Clinical background
In cardiogenic shock the central clinical problem is to judge whether the heart is able to maintain adequate perfusion, and to what extent the treatment instituted (inotropes, vasopressors, mechanical circulatory support) actually improves pump function. Individual haemodynamic variables such as mean arterial pressure, cardiac output or cardiac index are all load-dependent and do not reflect the total work of the heart. Cardiac power output (CPO) combines pressure and flow into a single measure of hydraulic power and thereby gives a more integrated picture of pump function than any single parameter.
The tool serves two purposes: risk stratification at the time of assessment, and quantification of treatment response. In the SHOCK trial registry, CPO was the strongest independent haemodynamic correlate of in-hospital mortality in cardiogenic shock, which established its role as a prognostic measure in this population [1].
Calculating cardiac power output (CPO)
The calculator uses the simplified formula applied in the SHOCK trial registry:
The constant 451 converts the units to watts. Mean arterial pressure and cardiac output are obtained by invasive haemodynamic monitoring, usually with a pulmonary artery catheter. Normal resting CPO is approximately 1 W, based on a physiological reference with a blood pressure of 120/80 mmHg and a cardiac output of 5 L/min [2].
The original description of CPO, developed by Tan, included right atrial pressure (RAP) in the numerator: [3]. RAP was omitted in the simplified version used in the SHOCK registry, and it is this simplified formula that the calculator uses. This has practical consequences: when RAP is elevated, which is common in cardiogenic shock, the simplified formula overestimates the true systolic power of the heart, because diastolic filling force (work done on the heart, not by the heart) is included in the calculation [2,3].
Derivation cohort
CPO as a prognostic tool was derived from the SHOCK trial registry, which included 541 patients with cardiogenic shock enrolled between 1993 and 1997 [1]. The majority had shock complicating acute myocardial infarction. CPO was calculated at pulmonary artery catheterisation, in most cases during ongoing inotropic treatment and in a third of cases also during intra-aortic balloon counterpulsation. The outcome measure was in-hospital mortality.
In multivariable analysis adjusted for age and history of hypertension, CPO remained the strongest independent haemodynamic correlate of in-hospital mortality, with an odds ratio of 0.60 (95% CI 0.44 to 0.83) per 0.20 W increase in CPO (n = 181) [1]. This corresponds to each 0.20 W higher CPO being associated with 40% lower odds of in-hospital death. CPO outperformed cardiac output, cardiac index, stroke volume, left ventricular stroke work, ejection fraction and coronary perfusion pressure in prognostic strength.
Interpretation in practice
CPO is not a binary decision tool with a single threshold for treatment choice, but a continuous measure in which lower values indicate a worse prognosis. The following bands can serve as orientating benchmarks, based on published data:
| CPO (W) | Interpretation | Clinical action |
|---|---|---|
| ≥ 0.80 | Relatively preserved pump function | Continue monitoring; optimise underlying treatment |
| 0.60 to 0.79 | Moderately impaired pump function | Consider intensified inotropic/vasopressor support; assess treatment response with repeated measurements |
| < 0.60 | Severely impaired pump function, high mortality risk | Consider mechanical circulatory support; assess candidacy for advanced therapies |
The threshold of 0.60 W has been identified in several studies as a point at which mortality risk rises markedly [4]. In heart failure cohorts, cut-offs of 0.54 to 0.60 W have been reported, with significantly worse survival below this level [4]. In cardiogenic shock in the SHOCK registry, the median CPO of non-survivors was considerably lower than that of survivors [1].
An important application is to follow treatment response: if CPO does not rise after institution of inotropic or mechanical circulatory support, this suggests that the treatment is not improving effective cardiac pumping, and the strategy should be reconsidered.
Validation and performance
SHOCK registry (derivation). CPO was the strongest independent haemodynamic predictor of in-hospital mortality among 541 patients with cardiogenic shock, with an OR of 0.60 per 0.20 W after adjustment for age and hypertension [1].
Baldetti et al. (2022). In a cohort of 80 patients with SCAI stage B to D cardiogenic shock (66% acute decompensated heart failure, the remainder other aetiologies), CPI (CPO indexed to body surface area) achieved an AUC of 0.673 (95% CI 0.529 to 0.817) for in-hospital mortality [2]. A CPI threshold of 0.32 W/m² was used to separate low from high risk. Mortality was 36.4% in the low-CPI group versus 11.1% in the high-CPI group. The study was small and single-arm, which limits generalisability.
Belkin et al. (2022), ESCAPE subanalysis. In 157 patients with acute decompensated heart failure (not cardiogenic shock), CPO was not significantly associated with the primary outcome (freedom from LVAD, transplantation or death at 6 months): OR 0.32 (95% CI 0.08 to 1.29, p = 0.11) [3]. Median CPO was 0.70 W. A cut-off of 0.69 W did not significantly separate the groups in Kaplan–Meier analysis (76% versus 64%, p = 0.08). This indicates that CPO without RAP correction has poorer prognostic performance in less severely ill heart failure populations without frank shock.
Systematic review (Farshadmand et al., 2022). A review of 41 studies and 33,906 patients found that CPO and CPI had prognostic value in cardiogenic shock, heart failure, septic shock and after TAVR [4]. In heart failure, CPO cut-offs of 0.54 and 0.60 W were reported with significantly worse survival below these levels. In TAVR, CPI was the strongest predictor of one-year mortality, with cut-offs around 0.48 to 0.49 W/m².
In summary, CPO is best validated in cardiogenic shock, where discrimination is moderate to good. In heart failure populations without shock, the prognostic value of the simplified formula is weaker, particularly when RAP is elevated.
Limitations
The simplification without RAP. The formula used by this calculator (without RAP) overestimates the true systolic power of the heart when right atrial pressure is elevated, which is common in cardiogenic shock with venous congestion [2,3]. Two independent studies have shown that the original formula including RAP has better prognostic performance, particularly at RAP > 8 mmHg [2,3]. In severe right ventricular or biventricular failure, the simplified formula may systematically misclassify as high risk patients who in fact have preserved left ventricular pumping but high filling pressures.
Load dependence during treatment. CPO is often measured during ongoing inotropic or vasopressor treatment, or with mechanical circulatory support. The value then reflects not the intrinsic pump function of the heart, but the sum of residual cardiac function and external support. This was already the case in the SHOCK registry, where 95% of patients were on sympathomimetic amines and 27% on intra-aortic balloon counterpulsation [2]. Comparisons between patients on different intensities of treatment are therefore difficult to interpret.
Population. The derivation concerns patients with cardiogenic shock, predominantly complicating acute myocardial infarction. CPO is not validated for risk stratification in septic shock, although individual studies have shown an association [4]. Nor is it validated for outpatient heart failure or for patients without invasive haemodynamic monitoring.
Age and sex. In the SHOCK registry there was an inverse correlation between CPI and age (r = −0.334, p < 0.001), and women had lower CPI than men (0.29 versus 0.35 W/m², p = 0.005), even after adjustment for age [1]. This means that older and female patients systematically obtain lower values, in part independently of actual pump function, which must be taken into account in interpretation.
References
- Fincke R, Hochman JS, Lowe AM, et al. Cardiac power is the strongest hemodynamic correlate of mortality in cardiogenic shock: a report from the SHOCK trial registry. J Am Coll Cardiol. 2004;44(2):340-8. PMID: 15261929
- Baldetti L, Pagnesi M, Gallone G, et al. Prognostic value of right atrial pressure-corrected cardiac power index in cardiogenic shock. ESC Heart Fail. 2022;9(6):3920-3930. PMID: 35950538
- Belkin MN, Alenghat FJ, Besser SA, et al. Improved prognostic performance of cardiac power output with right atrial pressure: a subanalysis of the ESCAPE trial. J Card Fail. 2022;28(5):866-869. PMID: 34774746
- Farshadmand J, Lowy Z, Hai O, et al. Utility of cardiac power hemodynamic measurements in the evaluation and risk stratification of cardiovascular conditions. Healthcare (Basel). 2022;10(12):2417. PMID: 36553940
