Written and maintained by CASRAI Editorial Board
Last updated
The Killip classification is a bedside tool for grading the severity of heart failure in a patient presenting with acute myocardial infarction (MI). It sorts a patient into one of four classes — I through IV — using only the physical exam: no labs, no imaging, no calculator. Developed by Thomas Killip and John Kimball in 1967 from a series of 250 coronary care unit admissions, it remains one of the most widely used prognostic tools in acute MI because it can be applied the moment a clinician reaches the bedside, before troponin or an echocardiogram is available.
Killip is often mentioned in the same breath as the TIMI score, and the two are frequently confused or conflated because both are point-based tools applied to the same patient population at the same moment — presentation with acute coronary syndrome. They are not the same tool and answer different questions. This page covers what Killip class actually measures, how it’s assigned, its prognostic value, and exactly how it differs from and complements TIMI.
The four Killip classes
Each class is defined entirely by physical-exam findings for signs of heart failure — there is no scoring arithmetic, just a categorical assignment to the highest class the findings support:
| Class | Clinical findings | What it represents |
|---|---|---|
| I | No rales, no S3 gallop, no evidence of heart failure | No clinical heart failure |
| II | Rales in the lower half of the lung fields (or less), S3 gallop, and/or elevated jugular venous pressure | Mild to moderate heart failure |
| III | Frank, overt pulmonary edema (rales throughout both lung fields) | Severe heart failure |
| IV | Cardiogenic shock — hypotension (systolic blood pressure typically <90 mmHg) with signs of hypoperfusion (cold extremities, altered mentation, oliguria) | Cardiogenic shock |
A patient is assigned to the highest class their findings support at the time of exam — Killip class is a snapshot, not a cumulative score, and it can change over a hospital stay as the patient improves or deteriorates. Because every input is obtainable at the bedside by inspection, auscultation, and blood pressure measurement, it requires no equipment beyond a stethoscope and can be reassigned repeatedly as a patient’s status changes, which is part of why it has stayed in routine use for nearly six decades.
Why the bedside-only design matters
Unlike scoring tools that incorporate ECG findings, cardiac biomarkers, or continuous variables like heart rate and creatinine, Killip class uses none of these. That’s a deliberate simplicity, not a limitation the original authors were trying to work around: Killip and Kimball designed it specifically to let a clinician stratify a patient’s heart-failure severity, and therefore short-term risk, in the first minutes of contact — before labs return, before imaging is scheduled, and without needing a calculator or reference chart. That same property is why it’s still embedded in chest-pain order sets and ACS documentation templates today, even though more statistically powerful tools exist for later, more complete risk stratification once additional data is available.
Prognostic value: what Killip class predicts
Killip class was developed as, and remains, a mortality-prognostic tool for acute MI — higher class correlates with substantially higher short-term mortality. In the original 1967 derivation cohort of 250 patients, in-hospital mortality rose steeply across the four classes: roughly 6% in class I, 17% in class II, 38% in class III, and 81% in class IV. Those figures reflect coronary care in the pre-reperfusion era and are not representative of outcomes today — contemporary registries, reflecting decades of advances in reperfusion therapy (fibrinolysis and primary PCI), antiplatelet/anticoagulant therapy, and critical care support, report substantially lower mortality at every class, though the same graded, step-wise relationship between class and mortality risk persists: higher-class patients still carry meaningfully worse outcomes than lower-class ones, and Killip class IV (cardiogenic shock) remains the strongest single predictor within the classification, still carrying a high risk of death despite modern reperfusion and mechanical circulatory support. That durability — the ordinal relationship holding up across nearly 60 years of otherwise-transformed MI care — is a large part of why the classification is still in active clinical and quality-reporting use rather than having been fully superseded.
Killip class vs. the TIMI score: different questions, same patient
Because both are commonly documented at presentation for the same acute-MI patient, Killip class and the TIMI score get conflated in practice — but they measure genuinely different things:
- What each one is. Killip class is a categorical grade of the patient’s current heart-failure severity, assigned from physical-exam findings alone. TIMI is a point-based ischemic risk score that predicts short-term adverse cardiac events (death, reinfarction, urgent revascularization for UA/NSTEMI; 30-day mortality for STEMI) from a combination of clinical history, presentation, and ECG/biomarker findings.
- What each one needs. Killip class needs nothing beyond a physical exam. TIMI needs history (age, risk factors, prior CAD, recent aspirin use), the presenting ECG, and cardiac biomarkers — inputs that may not all be available in the first minutes.
- How they relate, not compete. Killip class II–IV is itself one of the eight weighted variables inside the TIMI Risk Score for STEMI — meaning a patient’s Killip class doesn’t just sit alongside TIMI, it is a direct input into it. A patient can score identically on TIMI’s other seven STEMI variables and still land in a materially different risk category purely because of Killip class. The two tools are complementary layers of the same initial evaluation, not competing options to choose between: Killip characterizes current hemodynamic/heart-failure status, TIMI (and, more broadly, tools like the GRACE score) synthesize that alongside other variables into a forward-looking ischemic risk estimate.
For a hospital’s chest-pain protocol or quality-reporting workflow, this distinction is operationally important: a documentation or abstraction template that records only a TIMI point total without separately capturing Killip class is losing an interpretable clinical signal that TIMI only partially encodes (as a binary II–IV flag, not the full four-class grade) — and vice versa, a workflow that captures Killip class but never links it into a formal ischemic-risk tool leaves prognostic information on the table.
Why this matters for hospital patient-safety and quality programs
For patient-safety officers, quality directors, and cardiology/ICU staff building or auditing chest-pain and ACS protocols, Killip class shows up in a few recurring, practical places:
- Order-set and pathway design. Standardized chest-pain and STEMI/NSTEMI order sets frequently require a documented Killip class at admission as a discrete, auditable field — verifying it’s captured correctly, and captured early enough to be clinically useful, is a real order-set-validation question.
- Risk-adjustment and case-mix reporting. Because Killip class is one of the more powerful single predictors of in-hospital MI mortality, it’s a common covariate in quality-reporting and outcomes case-mix adjustment — a chart missing a Killip-class field, or one recorded inconsistently between clinicians, degrades the reliability of any mortality-adjusted comparison built on top of it.
- Escalation and triage thresholds. Because class IV (cardiogenic shock) carries a categorically different risk and resource need than classes I–III — often triggering mechanical circulatory support evaluation, ICU-level care, or transfer protocols — some institutions build explicit escalation triggers directly off the Killip class documented at presentation, making accurate, timely assignment a direct patient-safety issue, not just a documentation formality.
Related bedside scoring tools
Killip class is one of several validated, point-or-category-based bedside tools hospitals use to standardize acute-presentation decision-making. See the TIMI score for ischemic risk stratification in the same ACS population this page discusses, the APACHE II score for general ICU mortality prediction, and the Ranson Criteria for a comparably bedside-driven severity classification in a different acute condition. For the broader set of clinical-risk and patient-safety topics this page sits alongside, see the Patient Safety pillar.
Frequently asked questions
Does Killip class require any lab tests or imaging?
No. It is assigned entirely from the physical exam — auscultation for rales and an S3 gallop, inspection of jugular venous pressure, and blood pressure/perfusion assessment for shock. That’s a deliberate design choice: it lets clinicians grade heart-failure severity in the first minutes of contact, before labs or imaging results are available.
Is Killip class the same as the TIMI score?
No, and they aren’t interchangeable. Killip class is a four-category grade of current heart-failure severity from exam findings alone. TIMI is a separate point-based score predicting short-term ischemic risk (death, reinfarction, or urgent revascularization) from a combination of history, ECG, and biomarker data. Killip class II–IV is actually one input inside the TIMI Risk Score for STEMI, so the two are linked, not competing.
Can a patient’s Killip class change during a hospital stay?
Yes. Killip class reflects the patient’s status at the time of a given exam, not a fixed value assigned once at admission. A patient can move between classes as heart failure worsens or improves and is typically reassessed as clinical status changes, not just recorded once.
Is Killip class still clinically relevant given how much MI treatment has changed since 1967?
Yes. While the exact mortality rates from the original 1967 cohort no longer apply — contemporary reperfusion and critical-care advances have lowered mortality substantially at every class — the same graded, step-wise relationship between higher class and worse short-term outcome has held up across decades of otherwise-transformed acute MI care, and class IV (cardiogenic shock) remains one of the strongest single mortality predictors in acute MI.
