ECG criteria
- Concordant ST elevation ≥ 1 mm in a lead with a positive (concordant) QRS complex — any lead
- Concordant ST depression ≥ 1 mm in leads V1–V3 (original Sgarbossa criterion, retained in the modification)
- Discordant ST elevation: measure ST elevation at the J-point (in mm) and the depth of the preceding S wave (in mm) in the same lead and complex
- Calculate the ST/S ratio: divide the ST elevation (in mm) by the signed S wave amplitude (S below the isoelectric line is a negative number)
- The criterion is positive with ST elevation ≥ 1 mm and ST/S ≤ −0.25 (i.e., ST elevation is 25% or more of the S wave depth)
- Evaluate all leads; the ST/S criterion is most commonly found in V1–V3 due to deep negative S waves, but can be positive in any lead
- Absolute discordant ST elevation ≥ 5 mm (old Sgarbossa criterion) — low sensitivity, replaced in the Smith modification by the relative ST/S ≤ −0.25 criterion
- Evaluate each lead separately — one lead satisfying any of the criteria is sufficient
What it looks like on ECG
In LBBB and ventricular pacing, the QRS complex is wide (≥ 120 ms), and the ST segment and T wave are discordant to the main QRS vector — this is a normal variant for such morphology. The purpose of the Smith criteria is to identify ST changes that exceed the expected discordance.
Evaluate three patterns:
| Criterion | Where to look | Cut-off |
|---|---|---|
| Concordant ST elevation | any lead with a positive QRS | ≥ 1 mm |
| Concordant ST depression | V1–V3 | ≥ 1 mm |
| Excessive discordant ST elevation | lead with a deep S wave (usually V1–V3) | ST ≥ 1 mm and ST/S ≤ −0.25 |
Concordance means that ST elevation or depression is directed in the same direction as the main QRS complex in that lead — this combination is uncharacteristic of typical LBBB and, in the appropriate clinical context, significantly increases the probability of acute coronary occlusion.
Discordant ST elevation in LBBB is almost always present (especially in V1–V3, where QRS is negative due to a deep S wave) — the question is how pronounced it is. If the ST elevation is a quarter of the S wave depth or more (and at least 1 mm in absolute value), this is suspicious for MI rather than a simple repolarization phenomenon of the block.
During evaluation, measure the ST displacement at the J-point relative to the isoelectric line and the amplitude of the S wave in the same lead and complex (in mm), then divide the ST elevation by the signed S wave amplitude: ST above the isoelectric line is a positive value, S below the isoelectric line is a negative value. For example, with an ST elevation of +6 mm and an S wave of −20 mm, the ratio is 6/(−20) = −0.30.
Mechanism
In LBBB, ventricular depolarization is abnormal (via the right bundle branch and interventricular septum), so repolarization (ST segment, T wave) is expectedly discordant to QRS — this is an electrical, not an ischemic phenomenon. Similar discordance occurs with ventricular pacing because the impulse from the right ventricular endocardium propagates via an abnormal pathway resembling LBBB.
Myocardial ischemia or necrosis creates an additional injury current that either makes the ST segment concordant or significantly enhances the expected discordance. Concordant changes are not expected in typical isolated LBBB or ventricular pacing and therefore increase the probability of acute ischemia/occlusion, although they are not entirely specific and require clinical and instrumental correlation — concordant ST displacements occasionally occur in other conditions, measurement errors, or loading changes. The ST/S criterion accounts for the fact that the amplitude of discordant ST displacement in LBBB is proportional to the QRS amplitude (the deeper the S wave, the greater the expected discordant ST elevation) — therefore, an absolute threshold in mm performs poorly, while a relative threshold (to the S wave amplitude) performs better.
Causes
- Acute coronary occlusion in the setting of baseline LBBB (chronic or acute)
- Acute coronary occlusion in a patient with a cardiac pacemaker
- Situations where standard STEMI criteria are not applicable due to baseline QRS morphology
Differential diagnosis
- Isolated LBBB without ischemia: discordant ST elevation in V1–V3 is present, but ST/S > −0.25, no concordant changes
- Ventricular pacing without ischemia: the same logic applies, Smith criteria apply just as in LBBB
- Hyperkalemia with QRS widening: true LBBB morphology is absent, T wave is peaked, different clinical context
- Wellens syndrome and de Winter waves: occur with narrow QRS, do not require Smith criteria
- Old Sgarbossa criteria (absolute discordant ST elevation ≥ 5 mm): less sensitive, miss some true occlusions that are caught by the relative ST/S criterion
Clinical significance and management
Positive Smith criteria in a patient with LBBB or ventricular pacing and ischemic symptoms indicate a high probability of acute coronary occlusion and require immediate evaluation for emergency coronary angiography/reperfusion. They do not replace clinical evaluation and are not an absolute indication for fibrinolysis on their own: the management strategy is determined by the combination of symptoms, hemodynamics, ECG dynamics, echocardiography, and troponin; if the probability of occlusion is high, immediate interventional cardiology consultation and, as a rule, emergency coronary angiography are necessary. If timely PCI is unavailable, fibrinolysis is considered according to general STEMI rules, but only with high clinical probability and absence of contraindications.
The absence of criteria does not completely rule out occlusion, so decisions are made taking into account the clinical picture, ECG dynamics, and biomarkers; when in doubt, repeat ECGs and an urgent cardiology/interventional consultation are warranted. Old LBBB without access to previous ECGs for comparison is a reason not to delay workup, but LBBB alone (without Smith criteria and without clear clinical symptoms) is not an automatic indication for emergency reperfusion, as was previously believed based on the "new or presumed new LBBB" rule.
Common interpretation errors
- Using the old absolute threshold of ST ≥ 5 mm instead of the relative ST/S ≤ −0.25 — missing some true occlusions
- Measuring the S wave and ST segment in different leads or different complexes
- Applying Smith criteria to a narrow QRS or right bundle branch block — the method is validated specifically for LBBB and ventricular pacing
- Searching only for concordant changes without calculating the ST/S ratio in leads with discordant elevation
- Regarded positive Smith criteria or any LBBB as an independent diagnosis of STEMI and an absolute indication for fibrinolysis without clinical and instrumental evaluation
1. In a patient with LBBB and chest pain, lead V2 shows an S-wave depth of 20 mm, and ST elevation in the same lead is 6 mm. Is the discordant Smith criterion positive?
6 mm / 20 mm = 0.3 in modulus, which exceeds the 0.25 threshold, and ST elevation is ≥ 1 mm — the criterion is positive.
2. Which of the following Smith criteria signs is considered most specific for probable acute coronary occlusion in the setting of LBBB?
ST concordance with the QRS direction is uncharacteristic of typical bundle branch block and, in the appropriate clinical context, indicates an additional ischemic injury current.
3. In a patient with ventricular pacing and ACS symptoms, Smith criteria showed no deviations. How should this be interpreted?
A negative result by Smith criteria reduces but does not rule out the probability of occlusion — decisions are made based on the clinical picture and dynamics.
Frequently asked questions
How do Smith's criteria differ from the original Sgarbossa criteria?
The original Sgarbossa criteria used a fixed discordant ST elevation threshold of ≥ 5 mm, which resulted in low sensitivity. The Smith criteria replaced this item with a relative measure — the ratio of ST elevation to S wave depth (ST ≥ 1 mm and ST/S ≤ −0.25), which significantly improves the detection of true occlusion.
Are Smith's criteria applicable in right bundle branch block?
No, the rule was developed and validated for LBBB and ventricular pacing, where baseline ST-T discordance occurs predictably. In RBBB, the ST-T morphology is different, and Smith's criteria are not used.
Is LBBB dangerous on its own when infarction is suspected?
LBBB itself does not prove acute myocardial infarction, but it is a clinically significant finding often associated with structural heart disease. In suspected ACS with LBBB, Smith's criteria and the clinical context are evaluated, rather than relying solely on the presence of the bundle branch block.
Is it necessary to compare the ECG with old tracings when MI is suspected in the presence of LBBB?
Comparison is helpful, but the absence of previous ECGs should not delay diagnosis. Smith's criteria can be applied to a single current ECG without comparison to older tracings.
How to calculate the ST/S criterion in practice?
Measure the ST segment and S wave in each lead with discordant ST elevation and check the ratio for each lead separately: the ST elevation at the J-point and the depth of the preceding S wave (both values in mm), then divide the ST elevation by the S-wave amplitude, taking signs into account. The criterion is positive if in at least one lead ST/S ≤ −0,25 with ST elevation ≥1 mm.
ECG Course
Interpretation from basics to myocardial infarction — learn to read any ECG using a systematic algorithm.
- Video lessons and theory for every topic
- Simulator case studies under every lesson
- Unlimited ECG simulator practice
- Study guide: "All ECGs in 25 minutes"
ECG Strip Simulator
ECG, radiology, histology, pathomorphology, pharmacology — thousands of flashcards with detailed explanations.
- Real ECG strips, not textbook drawings
- Instant answers and explanations
- Streaks and progress tracking in your dashboard
Related terms
More in the section "Criteria, Formulas, Algorithms"
Sources
- Surawicz B., Knilans T. Chou's Electrocardiography in Clinical Practice
- Hampton J. The ECG in Practice
- ESC Guidelines for the management of acute coronary syndromes
Educational material for medical students and physicians. Does not replace clinical guidelines and physical examination; the ECG must always be interpreted in conjunction with the clinical picture.