Critical illness: right heart catheterisation was associated with more deaths.

Clinical bottom line (level 3b)

  1. Critically ill medical patients who had right heart catheterisation (Swan-Ganz catheter) were more likely to die than those who did not (NNH = 21 at 30 days) .
  2. Costs of care, intensity of care and length of ICU stay were higher in patients managed by right heart catheterisation.
Connors et al: Journal of the American Medical Association 1996; 276 (11): 889-897
Expires July 2003

The study

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

Setting: intensive care units of five teaching hospitals, USA

5735 patients (aged range <50 to 80 years, 56% male) critically ill patients with severity scores suggestive of aggregate 6 month mortality in excess of 50%, and the presence of one or more of nine disease categories:
  • acute respiratory failure
  • chronic obstructive pulmonary disease
  • congestive heart failure
  • cirrhosis
  • nontraumatic coma
  • colon cancer metastatic to liver
  • non-small cell lung cancer stage III or IV
  • multisystem organ failure with malignancy or sepsis


Excluded if
  • aged <18 years
  • death or discharge within 48 hours
  • inability to speak English
  • acute psychiatric disorders
  • AIDS
  • acute burns
  • head trauma or other trauma unless MOFS developed later


    • Cases: 2184 patients (% male, mean age ): right heart catheterisation within the first 24 hours
    Controls: 3551 patients (% male, mean age ): no right heart catheterisation within the first 48 hours

    Factors studied:
  • death


    • Factors summarised:
  • right heart catheterisation


    • Two analyses were performed, the first with the use of a prognostic index (propensity score) as matching. The second used a multivariable regression analysis to adjust for age, sex, comorbid illness, disease category, severity of illness, Glasgow Coma Score, two month prognosis on day one, activities of daily living and Duke Activity Status Index.

    Outcomes studied:
  • death

    • The cohort with Swan-Gantz catheterisation were matched with controls with a similar prognostic score (n=2106) to create a comparator group.

    The evidence

    Patient expected event rate for death: 46%
    risk factor for
    death     		adjusted OR
    (95% CI)    		 NNH
    (95% CI)
    right heart catheterisation 1.21
    (1.09 to 1.25)     		21
    (18 to 47)

    • Total cost of treatment for the control group was $35, 700 and for the experimental group $49,300 - a mean difference of $79, 000 (p<0.01).
    • Mean Therapeutic Index Scoring System (measure of intensity of care) score was 30 in the control group and 34 in the experimental group- mean difference of 4 (p<0.01).
    • Mean length of stay in the control group was 13 days and in the experimental group was 14.5 days- mean difference 1.5 days (p<0.01).

    Comments

    1. Sensitivity analysis was used to determine whether a missing variable associated with both worse prognosis and increased probability of right heart catheterisation might explain the results. This demonstrated that such a missing variable would have to increase risk of death six-fold and increase probability of right heart catheterisation six-fold for a true relative risk of 0.8 to be misrepresented as a relative hazard of 1.21. None of the variables studied had such a large effect.
    2. Possible explanations for this study include:
      • risks of right heart catheterisation (RHC) truly outweigh benefits
      • RHC may be a marker for a more aggressive style of care that is deleterious to patient outcomes
      • changes in therapy (in particular, use of fluid and inotropes to achieve supranormal values of cardiac output) may lead to higher mortality
      • RHC is actually beneficial and this benefit is masked by a confounding variable that is not included in the analysis
    3. A randomised trial needs to be performed to help make the results more conclusive.

    Citation

    1. Connors AF, Speroff T, Dawson NV, et al: The effectiveness of right heart catheterization in the initial care of critically ill patients. Journal of the American Medical Association 1996; 276 (11): 889-897
    Search Terms: author's files
    Contributor: Catherine Clase, Chris Ball and Clare Wotton, April 2000
    Reviewer: Luis Ruiz Del Fresno

    Clinical Question.
    Patient critically ill
    Intervention or Exposure right heart catheterisation
    Outcome mortality