 patients, 51% women and 49% men, were hospitalized for ischemic stroke between 2002 and 2009. Table 1 shows the main demographic characteristics, risk factors, and pharmacologic profile at baseline of all patients with ischemic stroke. Mean age was 75.2 years (standard deviation, 612.5), women were older than
Figure 1.
Incidence of ischemic stroke, by age.
P. SANTALUCIA ET AL.
4
Figure 2. Recurrence of ischemic stroke in the year after first event. Recurrence rate refers to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Figure 4. Prescriptions of recommended drugs to all patients with ischemic stroke in the year after the first event. Prescription rates refer to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
apy (Fig 5), with an overall increase of 24% (from 35.3% to 43.9%; P , .0001). There was a 13% rise, although not statistically significant (from 80.6% to 91.2%, P 5 .336), in the age group up to 64 years, 11% (from 64.8% to 72.1%, P 5 .004) in the 65-74 year olds and 30% in the 75-84 years class (from 40.2% to 53.7%, P , .0001). The prescription rate of anticoagulants in patients 851 years was 10.8% in 2002 to 17.3% in 2010 (P , .0001). Multivariable analysis of prescriptions of recommended drugs showed a 3% yearly increase for antihypertensives (OR, 1.03; 95% CI, 1.02-1.04), 5% for antiplatelets (OR, 1.05; 95% CI, 1.04-1.06), 4% for anticoagulants (OR, 1.04; 95% CI, 1.03-1.06), and 14% for lipid-lowering drugs (OR, 1.14; 95% CI, 1.13-1.16).
1000 to 1.1 of 1000 (P ,.0001) the incidence fell from 3.2 of 1000 to 2.4 of 1000 (P , .001) in the 65-74 years age group, from 7.1 of 1000 to 5.3 of 1000 (P , .001) in those aged 75-84 years, and from 11.9 of 1000 to 9.4 of 1000 (P , .001) in age 85 years and older group. Multivariable analysis showed a 3% yearly reduction (OR, .97; 95% CI, .97-.98). The 1-year stroke recurrence rate fell 30% (from 10.0% to 7.0%, [P , .001] between 2002 and 2010 [Fig 2]). Multivariable analysis indicated a 6% yearly reduction (OR, .94; 95% CI, .92-.95). The 30-day mortality declined from 13.3 of 1000 to 12.1 of 1000 (P , .0001; Fig 3). The reduction was larger in 75-84 and 85 years or older patients, although between 2002 and 2010 there was a slight decline in all age classes. Multivariable analysis indicated 5% reduction (OR, .95; 95% CI, .94-.97). Figure 4 shows prescription patterns in the years after the index event for the drug classes most recommended for secondary stroke prevention. Between 2002 and 2010, the increase was linear for lipid-lowering and antiplatelet prescriptions, respectively, from 19.1% to 34.6% (P ,.001) and from 60.2% to 65.0% (P ,.001). The proportion of patients using antihypertensive therapies also rose slightly, from 65.8% to 68.3%. The prescription rate of anticoagulants rose from 11.8% to 14.1% (P , .001). Patients with AF were more likely to receive anticoagulant ther-
This study showed a significant linear decrease of stroke incidence and recurrence rates in this large cohort of patients over a decade of observation. There was also a positive trend in prescription rates of the main drugs recommended for secondary stroke prevention. Over the past decade, a decline has been reported in stroke incidence and its sequelae, probably because of the overall improvement of control and care of risk factors and better management of acute stroke patient.12-14 Stroke incidence rates in Italy are known mainly from epidemiologic registries in different regions from north to south.6 Three
Figure 3. Thirty-day mortality rate, by age. Mortality rate refers to the thirty days after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Figure 5. Prescriptions of anticoagulants to patients with stroke and atrial fibrillation in the year after the first event. Prescription rates refer to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Discussion
STROKE TRENDS FROM 2002 TO 2010
registers reported data valuable for examining stroke incidence trends.6 The L’Aquila register, the largest one including over 4000 patients, based on a 5-year data set, showed a mean rise in annual stroke incidence of 4.28% mainly due to the higher incidence in women, particularly those aged 851 years. Disparities between our results and epidemiologic data on stroke incidence from Italian registers might be variously explained. Possible explanations could be that Italian registers have been conducted in limited geographic areas not representative of the general Italian population and in different periods, the decade before our years of observation. Besides, there are substantial differences in patients number, smaller in the registers, although direct comparisons are not feasible. Despite differences in the Italian context, improvements in lifestyle and pharmacologic approaches to the management of stroke patients in primary and secondary prevention have led to an overall improvement in epidemiologic trends. Furthermore, in Western countries, there has been a short-term reduction in mortality after stroke, very likely due to better care in the acute setting.4,5,15-19 Our analysis too showed a decline in 30-day mortality after acute ischemic stroke, reflecting the development in the last decade in Italy and across all Europe of standardized acute stroke care based on dedicated stroke units20 applying specific acute stroke treatments.21,22 Although there are no systematic data on stroke recurrence trends, the decline over the years may possibly be because of generally closer adherence to guidelines and increased prescription of drugs recommended for secondary stroke prevention.23 Again, our analysis showed a decline in stroke recurrences from 2002 to 2010 and linear increases in the drug classes most recommended for secondary stroke prevention, mainly antiplatelet and lipid-lowering therapies. The Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS) trial provided additional evidence of the efficacy and safety of antiplatelet molecules other than aspirin alone in long-term treatment.24 The Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial contributed to the prescription of statins for secondary stroke prevention regardless of the patient’s lipid profile.25 In our cohort, prescription rates and patterns of oral anticoagulant therapy for secondary prevention of stroke of cardioembolic origin showed a well-known age-related distribution. Prescriptions tended to increase in patients up to age 74 years, whereas less than 20% of those in the 851 group were prescribed anticoagulants. People aged 801 years have the highest prevalence of AF in the general population and are therefore those who would benefit most from anticoagulant treatment on account of the high risk of peripheral thromboembolism and cardioembolic stroke.26 However, they are also the most fragile age group in terms of potential bleeding risk.27,28 This represents one of the reason for the low prescription rate of anticoagulants in elderly despite
5 29
evidence-based efficacy and guideline indications. The large trials on secondary prevention of cardioembolic stroke have shown the superiority of warfarin over aspirin,30,31 and furthermore, in elderly patients, aspirin has a harmful profile because of progressively reduced efficacy along with an increased bleeding risk.32 Dualantiplatelet therapy in patients unsuitable for anticoagulant therapy, although superior to aspirin alone, also carries a higher bleeding risk.33 Our analysis covers the years before approval of direct oral anticoagulants in clinical practice that, at least as effective as warfarin and with a favorable safety profile, with no need for frequent anticoagulation monitoring, low potential for food and drug interactions, and fast onset and offset of action, have already influenced prescription patterns, also in elderly people.11 In addition, closer attention to the selection of candidates for anticoagulant therapy34 and better adherence to guidelines has helped boost physician’s awareness in the field of thromboembolism prevention in patients with AF. Our cohort is strictly representative of a specific area, as the data come from administrative databases for a single Italian region, and the results of our analysis can be considered to correspond to a uniform standard of care. The biggest limitation of analyses conducted on administrative data is the unavailability of specific clinical data so that some variables cannot be taken into account. Nevertheless, information on demographics, diagnostic procedures, admission and discharge dates, length of hospital stay, and destination at discharge are all included in these data sets. Data on all drugs, subject to prescription, primary, and refills, are also available and can be used to observe prescription trends. Besides, administrative data are feasible, timely and inexpensive, and involve a large number of unselected people with wide coverage and continuity. The availability of large temporal series makes it possible to study long-term outcomes and epidemiologic trends. Because stroke is a hospital-managed disease, administrative health databases may be a valuable source of information for epidemiologic evaluations in a large setting of unselected patients. Potential benefits of administrative health data sets analysis focused on diseases at high epidemiologic impact, such as stroke, is the generalizability of results to the real world and the possibility of evaluating any change in the health care and prescriptions trends and how they correlate with clinical outcomes.
Conclusion Our study offers a systematic overview of stroke epidemiologic patterns over the last decade in a large area of northern Italy. The analysis showed a trend of decrease of stroke, recurrence, and mortality and concurrent increase in prescriptions of the most recommended drugs for secondary stroke prevention. We also observed,
P. SANTALUCIA ET AL.
6
according with previous studies, a reduction of stroke incidence. Although stroke incidence trends could be expected to arise due to the ageing of the population worldwide, however, improvements in global stroke care have certainly contributed and will very likely further reduce the overall burden of stroke. Acknowledgment: The authors would like to thank Igor Monti from IRCCS–Istituto di Ricerche Farmacologiche Mario Negri, Carolina Vastola, Simone Schiatti and Giovanna Rigotti from LombardiaInformatica S.p.A, and Alfredo Bevilacqua from SANTER Reply S.p.A for their assistance in data acquisition and programming support. The authors are grateful to Guya Sgaroni and Fiorenza Clerici for secretarial assistance and to J.D. Baggott for kindly editing the article.
References 1. Fergin VL, Forouzanfar MH, Krishnamurthi R, et al. Global and regional burden of stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet 2014;383:245-255. 2. Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol 2009;8:1006-1018. 3. Di Carlo A. Human and economic burden of stroke. Age and Ageing 2009;38:4-5. 4. Feigin VL, Lawes CM, Bennett DA, et al. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 2009;8:355-369. 5. Krishnamurthi RV, Feigin VL, Forouzanfar MH, et al. Global and regional burden of first-ever ischaemic and haemorrhagic stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet Glob Health 2013;1:e259-e281. 6. Sacco S, Stracci F, Cerone D, et al. Epidemiology of stroke in Italy. Int J Stroke 2011;6:219-227. 7. Goldstein LB, Bushnell CD, Adams R, et al. Guidelines for the primary prevention of stroke. A guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42: 517-584. 8. SPREAD 2012 – Stroke prevention and educational awareness diffusion. Ictus cerebrale: Linee guida italiane. Available at www.siapav.it/pdf/SPREAD%202012.pdf. 9. Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working toward a prioritized world agenda. Int J Stroke 2010;5:238-256. 10. Kernan WN, Ovbiagele B, Black HR, et al, on behalf of the American Heart Association Stroke Council, Council on Cardiovascular and Stroke Nursing, Council on Clinical Cardiology, and Council on Peripheral Vascular Disease. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45: 2160-2236. 11. Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation. An update of the 2010 ESC Guidelines for the management of atrial fibrillation. Eur Heart J 2012;33:2719-2747.
12. Lee S, Shafe AC, Cowie MR. UK stroke incidence, mortality and cardiovascular risk management 1999-2008: time-trend analysis from the General Practice Research Database. BMJ Open 2011;1:e000269. http://dx.doi.org/ 10.1136/bmjopen-2011-000269. 13. Allen NB, Holford TR, Bracken MB, et al. Trends in oneyear recurrent ischemic stroke among the elderly in the USA: 1994-2002. Cerebrovasc Dis 2010;30:525-532. 14. Wieberdink RG, Ikram MA, Hofman A, et al. Trends in stroke incidence rates and stroke risk factors in Rotterdam, the Netherlands from 1990 to 2008. Eur J Epidemiol 2012;27:287-295. 15. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev 2013;9:CD000197. http://dx.doi.org/10.1002/14651858.CD000197.pub3. 16. Meretoja A, Roine RO, Kaste M, et al. Effectiveness of primary and comprehensive stroke centers PERFECT stroke: a nationwide observational study from Finland. Stroke 2010;41:1102-1107. 17. Nimptsch U, Mansky T. Stroke unit care trends of in-hospital mortality for stroke in Germany 2005-2010. Int J Stroke 2014;9:260-265. 18. Schmidt M, Jacobsen JB, Johnsen SP, Bøtker HE, Sørensen HT. Eighteen-year trends in stroke mortality and the prognostic influence of comorbidity. Neurology 2014;82:340-350. 19. Lackland DT, Roccella EJ, Deutsch AF, et al. Factors influencing the decline in stroke mortality. A statement from the American Heart Association/American Stroke Association. Stroke 2014;45:315-353. 20. Candelise L, Gattinoni M, Bersano A, et al, on the behalf of the PROSIT Study Group. Stroke unit care for acute stroke patients: an observational follow-up study. Lancet 2007;369:299-305. 21. The NINDS rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke r-tPA Stroke Study Group. N Engl J Med 1995;333:1581-1587. 22. Del Zoppo GJ, Saver JL, Jauch EC, et al. Expansion of the time window for treatment of acute ischemic stroke with intravenous tissue plasminogen activator: a science advisory from the American Heart Association/American Stroke Association. Stroke 2009;40:2945-2948. 23. Kernan WN, Ovbiagele B, Black HR, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guidelines for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45:2160-2236. 24. Sacco RL, Diener HC, Yusuf S, et al, for the PRoFESS Study Group. Aspirin and extended-release dipyridamole versus clopidogrel for recurrent stroke. N Engl J Med 2008;359:1238-1251. 25. Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) Investigators. High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med 2006;355:549-559. 26. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a major contributor to stroke in the elderly: the Framingham Study. Arch Intern Med 1987;147:1561-1564. 27. Lip GY, Andreotti F, Fauchier L, et al. Bleeding risk assessment and management in atrial fibrillation patients: a position document from the European Heart Rhythm Association, endorsed by the European Society of Cardiology Working Group on Thrombosis. Europace 2011;13:723-746. 28. Poli D, Antonucci E, Testa S, et al. Bleeding risk in very old patients on vitamin K antagonist treatment: results
STROKE TRENDS FROM 2002 TO 2010 of a prospective collaborative study on elderly patients followed by Italian Centres for Anticoagulation. Circulation 2011;124:824-829. 29. Bungard TJ, Ghali WA, Teo KK, et al. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000;160:41-46. 30. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857-867. 31. Go AS, Hylek EM, Chang Y, et al. Anticoagulant therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical trial? JAMA 2003;290:2685-2692.
7 32. Olesen JB, Lip GY, Lindhardsen J, et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: a net clinical benefit using a ‘‘real world’’ nationwide cohort study. Thromb Haemost 2011;106: 739-749. 33. ACTIVE Investigators. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 2009;360:2066-2078. 34. Lip GY, Nieuwlaat R, Pisters R, et al. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the EuroHeart Survey on atrial fibrillation. Chest 2010;137:263-272.
Background: To describe the incidence of ischemic stroke, short-term mortality, recurrences, and prescription patterns. Methods: Data from administrative health databases of the Lombardy Region from 2002 to 2010 (about 4 million people) were analyzed for stroke incidence and recurrence, mortality, and drug prescriptions after an ischemic stroke. Results: A total of 43,352 patients with a first hospital admission for ischemic stroke were identified. During 8 years, stroke incidence decreased from 3.2 of 1000 to 2.4 of 1000 (P , .001) in people aged 65-74 years, from 7.1 of 1000 to 5.3 of 1000 (P , .001) at ages 75-84 years and from 11.9 of 1000 to 9.4 of 1000 (P , .001) at age 85 years or older. Stroke recurrences dropped by 30% (from 10.0% to 7.0%, P , .001) and 30-day mortality rate also decreased. Prescription trends showed linear increase in antiplatelets and lipid-lowering drugs, respectively, from 60.2% to 65.0% (P , .001) and from 19.1% to 34.6% (P , .001), whereas antihypertensive prescriptions did not change appreciably. Anticoagulant prescription increased in patients with atrial fibrillation, from 64.8% to 72.1% in the 65-74 years age group, (P 5 .004) and from 40.2% to 53.7% in the 75-84 years age group (P , .001); less than 20% of the 85 years or older age group were treated with anticoagulants (P , .0001). Conclusions: Stroke incidence, recurrence, and 30-day mortality decreased from 2002 to 2010 concomitant with an increase in prescriptions of secondary stroke prevention drugs. Key Words: Stroke incidence— recurrences—mortality—secondary stroke prevention—administrative database. Ó 2015 by National Stroke Association
Stroke is a major public health problem worldwide, causing about 6 million deaths each year.1 Stroke is also considered the main cause of disability-adjusted life years with a total annual cost of 27 billion Euros.2,3 The past 2 decades showed a worldwide stroke incidence, and
mortality rates decrease in high-income countries, whereas it has been observed an increase in middle-tolow income countries.4,5 Age and aging are expected to be the most significant population factors affecting the epidemiologic
From the *Laboratory for Quality Assessment of Geriatric Therapies and Services, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; †Scientific Direction and Emergency Medicine Department, Fondazione IRCCS Ca Granda Ospedale Maggiore Policlinico, Milan; ‡Laboratory of General Practice Research, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; xLaboratory of Geriatric Neuropsychiatry, IRCCS-Istituto di Ricerche Farmacologiche Mario Negri, Milan; and jjRegional Health Ministry, Milan, Italy.
Received January 8, 2015; revision received March 10, 2015; accepted May 6, 2015. There are no conflicts of interest. Address correspondence to Paola Santalucia, MD, IRCCS–Istituto di Ricerche Farmacologiche Mario Negri, Via Giuseppe La Masa 19, Milan 20156, Italy. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.05.008
Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2015: pp 1-7
1
P. SANTALUCIA ET AL.
2
facts and figures of stroke. In Europe, Italy is one of the countries with the highest percentage of people aged older than 65 years, amounting to about 20% of the whole population. By 2020, assuming stable incidence rates, 195,000 new stroke cases per year are expected because of age.3 Annual stroke incidence rates, standardized to the Italian population, range from 175 of 100,000 to 360 of 100,000 in men and from 130 of 100,000 to 273 of 100,000 in women; 30-day mortality for all strokes ranges from 18.1% to 33.0% and 1-year mortality rates from 37.9% to 40.2%.6 This epidemic and economic burden has, over the years, fostered joint action by scientific societies and public institutions to promote health policies and facilitate risk factor control and stroke prevention campaigns.7-9 Secondary stroke prevention includes conventional lifestyle and pharmacologic approaches as well as more specific interventions such as anticoagulation for prevention of cardioembolic stroke due to atrial fibrillation.10,11 The aim of this study was to describe the incidence of ischemic stroke and short-term mortality, including prescription trends in relation to recurrences, in a population of more than 4 million people yearly from 2002 to 2010. Population-based administrative database was used as a source of reliable data for estimating stroke epidemiology and prescription patterns.
Method Data Source We used the linkable administrative health databases of Lombardy, which include a population registry with the demographic data of all residents and detailed information on drug prescriptions and hospital admissions. Data were available from 2000 to 2010 for more than 4 million inhabitants per year. Lombardy is the most heavily populated Italian region, comprising urban, industrial, and rural areas. Health care in Italy is publicly funded for all residents, irrespective of social class or employment, and everyone is assigned a personal identification number kept in the National Civil Registration System. Citizens are cared for by general practitioners as part of the National Health System. The pharmacy prescription database contains the medication name and anatomic therapeutic chemical classification code, quantity, and dispensation date. The hospital discharge database records the date of hospital admission, date of discharge or death, diagnosis, and procedures done. All these databases are linked anonymously with the demographic data using unique encrypted patient codes, in accordance with Italian laws for the treatment of confidential data. Approval from an ethics committee is not required to analyze encrypted administrative data.
Study Population Patients with ischemic stroke were identified as new cases (incidence cases) if they were discharged with a diagnosis of an acute cerebrovascular event (primary diagnosis International Classification of Disease, Ninth Revision [ICD-9] code 434.xx, 436) between 1 January, 2002, and 31 December, 2009, with no previous stroke diagnosis in the 2 years before entering the cohort.
Stroke Incidence, Recurrence, and Mortality Stroke incidence figures, according to age, were calculated for 2002-2009. Hospital admissions for stroke recurrences, according to discharge code of acute cerebrovascular disease, were identified in the year after the first event. Thirty-day and mortality rates were also recorded.
Pharmacologic Treatments Prescription rates and trends for antihypertensive drugs (C02, C03, C07, C08, and C09), lipid-lowering drugs (C10), antiplatelets (B01AC), and anticoagulant drugs (B01AA03 and B01AA07) were taken into account. Data on prescriptions of at least one of the previously mentioned classes of drugs during the year before stroke were considered for the description of the patients characteristics at baseline. Prescriptions for secondary prevention were considered in the year after the first hospital admission for stroke.
Statistical Analysis Variables were classified as follows. Patients were grouped by age (0-64, 65-74, 75-84, and $85 years) taking the largest group as reference (65-74 years). Males were the reference group for sex. Total numbers of drugs except antihypertensives, antiplatelet drugs, anticoagulants, and lipid-lowering drugs were grouped in classes with similar frequencies; the reference group was the group that took fewer than 2 concomitant drugs (as a proxy for the healthiest subjects). Patients were considered censored at death (except for analyses in which death was the outcome), emigration, or admission to a nursing home. Atrial fibrillation (AF, ICD-9 code 42731, 42732), myocardial infarction (ICD-9 code 410.xx), and heart failure (ICD-9 code 428.x, 7855.x) referred to hospital admission for these causes in the 2 years preceding the stroke event. Diabetes was defined if the patient was treated with antidiabetic drugs (Anatomic Therapeutic Chemical classification code A10), hospitalized with diagnostic code of diabetes (ICD-9 code 250.xx), or recorded in the exemption registry for diabetes. The crude annual stroke incidence was calculated from the number of patients discharged with a stroke diagnosis and the personyears contributed by the residents, per 1000 person-years, by age group. Stroke recurrences were calculated on patients discharged alive after a stroke. Crude mortality rate at 30 days was the number of patients who died as a
STROKE TRENDS FROM 2002 TO 2010
3
Table 1. Characteristics of patients with ischemic stroke according to sex Total no. 43,352 Mean age 6 SD, y 75.2 6 12.5 Age classes (y) 0-64 7298 65-74 10,350 75-84 16,067 $85 9637 Treatments Antihypertensives* 30,953 Antiplatelet drugs* 18,014 Anticoagulants* 3136 Lipid-lowering drugs* 7214 Number of other treatmentsy 0-1 11,718 2-3 11,086 4-5 8281 6-7 5267 81 7000 Atrial fibrillation 6539 Diabetes 9451 Myocardial infarction 1104 Heart failure 3308
% 100
Females
%
Males
%
22,089 78.0 6 12.2
50.9
21,263 72.3 6 12.1
49.0
P value (F versus M) ,.0001
16.8 23.9 37.1 22.2
2485 4061 8751 6792
11.2 18.4 39.6 30.7
4813 6289 7316 2845
22.6 29.6 34.4 13.4
,.0001
71.4 41.5 7.2 16.6
16,680 8967 1671 3247
75.5 40.6 7.6 14.7
14,273 9047 1465 3967
67.1 42.5 6.9 18.7
,.0001 ,.0001 ,.0067 ,.0001
27.0 25.6 19.1 12.1 16.1 15.1 21.8 2.5 7.6
5178 5556 4394 2960 4001 4090 4464 531 1928
23.4 25.1 19.9 13.4 18.1 18.5 20.2 2.4 8.7
6540 5530 3887 2307 2999 2449 4987 573 1380
30.8 26.0 18.3 10.8 14.1 11.5 23.4 2.7 6.5
,.0001
,.0001 ,.0001 .0546 ,.0001
Abbreviations: F, female; M, male; SD, standard deviation. *In the year before entering the study cohort. yExcluding antihypertensives, antiplatelet drugs, anticoagulants, and lipid-lowering drugs.
percentage of the total number with stroke. Prescription rates referred to the year after the stroke event. Characteristics of males and females were compared using the chisquare test. The relationship between calendar year (treated as continuous variable) and stroke incidence, stroke recurrences, 30-day mortality, and each drug treatment were analyzed using multivariable logistic regression models adjusting for sex, age class, and number of concomitant medications. We also repeated the analyses using dummy year variables to explore possible nonlinearities in the trends. Since most subjects were present multiple times during the years of observation, a clustered sandwich estimator was used to correct for nonindependence among data. The results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). Statistical analyses were done using STATA 12.0 (StataCorp LP, College Station, TX) and JMP 11.1 (SAS Institute Inc, Cary, NC).
men (78.0 6 12.2 versus 72.3 6 12.1, P , .001) and consistently more in the 851 age group (30.7% versus 13.4%, P , .001). Among major stroke risk factors, AF at the time of first cerebrovascular event was present in 15.1% of the study cohort, more frequent in women than men, diabetes in 21,8%, more frequent in men than women, history of myocardial infarction in 2.5% equally represented between sexes, and heart failure in 7.6% of the population, more frequent in women than men. During the year before the index event, 71.4% of the subjects were taking antihypertensive drugs, 41.5% were on antiplatelets, 7.2% on anticoagulant therapies, and 16.6% subjects were on lipid-lowering therapy. Figure 1 shows the stroke incidence between 2002 and 2009, by age the overall incidence decreased from 1.4 of
Results Out of a population of more than 4 million per year, 43,352 patients, 51% women and 49% men, were hospitalized for ischemic stroke between 2002 and 2009. Table 1 shows the main demographic characteristics, risk factors, and pharmacologic profile at baseline of all patients with ischemic stroke. Mean age was 75.2 years (standard deviation, 612.5), women were older than
Figure 1.
Incidence of ischemic stroke, by age.
P. SANTALUCIA ET AL.
4
Figure 2. Recurrence of ischemic stroke in the year after first event. Recurrence rate refers to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Figure 4. Prescriptions of recommended drugs to all patients with ischemic stroke in the year after the first event. Prescription rates refer to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
apy (Fig 5), with an overall increase of 24% (from 35.3% to 43.9%; P , .0001). There was a 13% rise, although not statistically significant (from 80.6% to 91.2%, P 5 .336), in the age group up to 64 years, 11% (from 64.8% to 72.1%, P 5 .004) in the 65-74 year olds and 30% in the 75-84 years class (from 40.2% to 53.7%, P , .0001). The prescription rate of anticoagulants in patients 851 years was 10.8% in 2002 to 17.3% in 2010 (P , .0001). Multivariable analysis of prescriptions of recommended drugs showed a 3% yearly increase for antihypertensives (OR, 1.03; 95% CI, 1.02-1.04), 5% for antiplatelets (OR, 1.05; 95% CI, 1.04-1.06), 4% for anticoagulants (OR, 1.04; 95% CI, 1.03-1.06), and 14% for lipid-lowering drugs (OR, 1.14; 95% CI, 1.13-1.16).
1000 to 1.1 of 1000 (P ,.0001) the incidence fell from 3.2 of 1000 to 2.4 of 1000 (P , .001) in the 65-74 years age group, from 7.1 of 1000 to 5.3 of 1000 (P , .001) in those aged 75-84 years, and from 11.9 of 1000 to 9.4 of 1000 (P , .001) in age 85 years and older group. Multivariable analysis showed a 3% yearly reduction (OR, .97; 95% CI, .97-.98). The 1-year stroke recurrence rate fell 30% (from 10.0% to 7.0%, [P , .001] between 2002 and 2010 [Fig 2]). Multivariable analysis indicated a 6% yearly reduction (OR, .94; 95% CI, .92-.95). The 30-day mortality declined from 13.3 of 1000 to 12.1 of 1000 (P , .0001; Fig 3). The reduction was larger in 75-84 and 85 years or older patients, although between 2002 and 2010 there was a slight decline in all age classes. Multivariable analysis indicated 5% reduction (OR, .95; 95% CI, .94-.97). Figure 4 shows prescription patterns in the years after the index event for the drug classes most recommended for secondary stroke prevention. Between 2002 and 2010, the increase was linear for lipid-lowering and antiplatelet prescriptions, respectively, from 19.1% to 34.6% (P ,.001) and from 60.2% to 65.0% (P ,.001). The proportion of patients using antihypertensive therapies also rose slightly, from 65.8% to 68.3%. The prescription rate of anticoagulants rose from 11.8% to 14.1% (P , .001). Patients with AF were more likely to receive anticoagulant ther-
This study showed a significant linear decrease of stroke incidence and recurrence rates in this large cohort of patients over a decade of observation. There was also a positive trend in prescription rates of the main drugs recommended for secondary stroke prevention. Over the past decade, a decline has been reported in stroke incidence and its sequelae, probably because of the overall improvement of control and care of risk factors and better management of acute stroke patient.12-14 Stroke incidence rates in Italy are known mainly from epidemiologic registries in different regions from north to south.6 Three
Figure 3. Thirty-day mortality rate, by age. Mortality rate refers to the thirty days after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Figure 5. Prescriptions of anticoagulants to patients with stroke and atrial fibrillation in the year after the first event. Prescription rates refer to the year after the stroke event between 2002 and 2010, based, for each patient, on the year of entry in the study cohort.
Discussion
STROKE TRENDS FROM 2002 TO 2010
registers reported data valuable for examining stroke incidence trends.6 The L’Aquila register, the largest one including over 4000 patients, based on a 5-year data set, showed a mean rise in annual stroke incidence of 4.28% mainly due to the higher incidence in women, particularly those aged 851 years. Disparities between our results and epidemiologic data on stroke incidence from Italian registers might be variously explained. Possible explanations could be that Italian registers have been conducted in limited geographic areas not representative of the general Italian population and in different periods, the decade before our years of observation. Besides, there are substantial differences in patients number, smaller in the registers, although direct comparisons are not feasible. Despite differences in the Italian context, improvements in lifestyle and pharmacologic approaches to the management of stroke patients in primary and secondary prevention have led to an overall improvement in epidemiologic trends. Furthermore, in Western countries, there has been a short-term reduction in mortality after stroke, very likely due to better care in the acute setting.4,5,15-19 Our analysis too showed a decline in 30-day mortality after acute ischemic stroke, reflecting the development in the last decade in Italy and across all Europe of standardized acute stroke care based on dedicated stroke units20 applying specific acute stroke treatments.21,22 Although there are no systematic data on stroke recurrence trends, the decline over the years may possibly be because of generally closer adherence to guidelines and increased prescription of drugs recommended for secondary stroke prevention.23 Again, our analysis showed a decline in stroke recurrences from 2002 to 2010 and linear increases in the drug classes most recommended for secondary stroke prevention, mainly antiplatelet and lipid-lowering therapies. The Prevention Regimen for Effectively Avoiding Second Strokes (PRoFESS) trial provided additional evidence of the efficacy and safety of antiplatelet molecules other than aspirin alone in long-term treatment.24 The Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial contributed to the prescription of statins for secondary stroke prevention regardless of the patient’s lipid profile.25 In our cohort, prescription rates and patterns of oral anticoagulant therapy for secondary prevention of stroke of cardioembolic origin showed a well-known age-related distribution. Prescriptions tended to increase in patients up to age 74 years, whereas less than 20% of those in the 851 group were prescribed anticoagulants. People aged 801 years have the highest prevalence of AF in the general population and are therefore those who would benefit most from anticoagulant treatment on account of the high risk of peripheral thromboembolism and cardioembolic stroke.26 However, they are also the most fragile age group in terms of potential bleeding risk.27,28 This represents one of the reason for the low prescription rate of anticoagulants in elderly despite
5 29
evidence-based efficacy and guideline indications. The large trials on secondary prevention of cardioembolic stroke have shown the superiority of warfarin over aspirin,30,31 and furthermore, in elderly patients, aspirin has a harmful profile because of progressively reduced efficacy along with an increased bleeding risk.32 Dualantiplatelet therapy in patients unsuitable for anticoagulant therapy, although superior to aspirin alone, also carries a higher bleeding risk.33 Our analysis covers the years before approval of direct oral anticoagulants in clinical practice that, at least as effective as warfarin and with a favorable safety profile, with no need for frequent anticoagulation monitoring, low potential for food and drug interactions, and fast onset and offset of action, have already influenced prescription patterns, also in elderly people.11 In addition, closer attention to the selection of candidates for anticoagulant therapy34 and better adherence to guidelines has helped boost physician’s awareness in the field of thromboembolism prevention in patients with AF. Our cohort is strictly representative of a specific area, as the data come from administrative databases for a single Italian region, and the results of our analysis can be considered to correspond to a uniform standard of care. The biggest limitation of analyses conducted on administrative data is the unavailability of specific clinical data so that some variables cannot be taken into account. Nevertheless, information on demographics, diagnostic procedures, admission and discharge dates, length of hospital stay, and destination at discharge are all included in these data sets. Data on all drugs, subject to prescription, primary, and refills, are also available and can be used to observe prescription trends. Besides, administrative data are feasible, timely and inexpensive, and involve a large number of unselected people with wide coverage and continuity. The availability of large temporal series makes it possible to study long-term outcomes and epidemiologic trends. Because stroke is a hospital-managed disease, administrative health databases may be a valuable source of information for epidemiologic evaluations in a large setting of unselected patients. Potential benefits of administrative health data sets analysis focused on diseases at high epidemiologic impact, such as stroke, is the generalizability of results to the real world and the possibility of evaluating any change in the health care and prescriptions trends and how they correlate with clinical outcomes.
Conclusion Our study offers a systematic overview of stroke epidemiologic patterns over the last decade in a large area of northern Italy. The analysis showed a trend of decrease of stroke, recurrence, and mortality and concurrent increase in prescriptions of the most recommended drugs for secondary stroke prevention. We also observed,
P. SANTALUCIA ET AL.
6
according with previous studies, a reduction of stroke incidence. Although stroke incidence trends could be expected to arise due to the ageing of the population worldwide, however, improvements in global stroke care have certainly contributed and will very likely further reduce the overall burden of stroke. Acknowledgment: The authors would like to thank Igor Monti from IRCCS–Istituto di Ricerche Farmacologiche Mario Negri, Carolina Vastola, Simone Schiatti and Giovanna Rigotti from LombardiaInformatica S.p.A, and Alfredo Bevilacqua from SANTER Reply S.p.A for their assistance in data acquisition and programming support. The authors are grateful to Guya Sgaroni and Fiorenza Clerici for secretarial assistance and to J.D. Baggott for kindly editing the article.
References 1. Fergin VL, Forouzanfar MH, Krishnamurthi R, et al. Global and regional burden of stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet 2014;383:245-255. 2. Pendlebury ST, Rothwell PM. Prevalence, incidence, and factors associated with pre-stroke and post-stroke dementia: a systematic review and meta-analysis. Lancet Neurol 2009;8:1006-1018. 3. Di Carlo A. Human and economic burden of stroke. Age and Ageing 2009;38:4-5. 4. Feigin VL, Lawes CM, Bennett DA, et al. Worldwide stroke incidence and early case fatality reported in 56 population-based studies: a systematic review. Lancet Neurol 2009;8:355-369. 5. Krishnamurthi RV, Feigin VL, Forouzanfar MH, et al. Global and regional burden of first-ever ischaemic and haemorrhagic stroke during 1990-2010: findings from the Global Burden of Disease Study 2010. Lancet Glob Health 2013;1:e259-e281. 6. Sacco S, Stracci F, Cerone D, et al. Epidemiology of stroke in Italy. Int J Stroke 2011;6:219-227. 7. Goldstein LB, Bushnell CD, Adams R, et al. Guidelines for the primary prevention of stroke. A guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2011;42: 517-584. 8. SPREAD 2012 – Stroke prevention and educational awareness diffusion. Ictus cerebrale: Linee guida italiane. Available at www.siapav.it/pdf/SPREAD%202012.pdf. 9. Hachinski V, Donnan GA, Gorelick PB, et al. Stroke: working toward a prioritized world agenda. Int J Stroke 2010;5:238-256. 10. Kernan WN, Ovbiagele B, Black HR, et al, on behalf of the American Heart Association Stroke Council, Council on Cardiovascular and Stroke Nursing, Council on Clinical Cardiology, and Council on Peripheral Vascular Disease. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45: 2160-2236. 11. Camm AJ, Lip GY, De Caterina R, et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation. An update of the 2010 ESC Guidelines for the management of atrial fibrillation. Eur Heart J 2012;33:2719-2747.
12. Lee S, Shafe AC, Cowie MR. UK stroke incidence, mortality and cardiovascular risk management 1999-2008: time-trend analysis from the General Practice Research Database. BMJ Open 2011;1:e000269. http://dx.doi.org/ 10.1136/bmjopen-2011-000269. 13. Allen NB, Holford TR, Bracken MB, et al. Trends in oneyear recurrent ischemic stroke among the elderly in the USA: 1994-2002. Cerebrovasc Dis 2010;30:525-532. 14. Wieberdink RG, Ikram MA, Hofman A, et al. Trends in stroke incidence rates and stroke risk factors in Rotterdam, the Netherlands from 1990 to 2008. Eur J Epidemiol 2012;27:287-295. 15. Organised inpatient (stroke unit) care for stroke. Cochrane Database Syst Rev 2013;9:CD000197. http://dx.doi.org/10.1002/14651858.CD000197.pub3. 16. Meretoja A, Roine RO, Kaste M, et al. Effectiveness of primary and comprehensive stroke centers PERFECT stroke: a nationwide observational study from Finland. Stroke 2010;41:1102-1107. 17. Nimptsch U, Mansky T. Stroke unit care trends of in-hospital mortality for stroke in Germany 2005-2010. Int J Stroke 2014;9:260-265. 18. Schmidt M, Jacobsen JB, Johnsen SP, Bøtker HE, Sørensen HT. Eighteen-year trends in stroke mortality and the prognostic influence of comorbidity. Neurology 2014;82:340-350. 19. Lackland DT, Roccella EJ, Deutsch AF, et al. Factors influencing the decline in stroke mortality. A statement from the American Heart Association/American Stroke Association. Stroke 2014;45:315-353. 20. Candelise L, Gattinoni M, Bersano A, et al, on the behalf of the PROSIT Study Group. Stroke unit care for acute stroke patients: an observational follow-up study. Lancet 2007;369:299-305. 21. The NINDS rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. The National Institute of Neurological Disorders and Stroke r-tPA Stroke Study Group. N Engl J Med 1995;333:1581-1587. 22. Del Zoppo GJ, Saver JL, Jauch EC, et al. Expansion of the time window for treatment of acute ischemic stroke with intravenous tissue plasminogen activator: a science advisory from the American Heart Association/American Stroke Association. Stroke 2009;40:2945-2948. 23. Kernan WN, Ovbiagele B, Black HR, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guidelines for healthcare professionals from the American Heart Association/American Stroke Association. Stroke 2014;45:2160-2236. 24. Sacco RL, Diener HC, Yusuf S, et al, for the PRoFESS Study Group. Aspirin and extended-release dipyridamole versus clopidogrel for recurrent stroke. N Engl J Med 2008;359:1238-1251. 25. Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) Investigators. High-dose atorvastatin after stroke or transient ischemic attack. N Engl J Med 2006;355:549-559. 26. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation: a major contributor to stroke in the elderly: the Framingham Study. Arch Intern Med 1987;147:1561-1564. 27. Lip GY, Andreotti F, Fauchier L, et al. Bleeding risk assessment and management in atrial fibrillation patients: a position document from the European Heart Rhythm Association, endorsed by the European Society of Cardiology Working Group on Thrombosis. Europace 2011;13:723-746. 28. Poli D, Antonucci E, Testa S, et al. Bleeding risk in very old patients on vitamin K antagonist treatment: results
STROKE TRENDS FROM 2002 TO 2010 of a prospective collaborative study on elderly patients followed by Italian Centres for Anticoagulation. Circulation 2011;124:824-829. 29. Bungard TJ, Ghali WA, Teo KK, et al. Why do patients with atrial fibrillation not receive warfarin? Arch Intern Med 2000;160:41-46. 30. Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med 2007;146:857-867. 31. Go AS, Hylek EM, Chang Y, et al. Anticoagulant therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical trial? JAMA 2003;290:2685-2692.
7 32. Olesen JB, Lip GY, Lindhardsen J, et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: a net clinical benefit using a ‘‘real world’’ nationwide cohort study. Thromb Haemost 2011;106: 739-749. 33. ACTIVE Investigators. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med 2009;360:2066-2078. 34. Lip GY, Nieuwlaat R, Pisters R, et al. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor-based approach: the EuroHeart Survey on atrial fibrillation. Chest 2010;137:263-272.