AHA/ACCF/HRS Recommendations for the Standardization and Interpretation of the ECG — Part IV: ST Segment, T and U Waves, and QT Interval (2009)

Authors, Journal, Affiliations, Type, DOI

• Pentti M. Rautaharju, Borys Surawicz, Leonard S. Gettes; on behalf of the AHA Electrocardiography and Arrhythmias Committee writing group (includes Bailey, Childers, Deal, Gorgels, Hancock, Josephson, Kligfield, Kors, Macfarlane, Mason, Mirvis, Okin, Pahlm, van Herpen, Wagner, Wellens)
• Journal of the American College of Cardiology, Vol. 53, No. 11, March 17, 2009; pp. 982–91; copublished in Circulation
• Endorsed by the International Society for Computerized Electrocardiology
• Type: Scientific statement / consensus standardization document (Part IV of a 6-part AHA/ACCF/HRS ECG standardization series)
• DOI: 10.1016/j.jacc.2008.12.014

Overview

Part IV of the AHA/ACCF/HRS ECG standardization series establishes foundational measurement standards and interpretive frameworks for the ST segment, T and U waves, and QT interval — the ECG components reflecting ventricular repolarization. The document formally defines primary versus secondary repolarization abnormalities, provides numeric ST elevation and depression thresholds stratified by sex/age/race, introduces a quantitative T-wave classification system, clarifies U wave norms and clinical significance, and addresses the major challenges of QT interval measurement and rate correction. Critically, it recommends linear regression functions (not Bazett's formula) for QT-rate correction and sets practical clinical thresholds for prolonged QT (women ≥460 ms; men >450 ms) and short QT (≤390 ms). It also explicitly recommends against including QT dispersion in routine ECG reports due to fundamental methodological problems.

Keywords

Electrocardiography, ST segment, T wave, U wave, QT interval, repolarization, primary repolarization abnormality, secondary repolarization abnormality, Bazett formula, Fridericia formula, QT correction, QT dispersion, T-wave alternans, J point, early repolarization

Key Takeaways

Primary vs Secondary Repolarization Abnormalities

• Primary repolarization abnormalities: Changes in ST/T wave caused by alterations in the shape and/or duration of repolarization phases of the transmembrane action potential, occurring without changes in depolarization (normal QRS). Causes include ischemia, myocarditis, drugs, toxins, electrolyte abnormalities (especially Ca²⁺ and K⁺), abrupt heart rate changes, hyperventilation, body position changes, catecholamines, sympathetic activation/ablation, and temperature changes
• Secondary repolarization abnormalities: ST/T wave changes that arise directly from changes in the sequence and/or duration of ventricular depolarization (abnormal QRS). Do NOT require changes in individual cell action potentials; instead reflect voltage gradients normally canceled that become manifest due to altered depolarization sequence. Examples: LBBB (ST-T vectors opposite to mean QRS vector), RBBB (ST-T opposite to slow terminal QRS component), ventricular preexcitation (ST-T opposite to delta wave), ectopic ventricular complexes, paced rhythms
• Primary and secondary repolarization abnormalities may coexist (e.g., ventricular hypertrophy with both primary action potential changes and secondary QRS-related changes)
• Cardiac memory: Secondary changes from prolonged pacing may take hours to days to develop and dissipate (unlike acute ectopic beats which revert promptly)
• Recommendation: designating ST-T abnormalities as primary or secondary is clinically relevant and should be incorporated into automated ECG algorithms

ST-Segment Abnormalities — Measurement

• ST and T-wave amplitudes are referenced against the TP or PR segment
• Low-frequency filtering (to remove baseline drift) means that true DC voltage levels are inaccessible; ST elevation may reflect PR/TP depression, true ST elevation, or both; ST depression may reflect PR/TP elevation, true ST depression, or both
• J point: displacement of ST segment is measured at the junction of QRS end and ST segment onset ("J point"); in exercise testing, also at J+40 and up to J+80 ms
• ST segment can be elevated, depressed, upsloping, horizontal, or downsloping; depressed ST may be further described as horizontal, downsloping, upsloping (rapidly/slowly)
• Total QRS amplitude affects ST-segment amplitude abnormalities — should be considered

ST-Segment Abnormalities — Normal Limits and Thresholds

• Normal ST elevation is greater in young/middle-aged males than females; greater in Blacks than Whites; most pronounced in V2
• Normal J-point elevation upper limits (98th percentile) in V2:
  • White men <40 years: ~0.3 mV (up to 0.33 mV age 24–29)
  • White men ≥40 years: ~0.25 mV
  • White women (all ages): ~0.15 mV
  • Black men ≥40 years: ~0.20 mV; Black women ≥40 years: ~0.15 mV
• Threshold for abnormal J-point elevation (per Part VI of this series, acute ischemia/infarction):
  • V2 and V3: ≥0.2 mV in men ≥40; ≥0.25 mV in men <40; ≥0.15 mV in women
  • All other standard leads: ≥0.1 mV in men and women
• ST depression threshold: ≥0.1 mV (noted as meaningful for ECG report); Part VI threshold for abnormal J-point depression: ≥0.05 mV in V2–V3; ≥0.1 mV in all other leads
• Three main causes of ST elevation: (1) normal variant "early repolarization" — J-point elevation with rapidly upsloping or normal ST segment; (2) injury currents in acute ischemia/ventricular dyskinesis; (3) injury currents in pericarditis
• Normal J-point elevation in V1–V2 typically shows steep downsloping ST segment; ischemia shows more horizontal ST
• Race-, age-, and sex-stratified normal limits should be incorporated into automated ECG algorithms to avoid misclassifying normal ST variants as injury patterns

T-Wave Abnormalities — Normal Values

• Normal T-wave polarity in adults ≥20 years: inverted in aVR; upright or inverted in aVL, III, V1; upright in leads I, II, V3–V6
• T-wave commonly inverted in V1–V3 in children >1 month; may be slightly inverted in aVF and V2 in adolescents/young adults <20 years
• Normal T-wave amplitude in V2 (upper thresholds, men): 1.0–1.4 mV (up to 1.6 mV in men 18–29 years)
• Normal T-wave amplitude in V2 (upper thresholds, women): 0.7–1.0 mV
• T-wave amplitude in lateral V5–V6: slightly negative (<0.1 mV) in 2% of White men/women ≥60; 2% of Black men/women ≥40; negative ≥0.1 mV in 5% of Black men/women ≥60

T-Wave Abnormalities — Quantitative Descriptors

• Qualitative descriptors: peaked, symmetrical, biphasic, flat, inverted
• Quantitative classification for T wave in leads I, II, aVL, V2–V6:
  • Inverted: amplitude −0.1 to −0.5 mV
  • Deep negative: amplitude −0.5 to −1.0 mV
  • Giant negative: amplitude < −1.0 mV
  • Low: amplitude <10% of R-wave amplitude in same lead
  • Flat: peak T-wave amplitude between +0.1 and −0.1 mV in leads I, II, aVL (with R wave >0.3 mV), and V4–V6
• Giant T-wave inversion is generally limited to: hypertrophic cardiomyopathies, NSTEMI (non–ST-segment elevation myocardial infarction), and neurological events (particularly intracranial hemorrhage)
• Isolated minor T-wave abnormalities: classify as "slight" or "indeterminate"; list most common causes; compare with prior ECGs
• T-wave notching vs U wave: interval between 2 summits of a notched T wave is usually <150 ms; interval between T-wave peak and U wave usually exceeds 150 ms at HR 50–100 bpm

T-Wave Alternans

• T-wave alternans = T-wave amplitude variations alternating every second beat
• Typically observed at microvolt level (microvolt T-wave alternans)
• Indicates latent instability of repolarization, predictive of malignant arrhythmias
• Generally NOT present at rest even in high-risk patients; provoked by exercise or pharmacological stress or pacing
• Role not fully defined at the time of this document but holds "substantial potential" for identifying patients at high arrhythmic risk

The U Wave

• Mechanoelectric phenomenon: low-amplitude, low-frequency deflection following the T wave; most evident in V2 and V3
• Normal amplitude: ~0.33 mV or ~11% of T-wave amplitude in V2–V3
• Heart-rate dependent: rarely present at HR >95 bpm; present in 90% of cases at HR <65 bpm (enhanced by bradycardia)
• Apparent increased U-wave amplitude with hypokalemia (classically K⁺ <2.7 mmol/L) more likely represents fusion of U wave with T wave rather than true U-wave amplitude increase
• U-T fusion also occurs with sympathetic activation and in markedly prolonged QT (congenital and acquired LQTS)
• Inverted U wave in V2–V5 is abnormal: may appear transiently during acute ischemia or in hypertension
• Abnormal U waves are subtle and frequently overlooked by readers and automated systems
• Recommendation: include U wave statements when U wave is inverted, merged with T, or exceeds T-wave amplitude

QT Interval — Measurement

• Defined: onset of QRS complex to end of T wave (earliest indication of ventricular depolarization to latest indication of repolarization)
• Measurement challenges: (1) identifying QRS onset and T-wave end; (2) determining appropriate lead(s); (3) adjusting for QRS duration, gender, and rate
• Lead selection: use lead showing longest QT (usually V2 or V3); if this value differs by >40 ms from adjacent leads, measurement may be in error — use adjacent leads
• When T and U waves are superimposed and inseparable: measure QT in leads not showing U waves (often aVR and aVL), or extend downslope tangent of T to TP segment — both methods may underestimate QT
• QRS onset up to 20 ms earlier in V2–V3 than limb leads
• Normal QT dispersion between leads: differences up to 50 ms (some sources up to 65 ms); less in women than men
• Digital simultaneous multilead recording: allows temporal alignment; automatically measured QT often longer than any individual lead; established "normal" limits from single-channel sequential recordings may no longer be valid
• Recommendation: provide optional display of temporally aligned superimposed ECG leads; always visually validate computer-reported QT prolongation

QT Interval — Rate Correction

• Many formulas exist; most widely used:
  • Bazett (1920): QTc = QT / √RR — leaves strong positive residual correlation (r = 0.32) with heart rate; may substantially over-correct QT at high heart rates
  • Fridericia (1920): QTc = QT / ∛RR — leaves negative residual correlation (r = −0.26 to −0.32) with heart rate
• Linear regression formulas: effectively remove rate dependence of adjusted QT; clearly preferable to both Bazett and Fridericia
• Recommendation: use linear regression functions (not Bazett) for QT-rate correction; identify the correction method used in ECG analysis reports; do NOT attempt rate correction when RR variability is large (e.g., atrial fibrillation) or when T-wave end identification is unreliable

QT Interval — Normal Thresholds (Gender and Rate Corrected)

• Prolonged QT: women ≥460 ms; men >450 ms
• Short QT: women and men ≤390 ms
• Gender difference: QT longer in young/middle-aged females than males by 6–15 ms; difference appears at adolescence (testosterone shortens QT in boys but not girls); gender difference becomes small after age 40 and practically disappears in older adults
• FDA drug evaluation severity levels for prolonged QTc: >350 ms, >480 ms, >500 ms
• QT adjusted by Bazett's formula may produce false QT prolongations — particularly at high heart rates

QT Interval — Correction for QRS Duration

• QT interval prolongs in ventricular conduction defects; QRS duration adjustment is necessary
• Best accomplished by incorporating QRS duration and RR interval as covariates in QT adjustment formula, or by using JT interval (QT – QRS duration)
• If JT interval is used, apply normal standards specifically established for JT (not QT)

QT Dispersion — NOT Recommended

• QT dispersion = difference between longest and shortest QT intervals across leads; introduced 1990 for LQTS risk identification
• Despite 670 publications and widespread teaching, fundamental methodological problems identified:
  • QT differences between leads largely reflect measurement artifact and T-loop morphology, not regional repolarization heterogeneity
  • Cannot extract non-dipolar signal information beyond X, Y, Z ECG components
  • Concept that QT dispersion measures localized repolarization heterogeneity has NOT been validated
• Recommendation: QT dispersion should NOT be included in routine ECG reports
• Continued research into markers of increased dispersion of myocardial repolarization on body surface ECG encouraged

Limitations of the Document

• Published 2009 — some normative data and threshold recommendations pre-date large-scale multiethnic population studies
• Linear regression QT correction recommendation not universally adopted in clinical practice; Bazett formula remains dominant despite its known limitations
• Normal ST elevation limits primarily derived from White populations; data for other racial groups more limited
• QT thresholds established without uniform correction methodology — different studies used different correction formulas, making precise thresholds uncertain
• Part of a 6-document series; does not address acute ischemia/infarction in detail (covered in Part VI) or exercise stress testing
• Pediatric QT normal limits available only for Bazett's formula at time of publication

Key Concepts Mentioned

• concepts/Cardiac-Repolarization — primary subject; QT measurement standards and thresholds directly update this page
• concepts/ST-T-Changes — primary subject; ST thresholds, T-wave descriptors, U wave standards add measurement framework
• concepts/Cardiac-Action-Potential — physiological basis of ST/T wave generation discussed

Key Entities Mentioned

• entities/Long-QT-Syndrome — QT thresholds and pattern descriptions (LQT1/2/3 ST-T patterns) referenced
• entities/HCM — one of the three entities causing giant T-wave inversion

Wiki Pages Updated

• wiki/sources/ecg-sttu-aha-2009.md — created
• wiki/concepts/ST-T-Changes.md — ST elevation/depression thresholds, T-wave quantitative descriptors and normal polarity, U wave section added; source_count updated
• wiki/concepts/Cardiac-Repolarization.md — QT measurement standards, linear regression recommendation, normal thresholds, QT dispersion recommendation added; source_count updated
• wiki/sourceindex.md — new entry added
• wiki/wikiindex.md — ST-T-Changes and Cardiac-Repolarization descriptions updated