Advertisement

Risk of haemorrhagic and ischaemic stroke in patients with cancer: A nationwide follow-up study from Sweden

  • Bengt Zöller
    Correspondence
    Corresponding author: Mobile: +46 70 6691476; fax: +46 40 391370.
    Affiliations
    Center for Primary Health Care Research, Lund University/Region Skåne, Clinical Research Centre, Floor 11, Building 28, Entrance 72, Skåne University Hospital, 205 02 Malmö, Sweden
    Search for articles by this author
  • Jianguang Ji
    Affiliations
    Center for Primary Health Care Research, Lund University/Region Skåne, Clinical Research Centre, Floor 11, Building 28, Entrance 72, Skåne University Hospital, 205 02 Malmö, Sweden
    Search for articles by this author
  • Jan Sundquist
    Affiliations
    Center for Primary Health Care Research, Lund University/Region Skåne, Clinical Research Centre, Floor 11, Building 28, Entrance 72, Skåne University Hospital, 205 02 Malmö, Sweden

    Stanford Prevention Research Centre, Stanford University School of Medicine, Medical School Office Building, 251 Campus Drive, Mail Code 5411, Stanford, CA 94305-5411, USA
    Search for articles by this author
  • Kristina Sundquist
    Affiliations
    Center for Primary Health Care Research, Lund University/Region Skåne, Clinical Research Centre, Floor 11, Building 28, Entrance 72, Skåne University Hospital, 205 02 Malmö, Sweden
    Search for articles by this author
Published:February 01, 2012DOI:https://doi.org/10.1016/j.ejca.2012.01.005

      Abstract

      Background

      Stroke is common in cancer patients, but risk estimates for different cancer sites/types have not been determined. The aim of this nationwide study was to examine whether there is an association between cancer and first hospitalisation for haemorrhagic or ischaemic stroke.

      Methods

      All 820,491 individuals in Sweden with a diagnosis of cancer between 1st January 1987 and 31st December 2008 were followed for first hospitalisation for haemorrhagic or ischaemic stroke. The reference population was the total population of Sweden without cancer. Standardised incidence ratios (SIRs) for haemorrhagic and ischaemic strokes were calculated.

      Results

      Overall risk of haemorrhagic stroke and ischaemic stroke during the first 6 months after diagnosis of cancer was 2.2 (95% confidence interval (CI) = 2.0–2.3) and 1.6 (CI = 1.5–1.6), respectively. For 18 and 20 of the 34 cancers studied, respectively, risk of haemorrhagic and ischaemic strokes was increased. Overall stroke risk decreased rapidly, but remained elevated, even 10+ years after diagnosis of cancer 1.2 (CI = 1.1–1.3) for haemorrhagic stroke and 1.1 (CI = 1.1–1.2) for ischaemic stroke. The risk of stroke was highest during the first 6 months after diagnosis of cancer of the nervous system (29 (CI = 25–34) for haemorrhagic stroke and 4.1 (CI = 3.4–4.8) for ischaemic stroke)) or leukaemia (13 (CI = 10–16) for haemorrhagic stroke and 3.0 (CI = 2.5–3.7) for ischaemic stroke)). Metastasis was associated with an increased risk of haemorrhagic stroke 2.2 (CI = 1.8–2.7) and ischaemic stroke 1.5 (CI = 1.3–1.7).

      Interpretation

      Several cancer sites/types are associated with an increased risk of haemorrhagic and ischaemic strokes.

      Keywords

      1. Introduction

      Haemorrhagic and ischaemic strokes are major causes of morbidity and mortality worldwide.
      • Donnan G.A.
      • Fisher M.
      • Macleod M.
      • Davis S.M.
      Stroke.
      In an autopsy study from 1985, 14.6% of cancer patients suffered from ischaemic stroke.
      • Graus F.
      • Rogers L.R.
      • Posner J.B.
      Cerebrovascular complications in patients with cancer.
      However, the clinical significance of these findings is not clear, as the study included silent infarcts and diffuse vascular encephalopathies.
      • Graus F.
      • Rogers L.R.
      • Posner J.B.
      Cerebrovascular complications in patients with cancer.
      Cancer is also overrepresented in stroke patients.
      • Lindvig K.
      • Moller H.
      • Mosbech J.
      • Jensen O.M.
      The pattern of cancer in a large cohort of stroke patients.
      Notably, a more than 10-fold increase in the risk of brain tumours within the first year after diagnosis of stroke was reported in a previous study.
      • Lindvig K.
      • Moller H.
      • Mosbech J.
      • Jensen O.M.
      The pattern of cancer in a large cohort of stroke patients.
      However, diagnostic misinterpretation of a brain tumour as a stroke may have been possible with the study design use.
      • Lindvig K.
      • Moller H.
      • Mosbech J.
      • Jensen O.M.
      The pattern of cancer in a large cohort of stroke patients.
      Minor increases in the risk of cancer at other sites were also found in the first year of follow-up.
      • Lindvig K.
      • Moller H.
      • Mosbech J.
      • Jensen O.M.
      The pattern of cancer in a large cohort of stroke patients.
      In a retrospective study of 96 patients with confirmed stroke, lung cancer was the most common primary tumour (30%), followed by brain malignancy and prostate cancer (9% each).
      • Cestari D.M.
      • Weine D.M.
      • Panageas K.S.
      • Segal A.Z.
      • DeAngelis L.M.
      Stroke in patients with cancer: incidence and etiology.
      Cerebral haemorrhage has been reported frequently in cancer patients, especially those with haematological malignancies. In patients with leukaemia and lymphoma, 72% and 36% of strokes, respectively, were haemorrhagic.
      • Rogers L.R.
      Cerebrovascular complications in cancer patients.
      A number of cancer-related causes of stroke have been suggested,
      • Graus F.
      • Rogers L.R.
      • Posner J.B.
      Cerebrovascular complications in patients with cancer.
      • Nguyen T.
      • DeAngelis L.M.
      Stroke in cancer patients.
      • Haller S.
      • Lyrer P.
      Malignancy and stroke.
      including treatment-related side-effects, direct tumour compressing or invasion of blood vessels, tumour-induced haemostatic activation
      • Franchini M.
      • Montagnana M.
      • Favaloro E.J.
      • Lippi G.
      The bidirectional relationship of cancer and hemostasis and the potential role of anticoagulant therapy in moderating thrombosis and cancer spread.
      or an alteration in blood viscosity, and non-bacterial thrombotic endocarditis.
      • Graus F.
      • Rogers L.R.
      • Posner J.B.
      Cerebrovascular complications in patients with cancer.
      • Nguyen T.
      • DeAngelis L.M.
      Stroke in cancer patients.
      • Haller S.
      • Lyrer P.
      Malignancy and stroke.
      Tumour cells produce various cytokines and chemokines that attract leucocytes, which may trigger an inflammatory response.
      • Coussens L.M.
      • Werb Z.
      Inflammation and cancer.
      This may in turn have prothrombotic and atherosclerotic effects.
      • Libby P.
      Inflammation in atherosclerosis.
      • van Leuven S.I.
      • Franssen R.
      • Kastelein J.J.
      • et al.
      Systemic inflammation as a risk factor for atherothrombosis.
      However, there is no consensus as to whether cancer is an independent risk factor for stroke like traditional risk factors such as hypertension and smoking. It has been proposed that conventional stroke etiologies account for the majority of cerebral ischaemic events in the adult cancer population.
      • Cestari D.M.
      • Weine D.M.
      • Panageas K.S.
      • Segal A.Z.
      • DeAngelis L.M.
      Stroke in patients with cancer: incidence and etiology.
      • Chaturvedi S.
      • Ansell J.
      • Recht L.
      Should cerebral ischemic events in cancer patients be considered a manifestation of hypercoagulability?.
      Moreover, cancer and ischaemic stroke share a common risk factor (tobacco smoke), and an increased risk of ischaemic stroke would be expected among patients with smoking-related cancers (i.e. cancers of the lung, larynx, oesophagus, mouth and tongue, pharynx, urinary bladder, pancreas and kidney).
      • Dreyer L.
      • Olsen J.H.
      Cancer risk of patients discharged with acute myocardial infarct.
      • Dreyer L.
      • Olsen J.H.
      Risk for non-smoking-related cancer in atherosclerotic patients.
      • Horn-Ross P.L.
      • Ljung B.M.
      • Morrow M.
      Environmental factors and the risk of salivary gland cancer.
      • Fernberg P.
      • Odenbro A.
      • Bellocco R.
      • et al.
      Tobacco use, body mass index, and the risk of leukemia and multiple myeloma: a nationwide cohort study in Sweden.
      Other suspected smoking-related cancers sites include lip, liver, cervix, stomach and salivary gland; leukaemia is also suspected to be cancer-related Thus, cancer and ischaemic stroke share a number of risk factors, and an increased risk of ischaemic stroke among cancer patients might be expected.
      We hypothesised that cancer may affect risk of haemorrhagic and ischaemic strokes. In a nationwide follow-up study of data from 1987 to 2008, we estimated risk of hospitalisation for haemorrhagic and ischaemic strokes in patients with different cancers. An important novel contribution of the study is the use of nationwide validated registers such as the Swedish Cancer Registry and the Hospital Discharge Register, thereby eliminating the risk of selection and recall bias.

      2. Methods

      2.1 MigMed 2 Database

      This study was approved by the Ethics Committee of Lund University, Sweden. Data used in this study were retrieved from the MigMed 2 Database (an updated version of the original MigMed database), maintained at the Center for Primary Health Care Research, Lund University/Region Skåne, Malmö. MigMed 2 contains data on all individuals registered as residents of Sweden. It contains individual-level information on age, sex, occupation, geographic region of residence, hospital diagnoses and dates of hospital admissions in Sweden, as well as country of birth, parents’ country of birth, date of emigration and date and cause of death. The database was constructed using several national Swedish data registers (reviewed by Rosen and Hakulinen),
      • Rosen M.
      • Hakulinen T.
      Use of disease registers.
      including, but not limited to, the Swedish Cancer Registry,

      Cancer incidence in Sweden 2005. Stockholm: The National Board of Health and Welfare; 2007.

      the Swedish National Population and Housing Census, the Total Population Register, the Multi-Generation Register

      The multi-generation register 2009. A description of contents and quality. Statistics Sweden. Available from: <http://www.scb.se/statistik/_publikationer/BE9999_2009A01_BR_BE96BR1003.pdf> [accessed 22.03.11].

      and the Swedish Hospital Discharge Register.

      The Swedish Hospital Discharge Register 1987–1996: quality and contents. Stockholm: The National Board of Health and Welfare; 1998.

      Information retrieved from the various registers in the MigMed 2 Database is linked, at the individual level, via the 10-digit personal identification number assigned to each resident of Sweden for his or her lifetime. Registration numbers were replaced by serial numbers to preserve anonymity. As well as being used to track all records in the database at the individual level, these serial numbers were used to check that individuals with hospital diagnoses of haemorrhagic and ischaemic strokes appeared only once in the dataset (for the first hospital diagnosis of haemorrhagic or ischaemic stroke during the study period).
      The follow-up period for analysis of data in the present study started on 1st January 1987 and continued until hospitalisation for haemorrhagic or ischaemic stroke, death, emigration or the end of the study period (31st December 2008). Data for first hospitalisation for haemorrhagic and ischaemic strokes during the study period were retrieved from the Hospital Discharge Register (1987–2008). This register does not include data for hospital outpatients or patients treated at primary health care centres.

      2.2 Predictor variable

      The predictor variable was diagnosis of cancer in the Swedish Cancer Registry. Cancer site/type was identified according to the 7th revision of the International Classification of Diseases (ICD-7).

      Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

      The Swedish Cancer Registry records all new cases of cancer. Close to 100% of all cases nationwide have been histologically or cytologically confirmed.

      Cancer incidence in Sweden 2005. Stockholm: The National Board of Health and Welfare; 2007.

      Information on metastasis has been included in the Swedish Cancer Registry since 2002.

      2.3 Outcome variables

      Diagnoses of haemorrhagic and ischaemic strokes were based on the 9th and 10th revisions of the International Classification of Diseases (ICD-9 and ICD-10). Cases of ischaemic stroke were identified using the following ICD codes: 433, 434, 435, 437.0, and 437.1 (ICD-9) and I63 (not I636), I65, I66, I67.2 and I67.8 (ICD-10). Cases of haemorrhagic stroke were identified using the following ICD codes: 431 and 432 (ICD-9) and I61 and I62 (ICD-10).

      2.4 Individual-level variables

      The individual-level variables were sex, age, time period, geographic region of residence, socioeconomic status (SES) and comorbidity.
      Sex: male or female.
      Age was divided into 5-year categories. Subjects of all ages were included in the study.
      Time period was divided into five categories: 1987–1990, 1991–1994, 1995–1999, 2000–2004 and 2005–2008.
      Geographic region of residence was included as an individual-level variable to allow adjustment for possible differences in hospital admissions for haemorrhagic and ischaemic strokes between different geographic regions in Sweden. It was categorised as (1) large city (city with a population of >200,000 (i.e. Stockholm, Gothenburg or Malmo)); (2) Southern Sweden (rural and urban); (3) Northern Sweden (rural and urban); or (4) unknown (if geographic region of residence could not be determined). Note: the three ‘large cities’ all lie in Southern Sweden, but were excluded from the Southern Sweden category.
      Occupation was used as a proxy for socioeconomic status (SES). Occupational data were retrieved from national census records in the MigMed 2 Database. We classified each individual’s occupation into one of six categories: (1) blue-collar worker, (2) white-collar worker, (3) professional, (4) self-employed, (5) farmer and (6) non-employed (individuals without paid employment). Students without an occupation were categorised on the basis of their father’s or mother’s occupation. If that was not possible, they were included in the ‘non-employed’ category. For individuals aged <20 years, parental occupation was used.
      Comorbidity was defined as first hospital diagnosis at follow-up (1987–2008) for the following conditions: (1) chronic obstructive pulmonary disease (COPD) (490–496 (ICD-9) and J40–J49 (ICD-10)); (2) obesity (278A (ICD-9) and E65–E68 (ICD-10)); (3) alcoholism (291 and 303 (ICD-9) and F10 (ICD-10)); (4) diabetes mellitus (250 (ICD-9) and E10–E14 (ICD-10)) and (5) hypertension (401–405 (ICD-9) and I10-I15 (ICD-10)).

      2.5 Statistical analysis

      Person-years of risk (i.e. number of persons at risk multiplied by time at risk) were calculated from the time at which subjects were included in the study (1987 or later) until first hospitalisation for haemorrhagic or ischaemic stroke, death and emigration or the end of the study period. The expected number of cases was based on the number of cases in the reference group. Standardised incidence ratios (SIRs) were calculated as the ratio of observed (O) and expected (E) number of haemorrhagic or ischaemic stroke cases using the indirect standardisation method.
      • Rothman K.J.
      • Greenland S.
      Modern epidemiology.
      Expected numbers were calculated for anyone without cancer, that is, for the whole Swedish population without cancer. The expected numbers were calculated as age (5-year groups), sex, period (5-year groups), region and socioeconomic status (=SES)-specific standard incidence rates. An additional adjustment was made for hospitalisation for comorbidities: (1) chronic obstructive pulmonary disease (COPD); (2) obesity; (3) alcoholism; (4) diabetes mellitus and (5) hypertension. Ninety-five percent confidence intervals (95% CI) were calculated assuming a Poisson distribution,
      • Rothman K.J.
      • Greenland S.
      Modern epidemiology.
      All analyses were performed using SAS version 9.2 (SAS Institute, Cary, NC, USA).

      3. Results

      Totally 820,491 cancer cases were identified between 1987 and 2008, 51.9% males (n = 425899) and 48.1% females (n = 394.592).

      Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

      The four most common cancers among males were prostate cancer (n = 139,510), lung cancer (n = 36,424), urinary bladder (n = 29,705) and colon cancer (29,633).

      Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

      The four most common cancers among females were breast cancer (115,705), colon cancer (32,169), endometrium cancer (n = 23,479) and lung cancer (23,220).

      Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

      Table 1 shows the basic characteristics of patients with and without cancer who were hospitalised with haemorrhagic or ischaemic stroke during the study period. A total of 77,031 and 396,702 individuals without cancer were hospitalised with a main diagnosis of haemorrhagic and ischaemic strokes, respectively (Table 1), while 6926 and 31,524 individuals with cancer were subsequently hospitalised for haemorrhagic and ischaemic strokes (Table 1), respectively.
      Table 1Basic characteristics of the stroke patients.
      ParameterIschaemic strokeHaemorrhagic stroke
      With cancerWithout cancerWith cancerWithout cancer
      Number%Number%Number%Number%
      Age (years)
      <6011063.537,8369.55307.712,54416.3
      60–69365811.666,30216.7100014.413,96918.1
      70–7911,00234.9136,37634.4237334.325,14732.6
      ⩾8015,75850.0156,18839.4302343.625,37132.9
      Sex
      Male17,67156.1200,14450.5428261.842,58655.3
      Female13,85343.9196,55849.5264438.234,44544.7
      Time period
      1987–19898482.748,43712.22063.0835210.8
      1990–1994466314.892,02623.289212.916,29221.1
      1995–1999791725.199,14025.0164423.718,63424.2
      2000–2004933429.688,70822.4209130.218,84324.5
      2005–2008876227.868,39117.2209330.214,91019.4
      Region of residence
      Large city11,21135.6127,89832.2234133.824,09031.3
      South13,51542.9162,45641.0297142.929,86838.8
      North611219.485,19321.5136819.817,07722.2
      Unknown6862.221,1465.32463.659967.8
      All31,524100.0396,702100.06926100.077,031100.0

      3.1 Ischaemic stroke

      Risk of ischaemic stroke was increased during the first 6 months after diagnosis of 23 of the 34 cancers studied (Tables 2). Overall risk of ischaemic stroke during the first 6 months after diagnosis of cancer was 1.6 (95% CI 1.51.6). It decreased thereafter, but remained relatively constant after 6 months over time (SIR 1.1 after 612 months (95% CI 1.1–1.2), 1.1 after 1–5 years (95% CI 1.1–1.2), 1.2 after 5–10 years (95% CI 1.2–1.2) and 1.1 after 10+ years (95% CI 1.1–1.2)).
      Table 2Standardised incidence ratio (SIRs) for subsequent ischaemic stroke in cancer patients by follow-up time.
      Cancer site/typeFollow-up interval
      <6 months6–12 months1–5 years5–10 years10+ yearsAll1+ years
      OSIR95% Confidence interval (CI)OSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CI
      Upper aerodigestive tract671.31.0
      Bold type, 95% CI does not include 1.00.
      1.7411.00.71.3272
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.32481.51.31.71481.51.21.77761.31.21.46681.31.21.4
      Salivary gland71.50.63.251.2042.9251.10.71.6211.30.82.0201.71.12.778
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.666
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.7
      Oesophagus311.51.12.2111.10.51.9210.80.51.3121.00.51.861.20.42.5811.10.91.4390.90.71.3
      Stomach1201.81.52.2401.00.71.41581.11.01.3101
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.5521.00.81.44711.21.11.4311
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.3
      Small intestine212.21.43.4101.40.62.5330.80.61.2210.90.51.3141.00.51.7991.10.91.3680.90.71.1
      Colon3381.61.51.81941.11.01.311261.21.11.37001.21.11.34281.31.21.427861.31.21.322541.21.21.3
      Rectum1811.61.31.81061.10.91.35291.00.91.13851.21.11.3229
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.414301.11.11.21143
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.2
      Anus81.50.63.061.30.52.8241.00.61.5150.90.51.5131.20.62.1661.10.81.4521.00.81.3
      Liver551.31.01.7291.81.22.6380.90.61.2231.10.71.7131.00.51.71581.21.01.4741.00.81.2
      Pancreas902.21.82.7141.00.61.7281.40.92.060.90.32.010.30.01.61391.61.41.9351.10.81.6
      Nose41.10.32.841.30.33.2201.20.71.8171.50.92.4142.11.23.6591.41.11.8511.51.11.9
      Lung3132.21.92.4991.31.01.62301.10.991.31041.10.91.3631.31.01.68091.41.31.5397
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.3
      Breast3581.51.31.62661.21.01.315881.11.01.11316
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.11023
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.145511.11.11.13927
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.1
      Cervix211.61.02.4151.30.82.2741.21.01.6601.31.01.6471.10.81.42171.21.11.4181
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.4
      Endometrium1091.71.42.0611.00.81.34071.00.91.13351.00.91.12631.00.91.11175
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.110051.01.01.1
      Ovary611.81.42.4190.70.41.11210.90.71.1941.00.81.2971.21.01.53921.11.01.23121.00.91.1
      Other female genital201.71.12.7192.01.23.1501.10.81.4411.31.01.8181.00.61.61481.31.11.51091.20.91.4
      Prostate6881.21.11.3596
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.236401.21.11.219351.21.21.36361.21.11.374951.21.21.262111.21.21.2
      Testis32.00.45.921.40.15.190.80.41.660.50.21.0120.70.41.2320.70.51.0270.70.41.0
      Other male genital102.01.03.830.70.12.1271.20.761.7181.20.71.860.70.21.4641.10.91.4511.10.81.4
      Kidney831.51.21.9461.10.81.42641.11.01.22061.21.11.41040.90.81.17031.11.11.2574
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.2
      Urinary bladder2531.71.51.91601.21.11.48741.21.11.36051.21.11.33771.21.11.422691.21.21.318561.21.11.2
      Melanoma90
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.6610.90.71.23981.00.91.13461.11.01.22290.90.81.111241.01.01.19731.00.91.1
      Skin, squamous cell201
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.4185
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.411921.41.31.46781.31.21.4293
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.325491.31.21.321631.31.21.4
      Eye71.30.52.740.80.22.0250.80.51.2221.10.71.6100.70.31.3680.90.71.1570.90.71.1
      Nervous system1404.13.44.8321.30.91.91811.31.41.51401.10.91.31491.31.11.66421.41.31.64701.21.11.3
      Thyroid gland141.60.82.6101.30.62.5591.20.91.5461.00.81.4421.00.71.31711.11.01.31471.10.91.2
      Endocrine glands642.11.62.7391.30.91.82661.21.11.42771.41.21.51851.10.91.38311.31.21.47281.21.11.3
      Bone21.20.14.3040.50.11.391.40.62.650.90.32.1200.80.51.3180.90.51.4
      Connective tissue21
      Bold type, 95% CI does not include 1.00.
      1.6
      1.02.5101.00.41.8631.10.91.4411.10.81.5200.80.51.21551.10.91.31241.10.91.2
      Non-Hodgkin lymphoma1641.61.41.8941.10.91.34831.00.91.1322
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.31751.21.11.412381.11.11.2980
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.2
      Hodgkin’s disease51.30.43.072.20.94.6241.30.81.9120.80.41.4181.50.92.4661.31.01.6541.20.91.6
      Myeloma541.51.12.0250.80.51.21361.00.91.2451.00.81.4161.10.61.82761.11.01.21971.00.91.2
      Leukaemia1053.02.53.7301.20.81.71251.10.91.3881.51.21.8371.30.91.83851.51.31.62501.21.11.4
      All37081.61.51.622431.11.11.2125151.11.11.282951.21.21.247631.11.11.2315241.21.21.2255731.21.11.2
      low asterisk Bold type, 95% CI does not include 1.00.
      Risk of ischaemic stroke was ⩾2 during the first 6 months after diagnosis of cancers of six sites/types: small intestine, pancreas, lung, nervous system, endocrine glands and leukaemia (Table 2). For 11 cancer sites/types—upper aerodigestive tract, salivary gland, colon, rectum, nose, breast, prostate, urinary bladder, skin (squamous cell), nervous system and non-Hodgkin lymphoma—risk of ischaemic stroke was increased 10+ years after hospitalisation (Table 2).

      3.2 Haemorrhagic stroke

      Risk of haemorrhagic stroke was increased during the first 6 months after diagnosis of 15 of the 34 cancers studied (Table 3). Overall risk of haemorrhagic stroke during the first 6 months after diagnosis of cancer was 2.2 (95% CI 2.0–2.3). It was much lower between 6 and 12 months, but thereafter remained relatively constant over time (SIR 1.4 after 6–12 months (95% CI 1.3–1.5), 1.3 after 1–5 years (95% CI 1.2–1.3), 1.3 after 5–10 years (95% CI 1.2–1.4), and 1.2 after 10+ years (95% CI 1.1–1.3)).
      Table 3SIRs for subsequent haemorrhagic stroke in cancer patients by follow-up time.
      Cancer site/typeFollow-up interval
      <6 months6–12 months1–5 years5–10 years10+ yearsAll1+ years
      OSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CI
      Upper aerodigestive tract90.90.41.860.70.31.661
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.7421.30.91.8301.41.02.01481.31.11.51331.31.116
      Salivary gland22.40.28.7081.90.83.761.90.74.210.40.02.5171.50.92.4151.6092.6
      Oesophagus82.00.94.131.50.34.481.60.73.220.90.13.211.00.05.4221.61.02.4111.30.72.4
      Stomach191.50.92.491.20.62.4411.61.12.2382.51.83.490.90.41.71161.61.42.0881.71.42.1
      Small intestine63.41.27.4091.20.62.391.90.93.741.40.43.628
      Bold type, 95% CI does not include 1.00.
      1.6
      1.02.2221.50.92.2
      Colon551.51.11.9511.61.22.12281.41.21.51531.41.21.7801.21.01.55671.41.31.54611.41.21.5
      Rectum200.90.61.4281.51.02.21051.10.91.3961.51.21.9401.10.81.42891.21.11.42411.21.11.4
      Anus011.20.06.881.80.83.682.71.25.30171.50.92.4161.71.02.8
      Liver182.41.43.851.70.54.0101.20.62.330.80.22.331.20.23.6391.61.12.1161.10.61.8
      Pancreas91.20.62.431.20.23.661.60.63.521.60.25.911.40.08.2211.40.82.191.60.73.0
      Nose34.20.81211.60.09.441.30.33.362.81.06.132.10.46.2172.11.23.413
      Bold type, 95% CI does not include 1.00.
      1.9
      1.03.3
      Lung491.81.32.3181.20.71.9441.10.81.5110.60.31.190.90.41.8131
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.4641.00.71.2
      Breast370.90.61.2330.80.61.22420.90.81.12311.11.01.21781.10.91.27211.00.91.16511.00.91.1
      Cervix52.20.75.131.50.34.590.80.41.6111.20.62.250.60.21.4331.00.71.4250.90.61.3
      Endometrium151.40.82.3111.10.62.0631.00.71.2531.00.71.3441.00.71.31861.00.91.21601.00.81.1
      Ovary81.40.62.840.80.22.2261.10.71.6171.10.61.7130.90.51.6681.10.81.4561.00.81.4
      Other female genitals10.50.03.00101.30.62.481.60.73.120.60.12.3211.10.71.7201.30.81.9
      Prostate1641.41.21.61421.31.11.68071.31.21.44201.31.21.41511.31.11.616841.31.21.413781.31.21.4
      Testis0051.70.53.951.40.53.461.30.52.9161.40.82.2161.40.82.4
      Other male genitals11.00.05.8071.50.63.120.60.12.310.50.02.8110.90.51.7101.00.51.8
      Kidney414.12.95.6162.01.13.2912.01.62.5541.61.22.2351.61.12.22372.01.82.31801.81.62.1
      Urinary bladder321.10.81.6371.51.12.11601.11.01.31171.21.01.4891.41.11.74351.21.11.33661.21.11.3
      Melanoma261.91.32.8221.71.12.61521.91.62.2961.51.21.9631.21.01.63591.61.51.83111.61.41.8
      Skin, squamous cell43
      Bold type, 95% CI does not include 1.00.
      1.4
      1.01.9371.30.91.82011.21.11.4122
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.4731.41.11.84761.31.21.43961.21.11.4
      Eye22.00.27.311.00.05.991.50.72.951.30.43.010.40.02.1181.20.72.0151.20.72.0
      Nervous system199292534265.53.68.0511.81.42.4592.31.72.9331.41.02.03684.23.84.61431.91.62.2
      Thyroid gland63.61.37.910.70.04.160.60.21.4101.20.62.291.10.52.1321.10.81.6251.00.61.4
      Endocrine glands305.63.88.091.80.83.3591.51.22.0671.81.42.3361.10.81.52011.71.52.01621.51.31.8
      Bone0031.90.45.721.50.25.743.40.98.891.90.93.79
      Bold type, 95% CI does not include 1.00.
      2.2
      1.04.2
      Connective tissue83.21.46.542.00.55.2131.20.62.181.10.52.230.60.11.6361.30.91.8241.00.71.6
      Non-Hodgkin lymphoma452.31.73.1140.90.51.5961.10.91.370
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.6431.51.12.02681.31.21.52091.21.11.4
      Hodgkin’s disease22.60.29.611.60.09.282.10.94.241.30.43.420.70.12.7171.60.92.5141.50.82.5
      Myeloma192.91.84.6112.01.03.6471.91.42.6111.30.72.451.70.54.0932.01.62.4631.81.42.3
      Leukaemia82131016316.54.49.3733.32.64.2211.81.12.7162.61.54.32234.43.85.01102.82.33.3
      All9652.22.02.35281.41.31.526711.31.21.317691.31.21.49931.21.11.369261.41.31.454331.31.21.3
      low asterisk Bold type, 95% CI does not include 1.00.
      Risk of haemorrhagic stroke was ⩾2 during the first 6 months after diagnosis of cancers of 10 sites/types: small intestine, liver, kidney, nervous system, thyroid gland, endocrine glands, connective tissue, non-Hodgkin lymphoma, myeloma and leukaemia (Table 2). For six cancer sites/types—prostate, kidney, urinary bladder, skin (squamous cell), non-Hodgkin lymphoma and leukaemia—risk of CHD was increased 10+ years after hospitalisation (Table 3).

      3.3 Cancer sites unrelated to tobacco smoking

      Several cancer sites/types unrelated to tobacco smoking—small intestine, colon, rectum, breast, endometrium, ovary, other female genital, prostate, melanoma, skin (squamous cell), nervous system, endocrine glands, connective tissue, non-Hodgkin lymphoma and myeloma—were associated with increased ischaemic stroke risk during the first 6 months after cancer diagnosis (Table 2). Several cancer sites/types unrelated to tobacco smoking—small intestine, colon, prostate, melanoma, skin (squamous cell), nervous system, thyroid gland, endocrine glands, connective tissue, non-Hodgkin lymphoma and myeloma—were associated with increased haemorrhagic stroke risk the first 6 months after cancer diagnosis (Table 3).

      3.4 Metastasis and haemorrhagic and ischaemic strokes’ risk

      The Swedish Cancer Registry only contains data on metastasis since 2002. Between 2002 and 2008, overall risk of haemorrhagic and ischaemic strokes was higher in cancer patients with metastasis than in those without metastasis (Table 4).
      Table 4SIRs for subsequent stroke in cancer patients by metastasis.
      Cancer site/typeIschaemic strokeHaemorrhagic stroke
      No metastasisWith metastasisNo metastasisWith metastasis
      OSIR95% CIOSIR95% CIOSIR95% CIOSIR95% CI
      Upper aerodigestive tract561.51.11.9060.70.31.60
      Oesophagus61.10.42.332.00.46.021.60.26.00
      Stomach201.40.92.272.30.94.851.70.54.011.50.08.7
      Colon2301.51.31.7351.81.32.545
      Bold type, 95% CI does not include 1.00.
      1.4
      1.01.882.00.94.0
      Rectum841.00.81.390.90.41.7120.710.41.210.50.02.7
      Liver101.60.83.031.80.45.453.91.29.10
      Pancreas102.01.03.7165.83.39.422.00.27.211.70.09.9
      Nose52.40.85.6012.20.0130
      Lung1361.61.41.9642.11.62.6201.10.71.8182.81.64.4
      Breast310
      Bold type, 95% CI does not include 1.00.
      1.1
      1.01.381.10.52.1440.80.61.10
      Prostate2941.00.81.1971.10.91.4781.10.91.4472.41.83.2
      Other male genitals143.62.06.1011.10.06.30
      Kidney281.20.81.751.20.42.771.40.62.977.63.016
      Urinary bladder61
      Bold type, 95% CI does not include 1.00.
      1.3
      1.01.731.60.34.660.60.21.30
      Melanoma390.90.61.322.90.310202.21.43.40
      Skin, squamous cell1531.21.01.413.30.019441.61.22.20
      Thyroid gland71.20.52.523.00.31110.80.04.60
      Endocrine glands0000
      Connective tissue102.00.93.6010.90.05.10
      All14991.21.11.32581.51.31.7304
      Bold type, 95% CI does not include 1.00.
      1.2
      1.01.3832.21.82.7
      low asterisk Bold type, 95% CI does not include 1.00.

      4. Discussion

      The present study is the first nationwide study of cancer and risk of haemorrhagic and ischaemic strokes to present precise risk estimates for multiple different cancer sites/types. The results indicate that many cancer sites/types are associated with an increased risk of hospitalisation for haemorrhagic and ischaemic strokes. This is consistent with previous studies that have shown stroke to be common among cancer patients.
      • Graus F.
      • Rogers L.R.
      • Posner J.B.
      Cerebrovascular complications in patients with cancer.
      • Lindvig K.
      • Moller H.
      • Mosbech J.
      • Jensen O.M.
      The pattern of cancer in a large cohort of stroke patients.
      • Cestari D.M.
      • Weine D.M.
      • Panageas K.S.
      • Segal A.Z.
      • DeAngelis L.M.
      Stroke in patients with cancer: incidence and etiology.
      • Rogers L.R.
      Cerebrovascular complications in cancer patients.
      • Nguyen T.
      • DeAngelis L.M.
      Stroke in cancer patients.
      • Haller S.
      • Lyrer P.
      Malignancy and stroke.
      Moreover, the association is not limited to smoking-related cancers. Several non-smoking-related cancers were also associated with increased risk of haemorrhagic and ischaemic strokes (Table 2, Table 3). Although it declines rapidly after 6 months, the overall risk of haemorrhagic and ischaemic strokes remains slightly raised for 10 or more years. Overall risk of haemorrhagic and ischaemic strokes is additionally related to the presence of metastasis (Table 4). The results of the present study suggest that there is a general association between cancer, especially newly diagnosed cancer, and increased risk of subsequent haemorrhagic and ischaemic strokes. We have recently reported increased risk of coronary heart disease and myocardial infarction in cancer patients.

      Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

      Thus, arterial cardiovascular disease just like venous thromboembolism appears to be associated with cancer.
      • Franchini M.
      • Montagnana M.
      • Favaloro E.J.
      • Lippi G.
      The bidirectional relationship of cancer and hemostasis and the potential role of anticoagulant therapy in moderating thrombosis and cancer spread.
      Some cancer sites/types, notably the nervous system and leukaemia, are especially strongly associated with the development of ischaemic and haemorrhagic strokes, and it is of especial clinical relevance to consider them.
      The increased risks of haemorrhagic and ischaemic strokes for different cancer sites/types may have different underlying causes. Unfortunately, we were unable to determine the mechanisms involved due to lack of access to laboratory and treatment data. One possible cause of the increased stroke risk among cancer patients might be that cardiovascular risk factors like hypertension and diabetes and other comorbidities have been reported to be common among cancer patients.
      • Coebergh J.W.
      • Janssen-Heijnen M.L.
      • Razenberg P.P.
      Prevalence of co-morbidity in newly diagnosed patients with cancer: a population-based study.
      • Janssen-Heijnen M.L.
      • Houterman S.
      • Lemmens V.E.
      • et al.
      Prognostic impact of increasing age and co-morbidity in cancer patients: a population-based approach.
      • Janssen-Heijnen M.L.
      • Szerencsi K.
      • van de Schans S.A.
      • et al.
      Cancer patients with cardiovascular disease have survival rates comparable to cancer patients within the age-cohort of 10 years older without cardiovascular morbidity.
      Since inflammation has been linked to both atherosclerosis and haemostatic activation, cancer-associated inflammation/haemostatic activation may constitute a general link between cancer and haemorrhagic and ischaemic strokes.
      • Franchini M.
      • Montagnana M.
      • Favaloro E.J.
      • Lippi G.
      The bidirectional relationship of cancer and hemostasis and the potential role of anticoagulant therapy in moderating thrombosis and cancer spread.
      • Coussens L.M.
      • Werb Z.
      Inflammation and cancer.
      • Libby P.
      Inflammation in atherosclerosis.
      • van Leuven S.I.
      • Franssen R.
      • Kastelein J.J.
      • et al.
      Systemic inflammation as a risk factor for atherothrombosis.
      The observed association of tumour metastasis with haemorrhagic and ischaemic strokes’ risk indicates that this hypothesis is plausible (i.e. increased inflammation and haemostatic activation) due to presence of metastasis (i.e. increased tumour burden) (Table 3). As well as reducing tumour size, efficient cancer treatment might also reduce associated inflammation, haemostatic activation and, thus, risk of haemorrhagic and ischaemic strokes. The fact that the risk of haemorrhagic and ischaemic strokes decreases rapidly suggests that haemorrhagic and ischaemic strokes’ risk could be linked to tumour size, which is likely to decrease over time due to treatment. However, as we lack treatment data, we cannot test this hypothesis. The decreased risk after 6 months could also be due to the cessation of treatment with cytostatics.
      • Yeh E.T.
      • Bickford C.L.
      Cardiovascular complications of cancer therapy: incidence, pathogenesis, diagnosis, and management.
      Cytostatics could not only promote thrombosis, but also bleeding due to thrombocytopenia.
      • Yeh E.T.
      • Bickford C.L.
      Cardiovascular complications of cancer therapy: incidence, pathogenesis, diagnosis, and management.
      The rapid decrease could also be due to death of patients with cerebral metastasis or successful treatment of cerebral metastasis. Indeed, the risks for both haemorrhagic and ischaemic strokes were highest in patients with primary brain tumour (i.e. nervous system tumours). In smoking-related cancers, the increased haemorrhagic and ischaemic strokes’ risk may be due to tobacco smoking.
      • Dreyer L.
      • Olsen J.H.
      Cancer risk of patients discharged with acute myocardial infarct.
      • Dreyer L.
      • Olsen J.H.
      Risk for non-smoking-related cancer in atherosclerotic patients.
      • Horn-Ross P.L.
      • Ljung B.M.
      • Morrow M.
      Environmental factors and the risk of salivary gland cancer.
      • Fernberg P.
      • Odenbro A.
      • Bellocco R.
      • et al.
      Tobacco use, body mass index, and the risk of leukemia and multiple myeloma: a nationwide cohort study in Sweden.
      Thus, smoking cessation—a confounder for which we could not adjust—may also contribute to the rapid decrease in the incidence of haemorrhagic and ischaemic strokes after diagnosis of cancer. Thus, a number of mechanisms are possible.
      The effects of treatment (cerebral or neck irradiation)
      • Dubec J.J.
      • Munk P.L.
      • Tsang V.
      • et al.
      Carotid artery stenosis in patients who have undergone radiation therapy for head and neck malignancy.
      • Morris B.
      • Partap S.
      • Yeom K.
      • et al.
      Cerebrovascular disease in childhood cancer survivors: a Children’s Oncology Group Report.
      may also contribute to the identified associations, especially in the case of primary brain tumour or cerebral metastasis. Risk for stroke was previously shown to be increased in survivors of paediatric CNS tumours, Hodgkin’s disease and acute lymphoblastic leukaemia,
      • Morris B.
      • Partap S.
      • Yeom K.
      • et al.
      Cerebrovascular disease in childhood cancer survivors: a Children’s Oncology Group Report.
      in agreement with our study, in which the highest risks were observed for brain tumours and leukaemia. However, the stroke risk was not increased in patients with Hodgkin’s disease in the present study.
      Another potential confounder that could contribute to the increased haemorrhagic and ischaemic strokes’ risk during the first 6 months after diagnosis is the stress of having cancer. Although we were unable to evaluate the impact of psychosocial stress, psychosocial factors have been suggested to be risk factors for cardiovascular disease.
      • Rozanski A.
      • Blumenthal J.A.
      • Davidson K.W.
      • Saab P.G.
      • Kubzansky L.
      The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology.
      The present study has certain limitations. For example, we had no data on general cardiovascular disease risk factors such as weight, smoking and diet. It is unrealistic to gather such data for an entire national population. However, we did adjust for socioeconomic status, which is associated with risk factors such as smoking and alcohol. Adjustment was also made for comorbidities (COPD, obesity, alcoholism, hypertension and diabetes mellitus). A further limitation is that we had no access to outpatient data, which means that only the most severe cases of haemorrhagic and ischaemic strokes (i.e. those requiring hospitalisation) were included in the analyses. However, almost all cases of haemorrhagic and ischaemic strokes should, according to official guidelines, be treated at hospitals in Sweden.

      National guidelines for stroke care 2005. Basis for prioritisation. Stockholm: The National Board of Health and Welfare; 2005 [in Swedish].

      Computed tomography of the brain should also be performed in patients with stroke symptoms.

      National guidelines for stroke care 2005. Basis for prioritisation. Stockholm: The National Board of Health and Welfare; 2005 [in Swedish].

      Moreover, incidence rates were calculated for the whole follow-up period, divided into five time periods, and adjustments were made for possible changes in incidence rates over time. A further limitation was our lack of access to treatment data.
      The study also had a number of strengths. For instance, the study population included all individuals in Sweden diagnosed with cancer and hospitalised with haemorrhagic or ischaemic stroke during the study period. Because of the personal identification number assigned to each resident in Sweden, it was possible to trace every subject for the whole follow-up period. Data on occupation were 99.2% complete (1980 and 1990 censuses), which enabled us to adjust our models for socioeconomic status. Moreover, the Swedish Cancer Registry records all new cases of cancer, with almost all cases being histologically or cytologically confirmed.

      Cancer incidence in Sweden 2005. Stockholm: The National Board of Health and Welfare; 2007.

      A further strength of the present study was the use of validated hospital discharge data.
      • Rosen M.
      • Hakulinen T.
      Use of disease registers.

      Validity of the diagnoses from the Swedish in-care register 1987 and 1995. Stockholm: The National Board of Health and Welfare; 2000 [in Swedish].

      • Lindblad U.
      • Råstam L.
      • Ranstam J.
      • Peterson M.
      Validity of register data on acute myocardial infarction and acute stroke: the Skaraborg Hypertension Project.
      The Hospital Discharge Register has high validity,
      • Rosen M.
      • Hakulinen T.
      Use of disease registers.

      Validity of the diagnoses from the Swedish in-care register 1987 and 1995. Stockholm: The National Board of Health and Welfare; 2000 [in Swedish].

      • Lindblad U.
      • Råstam L.
      • Ranstam J.
      • Peterson M.
      Validity of register data on acute myocardial infarction and acute stroke: the Skaraborg Hypertension Project.
      especially for cardiovascular disorders such as haemorrhagic and ischaemic strokes, for which approximately 95% of diagnoses have been shown to be correct.

      Validity of the diagnoses from the Swedish in-care register 1987 and 1995. Stockholm: The National Board of Health and Welfare; 2000 [in Swedish].

      • Lindblad U.
      • Råstam L.
      • Ranstam J.
      • Peterson M.
      Validity of register data on acute myocardial infarction and acute stroke: the Skaraborg Hypertension Project.
      Another advantage was that the exclusive use of hospital diagnoses eliminated recall bias.
      In summary, risk of hospitalisation for haemorrhagic and ischaemic strokes was, for a number of cancer sites/types, found to be significantly increased during the first 6 months after diagnosis of cancer. The risk of haemorrhagic and ischaemic strokes decreased rapidly thereafter, but for many cancer sites remained elevated for more than 10 years. Overall risk of haemorrhagic and ischaemic strokes was related to the presence of metastasis, suggesting that haemorrhagic and ischaemic strokes’ risk is affected by the severity of cancer. The findings of the present study indicate that newly diagnosed cancer in general is associated with increased risk of ischaemic and haemorrhagic strokes.

      Role of the funding source

      None.

      Contributors

      All authors contributed to the conception and design of the study; JS and KS acquired; all authors contributed to the analysis and interpretation of data; BZ drafted the manuscript and all authors revised it critically and approved the final version. All authors had full access to all of the data (including statistical reports and tables) and take responsibility for their integrity and the accuracy of their analysis.

      Conflict of interest statement

      None declared.

      Acknowledgements

      The authors wish to thank the CPF’s Science Editor Stephen Gilliver for his useful comments on the text. The registers used in the present study are maintained by Statistics Sweden and the National Board of Health and Welfare. This work was supported by grants awarded to Kristina and Jan Sundquist by the Swedish Research Council (2008-3110 and 2008-2638), the Swedish Council for Working Life and Social Research (2006-0386, 2007-1754 and 2007-1962), Formas (2006-4255-6596-99 and 2007-1352), and to Bengt Zöller by the Swedish Heart and Lung Foundation and Region Skåne (REGSKANE-124611).

      References

        • Donnan G.A.
        • Fisher M.
        • Macleod M.
        • Davis S.M.
        Stroke.
        Lancet. 2008; 371: 1612-1623
        • Graus F.
        • Rogers L.R.
        • Posner J.B.
        Cerebrovascular complications in patients with cancer.
        Medicine (Baltimore). 1985; 64: 16-35
        • Lindvig K.
        • Moller H.
        • Mosbech J.
        • Jensen O.M.
        The pattern of cancer in a large cohort of stroke patients.
        Int J Epidemiol. 1990; 19: 498-504
        • Cestari D.M.
        • Weine D.M.
        • Panageas K.S.
        • Segal A.Z.
        • DeAngelis L.M.
        Stroke in patients with cancer: incidence and etiology.
        Neurology. 2004; 62: 2025-2030
        • Rogers L.R.
        Cerebrovascular complications in cancer patients.
        Oncology (Williston Park). 1994; 8: 23-30
        • Nguyen T.
        • DeAngelis L.M.
        Stroke in cancer patients.
        Curr Neurol Neurosci Rep. 2006; 6: 187-192
        • Haller S.
        • Lyrer P.
        Malignancy and stroke.
        Semin Cerebrovasc Dis Stroke. 2005; 5: 47-54
        • Franchini M.
        • Montagnana M.
        • Favaloro E.J.
        • Lippi G.
        The bidirectional relationship of cancer and hemostasis and the potential role of anticoagulant therapy in moderating thrombosis and cancer spread.
        Semin Thromb Hemost. 2009; 35: 644-653
        • Coussens L.M.
        • Werb Z.
        Inflammation and cancer.
        Nature. 2002; 420: 860-867
        • Libby P.
        Inflammation in atherosclerosis.
        Nature. 2002; 420: 868-874
        • van Leuven S.I.
        • Franssen R.
        • Kastelein J.J.
        • et al.
        Systemic inflammation as a risk factor for atherothrombosis.
        Rheumatology. 2008; 47: 3-7
        • Cestari D.M.
        • Weine D.M.
        • Panageas K.S.
        • Segal A.Z.
        • DeAngelis L.M.
        Stroke in patients with cancer: incidence and etiology.
        Neurology. 2004; 62: 2025-2030
        • Chaturvedi S.
        • Ansell J.
        • Recht L.
        Should cerebral ischemic events in cancer patients be considered a manifestation of hypercoagulability?.
        Stroke. 1994; 25: 1215-1218
        • Dreyer L.
        • Olsen J.H.
        Cancer risk of patients discharged with acute myocardial infarct.
        Epidemiology. 1998; 9: 178-183
        • Dreyer L.
        • Olsen J.H.
        Risk for non-smoking-related cancer in atherosclerotic patients.
        Cancer Epidemiol Biomarkers Prev. 1999; 8: 915-918
        • Horn-Ross P.L.
        • Ljung B.M.
        • Morrow M.
        Environmental factors and the risk of salivary gland cancer.
        Epidemiology. 1997; 8: 414-419
        • Fernberg P.
        • Odenbro A.
        • Bellocco R.
        • et al.
        Tobacco use, body mass index, and the risk of leukemia and multiple myeloma: a nationwide cohort study in Sweden.
        Cancer Res. 2007; 67: 5983-5986
        • Rosen M.
        • Hakulinen T.
        Use of disease registers.
        in: Ahrens W. Pigeot I. Handbook of epidemiology. Springer-Verlag, Berlin2005: 231-252
      1. Cancer incidence in Sweden 2005. Stockholm: The National Board of Health and Welfare; 2007.

      2. The multi-generation register 2009. A description of contents and quality. Statistics Sweden. Available from: <http://www.scb.se/statistik/_publikationer/BE9999_2009A01_BR_BE96BR1003.pdf> [accessed 22.03.11].

      3. The Swedish Hospital Discharge Register 1987–1996: quality and contents. Stockholm: The National Board of Health and Welfare; 1998.

      4. Zöller B, Ji J, Sundquist J, Sundquist K. Risk of coronary heart disease in patients with cancer: a nationwide follow-up study from Sweden. Eur J Cancer 2012; 48:121–8.

        • Rothman K.J.
        • Greenland S.
        Modern epidemiology.
        2nd ed. Lippincott-Raven, Philadelphia1998
        • Coebergh J.W.
        • Janssen-Heijnen M.L.
        • Razenberg P.P.
        Prevalence of co-morbidity in newly diagnosed patients with cancer: a population-based study.
        Crit Rev Oncol Hematol. 1998; 27: 97-100
        • Janssen-Heijnen M.L.
        • Houterman S.
        • Lemmens V.E.
        • et al.
        Prognostic impact of increasing age and co-morbidity in cancer patients: a population-based approach.
        Crit Rev Oncol Hematol. 2005; 55: 231-240
        • Janssen-Heijnen M.L.
        • Szerencsi K.
        • van de Schans S.A.
        • et al.
        Cancer patients with cardiovascular disease have survival rates comparable to cancer patients within the age-cohort of 10 years older without cardiovascular morbidity.
        Crit Rev Oncol Hematol. 2010; 76: 196-207
        • Yeh E.T.
        • Bickford C.L.
        Cardiovascular complications of cancer therapy: incidence, pathogenesis, diagnosis, and management.
        J Am Coll Cardiol. 2009; 53: 2231-2247
        • Dubec J.J.
        • Munk P.L.
        • Tsang V.
        • et al.
        Carotid artery stenosis in patients who have undergone radiation therapy for head and neck malignancy.
        Br J Radiol. 1998; 7: 872-875
        • Morris B.
        • Partap S.
        • Yeom K.
        • et al.
        Cerebrovascular disease in childhood cancer survivors: a Children’s Oncology Group Report.
        Neurology. 2009; 73: 1906-1913
        • Rozanski A.
        • Blumenthal J.A.
        • Davidson K.W.
        • Saab P.G.
        • Kubzansky L.
        The epidemiology, pathophysiology, and management of psychosocial risk factors in cardiac practice: the emerging field of behavioral cardiology.
        J Am Coll Cardiol. 2005; 45: 637-651
      5. National guidelines for stroke care 2005. Basis for prioritisation. Stockholm: The National Board of Health and Welfare; 2005 [in Swedish].

      6. Validity of the diagnoses from the Swedish in-care register 1987 and 1995. Stockholm: The National Board of Health and Welfare; 2000 [in Swedish].

        • Lindblad U.
        • Råstam L.
        • Ranstam J.
        • Peterson M.
        Validity of register data on acute myocardial infarction and acute stroke: the Skaraborg Hypertension Project.
        Scand J Soc Med. 1993; 21: 3-9