QT interval corrected for heart rate — five formulas.
Applies to / doesn’t apply to2
Applies to: Any patient with a measurable QT interval and heart rate on ECG.
- Not a diagnostic test for congenital long-QT or short-QT syndrome. Those criteria are a different question from rate correction and are not applied here.
- Bazett specifically is unreliable outside 60–100 bpm — it over-corrects above 100 and under-corrects below 60.
Fridericia (QTcF) is the default here — it is the ICH E14 convention for drug-safety QT monitoring and was the most accurate across heart rates in a systematic review. Bazett is what most automated ECG readouts print, but it over-corrects above 100 bpm and under-corrects below 60 — the 2013 HRS/EHRA/APHRS consensus advises against relying on it at rate extremes. Bazett is shown alongside the chosen result whenever the two diverge materially.
Measured on the ECG in milliseconds, from the start of the QRS to the end of the T wave — not including a U wave. Use the tangent method.
Every formula is derived from this. RR interval is taken as 60 ÷ heart rate.
A prior QTc for the same patient, in milliseconds. When given, the change from baseline (ΔQTc) is computed — a rise of more than 60 ms during a QT-prolonging therapy is the drug-safety threshold that warrants attention.
What this score estimates
Corrects the measured QT interval for heart rate, using one of five published formulas. The corrected value is reported in milliseconds.
Pearls & pitfalls5
- Formula choice changes the result on every ECG. Bazett over-corrects above 100 bpm and under-corrects below 60; Fridericia was the most accurate in a systematic review of ten studies, though that review covered athletes and people aged 14–35 rather than a general population (Mahendran 2023); Framingham is not reliably better than Bazett at extremes; Hodges has the weakest evidence base. ACEP has recommended Rautaharju at a 477 ms cut-off, so the authorities do not agree on a default.
- No risk bands are shown. The commonly quoted 440 ms threshold is exceeded by 10–20% of the healthy population, and the authorities that define thresholds do not agree with each other.
- Congenital long-QT and short-QT diagnostic criteria are a different question from rate correction and are not applied here. For reference only: the 2013 HRS/EHRA/APHRS consensus sets long-QT at 480 ms or above on repeated ECG, or 460–479 ms with unexplained syncope.
- Measure the QT without the U wave. Including it is the commonest measurement error; use the tangent method and, if a U wave is present, re-measure and consider why it is there.
- Rautaharju’s correction factor was not empirically derived from its near-all-male derivation cohort but borrowed from earlier estimates.
What the guidelines say1
US scientific statements use QTc to monitor for drug-induced QT prolongation; QTc >500 ms and a rise >60 ms from baseline are the values they cite as warranting attention.
Guideline positions are stated as published. Curie.MD Calc does not publish treatment direction.
Where this comes from8
- Bazett HC. An analysis of the time-relations of electrocardiograms. Heart. 1920;7:353-70. Predates online indexing — no PubMed record
- Fridericia LS. Die Systolendauer im Elektrokardiogramm bei normalen Menschen und bei Herzkranken. Acta Med Scand. 1920;53:469-86. English translation: Ann Noninvasive Electrocardiol. 2003;8(4):343-51. View source ↗
- Hodges M, Salerno D, Erlien D. Bazett’s QT correction reviewed. J Am Coll Cardiol. 1983;1:694. Conference abstract — not indexed
- Sagie A, Larson MG, Goldberg RJ, et al. Am J Cardiol. 1992;70(7):797-801. View source ↗
- Rautaharju PM, Mason JW, Akiyama T. Int J Cardiol. 2014;174(3):535-40. View source ↗
- Mahendran S, Vijayakumar S, Chan JS, et al. Clin Cardiol. 2023;46(9):1106-15. View source ↗
- Priori SG, Wilde AA, Horie M, et al. Heart Rhythm. 2013;10(12):1932-63. View source ↗
- Drew BJ, Ackerman MJ, Funk M, et al. Circulation. 2010;121(8):1047-60. View source ↗