Venous thromboembolism: patients were more likely to have cancer

Clinical bottom line (level 4)

  1. Patients with venous thromboembolism were at increased risk for having a cancer diagnosed (NNF = 39 for 12 months)
  2. Comparing the age-standardised rates of cancer in patients with the general population, those younger than 65y were at a greater risk (SIR ratio 6.7 compared with 3.9)
  3. However, the absolute risk was higher in patients > 65 (NNF = 40 for 12 months) , and patients with > 2 admissions with VTE (NNF = 20 for 12 months)
Baron et al: Lancet 1998; 351: 1077-1080
Expires February 2003

The study

Outcome study with objective outcomes, not adjusted for confounding factors, not validated in an independent set of patients.

Setting: population study, Sweden

61998 patients (aged ?, ?% male) venous thromboembolic disease (by discharge coding)

Excluded if
  • incomplete or inconsistent data entries




    • Cancer rates also compared with national standardised incidence rates

    Overall follow-up mean 7.7y, outcomes shown with 98% followed for 1 year
    Outcomes studied:
  • cases of venous thromboembolic disease

    • The evidence

    outcome time to outcome number of patients/total number %
    (95% CI)    		 NNF
    (95% CI)
    cases of venous thromboembolic disease 12 months 2509/61988 4%
    (% to %)    		 39
    (39 to 39)

    prognostic factor for
    cases of venous thromboembolic disease     		time to outcome unadjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    aged >65y 1 year 1.95
    (1.75 to 2.12)     		40
    (33 to 47)
    male 1 year 1.18
    (1.09 to 1.27)     		115
    (101 to 307)
    >1 episode of VTE 1 year 2.72
    (2.52 to 2.94)     		20
    (18 to 23)

    prognostic factor outcome present outcome absent unadjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    aged >65y 1766 32270 1.95
    (1.75 to 2.12)     		40
    (33 to 47)
    no aged >65y 743 27219

    prognostic factor outcome present outcome absent unadjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    male 1356 30140 1.18
    (1.09 to 1.27)     		115
    (101 to 307)
    no male 1153 29349

    prognostic factor outcome present outcome absent unadjusted RR
    (95% CI)    		 NNF+
    (95% CI)
    >1 episode of VTE 1120 13047 2.72
    (2.52 to 2.94)     		20
    (18 to 23)
    no >1 episode of VTE 1389 46442

    • The risks above demonstrate the absolute differences in risk. When compared with the age-specific national cancer rates for diagnosis in the first year of follow-up (standardised incidence rate: SIR)
      • younger (aged <65y): SIR = 6.7 (95% CI 6.2 - 7.2)
      • older (aged >65y): SIR = 3.9 (95% CI 3.7 - 4.0)
      • two or more VTE admissions: SIR = 8.2 (95% CI 7.7 - 8.6)
      • male: SIR = 4.2 (95% CI 4.0 - 4.5)
      • female: SIR = 4.7 (95% CI 4.5 - 5.0)
    • Cancers with greatest increase in risk: (RR > 6.0): liver, pancreas, ovarian, brain, Hodgkin's lymphoma

    Comments

    1. Comparisons with standardised incidence rates may distort figures for rare cancers
    2. The problems with using discharge coding (error, failure to collect correct information) will affect the data.
    3. A strategy of age and gender specific investigation for cancer appears warranted.

    Citation

    1. Baron JA, Gridley G, Weiderpass E, et al: Venous thromboembolism and cancer. Lancet 1998; 351: 1077-1080
    Contributor: Chris Ball and Bob Phillips, April 1998
    Reviewer: Alan Forster

    Clinical Question.
    Patient venous thromboembolism
    Intervention or Exposure no VTE
    Outcome cancer