Killip Classification — Post-MI Heart Failure
Select the class matching the patient's clinical presentation after acute MI.
The Killip Classification is a four-class clinical scoring system that stratifies the severity of cardiac failure complicating acute myocardial infarction (AMI). It was developed by Thomas Killip and John Kimball at New York Hospital and published in the American Journal of Cardiology in 1967, based on a prospective series of 250 AMI patients. The classification uses simple bedside clinical findings — the presence or absence of a third heart sound (S3 gallop), pulmonary crackles, frank pulmonary oedema, and signs of cardiogenic shock — to predict in-hospital mortality. Killip Class I describes patients with no clinical evidence of heart failure: no S3, no crackles, and no signs of hypoperfusion. In the original series, these patients had an in-hospital mortality of approximately 6%. Class II identifies patients with mild-to-moderate left ventricular failure, evidenced by an S3 gallop and/or basal pulmonary crackles in less than 50% of the lung fields, associated with a mortality of approximately 17%. Class III represents acute pulmonary oedema with crackles heard throughout the lung fields, and carries a mortality of approximately 38%. Class IV is cardiogenic shock — defined as a systolic blood pressure below 90 mmHg with clinical signs of peripheral hypoperfusion — and was associated with mortality of 61–81% in the original series. Though derived from the pre-thrombolytic era, the Killip Classification remains highly relevant. Contemporary data from the Global Registry of Acute Coronary Events (GRACE) and other registries confirm that Killip class remains an independent predictor of in-hospital mortality even in the modern reperfusion era, and it is embedded in multiple validated risk scores including GRACE and TIMI-STEMI. The classification guides decisions regarding ICU admission, haemodynamic monitoring, and initiation of vasopressor or mechanical circulatory support.
Killip Class I: No signs of heart failure (no S3, no crackles). In-hospital mortality ~6%. Killip Class II: S3 gallop and/or basal rales/crackles in <50% of lung fields. Mortality ~17%. Killip Class III: Acute pulmonary oedema — crackles throughout >50% of lung fields. Mortality ~38%. Killip Class IV: Cardiogenic shock — SBP <90 mmHg with signs of hypoperfusion (oliguria, cyanosis, cool extremities, altered consciousness). Mortality ~61–81%.
- 1Perform a focused clinical assessment at presentation and at each reassessment: auscultate the heart for a third heart sound (S3 gallop), auscultate the lung fields systematically for crackles, and assess peripheral perfusion.
- 2Class I: No S3 heard, lung fields clear bilaterally, no signs of hypoperfusion. The patient has acute MI without clinical heart failure.
- 3Class II: Auscultate an S3 at the cardiac apex (best heard in left lateral decubitus position with the bell of the stethoscope) and/or detect basal crackles in less than half of the lung fields. This represents mild-to-moderate left ventricular dysfunction.
- 4Class III: Bilateral crackles present in more than 50% of the lung fields, consistent with acute pulmonary oedema from elevated left atrial and pulmonary capillary wedge pressures. Patients typically present with severe dyspnoea, orthopnoea, and hypoxaemia.
- 5Class IV: Cardiogenic shock — systolic blood pressure persistently below 90 mmHg despite adequate filling, combined with clinical evidence of end-organ hypoperfusion (cold extremities, oliguria <30 mL/h, altered sensorium, peripheral cyanosis).
- 6Use the Killip class to guide immediate management: Classes I–II are typically managed in a coronary care unit with close monitoring; Class III requires non-invasive ventilation and urgent diuresis; Class IV requires ICU admission, vasopressors, inotropes, and consideration of intra-aortic balloon pump (IABP) or percutaneous ventricular assist device.
- 7Re-classify throughout the hospital admission — patients can deteriorate from Class I to Class III after recurrent ischaemia, or improve from Class III to Class I after successful PCI and diuresis.
Immediate reperfusion therapy (primary PCI) is the priority.
The absence of any clinical heart failure signs places this patient in Class I. Despite a large anterior MI, immediate reperfusion with primary PCI may prevent progression to higher Killip classes. He should be admitted to the coronary care unit with continuous haemodynamic monitoring.
S3 and basal crackles indicate mild-to-moderate left ventricular dysfunction.
An S3 gallop reflects a non-compliant or volume-overloaded left ventricle. Basal crackles represent early pulmonary venous congestion. This patient requires supplemental oxygen, careful fluid management, and urgent primary PCI. Loop diuretics may be needed cautiously if volume overload is clinically significant.
Bilateral widespread crackles with hypoxia confirm pulmonary oedema.
Massive LV dysfunction has caused acute pulmonary oedema from dramatically elevated pulmonary capillary wedge pressure. This patient requires immediate non-invasive ventilation (CPAP or BiPAP), intravenous loop diuretics, and urgent primary PCI. ICU admission is required. If blood pressure allows, nitrates may reduce preload and improve oxygenation while reperfusion is arranged.
SBP <90 mmHg with multiple signs of end-organ hypoperfusion despite fluid challenge.
Cardiogenic shock complicates 5–10% of STEMIs and carries the highest mortality of any Killip class. Immediate management: vasopressors (noradrenaline first-line), consideration of inotropes (dobutamine), emergent mechanical circulatory support (IABP or Impella), and immediate primary PCI regardless of time from symptom onset. The SHOCK trial demonstrated a survival benefit from early revascularisation in cardiogenic shock.
Stratifying in-hospital mortality risk at AMI presentation for triage, ICU allocation, and treatment escalation decisions.
Determining whether a patient with STEMI or NSTEMI should be admitted to a general coronary care unit versus an intensive care unit.
Guiding haemodynamic support decisions: diuretics and non-invasive ventilation in Class III; vasopressors, inotropes, and mechanical circulatory support in Class IV.
Informing the urgency of primary PCI — even in Class IV cardiogenic shock, immediate revascularisation is prioritised based on the SHOCK trial evidence.
Contributing to composite risk scores (GRACE, TIMI-STEMI) used to predict 30-day and 6-month mortality after acute coronary syndrome.
Right ventricular infarction and Killip classification
Right ventricular (RV) infarction, occurring in 30–50% of inferior STEMIs, can produce signs of hypoperfusion (cardiogenic shock pattern — Killip Class IV) with clear lung fields rather than pulmonary oedema. This occurs because the failing RV causes elevated right-sided filling pressures and reduced left-sided output without pulmonary congestion. The classic triad is hypotension, clear lungs, and elevated jugular venous pressure. Treatment differs critically from left-sided cardiogenic shock: these patients require volume loading rather than diuretics.
Mechanical complications of MI
Acute mitral regurgitation (from papillary muscle rupture) or ventricular septal defect (VSD) can cause sudden deterioration to Killip Class III–IV hours to days after the initial MI, even if the patient was initially Class I. A new loud systolic murmur in a patient with worsening haemodynamics post-MI should prompt urgent echocardiography to exclude these mechanical complications, which require surgical repair or catheter-based intervention rather than medical management alone.
Killip class versus haemodynamic subsets (Forrester Classification)
The Forrester haemodynamic classification provides greater physiological granularity by using invasive pulmonary capillary wedge pressure (PCWP) and cardiac index (CI) measurements via pulmonary artery catheter. Killip class correlates with but does not perfectly predict Forrester subset. A Killip Class III patient may have either a low or normal cardiac index depending on the degree of compensatory increase in peripheral resistance. In ICU settings, Forrester classification guides specific therapy but requires invasive monitoring; Killip class is used for rapid, non-invasive risk stratification.
NSTEMI and Killip classification
Although the Killip Classification was originally developed for STEMI, it is also routinely applied to NSTEMI and used in the GRACE score for non-ST elevation acute coronary syndromes. The same clinical criteria and prognostic relationships apply. Patients with NSTEMI presenting in Killip Class III–IV have similar short-term mortality to those with STEMI in equivalent classes and should be managed with equivalent urgency.
Inter-observer variability in S3 detection
Detection of the third heart sound (S3) — a key criterion separating Killip Class I from Class II — is notoriously difficult and subject to significant inter-observer variability, even among experienced cardiologists. Studies have shown kappa statistics as low as 0.3–0.4 for S3 detection. Clinicians should be trained to auscultate specifically for S3, use the bell of the stethoscope at the cardiac apex in the left lateral decubitus position, and correlate with other clinical signs when uncertain.
| Class | Clinical Findings | Original In-Hospital Mortality (1967) | Contemporary Mortality (Post-PCI Era) |
|---|---|---|---|
| I | No signs of heart failure — clear lungs, no S3 | ~6% | ~4–6% |
| II | S3 gallop and/or basal crackles (<50% lung fields) | ~17% | ~8–12% |
| III | Acute pulmonary oedema — crackles >50% lung fields | ~38% | ~15–25% |
| IV | Cardiogenic shock — SBP <90 mmHg + hypoperfusion signs | ~61–81% | ~40–60% |
What is the Killip Classification?
The Killip Classification is a four-class system developed by Killip and Kimball in 1967 to categorise the severity of cardiac failure complicating acute myocardial infarction. It uses bedside clinical findings — heart sounds, lung auscultation, blood pressure, and perfusion signs — to estimate in-hospital mortality risk. Class I has no heart failure (~6% mortality), Class II has mild failure (~17%), Class III has pulmonary oedema (~38%), and Class IV is cardiogenic shock (~61–81%).
How is Killip class determined at the bedside?
Killip class requires only a stethoscope and blood pressure measurement. Auscultate the heart for a third heart sound (S3 gallop) using the bell of the stethoscope at the cardiac apex with the patient in the left lateral decubitus position. Systematically auscultate both lung fields for crackles. Measure blood pressure and assess peripheral perfusion (skin temperature, colour, urine output, consciousness). No laboratory tests or imaging are required for classification, though they support clinical decision-making.
Is the Killip Classification still valid in the modern reperfusion era?
Yes. Although the original mortality figures from 1967 reflect the pre-thrombolytic, pre-PCI era — when mortality was substantially higher — the relative risk relationships between classes remain valid. Contemporary data from the GRACE registry and other large cohorts confirm that higher Killip class is independently associated with increased in-hospital mortality even after adjustment for reperfusion therapy, time to treatment, and modern pharmacotherapy. Absolute mortality rates are lower, but the class still distinguishes risk.
What is the difference between Killip Class III and Class IV?
Both involve severe left ventricular failure, but the distinction is haemodynamic. Class III patients have acute pulmonary oedema (widespread crackles, hypoxia) but maintain a systolic blood pressure above 90 mmHg with adequate peripheral perfusion. Class IV patients (cardiogenic shock) have SBP persistently below 90 mmHg combined with clinical signs of end-organ hypoperfusion — oliguria, altered consciousness, cold and mottled extremities — indicating that cardiac output is insufficient to maintain vital organ perfusion.
How does Killip class guide ICU admission and treatment decisions?
Class I and II patients are typically managed in a coronary care unit with continuous cardiac monitoring. Class III requires ICU-level care with non-invasive ventilation support (CPAP/BiPAP) and careful haemodynamic monitoring. Class IV (cardiogenic shock) mandates immediate ICU admission with invasive haemodynamic monitoring, vasopressor and inotrope support, and consideration of mechanical circulatory support devices (IABP, Impella, ECMO). Urgent revascularisation (primary PCI) is the cornerstone of treatment across all classes.
What role does Killip class play in the GRACE score?
The GRACE (Global Registry of Acute Coronary Events) score incorporates Killip class as one of eight variables to predict in-hospital and 6-month mortality after acute coronary syndrome. In the GRACE score, Killip class is assigned a weight of 0 points (Class I), 20 points (Class II), 39 points (Class III), or 59 points (Class IV), reflecting the steep mortality gradient between classes and its strong independent prognostic contribution.
Can a patient's Killip class change during their hospital admission?
Yes — in both directions. A patient who presents as Class I may deteriorate to Class III or IV if they develop recurrent ischaemia, mechanical complications (papillary muscle rupture, VSD), or arrhythmias. Conversely, a patient presenting as Class III may respond well to primary PCI, diuretics, and non-invasive ventilation and be reclassified as Class I or II within 24–48 hours. Serial reassessment is therefore essential.
What is the definition of cardiogenic shock in Killip Class IV?
Cardiogenic shock (Killip Class IV) requires a systolic blood pressure persistently below 90 mmHg (or a drop of >30 mmHg from baseline) despite adequate intravascular volume, combined with signs of reduced tissue perfusion: urine output below 30 mL/hour, cold and clammy extremities, peripheral cyanosis, and/or altered mental status. This definition aligns with ESC and AHA/ACC cardiogenic shock criteria used in contemporary clinical guidelines.
Pro Tip
When auscultating for the S3 gallop that distinguishes Killip Class I from Class II, position the patient in the left lateral decubitus position, use the bell of the stethoscope lightly applied to the cardiac apex, and listen during early diastole immediately after the S2. The S3 is a low-pitched sound best heard at the end of expiration. In noisy emergency departments, pressing the bell firmly (which converts it to a diaphragm) eliminates the S3 — this common error leads to underclassification.
Did you know?
When Thomas Killip published his classification in 1967, the coronary care unit had only existed for about five years — it was introduced by Desmond Julian in Edinburgh in 1961. The classification was developed before thrombolytics, before primary PCI, and before aspirin was a standard therapy. Despite being derived from a cohort of just 250 patients managed with bed rest and oxygen, it has been validated in millions of patients across more than 50 years and across every continent, making it one of the most enduring prognostic tools in all of cardiology.
References
- ›Killip T, Kimball JT — Treatment of Myocardial Infarction in a Coronary Care Unit (Am J Cardiol, 1967)
- ›2023 ESC Guidelines for the Management of Acute Coronary Syndromes
- ›Hochman JS et al. — Early Revascularisation in Acute MI Complicated by Cardiogenic Shock — SHOCK Trial (NEJM, 1999)
- ›Eagle KA et al. — GRACE Score — A Simple Approach for Risk Stratification of Patients with ACS (JAMA, 2004)
- ›De Luca G et al. — Killip Class and Short-Term Prognosis in AMI in the Modern PCI Era (Heart, 2009)