Advertisement

Incident cardiovascular disease in patients diagnosed with multiple sclerosis: A multi-database study

Published:October 01, 2019DOI:https://doi.org/10.1016/j.msard.2019.101423

      Highlights

      • MS patients have higher incidence of CVD after MS diagnosis than matched patients.
      • The rate of venous thomboembolism is 2–3 times higher.
      • The rate of peripheral vascular disease is >2 times higher.
      • The rate of myocardial infarction is increased among female patients with MS.

      Abstract

      Background

      Recent data on rates of cardiovascular disease (CVD) in patients after MS diagnosis are sparse.

      Objective

      To describe incident CVD in MS patients after diagnosis compared with a matched non-MS population.

      Methods

      We conducted a matched cohort study in two separate electronic medical databases, the United States Department of Defense military health care system and the United Kingdom's Clinical Practice Research Datalink GOLD. The study population included all patients with a first recorded diagnosis of MS and no history of CVD or selected measurable comorbidities associated with CVD and matched non-MS patients who were also free of CVD and the CVD associated comorbidities. We identified incident CVD outcomes first recorded after the MS diagnosis / matched date and calculated incidence rates and incidence rate ratios by type of CVD.

      Results

      Rates of venous thromboembolism and peripheral vascular disease were 2-fold higher among MS than non-MS patients in both databases and the risk of myocardial infarction was 2.5 times higher among female MS patients compared with non-MS females in both databases. Other CVD outcomes were not consistent between databases.

      Conclusion

      MS patients in the UK and the US have increased risk of venous thromboembolism and peripheral vascular disease. The risk of myocardial infarction is increased among female MS patients.

      Keywords

      1. Introduction

      Multiple sclerosis (MS) is the most prevalent permanently disabling neurological disease among young adults in Europe and North America and is associated with diminished quality of life and high socio-economic cost (
      • Hassan-Smith G.
      • Douglas M.R.
      Epidemiology and diagnosis of multiple sclerosis.
      ;
      • Noseworthy J.H.
      • Lucchinetti C.
      • Rodriguez M.
      • et al.
      Multiple sclerosis.
      ;
      • Pugliatti M.
      • Rosati G.
      • Carton H.
      • et al.
      The epidemiology of multiple sclerosis in Europe.
      ). A systematic review of prevalence and incidence of cardiac, cerebrovascular and peripheral vascular diseases in multiple sclerosis patients identified multiple studies that found higher prevalences of vascular comorbidities among MS patients compared with the general population, but few studies that specifically reported on the incidence of cardiovascular disease (CVD) (
      • Marrie R.A.
      • Reider N.
      • Cohen J.
      • et al.
      A systematic review of the incidence and prevalence of cardiac, cerebrovascular, and peripheral vascular disease in multiple sclerosis.
      ;
      • Marrie R.A.
      • Garland A.
      • Schaffer S.A.
      • et al.
      Traditional risk factors may not explain increased incidence of myocardial infarction in MS.
      ). Studies that reported the incidence of CVD after MS diagnosis observed inconsistent results, particularly for ischemic heart diseases. Prior studies have reported rates of myocardial infarction (MI) varying from 18% to 85% higher for MS patients than non-MS patients (
      • Marrie R.A.
      • Garland A.
      • Schaffer S.A.
      • et al.
      Traditional risk factors may not explain increased incidence of myocardial infarction in MS.
      ;
      • Jadidi E.
      • Mohammadi M.
      • Moradi T
      High risk of cardiovascular diseases after diagnosis of multiple sclerosis.
      ;
      • Christiansen C.F.
      • Christensen S.
      • Farkas D.K.
      • et al.
      Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study.
      ), while other studies that categorized CVD outcomes differently reported the rates of any ischemic heart disease to be the same (
      • Marrie R.A.
      • Fisk J.
      • Tremlett H.
      • et al.
      Differing trends in the incidence of vascular comorbidity in MS and the general population.
      ) or 12% lower (
      • Roshanisefat H.
      • Bahmanyar S.
      • Hillert J.
      • et al.
      Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.
      ) for MS compared with non-MS patients.
      Previous studies have found that patients have higher prevalence of many comorbidities strongly associated with CVD at MS diagnosis compared to an age- and sex-matched general population (
      • Marrie R.A.
      • Reider N.
      • Cohen J.
      • et al.
      A systematic review of the incidence and prevalence of cardiac, cerebrovascular, and peripheral vascular disease in multiple sclerosis.
      ;
      • Fromont A.
      • Binquet C.
      • Rollot F.
      • et al.
      Comorbidities at multiple sclerosis diagnosis.
      ;
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      ). In order to assess the risk of new CVD in MS patients, we conducted a study to follow MS and matched non-MS patients after MS diagnosis or matched date, among patients with no prior history of CVD or selected measurable comorbidities associated with CVD in two large databases from the United States (US) and the United Kingdom (UK).

      2. Methods

      The methods for this study have been previously described (
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      ). This study was conducted in two databases, one in the United States (US) and one in the United Kingdom (UK). The US study was conducted utilizing clinical and administrative data from the Department of Defense (DOD) military health care system. It is comprised of data contributed by members of the US DOD, retirees and dependents (14% active service members, 49% female) who are broadly representative of employed US citizens and their families. The UK study was conducted in Clinical Practice Research Datalink (CPRD) GOLD, a large, prospectively collected, anonymized medical record database encompassing over 500 general practices representative of the UK population as a whole (
      • Herrett E.
      • Gallagher A.M.
      • Bhaskaran K.
      • et al.
      Data resource profile: clinical practice research datalink (CPRD).
      ). The authors have used CPRD GOLD data for many studies, including many prior studies of MS (
      • Alonso A.
      • Jick S.S.
      • Olek M.J.
      • et al.
      Incidence of multiple sclerosis in the United Kingdom.
      ;
      • Hernán M.A.
      • Jick S.S.
      • Olek M.J.
      • et al.
      Recombinant hepatitis B vaccine and the risk of multiple sclerosis: a prospective study.
      ;
      • Alonso A.
      • Jick S.S.
      • Jick H.
      • et al.
      Antibiotic use and risk of multiple sclerosis.
      ;
      • Alonso A.
      • Jick S.S.
      • Hernan M.A
      Allergy, histamine 1 receptor blockers, and the risk of multiple sclerosis.
      ;
      • Á Alonso
      • SS Jick
      • Olek M.J.
      • et al.
      Recent use of oral contraceptives and the risk of multiple sclerosis.
      ;
      • Hernán M.A.
      • Jick S.S.
      • Logroscino G.
      • et al.
      Cigarette smoking and the progression of multiple sclerosis.
      ;
      • Jick S.S.
      • Li L.
      • Falcone G.J.
      • et al.
      Mortality of patients with multiple sclerosis: a cohort study in UK primary care.
      ;
      • Jick S.S.
      • Li L.
      • Falcone G.J.
      • et al.
      Epidemiology of multiple sclerosis: results from a large observational study in the UK.
      ).

      2.1 Study population

      Within each database we identified all people with a first recorded diagnosis of MS in years 2001–2016 (UK-CPRD) or 2004–2017 (US-DOD). To identify newly diagnosed cases of MS we required at least 1 year of enrollment in the database before the first MS diagnosis. Cases were then validated through multiple steps that varied according to database as reported previously (
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      ). All patients with MS were then matched to up to 10 patients without MS on age, sex, month, year of MS diagnosis/matched date in the database, and geography (same practice in the CPRD; same region in the DOD). From this study population, patients were excluded if they had any of the following before MS diagnosis/matched date: at least one (UK-CPRD) or at least five (US-DOD) diagnoses for a specific type of CVD or selected comorbidity (dyslipidemia, atrial fibrillation, and disorders related to peripheral vascular disease (PVD) (e.g. Raynaud's, Buerger's disease)) or had treated hypertension or treated type 2 diabetes (a diagnosis and treatment within 90 days of each other). See Supplemental Table 1. We only included medication-treated type 2 diabetes because it is difficult to distinguish between true diabetes diagnoses without treatment and screening in claims data. To control for presence of comorbidities strongly associated with CVD at cohort entry and their effects on the risk of subsequent CVD we excluded patients with these comorbidities before cohort entry. This reduced the inherent differences in CVD risk between MS and non-MS patients at MS diagnosis (or matched date).

      2.2 Study outcomes

      We identified an a priori list of CVD outcomes of interest: myocardial infarction (MI), stroke, cardiac arrest/sudden cardiac death, major adverse cardiac event (MACE), heart failure, angina and unspecified heart disease, transient ischemic attack (TIA) and unspecified cerebrovascular disease, VTE, PVD, pericardial disease, bradycardia and heart block, and other arrythmias (not including atrial fibrillation or atrial flutter). See Supplemental Table 1. A patient was considered to have a CVD outcome if the disease was recorded at least once (UK-CPRD) or at least five times (US-DOD). Chronic disease outcomes in the US-DOD require more recordings because the claims nature of the data includes rule-out diagnoses (
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      ). MACE was defined as the first of MI, stroke or cardiac arrest (US-DOD)/sudden cardiac death (UK-CPRD).
      Diagnoses were identified by ICD-9- and ICD-10-CM codes in the US-DOD and Read codes in the UK-CPRD GOLD. Medications were identified by National Drug Code and Healthcare Common Procedure Coding System codes in the US-DOD and GemScript codes in the UK-CPRD GOLD. See
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      for further details of MS and CVD definitions.

      2.3 Statistical analysis

      Using Byar's method, we estimated incidence rates (IRs) and incidence rate ratios (IRRs) of each CVD outcome with 95% confidence intervals (CIs), stratified by age and sex. We then provided cumulative incidence functions for each study outcome in both cohorts using the Kaplan Meier method. Each patient was followed from cohort entry (date of first MS diagnosis or matched date in the non-MS patients) to censor date. The censor date was the first of the following: 1) study outcome occurred (separately for each outcome), 2) loss of eligibility for US-DOD insurance coverage 3) end of patient record, 4) death or 5) end of data collection. Proportions were compared using a chi-square test or, where cell sizes were less than 5, a Fisher's exact test. Kaplan Meier curves were compared using a log rank test. Statistical analyses were carried out using SAS Release 9.3 (US-DOD study), SAS Release 9.4 (UK-CPRD study) (SAS Institute Inc., Cary NC, USA), and R v3.3.3 (both studies). Testing was not adjusted for multiplicity as this was a descriptive study of two independent large data sets exploring clinically relevant trends.

      2.4 Ethics review

      This study has been approved by review boards in the US and UK. The US-DOD study was approved by the Naval Medical Center, Portsmouth, VA, USA, institutional review board (IRB). All data derived from the Military Health System were de-identified prior to analysis and the research was conducted in compliance with federal and state laws, including the Health Insurance Portability and Accountability Act of 1996 (HIPAA) and Safe Harbor guidelines. The UK-CPRD study is based on data from the Clinical Practice Research Datalink obtained under license from the UK Medicines and Healthcare Products Regulatory Agency. The data is provided by patients and collected by the NHS as part of their care and support. This study was approved by the Independent Scientific Advisory Committee for Medicines and Healthcare Products Regulatory Agency database research (protocol no: 17_037RA). The interpretation and conclusions contained in this study are those of the authors alone.

      3. Results

      3.1 Study population at cohort entry

      From the overall study population described previously (
      • Persson R.
      • Lee S.
      • Ulcickas Yood M.
      • et al.
      Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
      ), we excluded 378 MS patients with prior CVD and 1911 with prior CVD-associated comorbidities from the US-DOD, and from the UK-CPRD we excluded 460 with prior CVD and 746 with prior comorbidities. The final study population included 6406 (US-DOD) and 5726 (UK-CPRD) MS patients matched, respectively, to 66,281 and 57,331 non-MS patients. Most MS patients were female (71%), and a large proportion was diagnosed before age 40 (55% in the US-DOD and 45% in the UK-CPRD); in both databases, more than 90% of MS patients were diagnosed before age 60. In the UK-CPRD MS patients were more likely than non-MS patients to be smokers and body mass indexes (BMIs) at cohort entry were similar. These life-style factors were missing in a large proportion of patients in the US-DOD. Median follow-up in the database after cohort entry was similar for MS and non-MS patients in the UK-CPRD (∼5 years) but was longer in MS patients compared to non-MS patients in the US-DOD (7.2 versus 5.3 years). See Table 1 for details.
      Table 1Patient characteristics at cohort entry (MS diagnosis /matching date), by database: 2004–2017 (US-DOD) and 2001–2016 (UK-CPRD GOLD).
      US - DODUK - CPRD
      MS patients N = 6406 (%)Non-MS patients N = 66,281 (%)MS patients N = 5726 (%)Non-MS patients N = 57,331 (%)
      Age
      Matching criteria.
       <403480 (54.3)36,399 (54.9)2582 (45.1)26,267 (45.8)
       40–592352 (36.7)25,239 (38.1)2705 (47.2)27,010 (47.1)
       60+574 (9.0)4643 (7.0)439 (7.7)4054 (7.1)
       Median (range)38 (7–85)38 (7–85)41 (1–87)41 (1–90)
      Sex
      Matching criteria.
       Female4547 (71.0)47,118 (71.1)4116 (71.9)40,973 (71.5)
       Male1859 (29.0)19,163 (28.9)1610 (28.1)16,358 (28.5)
      Calendar year of cohort entry
      Matching criteria.
      ,
      The first year of US-DOD study was 2004. The last year of the UK-CPRD study was 2016.
       2001–20051454 (22.7)14,467 (21.8)2040 (35.6)20,534 (35.8)
       2006–20102656 (41.5)27,454 (41.4)2005 (35.0)20,041 (35.0)
       2011–20172296 (35.8)24,360 (36.8)1681 (29.4)16,756 (29.2)
      Smoking
       Yes767 (12.0)
      p < 0.0001.
      5667 (8.6)1624 (28.4)
      p < 0.0001.
      13,182 (23.0)
       Former1231 (21.5)10,457 (18.2)
       No2907 (45.4)27,821 (42.0)2375 (41.5)27,118 (47.3)
       Unknown2732 (42.7)32,793 (49.5)496 (8.7)6574 (11.5)
      BMI
      Underweight < 18.5 kg/m2, normal weight 18.5–25 kg/m2, overweight 25–30 kg/m2, obese 30+ kg/m2.
       Underweight70 (1.1)
      p < 0.0001.
      526 (0.8)162 (2.8)1288 (2.3)
       Normal weight1280 (20.0)11,824 (17.8)2008 (35.1)19,984 (34.9)
       Overweight1410 (22.0)12,960 (19.6)1463 (25.6)14,669 (25.6)
       Obese904 (14.1)8294 (12.5)995 (17.4)10,385 (18.1)
       Unknown2742 (42.8)32,677 (49.3)1098 (19.2)11,005 (19.2)
      Length of record before cohort entry
       1–2 years2115 (33.0)
      p < 0.0001.
      19,158 (28.9)755 (13.2)7260 (12.7)
       3–5 years1934 (30.2)22,139 (33.4)943 (16.5)9810 (17.1)
       6–9 years1535 (24.0)17,166 (25.9)1216 (21.2)11,851 (20.7)
       10+ years822 (12.8)7818 (11.8)2812 (49.1)28,410 (49.6)
       Median (years)4.54.79.89.9
      Length of record after cohort entry
       <1 year376 (5.9)
      p < 0.0001.
      9270 (14.0)755 (13.2)7405 (12.9)
       1–2 years883 (13.8)12,742 (19.2)1170 (20.4)11,553 (20.2)
       3–5 years1369 (21.4)13,725 (20.7)1390 (24.3)13,797 (24.1)
       6–9 years1901 (29.7)15,783 (23.8)1366 (23.9)13,565 (23.7)
       10+ years1877 (29.3)14,761 (22.3)1045 (18.3)11,021 (19.2)
       Median5.37.24.95.0
      a Matching criteria.
      b The first year of US-DOD study was 2004. The last year of the UK-CPRD study was 2016.
      c p < 0.0001.
      d Underweight < 18.5 kg/m2, normal weight 18.5–25 kg/m2, overweight 25–30 kg/m2, obese 30+ kg/m2.

      3.2 Incidence of MI

      The incidence of MI was higher among MS patients than non-MS patients in the US-DOD (IRR 2.11; 95% CI 1.56–2.80), but not materially higher in the UK-CPRD (IRR 1.32; 95% CI 0.79–2.08) (Table 2 and Fig. 1). Among females, the risk of MI was more than 2.5 higher among MS patients than non-MS patients in both databases (US-DOD IRR 2.81; 95% CI 1.92–4.01 and UK-CPRD IRR 2.55; 95% CI 1.40–4.37). This effect was not observed among males although the number of cases was small. (Table 3)
      Table 2Incidence rates (IRs) and incidence rate ratios (IRRs) for cardiovascular events after cohort entry among MS and non-MS patients, by database: 2004–2017 (US-DOD) and 2001–2016 (UK-CPRD GOLD).
      MS PatientsNon-MS PatientsIRR (95% CI)
      # eventsPYIR (95% CI) per 10,000 PY# eventsPYIR (95% CI) per 10,000 PY
      MACEUS - DOD11844,87726.3 (21.9, 31.4)488388,50112.6 (11.5, 13.7)2.09 (1.70, 2.55)
      UK - CPRD4832,22614.9 (11.1, 19.6)349328,29910.6 (9.6, 11.8)1.40 (1.02, 1.87)
      MIUS - DOD5645,18512.4 (9.5, 16.0)229389,3705.9 (5.2, 6.7)2.11 (1.56, 2.80)
      UK - CPRD1932,3325.9 (3.7, 9.0)146328,9934.4 (3.8, 5.2)1.32 (0.79, 2.08)
      StrokeUS - DOD5945,14013.1 (10.0, 16.7)251389,4286.4 (5.7, 7.3)2.03 (1.51, 2.67)
      UK - CPRD2932,2999.0 (6.1, 12.7)205328,9626.2 (5.4, 7.1)1.44 (0.96, 2.09)
      Cardiac arrestUS - DOD745,3911.5 (0.7, 3.0)34390,1210.9 (0.6, 1.2)1.77 (0.72, 3.76)
      Sudden cardiac deathUK - CPRD032,4050 (0, 0.8)NR329,6640.1 (0.0, 0.2)
      VTEUS - DOD17144,55838.4 (32.9, 44.5)586387,93715.1 (13.9, 16.4)2.54 (2.14, 3.00)
      UK - CPRD6632,20020.5 (16.0, 25.9)346328,40010.5 (9.5, 11.7)1.95 (1.48, 2.51)
      PVDUS - DOD6745,03514.9 (11.6, 18.8)232388,9116.0 (5.2, 6.8)2.49 (1.89, 3.25)
      UK - CPRD1532,3524.6 (2.7, 7.5)65329,4242.0 (1.5, 2.5)2.35 (1.29, 4.00)
      Heart failureUS - DOD9144,95720.2 (16.4, 24.7)415388,53010.7 (9.7, 11.7)1.90 (1.50, 2.37)
      UK - CPRD1032,3723.1 (1.6, 5.5)107139,3523.2 (2.7, 3.9)0.95 (0.47, 1.73)
      Angina and ischemic heart diseaseUS - DOD12144,78827.0 (22.5, 32.2)614387,40815.8 (14.6, 17.1)1.70 (1.40, 2.06)
      UK - CPRD1632,3794.9 (2.9, 7.8)165328,9115.0 (4.3, 5.8)0.99 (0.57, 1.59)
      TIA and unspecified cerebrovascular diseaseUS - DOD9044,62020.2 (16.3, 24.7)294413,9937.1 (6.3, 7.9)2.84 (2.23, 3.58)
      UK - CPRD1632,3494.9 (2.9, 7.8)178328,9655.4 (4.7, 6.3)0.91 (0.53, 1.47)
      Pericardial diseaseUS - DOD1445,3353.1 (1.8, 5.0)50390,0501.3 (1.0, 1.7)2.41 (1.28, 4.24)
      UK - CPRDNR32,3910.9 (0.3, 2.5)28329,5540.8 (0.6, 1.2)1.09 (0.26, 3.08)
      Bradycardia & heart block
      ICD code for bradycardia code was first introduced in US in October 2015.
      US - DOD2345,3415.1 (3.3, 7.5)145389,5343.7 (3.2, 4.4)1.36 (0.86, 2.07)
      UK - CPRD932,3842.8 (1.4, 5.1)98329,2933.0 (2.4, 3.6)0.94 (0.44, 1.75)
      Other arrhythmias
      Ventricular fibrillation and flutter, premature beats, other specified and unspecified cardiac dysrhythmias, anomalous atrioventricular excitation, other specified and unspecified conduction disorders.
      US - DOD16044,43236.0 (30.8, 41.9)731386,19818.9 (17.6, 20.3)1.90 (1.60, 2.25)
      UK - CPRD5132,20315.8 (11.9, 20.6)407327,87912.4 (11.3, 13.7)1.28 (0.94, 1.69)
      PY=person-years, CI=confidence interval, NR=not reportable due to small cell size per CPRD data usage policy.
      a ICD code for bradycardia code was first introduced in US in October 2015.
      b Ventricular fibrillation and flutter, premature beats, other specified and unspecified cardiac dysrhythmias, anomalous atrioventricular excitation, other specified and unspecified conduction disorders.
      Fig. 1
      Fig. 1Cumulative incidence of myocardial infarction after cohort entry (MS diagnosis/matched date) among MS patients and non-MS patients in the US-DOD 2004–2017 (A) and in the UK-CPRD GOLD 2001–2016 (B).
      Table 3Incidence of cardiovascular events among MS and non-MS patients, by sex and database: 2004–2017 (US-DOD) and 2001–2016 (UK-CPRD GOLD).
      MS PatientsNon-MS PatientsIRR (95% CI)
      # eventsPYIR (95% CI) per 10,000 PY# eventsPYIR (95% CI) per 10,000 PY
      MACEUS - DODFemale8231,98325.6 (20.5, 31.7)289277,85810.4 (9.3, 11.7)2.47 (1.92, 3.13)
      Male3612,89427.9 (19.9, 38.2)199110,64318.0 (15.6, 20.6)1.55 (1.07, 2.18)
      UK - CPRDFemale2923,34512.4 (8.5, 17.6)188238,0967.9 (6.8, 9.1)1.57 (1.04, 2.29)
      Male19888121.4 (13.3, 32.7)16190,20317.8 (15.2, 20.8)1.20 (0.72, 1.88)
      MIUS - DODFemale3832,20611.8 (8.5, 16.0)117278,4064.2 (3.5, 5.0)2.81 (1.92, 4.01)
      Male1812,97913.9 (8.5, 21.4)112110,96410.1 (8.4, 12.1)1.37 (0.81, 2.20)
      UK - CPRDFemale1523,3866.4 (3.7, 10.3)60238,4922.5 (1.9, 3.2)2.55 (1.40, 4.37)
      MaleNR89464.5 (1.5, 10.6)8690,5009.5 (7.7, 11.7)0.47 (0.14, 1.13)
      StrokeUS - DODFemale4232,16513.1 (9.5, 17.5)166278,3396.0 (5.1, 6.9)2.19 (1.54, 3.04)
      Male1712,97513.1 (7.9, 20.5)85111,0897.7 (6.2, 9.4)1.71 (0.98, 2.81)
      UK - CPRDFemale1423,4136.0 (3.4, 9.8)128238,3625.4 (4.5, 6.4)1.11 (0.61, 1.86)
      Male15888616.9 (9.9, 27.1)7790,6008.5 (6.8, 10.6)1.99 (1.10, 3.35)
      VTEUS - DODFemale12231,71138.5 (32.1, 45.8)411277,20814.8 (13.4, 16.3)2.59 (2.11, 3.16)
      Male4912,84738.1 (28.5, 50.0)175110,72915.8 (13.6, 18.3)2.41 (1.74, 3.28)
      UK - CPRDFemale3823,34316.3 (11.7, 22.1)250237,83410.5 (9.3, 11.9)1.55 (1.08, 2.15)
      Male28885831.6 (21.5, 45.0)9690,56610.6 (8.6, 12.9)3.00 (1.92, 4.48)
      PVDUS - DODFemale4932,07915.3 (11.4, 20.0)154277,9395.5 (4.7, 6.5)2.76 (1.98, 3.77)
      Male1812,95613.9 (8.5, 21.5)78110,9727.0 (5.6, 8.7)1.98 (1.15, 3.22)
      UK - CPRDFemale723,4233.0 (1.3, 5.9)33238,6371.4 (1.0, 1.9)2.16 (0.88, 4.60)
      Male889299.0 (4.2, 16.9)3290,7873.5 (2.5, 4.9)2.54 (1.09, 5.25)
      CI=confidence interval, NR=not reportable due to small cell size per CPRD data usage policy, PY=person-years.

      3.3 Incidence of other major cardiac events

      Risks of stroke and MACE were also higher among MS compared to non-MS patients in both databases however these results may be biased by the misdiagnosis of MS symptoms as stroke symptoms. (Table 2 and Fig. 2) Stroke diagnoses were especially common in the US-DOD within the first weeks after first MS diagnosis (cohort entry) suggesting that the differential diagnosis process had not been completed in some cases. In order to address concerns of case misclassification, we performed a sensitivity analysis where we moved the cohort entry date to 6 months after first MS diagnosis or the matched date in the non-MS patients. The rate of stroke among MS patients was attenuated but still elevated compared to non-MS patients (US-DOD: IRR 1.63; 95% CI 1.16–2.24). The risk was elevated among females (US-DOD: IRR 1.96; 95% CI 1.32–2.82) and not males (US-DOD: IRR 1.01; 95% CI 0.47–1.90). Cardiac arrest and sudden cardiac death were rare in both databases (Table 2).
      Fig. 2
      Fig. 2Cumulative incidence of major adverse cardiac events (MACE) after cohort entry (MS diagnosis/matched date) among MS patients and non-MS patients in the US-DOD 2004–2017 (A) and in the UK-CPRD GOLD 2001–2016 (B).

      3.4 Incidence of VTE

      Risk of VTE was higher among MS patients than non-MS patients in both databases (US-DOD IRR 2.54; 95% CI 2.14–3.00 and UK-CPRD IRR 1.95; 95% CI 1.48–2.51). (Table 2 and Fig. 3) The risk was elevated among both sexes, although the risk was higher among males than females in the UK-CPRD (Table 3). Patients with MS had increased risk of VTE compared with non-MS patients across all age categories, although the trend across age groups was not consistent between databases (Table 4).
      Fig. 3
      Fig. 3Cumulative incidence of venous thromboembolism (VTE) after cohort entry (MS diagnosis/matched date) among MS patients and non-MS patients in the US-DOD 2004–2017 (A) and in the UK-CPRD GOLD 2001–2016 (B).
      Table 4Incidence of cardiovascular events among MS and non-MS patients, by age and database: 2004–2017 (US-DOD) and 2001–2016 (UK-CPRD GOLD).
      AgeMS PatientsNon-MS PatientsIRR (95% CI)
      # eventsPYIR (95% CI) per 10,000 PY# eventsPYIR (95% CI) per 10,000 PY
      MACEUS - DOD<40816,7424.8 (2.3, 9.0)20125,3971.6 (1.0, 2.4)3.00 (1.24, 6.56)
      40–595421,01425.7 (19.5, 33.3)199198,94610.0 (8.7, 11.5)2.57 (1.88, 3.44)
      60+56712178.6 (60.0, 101.3)26964,15841.9 (37.1, 47.2)1.88 (1.39, 2.48)
      UK - CPRD<40NR10,4192.9 (0.8, 7.7)16103,8961.5 (0.9, 2.4)1.87 (0.43, 5.61)
      40–592317,60313.1 (8.5, 19.3)174179,8589.7 (8.3, 11.2)1.35 (0.85, 2.04)
      60+22420352.3 (33.7, 77.8)15944,54635.7 (30.5, 41.6)1.47 (0.91, 2.24)
      MIUS - DOD<40216,7641.2 (0.2, 3.8)7125,4220.6 (0.2, 1.1)2.14 (0.32, 8.84)
      40–592421,13111.4 (7.5, 16.6)94199,2784.7 (3.8, 5.7)2.41 (1.50, 3.70)
      60+30728941.2 (28.3, 57.9)12864,67019.8 (16.6, 23.4)2.08 (1.37, 3.05)
      UK - CPRD<40010,4310 (0.0, 2.4)6103,9210.6 (0.2, 1.2)
      40–591317,6517.4 (4.1, 12.2)88180,1754.9 (3.9, 6.0)1.51 (0.80, 2.60)
      60+6425014.1 (5.9, 29.1)5244,89911.6 (8.7, 15.1)1.22 (0.47, 2.62)
      StrokeUS - DOD<40516,7503.0 (1.1, 6.5)12125,4221.0 (0.5, 1.6)3.12 (0.99, 8.41)
      40–592721,10712.8 (8.6, 18.3)91199,3374.6 (3.7, 5.6)2.80 (1.79, 4.24)
      60+27728337.1 (25.0, 53.1)14864,66922.9 (19.4, 26.8)1.62 (1.05, 2.40)
      UK - CPRD<40NR10,4192.9 (0.8, 7.7)10103,9041.0 (0.5, 1.7)2.99 (0.67, 9.78)
      40–591017,6395.7 (2.9, 10.1)87180,2264.8 (3.9, 5.9)1.17 (0.57, 2.15)
      60+16424137.7 (22.4, 59.8)10844,83424.1 (19.9, 29.0)1.57 (0.89, 2.57)
      VTEUS - DOD<404216,62225.3 (18.5, 33.8)97125,1667.7 (6.3, 9.4)3.26 (2.25, 4.65)
      40–597820,83837.4 (29.8, 46.5)298198,47915.0 (13.4, 16.8)2.49 (1.93, 3.18)
      60+51709871.9 (54.1, 93.7)19164,29229.7 (25.7, 34.2)2.42 (1.76, 3.27)
      UK - CPRD<401110,39710.6 (5.6, 18.3)62103,7646.0 (4.6, 7.6)1.77 (0.88, 3.22)
      40–593317,59418.8 (13.1, 26.0)180179,85810.0 (8.6, 11.6)1.87 (1.27, 2.68)
      60+22420952.3 (33.7, 77.7)10444,77923.2 (19.1, 28.0)2.25 (1.39, 3.49)
      PVDUS - DOD<40116,7650.6 (0.1, 2.8)5125,4240.4 (0.2, 0.9)
      40–592821,07613.3 (9.0, 18.9)78199,2823.9 (3.1, 4.9)3.39 (2.17, 5.16)
      60+38719552.8 (38.0, 71.7)14964,20523.2 (19.7, 27.2)2.28 (1.57, 3.21)
      UK - CPRD<40010,4310 (0.0, 2.4)0103,9290 (0.0, 0.2)
      40–59917,6675.1 (2.5, 9.3)31180,4491.7 (1.2, 2.4)2.97 (1.33, 5.98)
      60+6425414.1 (5.9, 29.1)3445,0487.5 (5.3, 10.4)1.87 (0.71, 4.13)
      CI=confidence interval, NR=not reportable due to small cell size per CPRD data usage policy, PY=person-years.

      3.5 Incidence of PVD

      MS patients had a higher risk of PVD across databases (Table 2) and sexes (Table 3). Few PVD cases were identified in patients younger than 40 years. In patients older than 40 years, the risk of PVD was consistently higher among MS patients compared with non-MS patients (Table 4).

      3.6 Incidence of other CVD

      Risks of other CVD diseases (heart failure, angina and ischemic heart disease, TIA and unspecified cerebrovascular disease, pericardial disease, and other arrhythmias) among MS patients compared with non-MS patients were increased in the US-DOD but not in the UK-CPRD. Risk of bradycardia/heart block was not elevated among MS patients in either database, although the number of events was small (Table 2).

      4. Discussion

      In these two studies of more than 12,000 MS patients with no history of CVD from two electronic health databases in US and the UK, rates of VTE and PVD were 2-fold higher among patients with MS than non-MS patients and rates of MI were 2.5 times higher among female MS patients compared with non-MS females. In the US-DOD, but not the UK-CPRD, MS patients had higher rates of stroke, heart failure, angina and ischemic heart disease, TIA and unspecified cerebrovascular disease, pericardial disease, and other arrhythmias compared with non-MS patients. Finally, the number of bradycardia and heart block events was small and rates were similar in MS and non-MS patients in both databases. The finding of higher VTE, PVD, and MI (among females) from two different countries and databases are noteworthy because the current paper excluded not only baseline CVD but also baseline diabetes, hypertension and dyslipidemia. The increased risk of incident PVD for patients with MS aged >40 years observed in both databases is a novel finding that, to our knowledge, has not previously been reported.
      The US-DOD database is a claims database, while the UK-CPRD GOLD database is based on electronic medical records (EMR) as recorded by general practitioners. As a result of the EMR nature of the UK-CPRD GOLD, infusions conducted outside of the general practioner's office are often not captured in the patient record, thus information on many MS treatments was not available for the UK-CPRD study and therefore we did not require that MS patients in the UK-CPRD receive treatment to be included in the study population. For this reason it is possible that we excluded some less severe MS patients from the US-DOD analysis and that this US cohort may comprise more severe MS patients compared to the UK cohort. This could in part explain why we observed higher rates of CVD in the US-DOD data compared with the UK-CPRD data. However, incidence rates of all cardiovascular events were higher in the US-DOD than in the UK-CPRD, both for MS and non-MS patients. The difference in cardiovascular risk between the US and UK may be a result of differences in the population in obesity, smoking or other unmeasured cardiovascular risks or due to differences in the health care system but this should not affect the IRRs. Differences may also be due to the claims nature of the US data, as diagnosis codes are used to bill for all tests and treatments for both confirmed and suspected conditions. To be classified as having CVD we used a threshold of 5 diagnoses on different days in the US-DOD to minimize misclassification, however the estimates of relative rates may be biased due to detection or surveillance bias where patients with MS may have more contact with the medical system. Another possible explanation for the difference is that the median follow-up in the database after cohort entry was similar for MS and non-MS patients in the UK-CPRD (∼5 years) but was longer in MS patients compared to non-MS patients in the US-DOD (7.2 versus 5.3 years). The longer follow-up in the US MS patients would result in an older population at the end of follow-up which could partially explain the higher rates of CVD in these patients. In addition the estimates of relative rates of cardiac events between MS and non-MS patients may be subject to differential bias due to misdiagnosis of MS symptoms, especially for stroke, or to increased thrombosis due to increased hospitalization in MS patients. Despite the differences in type of record and country, results from both databases suggest that patients with MS are at increased risk of VTE, PVD, and, among females, MI.
      This study included newly diagnosed MS, not new onset (intitial symptoms prior to actual diagnosis) and, in the US-DOD cohort, all MS patients received at least one disease-modifying treatment. Strictly speaking the results of this study are generalizable to the UK population and US members of the DOD and their family members, but there is no reason to suspect that these results would be materially different in other Western populations. The similarity of the results between the US and UK support this proposition.
      We found a 2–3 fold increased risk of incident VTE following MS diagnosis in both databases in our study, which is consistent with prior studies including one in the UK-CPRD (years 1987–2009) that reported relative risks (RRs) of 2.56 (95% CI 2.06–3.20) (
      • Peeters P.J.H.L.
      • Bazelier M.T.
      • Uitdehaag B.M.J.
      • et al.
      The risk of venous thromboembolism in patients with multiple sclerosis: the clinical practice research datalink.
      ), as well as in Sweden: RR 3.09 (95%CI: 2.30–4.16) for deep vein thrombosis (DVT) and RR 1.79 (95%CI 1.36–2.34) for pulmonary embolism (PE) (
      • Roshanisefat H.
      • Bahmanyar S.
      • Hillert J.
      • et al.
      Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.
      ). In Denmark, the IRRs for DVT and PE were 3.02 (95% CI 2.14–4.27) and 2.85 (95% CI 1.72–4.70), respectively, during the first year following MS diagnosis and slightly lower, 2.28 (95% CI 2.01–2.59) and 1.58 (95% CI 1.31–1.92) over 2–30 years of follow-up (
      • Christensen S.
      • Farkas D.K.
      • Pedersen L.
      • et al.
      Multiple sclerosis and risk of venous thromboembolism: a population-based cohort study.
      ). In a recent study from Finland, the odds ratio (OR) of PE for MS patients compared with population controls was 2.12 (95% CI 1.12–3.99) (
      • Murtonen A.
      • Kurki S.
      • Hänninen K.
      • et al.
      Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
      ). Lastly, a small study from Taiwan also reported a 7-fold increased risk of VTE among MS patients based on 8 cases among MS patients (
      • Chung W.-S.
      • Lin C.-L.
      • Tsai T.-C.
      • et al.
      Multiple sclerosis increases the risk of venous thromboembolism: a nationwide cohort analysis.
      ).
      A recent population based Canadian study found a small increased risk of incident MI in MS compared to non-MS patients, consistent with our results (
      • Marrie R.A.
      • Garland A.
      • Schaffer S.A.
      • et al.
      Traditional risk factors may not explain increased incidence of myocardial infarction in MS.
      ). A report from the population-based Danish National Registry also found a higher risk of MI compared with the general population. This report also found that the increased risk of MI was limited to the first few years after MS diagnosis which is not consistent with the findings of our study (Figs. 1 and 2) (
      • Christiansen C.F.
      • Christensen S.
      • Farkas D.K.
      • et al.
      Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study.
      ;
      • Christiansen C.F.
      Risk of vascular disease in patients with multiple sclerosis: a review.
      ). Our finding of a stronger relative risk of MI for women than men was also seen in a Swedish nationwide study of MS patients that observed an increased risk of first-time MI (IRR 1.85; 95% CI 1.5–2.16) compared with non-MS patients, with a stronger effect among women (IRR 2.01; 95% CI 1.63–2.49) than men (IRR 1.61; 95% CI 1.30–2.00) (
      • Jadidi E.
      • Mohammadi M.
      • Moradi T
      High risk of cardiovascular diseases after diagnosis of multiple sclerosis.
      ).
      Stroke and TIA/unspecified cerebrovascular disease were elevated in the US-DOD database but not in the UK-CPRD. Prior studies, including a study in the US-DOD (2006–2011), have also found an increased risk of stroke, TIA and cerebrovascular disease among MS patients (
      • Jadidi E.
      • Mohammadi M.
      • Moradi T
      High risk of cardiovascular diseases after diagnosis of multiple sclerosis.
      ;
      • Murtonen A.
      • Kurki S.
      • Hänninen K.
      • et al.
      Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
      ;
      • Thormann A.
      • Magyari M.
      • Koch-Henriksen N.
      • et al.
      Vascular comorbidities in multiple sclerosis: a nationwide study from Denmark.
      ;
      • Tseng C.-.H.H.
      • Huang W.-S.S.
      • Lin C.-L.L.
      • et al.
      Increased risk of ischaemic stroke among patients with multiple sclerosis.
      ;
      • Capkun G.
      • Dahlke F.
      • Lahoz R.
      • et al.
      Mortality and comorbidities in patients with multiple sclerosis compared with a population without multiple sclerosis: an observational study using the US department of defense administrative claims database.
      ). However, studies using the Swedish Multiple Sclerosis Register found that the risks of stroke and cerebrovascular diagnoses among MS patients were increased only for the first year following MS diagnosis (
      • Roshanisefat H.
      • Bahmanyar S.
      • Hillert J.
      • et al.
      Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.
      ;
      • Zöller B.
      • Li X.
      • Sundquist J.
      • et al.
      Risk of subsequent ischemic and hemorrhagic stroke in patients hospitalized for immune-mediated diseases: a nationwide follow-up study from Sweden.
      ). As MS, stroke and TIA may have similar symptoms, it is possible that there are more codes for these cerebrovascular outcomes among MS compared to non-MS patients as a consequence of the diagnostic work up inherent in the development of the differential diagnosis and the resultant billing required by US health insurers. As diagnosis codes for stroke were particularly common near the time of first MS diagnosis, we conducted an additional sensitivity analysis where we estimated the risk of stroke starting 6 months after cohort entry. In this analysis the risk of stroke attenuated but remained elevated among females, but not males. However, we cannot rule out the effect of differential misclassification of cerebrovascular outcomes in later years as symptoms increase with MS progression.
      As noted, heart failure, angina and ischemic heart disease, pericardial disease, and other arrhythmias (e.g. ventricular fibrillation and flutter, premature beats) were elevated for MS compared with non-MS patients in the US-DOD database but not the UK-CPRD. Increased risks of heart failure have been observed in Denmark, Sweden and Finland (
      • Jadidi E.
      • Mohammadi M.
      • Moradi T
      High risk of cardiovascular diseases after diagnosis of multiple sclerosis.
      ;
      • Christiansen C.F.
      • Christensen S.
      • Farkas D.K.
      • et al.
      Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study.
      ;
      • Murtonen A.
      • Kurki S.
      • Hänninen K.
      • et al.
      Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
      ). However, ischemic heart disease was not increased in MS patients in Denmark, Sweden or Finland (
      • Roshanisefat H.
      • Bahmanyar S.
      • Hillert J.
      • et al.
      Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.
      ;
      • Murtonen A.
      • Kurki S.
      • Hänninen K.
      • et al.
      Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
      ;
      • Thormann A.
      • Magyari M.
      • Koch-Henriksen N.
      • et al.
      Vascular comorbidities in multiple sclerosis: a nationwide study from Denmark.
      ). Prior data on the rates of pericardial disease and other arrythmias among MS patients are sparse (
      • Murtonen A.
      • Kurki S.
      • Hänninen K.
      • et al.
      Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
      ).
      Strengths of our study include the use of population based databases from two different countries, the large number of MS and non-MS patients, long-follow-up, and restriction to patients with no baseline CVD or comorbidities strongly associated with CVD including hypertension, diabetes and dyslipidemia. Study limitations are inherent in the use of electronic records, particularly claims data, for health studies. For example, neither database had information on physical activity, and, in the US-DOD database, smoking and BMI data was missing in a large proportion of the cohort, limiting the ability to evaluate these covariates in relation to differences in CVD rates. Although we did not use validated algorithms for our study outcomes in the US-DOD database, we required 5 or more diagnoses for each outcome to minimize misclassification. We did not have complete information on ethnicity, smoking, BMI, or physical activity in these databases and thus were not able to address whether MS is an independent risk factor for CVD or if it is associated with any of these covariates. Never-the-less it is useful to know that MS patients have a higher risk of CVD compared to non-MS patients.

      5. Conclusion

      Patients with MS have increased risks of VTE and PVD compared with non-MS patients after MS diagnosis. The risk of MI is also increased among female patients with MS.

      Disclaimer

      The views expressed in this work are those of the authors and do not necessarily reflect the official policy or position of the Department of the Navy, Department of Defense or the United States Government. Capt Michael Wagner is a military service member. This work was prepared as part of his official duties. Title 17 U.S.C. 105 provides that “Copyright protection under this title is not available for any work of the United States Government.” Title 17 U.S.C. 101 defines a United States Government work as a work prepared by a military service member or employee of the United States Government as part of their official duties. Research data was derived with oversight from Naval Medical Center, Portsmouth, Virginia IRB, NHRS identifier; number NMCP.2017.0039.

      Declaration of Competing Interest

      This study was supported by Celgene Corporation. SL, NM, SN and AL are employees of Celgene Corporation. AE is an employee of Health ResearchTx LLC, which has a business relationship with Celgene Corporation.

      Acknowledgment

      We would like to thank Dr. Ruth Dobson and Dr. Rachel Mackey for their review and valuable input on this manuscript.

      Appendix. Supplementary materials

      References

        • Á Alonso
        • SS Jick
        • Olek M.J.
        • et al.
        Recent use of oral contraceptives and the risk of multiple sclerosis.
        Arch. Neurol. 2005; 62: 1362
        • Alonso A.
        • Jick S.S.
        • Hernan M.A
        Allergy, histamine 1 receptor blockers, and the risk of multiple sclerosis.
        Neurology. 2006; 66: 572-575
        • Alonso A.
        • Jick S.S.
        • Jick H.
        • et al.
        Antibiotic use and risk of multiple sclerosis.
        Am. J. Epidemiol. 2006; 163: 997-1002
        • Alonso A.
        • Jick S.S.
        • Olek M.J.
        • et al.
        Incidence of multiple sclerosis in the United Kingdom.
        J. Neurol. 2007; 254: 1736-1741
        • Capkun G.
        • Dahlke F.
        • Lahoz R.
        • et al.
        Mortality and comorbidities in patients with multiple sclerosis compared with a population without multiple sclerosis: an observational study using the US department of defense administrative claims database.
        Mult. Scler. Relat. Disord. 2015; 4: 546-554
        • Christensen S.
        • Farkas D.K.
        • Pedersen L.
        • et al.
        Multiple sclerosis and risk of venous thromboembolism: a population-based cohort study.
        Neuroepidemiology. 2012; 38: 76-83
        • Christiansen C.F.
        Risk of vascular disease in patients with multiple sclerosis: a review.
        Neurol. Res. 2012; 34: 746-753
        • Christiansen C.F.
        • Christensen S.
        • Farkas D.K.
        • et al.
        Risk of arterial cardiovascular diseases in patients with multiple sclerosis: a population-based cohort study.
        Neuroepidemiology. 2010; 35: 267-274
        • Chung W.-S.
        • Lin C.-L.
        • Tsai T.-C.
        • et al.
        Multiple sclerosis increases the risk of venous thromboembolism: a nationwide cohort analysis.
        Eur. J. Clin. Invest. 2015; 45: 1228-1233
        • Fromont A.
        • Binquet C.
        • Rollot F.
        • et al.
        Comorbidities at multiple sclerosis diagnosis.
        J. Neurol. 2013; 260: 2629-2637
        • Hassan-Smith G.
        • Douglas M.R.
        Epidemiology and diagnosis of multiple sclerosis.
        Br. J. Hosp. Med. (Lond). 2011; 72: M146-M151
        • Hernán M.A.
        • Jick S.S.
        • Logroscino G.
        • et al.
        Cigarette smoking and the progression of multiple sclerosis.
        Brain. 2005; 128: 1461-1465
        • Hernán M.A.
        • Jick S.S.
        • Olek M.J.
        • et al.
        Recombinant hepatitis B vaccine and the risk of multiple sclerosis: a prospective study.
        Neurology. 2004; 63: 838-842
        • Herrett E.
        • Gallagher A.M.
        • Bhaskaran K.
        • et al.
        Data resource profile: clinical practice research datalink (CPRD).
        Int. J. Epidemiol. 2015; 44: 827-836
        • Jadidi E.
        • Mohammadi M.
        • Moradi T
        High risk of cardiovascular diseases after diagnosis of multiple sclerosis.
        Mult. Scler J. 2013; 19: 1336-1340
        • Jick S.S.
        • Li L.
        • Falcone G.J.
        • et al.
        Mortality of patients with multiple sclerosis: a cohort study in UK primary care.
        J. Neurol. 2014; 261: 1508-1517
        • Jick S.S.
        • Li L.
        • Falcone G.J.
        • et al.
        Epidemiology of multiple sclerosis: results from a large observational study in the UK.
        J. Neurol. 2015; 262: 2033-2041
        • Marrie R.A.
        • Fisk J.
        • Tremlett H.
        • et al.
        Differing trends in the incidence of vascular comorbidity in MS and the general population.
        Neurol. Clin. Pract. 2016; 6: 120-128
        • Marrie R.A.
        • Garland A.
        • Schaffer S.A.
        • et al.
        Traditional risk factors may not explain increased incidence of myocardial infarction in MS.
        Neurology. 2019; 92: e1624-e1633
        • Marrie R.A.
        • Reider N.
        • Cohen J.
        • et al.
        A systematic review of the incidence and prevalence of cardiac, cerebrovascular, and peripheral vascular disease in multiple sclerosis.
        Mult. Scler. J. 2015; 21: 318-331
        • Murtonen A.
        • Kurki S.
        • Hänninen K.
        • et al.
        Common comorbidities and survival in MS: risk for stroke, type 1 diabetes and infections.
        Mult. Scler. Relat. Disord. 2018; 19: 109-114
        • Noseworthy J.H.
        • Lucchinetti C.
        • Rodriguez M.
        • et al.
        Multiple sclerosis.
        N. Engl. J. Med. 2000; 343: 938-952
        • Peeters P.J.H.L.
        • Bazelier M.T.
        • Uitdehaag B.M.J.
        • et al.
        The risk of venous thromboembolism in patients with multiple sclerosis: the clinical practice research datalink.
        J. Thromb. Haemost. 2014; 12: 444-451
        • Persson R.
        • Lee S.
        • Ulcickas Yood M.
        • et al.
        Multi-database study of multiple sclerosis: identification, validation and description of MS patients in two countries.
        J. Neurol. 2019; 266 (Epub ahead of print 18 February): 1095-1106https://doi.org/10.1007/s00415-019-09238-8
        • Pugliatti M.
        • Rosati G.
        • Carton H.
        • et al.
        The epidemiology of multiple sclerosis in Europe.
        Eur. J. Neurol. 2006; 13: 700-722
        • Roshanisefat H.
        • Bahmanyar S.
        • Hillert J.
        • et al.
        Multiple sclerosis clinical course and cardiovascular disease risk - Swedish cohort study.
        Eur. J. Neurol. 2014; 21: 1353-1e88
        • Thormann A.
        • Magyari M.
        • Koch-Henriksen N.
        • et al.
        Vascular comorbidities in multiple sclerosis: a nationwide study from Denmark.
        J. Neurol. 2016; 263: 2484-2493
        • Tseng C.-.H.H.
        • Huang W.-S.S.
        • Lin C.-L.L.
        • et al.
        Increased risk of ischaemic stroke among patients with multiple sclerosis.
        Eur. J. Neurol. 2015; 22: 500-506
        • Zöller B.
        • Li X.
        • Sundquist J.
        • et al.
        Risk of subsequent ischemic and hemorrhagic stroke in patients hospitalized for immune-mediated diseases: a nationwide follow-up study from Sweden.
        BMC Neurol. 2012; 12: 41