CKS is no longer commissioned by the National Institute for Health and Clinical Excellence (NICE). NICE remains committed to providing a replacement service for CKS and is currently reviewing its options. In the meantime, although CKS content is now not being maintained, it still remains relevant and will continue to be made available. CKS content was generated under a programme of topic creation and update. To check if the topic you are viewing is current or out of date, please refer to the topic publication details by clicking on the 'How up-to-date is this topic?' link in the left hand menu on individual topic pages.
Heart failure - chronic - Evidence
Evidence on symptoms, signs, and medical history for diagnosing heart failure
No symptom, clinical sign, or item in the medical history has been shown to provide a reliable diagnosis of heart failure. However, several items provide clinically useful information, including overall clinical impression; breathlessness; dependent oedema; third or fourth heart sounds; abdominojugular reflux; jugular venous distention; pulmonary crackles; and history of myocardial infarction, coronary artery disease, hypertension, diabetes mellitus, and dyslipidaemia. The precision of clinical findings associated with heart failure varies greatly, and no studies have directly investigated the variability between different observers or the variability between different observations made by one observer.
Symptoms and signs
- The National Institute for Health and Clinical Excellence (NICE) found no evidence that any one clinical feature or combination of clinical features provides a reliable diagnosis of heart failure [National Clinical Guideline Centre for Acute and Chronic Conditions, 2010]. NICE identified three systematic reviews which assessed the accuracy of clinical features for diagnosing heart failure. Studies included in the systematic reviews assessed the accuracy of individual symptoms and signs (breathlessness, effort intolerance, increased jugular venous pressure, third heart sound, displaced apex beat, murmurs, oedema, fatigue) compared with a gold standard in the diagnosis of heart failure. No studies or reviews were identified which assessed the diagnostic accuracy of a collection of signs and symptoms. There was some overlap of the studies included in the three reviews, but NICE included all three reviews because each addressed a slightly different population or setting. Two of the reviews were considered to be of high quality and the third of moderate quality.
- The first review included 22 studies with 5237 people who presented to the emergency department with suspected heart failure and dyspnoea [Wang et al, 2005], the second review included 24 studies with 10,710 people who presented in primary care [Madhok et al, 2008], and the third review included 15 studies with 5666 participants from all settings [Mant et al, 2009].
- Table 1 summarizes the results from the systematic reviews identified by NICE for the specificity and sensitivity for each individual symptom. These data show that each of the clinical features has varying specificity and sensitivity, and that poor sensitivity limits the usefulness for the absence of any particular feature at ruling out a diagnosis of heart failure.The NICE guideline development group concluded that individual symptoms and signs are of limited use for diagnosing heart failure.
Table 1. Summary of diagnostic accuracy of symptoms and signs in people with suspected heart failure.
Diagnostic item | Sensitivity (%)* | Specificity (%)† | LR+‡
(95% CI) | LR–§
(95% CI) |
|---|
Dyspnoea | 83 | 54 | 1.79 (1.30 to 2.47) | 0.31 (0.12 to 0.79) |
Dyspnoea on exertion | 84 | 34 | 1.3 (1.2 to 1.4) | 0.48 (0.35 to 0.67) |
Orthopnoea | 44 50 | 89 77 | 3.91 (1.51 to 10.11) 2.2 (1.2 to 3.9) | 0.63 (0.53 to 0.74) 0.65 (0.45 to 0.92) |
Paroxysmal nocturnal dyspnoea | 41 | 84 | 2.6 (1.5 to 4.5) | 0.70 (0.54 to 0.91) |
Oedema | 53 51 | 72 76 | 3.91 (1.51 to 10.11) 2.1 (0.92 to 5.0) | 0.63 (0.53 to 0.74) 0.64 (0.39 to 0.91) |
Elevated jugular venous pressure | 52 39 | 70 92 | 1.73 (1.23 to 2.43) 5.1 (3.2 to 7.9) | 0.68 (0.56 to 0.84) 0.66 (0.57 to 0.77) |
Added heart sounds (all added sounds) | 11 | 99 | 12.1 (5.74 to 25.4) | 0.90 (0.82 to 0.99) |
Added third heart sound | 13 | 99 | 11 (4.9 to 25.0) | 0.88 (0.83 to 0.94) |
Added fouth heart sound | 5 | 97 | 1.6 (0.47 to 5.5) | 0.98 (0.93 to 1.0) |
Lung crepitations/ rales/abnormal breath sounds | 51 60 | 81 78 | 2.64 (1.86 to 3.74) 2.8 (1.9 to 4.1) | 2.64 (1.86 to 3.74) 0.51 (0.37 to 0.70) |
Fatigue | 31 | 70 | 1.0 (0.74 to 1.4) | 0.99 (0.85 to 1.1) |
Hepatomegaly/ hepatic congestion | 17 | 17 | — | — |
* Sensitivity: the proportion of people with a condition who have a positive test result. † Specificity: the proportion of people free of a condition who have a negative test result. ‡ LR+: the likelihood ratio for a positive test; how much the odds of the condition increase with a postive result. § LR–: the likelihood ratio for a negative test; how much the odds of the condition decrease with a negative result. |
|
Medical history
- NICE did not review the evidence on using the person's medical history to diagnose heart failure. CKS identified one systematic review which assessed the evidence on the diagnostic accuracy of medical history to diagnose heart failure in people with dyspnoea [Wang et al, 2005]; the results of which are summarized in Table 2. The table does not show 95% confidence intervals (CIs) for sensitivity and specificity, as these were not published. However, the table does show 95% CIs for likelihood ratios. In general, the 95% CIs are wide, which provides evidence of large variability in the accuracy of medical history for diagnosing heart failure.
Table 2. Summary of diagnostic accuracy of medical history in people with dyspnoea and suspected heart failure.
Diagnostic item | Sensitivity* | Specificity† | LR+‡
(95% CI) | LR–§
(95% CI) |
|---|
Myocardial infarction | 0.40 | 0.87 | 3.1 (2.0 to 4.9) | 0.69 (0.58 to 0.82) |
Coronary artery disease | 0.52 | 0.70 | 1.8 (1.1 to 2.8) | 0.68 (0.48 to 0.96) |
Dyslipidemia | 0.23 | 0.87 | 1.7 (0.43 to 6.9) | 0.89 (0.69 to 1.1) |
Diabetes mellitus | 0.28 | 0.83 | 1.7 (1.0 to 2.7) | 0.86 (0.73 to 1.0) |
Hypertension | 0.60 | 0.56 | 1.4 (1.1 to 1.7) | 0.71 (0.55 to 0.93) |
Smoker | 0.62 | 0.27 | 0.84 (0.58 to 1.2) | 1.4 (0.58 to 3.6) |
Chronic obstructive pulmonary disease | 0.34 | 0.57 | 0.81 (0.60 to 1.1) | 1.1 (0.95 to 1.4) |
* Sensitivity: the proportion of people with a condition who have a positive test result. † Specificity: the proportion of people free of a condition who have a negative test result. ‡ LR+: the likelihood ratio for a positive test; how much the odds of the condition increase with a postive result. § LR–: the likelihood ratio for a negative test; how much the odds of the condition decrease with a negative result. |
|
© NHS Institute for Innovation and Improvement