Framingham diagnostic criteria for heart failure

Clinical background

The Framingham diagnostic criteria for heart failure fill a specific role: a standardised clinical definition of congestive heart failure requiring only the history and examination. In everyday acute care, breathlessness and peripheral oedema may arise from the heart, lungs, kidneys or liver, and without a structured assessment both overdiagnosis and underdiagnosis are risks. The criteria were created to combine findings into a reproducible binary diagnosis independent of laboratory tests and imaging.

In modern practice, echocardiography and natriuretic peptides have taken over as the diagnostic cornerstones, and the 2021 ESC guidelines define heart failure as symptoms combined with objectively demonstrated structural or functional cardiac dysfunction together with a raised natriuretic peptide [6]. The Framingham criteria nonetheless remain the reference standard in epidemiological studies and a bedside tool when imaging and biomarkers are not immediately available.

Applying the Framingham diagnostic criteria

The criteria consist of eight major and eight minor criteria, all binary (yes/no). A diagnosis of congestive heart failure is made when at least two major criteria are met simultaneously, or when one major criterion is combined with at least two minor criteria:

$$\text{Heart failure} \iff \left(\text{Major} \geq 2\right) \lor \left(\text{Major} \geq 1 \land \text{Minor} \geq 2\right)$$

Minor criteria may be counted only if they cannot be explained by another condition. Weight loss of at least 4.5 kg over five days on treatment may count as either a major or a minor criterion, depending on which completes the diagnosis.

The major criteria are paroxysmal nocturnal dyspnoea or orthopnoea, jugular venous distension, crackles on auscultation of the lungs, cardiomegaly on radiography, acute pulmonary oedema, an S3 gallop, a venous pressure >16 cmH2O and a positive hepatojugular reflux.

The minor criteria are bilateral ankle oedema, nocturnal cough, dyspnoea on ordinary exertion, hepatomegaly, pleural effusion, a vital capacity reduced by one-third from the measured maximum, a heart rate >120 beats/min, and weight loss on treatment.

The criteria were derived from the Framingham Heart Study and published by McKee et al. in 1971 [1]. The cohort consisted of adult residents of Framingham, Massachusetts, and the criteria were formulated to allow longitudinal surveillance of heart failure incidence in a population study. They were developed decades before echocardiography or natriuretic peptides existed as routine methods.

Interpretation in practice

A positive result by the Framingham criteria does not establish the diagnosis on its own in modern practice. Their high sensitivity but limited specificity means that a substantial proportion of false positives is to be expected, particularly in patients with COPD, renal failure or anaemia, who share several minor criteria. A negative result, by contrast, markedly lowers the probability of heart failure and can direct the work-up towards other causes.

Validation and performance

In a Spanish validity study by Jimeno Sainz et al. of 224 patients with breathlessness in an emergency department, a sensitivity of 96% and a positive predictive value of 97% were measured against an ejection fraction ≤ 0.45 as the reference, but specificity was low, with a positive likelihood ratio of only 1.52 [2]. The negative likelihood ratio was 0.09, confirming that the absence of the criteria effectively lowers the probability of systolic heart failure.

Maestre et al. validated the criteria in 216 patients on a cardiology ward and reported a sensitivity of 92% and a specificity of 79% against echocardiographic left ventricular dysfunction as the reference [3]. The likelihood ratio for a negative result was 0.1 and for a positive result 4.3. The criteria were more reliable at excluding systolic heart failure (LR− 0.04) than diastolic (LR− 0.1), reflecting the fact that clinical findings in heart failure with preserved ejection fraction are more subtle.

In a primary care cohort from Portugal by Fonseca et al. of 5,434 people, sensitivity fell to 63% at a specificity of 93% [4]. This reflects the poorer performance of the criteria in a primary care population with a low baseline frequency of heart failure than on a selected cardiology ward.

The ARIC study by Rosamond et al. reviewed 705 hospitalisations in four American communities [5]. Among cases meeting the Framingham criteria, 68% were judged to be acute decompensated heart failure by independent reviewers, 10% chronic stable heart failure and 22% not heart failure at all. Approximately a quarter of hospitalisations that did not meet the Framingham criteria were nonetheless classified as acute decompensated heart failure by the review committee, indicating that false negatives also occur to a non-negligible extent. The criteria furthermore do not distinguish acute decompensated heart failure from chronic stable heart failure, which is a relevant limitation in acute management.

Limitations

The criteria cannot distinguish systolic from diastolic heart failure and do not demonstrate structural or functional cardiac dysfunction, which is the core of the modern definition of heart failure under the 2021 ESC guidelines [6]. They should not be used as the sole basis for treatment decisions when echocardiography and natriuretic peptides are available.

Several minor criteria are non-specific and may be present in a range of other conditions. Bilateral ankle oedema may be due to venous insufficiency or renal failure, nocturnal cough to COPD or gastro-oesophageal reflux, and hepatomegaly to liver disease. The rule that minor criteria count only if they cannot be explained by another condition introduces a subjective judgement that reduces reproducibility between assessors.

An S3 gallop and a positive hepatojugular reflux, among the most specific individual findings, occur infrequently and require experience to elicit reliably. A venous pressure >16 cmH2O requires specific equipment and is not part of modern bedside routine on most wards. Cardiomegaly on radiography requires access to a chest film, which partly defeats the purpose of an instrument that is by intent based on bedside findings.

The criteria were furthermore derived in a mid-twentieth-century cohort with a different disease pattern from today's, with a higher proportion of hypertensive and valvular heart failure and a lower proportion of heart failure with preserved ejection fraction. The diagnostic uncertainty the criteria entail is greatest precisely for heart failure with preserved ejection fraction, where the clinical findings are more subtle and echocardiography is decisive.

References

1. McKee PA, Castelli WP, McNamara PM, Kannel WB. The natural history of congestive heart failure: the Framingham study. N Engl J Med 1971;285(26):1441-6. PMID: 5122894
2. Jimeno Sainz A, Gil V, Merino J et al. Validity of Framingham criteria as a clinical test for systolic heart failure. Rev Clin Esp 2006;206(10):495-8. PMID: 17203567
3. Maestre A, Gil V, Gallego J et al. Diagnostic accuracy of clinical criteria for identifying systolic and diastolic heart failure: cross-sectional study. J Eval Clin Pract 2009;15(1):55-61. PMID: 19239582
4. Fonseca C, Oliveira AG, Mota T et al. Evaluation of the performance and concordance of clinical questionnaires for the diagnosis of heart failure in primary care. Eur J Heart Fail 2004;6(6):813-20. PMID: 15542422
5. Rosamond WD, Chang PP, Baggett C et al. Classification of heart failure in the Atherosclerosis Risk in Communities (ARIC) study: a comparison of diagnostic criteria. Circ Heart Fail 2012;5(2):152-9. PMID: 22271752
6. McDonagh TA, Metra M, Adamo M et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur J Heart Fail 2022;24(1):4-131. PMID: 35083827