Modeling the HIV epidemic in MSM in Cyprus: reaching only the 95-95-95 cascade of care targets fails to reduce HIV incidence by 90% in 2030.

National responses should be improved and accelerated to meet the target of ending the Acquired ImmunoDeficiency Syndrome (AIDS) epidemic by 2030. In the Republic of Cyprus, Men who have Sex with Men (MSM) are disproportionately affected by Human Immunodeficiency Virus (HIV), accounting approximately for half of all annual HIV diagnoses. This study assesses the evolution of HIV incidence in MSM in Cyprus until 2030 using a model calibrated to Cypriot epidemiological data. Four scenarios were examined: status quo, two scenarios focusing on introducing Pre-Exposure Prophylaxis (PrEP), and a 90% HIV incidence reduction scenario. Reaching only the 95-95-95 HIV cascade of care targets among MSM would reduce HIV incidence by 48.6% by 2030 compared to 2015. Initiating a PrEP intervention only for high risk MSM would cause a modest further reduction in HIV incidence. To meet the 90% reduction target, PrEP should be expanded to both high and medium risk MSM and, after 2025, behavioral interventions should be implemented so as high-risk MSM gradually move to the medium-risk category. Cyprus will not reach the HIV incidence reduction target by 2030 unless PrEP is gradually promoted and delivered to all high and medium risk MSM along with awareness and behavioral interventions.

Effect of public health interventions during the first epidemic wave of COVID-19 in Cyprus: a modelling study.

BACKGROUND: Cyprus addressed the first wave of SARS CoV-2 (COVID-19) by implementing non-pharmaceutical interventions (NPIs). The aims of this study were: a) to estimate epidemiological parameters of this wave including infection attack ratio, infection fatality ratio, and case ascertainment ratio, b) to assess the impact of public health interventions and examine what would have happened if those interventions had not been implemented. METHODS: A dynamic, stochastic, individual-based Susceptible-Exposed-Infected-Recovered (SEIR) model was developed to simulate COVID-19 transmission and progression in the population of the Republic of Cyprus. The model was fitted to the observed trends in COVID-19 deaths and intensive care unit (ICU) bed use. RESULTS: By May 8th, 2020, the infection attack ratio was 0.31% (95% Credible Interval [CrI]: 0.15, 0.54%), the infection fatality ratio was 0.71% (95% CrI: 0.44, 1.61%), and the case ascertainment ratio was 33.2% (95% CrI: 19.7, 68.7%). If Cyprus had not implemented any public health measure, the healthcare system would have been overwhelmed by April 14th. The interventions averted 715 (95% CrI: 339, 1235) deaths. If Cyprus had only increased ICU beds, without any social distancing measure, the healthcare system would have been overwhelmed by April 19th. CONCLUSIONS: The decision of the Cypriot authorities to launch early NPIs limited the burden of the first wave of COVID-19. The findings of these analyses could help address the next waves of COVID-19 in Cyprus and other similar settings.

High-risk behaviors and their association with awareness of HIV status among participants of a large-scale prevention intervention in Athens, Greece.

BACKGROUND: Aristotle was a seek-test-treat intervention during an outbreak of human immunodeficiency virus (HIV) infection among people who inject drugs (PWID) in Athens, Greece that started in 2011. The aims of this analysis were: (1) to study changes of drug injection-related and sexual behaviors over the course of Aristotle; and (2) to compare the likelihood of risky behaviors among PWID who were aware and unaware of their HIV status. METHODS: Aristotle (2012-2013) involved five successive respondent-driven sampling rounds of approximately 1400 PWID each; eligible PWID could participate in multiple rounds. Participants were interviewed using a questionnaire, were tested for HIV, and were classified as HIV-positive aware of their status (AHS), HIV-positive unaware of their status (UHS), and HIV-negative. Piecewise linear generalized estimating equation models were used to regress repeatedly measured binary outcomes (high-risk behaviors) against covariates. RESULTS: Aristotle recruited 3320 PWID (84.5% males, median age 34.2 years). Overall, 7110 interviews and blood samples were collected. The proportion of HIV-positive first-time participants who were aware of their HIV infection increased from 21.8% in round A to 36.4% in the last round. The odds of dividing drugs at least half of the time in the past 12 months with a syringe someone else had already used fell from round A to B by 90% [Odds Ratio (OR) (95% Confidence Interval-CI): 0.10 (0.04, 0.23)] among AHS and by 63% among UHS [OR (95% CI): 0.37 (0.19, 0.72)]. This drop was significantly larger (p = 0.02) among AHS. There were also decreases in frequency of injection and in receptive syringe sharing in the past 12 months but they were not significantly different between AHS (66 and 47%, respectively) and UHS (63 and 33%, respectively). Condom use increased only among male AHS from round B to the last round [OR (95% CI): 1.24 (1.01, 1.52)]. CONCLUSIONS: The prevalence of risky behaviors related to drug injection decreased in the context of Aristotle. Knowledge of HIV infection was associated with safer drug injection-related behaviors among PWID. This highlights the need for comprehensive interventions that scale-up HIV testing and help PWID become aware of their HIV status.

Act early, save lives: managing COVID-19 in Greece.

OBJECTIVES: The aim of the study was to assess the impact of social distancing interventions in Greece and to examine what would have happened if those interventions had not been implemented. STUDY DESIGN: A dynamic, discrete time, stochastic individual-based model was developed to simulate coronavirus disease 2019 (COVID-19) transmission. METHODS: The model was fitted to the observed trends in COVID-19 deaths and intensive care unit (ICU) bed use in Greece. RESULTS: If Greece had not implemented social distancing interventions, the healthcare system would have been overwhelmed between March 30 and April 4. The combined social distancing interventions and increase in ICU beds averted 4360 (95% credible interval: 3050, 5700) deaths and prevented the healthcare system from becoming overwhelmed. CONCLUSIONS: The quick and accurate interventions of the Greek government limited the burden of the COVID-19 outbreak.

A hepatitis C outbreak preceded the HIV outbreak among persons who inject drugs in Athens, Greece: Insights from a mathematical modelling study.

People who inject drugs (PWID) comprise one of the major transmission risk groups for human immunodeficiency virus (HIV) and hepatitis C virus (HCV). In 2011, Athens experienced a large HIV outbreak among PWID. Significant public health interventions were implemented in response to the HIV outbreak. The aims of this study were to estimate the indirect effects of the HIV interventions on HCV infection and to evaluate the concept of the association between HCV and HIV infections in the case of Athens. A dynamic, stochastic, individual-based model was developed to simulate HCV transmission among PWID. We calibrated the model to reproduce the observed HCV prevalence among PWID in Greece. Two years prior to the HIV outbreak, an undetected HCV outbreak has occurred. In 2009, the incidence of HCV infection increased from 640 (495, 842) cases in 2008 to 1260 (1060, 1500). The mean time from initiation of injecting drug use to HCV acquisition decreased from 29 months in 2008 to 13 months in 2009. After HIV interventions, HCV incidence declined by 64.8% in 2012, compared to 2009. The averted HCV incidence cases attributed to the HIV-implemented interventions were 2200 (1950, 2480), during 2012-2015. The cumulative number incident HCV cases in Athens during 2002-2015 was about 9900 (7800, 12 100). Our results highlight that before the 2011 HIV outbreak in Athens, an HCV outbreak occurred in 2009. Prevention measures for HIV that took place in the Athens metropolitan area in 2012 reduced significantly the incidence of HCV.

Is elimination of HCV possible in a country with low diagnostic rate and moderate HCV prevalence?: The case of Greece.

BACKGROUND AND AIM: The treatment of hepatitis C (HCV) with interferon (IFN)-free direct-acting antivirals (DAAs) is anticipated to change the future burden of disease. The aim of this study is to quantify the impact of IFN-free DAAs on HCV-related morbidity and mortality in Greece under different scenarios concerning treatment coverage and primary prevention, including the proposed by World Health Organization Global Hepatitis Strategy. METHODS: A previously described model was used to project the future disease burden up to 2030 under scenarios, which includes treatment based on the combination of pegylated-IFN with ribavirin (base case) and scenarios using DAAs therapies. RESULTS: Under the base case scenario, an increase in HCV-related morbidity and mortality is predicted in Greece (mortality in 2030: +23.6% compared with 2015). If DAAs are used with the same treatment coverage, the number of hepatocellular carcinoma cases and of liver related deaths are predicted to be lower by 4-7% compared with 2015. Under increased treatment coverage (from 2000 treated/year to approximately 5000/year in 2015-2020 and 2500/year subsequently), morbidity and mortality will decrease by 43-53% in 2030 compared with 2015. To achieve the WHO Global Hepatitis Strategy goals, a total number of 86 500 chronic hepatitis C patients will have to be treated during 2015-2030. CONCLUSIONS: Elimination of HCV in Greece by 2030 necessitates great improvements in primary prevention, implementation of large screening programs and high treatment coverage.

Estimating the time required to reach HPV vaccination targets across Europe.

BACKGROUND: Cervical cancer (CC) is one of the most common causes of cancer-related deaths in women. The World Health Organization (WHO) has called for the CC elimination as a public health priority and has urged countries to achieve a 90% vaccine coverage rate of human papilloma virus (HPV) vaccination among 15-year-old girls by 2030. RESEARCH DESIGN AND METHODS: Regression models were fitted to the WHO HPV vaccine coverage rate data to estimate when the 90% vaccine coverage rate target would be achieved in 22 European countries. RESULTS: The mean vaccine coverage rate of included countries was 62.2% (SD: 18.3). Nine countries (Iceland, Norway, Portugal, Ireland, Hungary, Spain, Sweden, Denmark, and Switzerland) are expected to achieve a 90% vaccine coverage rate by 2030. Six countries (Estonia, Cyprus, Netherlands, France, Germany, and Italy) are expected to reach a 90% vaccine coverage rate between 2030 and 2040 whereas seven countries (Belgium, Bulgaria, Finland, Latvia, Luxembourg, Malta, and Slovenia) are not expected to achieve the 90% vaccine coverage rate target by 2040. CONCLUSION: The majority of European countries are not on track to achieve 90% vaccine coverage rate by 2030. To achieve this, a significant increase in the annual vaccine coverage rate growth rate is required.

Prevalence of chronic HCV infection in EU/EEA countries in 2019 using multiparameter evidence synthesis.

Background: Epidemiological data are crucial to monitoring progress towards the 2030 Hepatitis C Virus (HCV) elimination targets. Our aim was to estimate the prevalence of chronic HCV infection (cHCV) in the European Union (EU)/European Economic Area (EEA) countries in 2019. Methods: Multi-parameter evidence synthesis (MPES) was used to produce national estimates of cHCV defined as: π = πrecρrec + πexρex + πnonρnon; πrec, πex, and πnon represent cHCV prevalence among recent people who inject drugs (PWID), ex-PWID, and non-PWID, respectively, while ρrec, ρex, and ρnon represent the proportions of these groups in the population. Information sources included the European Centre for Disease Prevention and Control (ECDC) national operational contact points (NCPs) and prevalence database, the European Monitoring Centre for Drugs and Drug Addiction databases, and the published literature. Findings: The cHCV prevalence in 29 of 30 EU/EEA countries in 2019 was 0.50% [95% Credible Interval (CrI): 0.46%, 0.55%]. The highest cHCV prevalence was observed in the eastern EU/EEA (0.88%; 95% CrI: 0.81%, 0.94%). At least 35.76% (95% CrI: 33.07%, 38.60%) of the overall cHCV prevalence in EU/EEA countries was associated with injecting drugs. Interpretation: Using MPES and collaborating with ECDC NCPs, we estimated the prevalence of cHCV in the EU/EEA to be low. Some areas experience higher cHCV prevalence while a third of prevalent cHCV infections was attributed to PWID. Further efforts are needed to scale up prevention measures and the diagnosis and treatment of infected individuals, especially in the east of the EU/EEA and among PWID. Funding: ECDC.

Impact of the COVID-19 Pandemic on HPV Vaccinations in Switzerland and Greece: Road to Recovery.

The COVID-19 pandemic has caused significant disruptions to healthcare, including reduced administration of routinely recommended HPV vaccines in a number of European countries. Because the extent and trends of accumulated vaccine dose deficits may vary by country, decision-makers need country-specific information regarding vaccine deficits to plan effective catch-up initiatives. To address this knowledge gap in Switzerland and Greece, this study used a previously published COVID-19 recovery calculator and historical vaccine sales data to quantify the cumulative number of missed doses and the catch-up rate required to clear the deficit in Switzerland and Greece. The resultant cumulative deficit in HPV doses for Switzerland and Greece were 24.4% and 21.7%, respectively, of the total number of doses disseminated in 2019. To clear the dose deficit by December 2025, monthly vaccination rates must be increased by 6.3% and 6.0% compared to 2019 rates in Switzerland and Greece, respectively. This study demonstrates that administration rates of routine HPV vaccines decreased significantly among Swiss and Greek adolescents during the COVID-19 pandemic and that a sustained increase in vaccination rates is necessary to recover the HPV dose deficits identified and to prevent long-term public health consequences.

Could the 2010 HIV outbreak in Athens, Greece have been prevented? A mathematical modeling study.

INTRODUCTION: In 2009 and 2010, Athens, Greece experienced a hepatitis C virus (HCV) and a Human Immunodeficiency Virus (HIV) outbreak among People Who Inject Drugs (PWID), respectively. The HCV outbreak was not detected, while that of HIV was identified in 2011. The integrated HIV-interventions, launched in early 2012, managed to reduce directly the HIV incidence and indirectly the HCV incidence. This study aims to assess what would have been the course of the HIV outbreak and its associated economic consequences if the 2009 HCV outbreak had been detected and integrated interventions had been initiated 1- or 2-years earlier. METHODS: The model was calibrated to reproduce the observed HIV epidemiological and clinical parameters among PWID of Athens, Greece. We examined the effect of the 1- or 2-years earlier detection scenarios, the 1-year later detection, the non-detection scenario, and compared them to the status quo scenario. RESULTS: Cumulative HIV cases, under the status-quo scenario during 2009-2019, were 1360 (90% Credible intervals: 290, 2470). If the HCV outbreak had been detected 1- or 2- years earlier, with immediate initiation of integrated interventions, 740 and 1110 HIV cases could be averted by 2019, respectively. Regarding the costs, if there was an efficient notification system to detect the HCV outbreak 1 or 2 years earlier, 35.2-53.2 million euros could be saved compared to the status quo by 2019. CONCLUSIONS: If the HCV outbreak had been detected and promptly addressed, the HIV outbreak would have been prevented and 35.2-53.2 million euros could have been saved.

The cost of a late-detected outbreak among people who inject drugs. A modeling study.

BACKGROUND: People who inject drugs (PWID) are at risk for human immunodeficiency virus (HIV) and hepatitis C virus (HCV). In 2009 and 2011, Athens, Greece experienced an HCV and an HIV outbreak among PWID, respectively. Of these, only the 2011 HIV outbreak was detected. However, the public health interventions implemented in response to the HIV outbreak tackled also indirectly the undetected HCV outbreak. The aim of this study is to highlight the potential benefits of an efficient notification system using as a case study the undetected 2009 HCV outbreak among PWID of Athens. More specifically, the study assesses whether an earlier implementation of the same public responses could diminish the scale of the HCV outbreak and estimates the potential cost-savings. METHODS: A previous dynamic, stochastic, individual-based model was used to simulate HCV transmission among PWID of Athens, Greece. We calibrated the model to reproduce the observed HCV prevalence. We examined the effect of the non-detection scenario, the 1- or 2-years earlier detection scenarios and compared them to the status quo scenario. RESULTS: Under the non-detection scenario, 2800 additional PWID would have been infected with HCV compared to the status quo by 2019. On the contrary, if the outbreak was detected 1- or 2- years earlier with immediate interventions, 440 and 970 HCV cases could be averted by 2019, respectively. Non-detection of the outbreak would cost an additional 43.2 (95% Credible interval: 2.7, 59.4) million euros to the healthcare system, compared to the status quo. On the other hand, if there was an efficient notification system to detect the HCV outbreak 1 or 2 years earlier, 6.8-15.6 million euros could have been saved by 2019. CONCLUSIONS: An efficient notification system among PWID is a cost-saving investment that could detect on time and contain future outbreaks, and save valuable resources of the healthcare system.

Planning the hepatitis C virus elimination in Cyprus: A modeling study.

BACKGROUND: Hepatitis C virus (HCV) infection is a major global public health problem. In the Republic of Cyprus, the estimated prevalence of chronic hepatitis C (CHC) among the general population is 0.6%, while the CHC prevalence among people who inject drugs (PWID) is estimated at 46%. Direct-acting antivirals that can eliminate HCV are not yet widely available in the Republic of Cyprus. However, when direct-acting antivirals become available, a long-term strategic plan to guide elimination efforts will be needed to maximize the effect of treatment. AIM: To determine the programmatic targets to eliminate HCV in the Republic of Cyprus. METHODS: A dynamic, stochastic, individual-based model of HCV transmission, disease progression, and cascade of care was calibrated to data from Cyprus. The model stratifies the population into the infected general population and the PWID population. A variety of test, prevention, and treatment strategies concerning the general population, PWID, or both were examined. The time horizon of the analysis was until 2034. RESULTS: Under the status quo scenario, the model predicted that 75 (95% confidence interval (CI): 60, 91) and 575 (95%CI: 535, 615) liver-related deaths and new infections would occur by 2034, respectively. Launching an expanded treatment program, without screening interventions, would cause modest outcomes regarding CHC prevalence (16.6% reduction in 2034 compared to 2020) and liver-related deaths (10 deaths would be prevented compared to the status quo scenario by 2034). Implementing a test and treat strategy among the general population but without any intervention in the PWID population would suffice to meet the mortality target but not the incidence target. To achieve HCV elimination in Cyprus, 3080 (95%CI: 3000, 3200) HCV patients need to be diagnosed and treated by 2034 (2680 from the general population and 400 from PWID), and harm reduction coverage among PWID should be increased by 3% per year (from 25% in 2020 to 67% in 2034). CONCLUSION: Elimination of HCV is a demanding public health strategy, which requires significant interventions both among the general population and high-risk groups.

Economic evaluation of the hepatitis C elimination strategy in Greece in the era of affordable direct-acting antivirals.

BACKGROUND: Hepatitis C virus (HCV) is a leading cause of worldwide liver-related morbidity and mortality. The World Health Organization released an integrated strategy targeting HCV-elimination by 2030. This study aims to estimate the required interventions to achieve elimination using updated information for direct-acting antiviral (DAA) treatment coverage, to compute the total costs (including indirect/societal costs) of the strategy and to identify whether the elimination strategy is cost-effective/cost-saving in Greece. AIM: To estimate the required interventions and subsequent costs to achieve HCV elimination in Greece. METHODS: A previously validated mathematical model was adapted to the Greek HCV-infected population to compare the outcomes of DAA treatment without the additional implementation of awareness or screening campaigns versus an HCV elimination strategy, which includes a sufficient number of treated patients. We estimated the total costs (direct and indirect costs), the disability-adjusted life years and the incremental cost-effectiveness ratio using two different price scenarios. RESULTS: Without the implementation of awareness or screening campaigns, approximately 20000 patients would be diagnosed and treated with DAAs by 2030. This strategy would result in a 19.6% increase in HCV-related mortality in 2030 compared to 2015. To achieve the elimination goal, 90000 patients need to be treated by 2030. Under the elimination scenario, viremic cases would decrease by 78.8% in 2030 compared to 2015. The cumulative direct costs to eliminate the disease would range from 2.1-2.3 billion euros (€) by 2030, while the indirect costs would be €1.1 billion. The total elimination cost in Greece would range from €3.2-3.4 billion by 2030. The cost per averted disability-adjusted life year is estimated between €10100 and €13380, indicating that the elimination strategy is very cost-effective. Furthermore, HCV elimination strategy would save €560-895 million by 2035. CONCLUSION: Without large screening programs, elimination of HCV cannot be achieved. The HCV elimination strategy is feasible and cost-saving despite the uncertainty of the future cost of DAAs in Greece.

HCV elimination among people who inject drugs. Modelling pre- and post-WHO elimination era.

BACKGROUND: Elimination of hepatitis C virus (HCV) among people who inject drugs (PWID) is a costly investment, so strategies should not only focus on eliminating the disease, but also on preventing disease resurgence. The aims of this study are to compute the minimum necessary antiviral therapies to achieve elimination with and without the additional expansion of harm reduction (HR) programs and to examine the sustainability of HCV elimination after 2030 if treatment is discontinued. METHOD: We considered two types of epidemic (with low (30%) and high (50%) proportion of PWID who engage in sharing equipment (sharers)) within three baseline chronic HCV (CHC) prevalence settings (30%, 45% and 60%), assuming a baseline HR coverage of 40%. We define sustainable elimination strategies, those that could maintain eliminations results for a decade (2031-2040), in the absence of additional treatment. RESULTS: The model shows that the optimum elimination strategy is dependent on risk sharing behavior of the examined population. The necessary annual treatment coverage to achieve HCV elimination under 45% baseline CHC prevalence, without the simultaneous expansion of HR programs, ranges between 4.7-5.1%. Similarly, under 60% baseline CHC prevalence the needed treatment coverage varies from 9.0-10.5%. Increasing HR coverage from 40% to 75%, reduces the required treatment coverage by 6.5-9.8% and 11.0-15.0% under 45% or 60% CHC prevalence, respectively. In settings with ≤45% baseline CHC prevalence, expanding HR to 75% could prevent the disease from rebounding after elimination, irrespective of the type of the epidemic. In high chronic HCV prevalence, counseling interventions to reduce sharing are also needed to maintain the HCV incident cases in low levels. CONCLUSIONS: Harm reduction strategies have a vital role in HCV elimination strategy, as they reduce the required number of treatments to eliminate HCV and they provide sustainability after the elimination. The above underlines that HCV elimination strategies should be built upon the existing HR services, and argue for HR expansion in countries without services.

Treatment and primary prevention in people who inject drugs for chronic hepatitis C infection: is elimination possible in a high-prevalence setting?

AIMS: To project the impact of scaling-up oral anti-viral therapy and harm reduction on chronic hepatitis C (CHC) prevalence and incidence among people who inject drugs (PWID) in Greece, to estimate the relationship between required treatment levels and expansion of harm reduction programmes to achieve specific targets and to examine whether hepatitis C virus (HCV) elimination among PWID is possible in this high-prevalence setting. DESIGN: A dynamic discrete time, stochastic individual-based model was developed to simulate HCV transmission among PWID incorporating the effect of HCV treatment and harm reduction strategies, and allowing for re-infection following treatment. SETTING/PARTICIPANTS: The population of 8300 PWID in Athens Metropolitan area. MEASUREMENTS: Reduction in HCV prevalence and incidence in 2030 compared with 2016. FINDINGS: Moderate expansion of HCV treatment (treating 4-8% of PWID/year), with a simultaneous increase of 2%/year in harm reduction coverage (from 44 to 72% coverage over 15 years), was projected to reduce CHC prevalence among PWID in Athens by 46.2-94.8% in 2030, compared with 2016. CHC prevalence would reduce to below 10% within the next 4-5 years if annual HCV treatment numbers were increased up to 16-20% PWID/year. The effect of harm reduction on incidence was more pronounced under lower treatment rates. CONCLUSIONS: Based on theoretical model projections, scaled-up hepatitis C virus treatment and harm reduction interventions could achieve major reductions in hepatitis C virus incidence and prevalence among people who inject drugs in Athens, Greece by 2030. Chronic hepatitis C could be eliminated in the next 4-5 years by increasing treatment to more than 16% of people who inject drugs per year combined with moderate increases in harm reduction coverage.

Are ALOX5AP gene SNPs a risk or protective factor for stroke?

ALOX5AP (5-lipoxygenase) has been recognized as a susceptibility gene for stroke. Using a case-control design, the whole coding and adjoining intronic regions of ALOX5AP were sequenced to study the role of SNPs and their interplay with other risk factors in Greek patients with stroke. Patients (n=213) were classified by the Trial of Org 10172 in Acute Stroke Treatment (TOAST). Their mean age of was 58.9 ± 14.64, comprising 145 males. The control group consisted of 210 subjects, ethnicity, sex and age matched, with no stroke history. Risk factors (hyperlipidemia, hypertension, atrial fibrillation, migraine, CAD, diabetes, smoking and alcohol consumption) were assessed as confounding factors and comparisons were done using logistic regression analysis. SNPs rs4769055, rs202068154 and rs3803277 located in intronic regions of the gene and according to in silico programs EX_SKIP and HSF possibly affecting splicing of exons 1 and 2 of ALOX5AP, showed significantly different frequencies between patients and controls. The genotype frequencies of rs4769055: AA, of rs202068154: AC and of rs3803277: CA were significantly higher (p<0.001, 0.058) in controls than in patients. The results were indicative of a protective role of the three SNPs either in homozygosity or heterozygosity for MAF and more specifically rs3803277: CA/AA genotypes were protective against SVO stroke subtype.