Hypercalcaemia: Primary hyperparathyroidism was associated with increased mortality

Clinical bottom line (level 3b)

  1. Patients with asymptomatic presumed hyperparathyroidism had a higher mortality rate than those with normal calcium levels. (NNH = 11 at 14 years)
  2. This effect was markedly reduced as age increased.
Palmer et al: Lancet 1987; i: 59-62
Expires December 2003

The study

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

Setting: community screening study in Sweden, 1969

516 patients (aged mean 59.3y, range 28-86, 85% female) cases with raised serum calcium (>2.60 mmol/L) on screening studies, two years apart, and matched controls

Excluded if
  • medication causing raised calcium
  • known malignant disease (diagnosed in part by 2-year interval as part of admission criteria)
  • parathyroid surgery during 14 year follow-up (n=18; 6 symptomatic)


    • Cases: 172 patients (85% female, mean age 59.3): raised serum calcium (presumed primary hyperparathyroidism)
    Controls: 344 patients (85% female, mean age 59.6): normal calcium level, matched for age, sex and date of screening examinations

    Factors studied:
  • Age, sex, calcium level, blood pressure, serum uric acid, serum glucose and serum cholesterol


    • Factors summarised:
  • hypercalcaemia (adjusted serum calcium >2.60 mmol/L)


    • Analysis was using a cox proportional hazards model, with risk estimates made using the Buckley-James technique

    Outcomes studied:
  • death (follow-up of Swedish national registries)

    • The evidence

    Patient expected event rate for death: 20%
    risk factor death
    present    		 death
    absent    		 unadjusted OR
    (95% CI)    		 NNH
    (95% CI)
    hypercalcaemia 56 116 1.65
    (1.10 to 12.47)     		11
    (5 to 66)
    no hypercalcaemia 78 266

    • In multivariate analyses, hypercalcaemia remained an independent predictor of mortality (cox beta value 5.0).
    • Other predictors of mortality were
      • age
      • sex
      • serum glucose
    • There was a significant interaction between calcium and glucose (both tended to increase similarly).
    • There was a significant interaction between age and calcium levels, with the effect of calcium levels been far less pronounced as age increased.
    • Compared to the population, hypercalcaemia was more prevalent in women (85% female vs 55% female)
    • 3% (6/172) patients suffered symptoms over 14 years and underwent surgery

    Comments

    1. The authors comment the raised mortality in asymptomatic hyperparathyroidism should not immediately prompt action as "parathyroid surgery has not been shown to reduce it".

    Citation

    1. Palmer M, Adam H-O, Bergstrom R, et al: Survival and renal function in untreated hypercalcaemia. Lancet 1987; i: 59-62
    Search Terms: reference from reviewer
    Contributor: Bob Phillips and Clare Wotton, December 2000
    Reviewer:

    Clinical Question.
    Patient primary hyperparathyroidism
    Intervention or Exposure epidemiology
    Outcome mortality