Clinical Overview

The U wave is the small, often overlooked sixth wave of the ECG, sitting on the baseline between the end of the T wave and the next P wave. A normal U wave is upright, shares the T wave’s polarity, and is visible in roughly half to three-quarters of all recordings, best seen in leads V2-V3 (Merck Manual Electrocardiography, 2026; LITFL U Wave, 2024). Its visibility is strongly rate-dependent: a normal U wave becomes more prominent as the heart rate slows, and is seen in around 90% of tracings once the rate falls to 65 bpm or below (Duque-González et al., Cardiovascular and Metabolic Science, 2021; LITFL U Wave, 2024). Because most people already carry a small, ordinary U wave, this dataset’s UW label cannot mean simply “a U wave is present” — this project’s own SNOMED cross-check found the code behind UW resolves to the “U wave abnormal” concept, so a strip carrying this label is annotated for something about the U wave that departs from normal, not for the wave’s mere existence. This page is scoped to that abnormal finding and covers its two distinct forms: an abnormally prominent (exaggerated, upright) U wave, and an inverted U wave.

A second, separate dataset entry, UWAb (“U Wave Abnormal”), has been treated as a merge candidate for this page rather than getting a page of its own — a reader may see UWAb referenced elsewhere on the site as a near-duplicate of this entry’s acronym. Its records, however, were deliberately not pulled into this page’s sample. The same SNOMED cross-check that confirmed UW’s code as correct found UWAb’s own code resolves instead to sinus arrest — an unrelated rhythm concept — and a review of how the two labels co-occur in the live dataset found essentially no overlap between them, which does not support treating UWAb’s records as though they follow the U-wave label rather than the sinus-arrest code they are actually mapped to. [CLINICAL REVIEW NEEDED: there is no documentation of what the original annotators intended UWAb to capture, so this page’s record sample is drawn from the UW pool only, and UWAb’s records are deliberately excluded rather than assumed to belong here.]

The mechanisms behind the two forms are different, which is why they are covered as separate findings rather than one. Genesis theories for the ordinary U wave include delayed repolarization of the Purkinje fibers or papillary muscles, prolonged repolarization of mid-myocardial M cells at slow rates, and a mechanoelectrical theory in which ventricular wall distension during rapid diastolic filling itself generates the deflection — no single theory is settled, and its origin “remains unknown and is subject to debate” as of a 2023 literature review (Kihlgren et al., Journal of Electrocardiology, 2023; Duque-González et al., 2021; Sattar and Chhabra, StatPearls Electrocardiogram, 2023). A prominent upright U wave is generally the ordinary wave simply exaggerated or unmasked — by a slow rate, by a drug, or by an electrolyte or repolarization disturbance that widens the gap between the T wave and the following P wave. An inverted U wave is a different, more specific problem: the leading pathophysiological explanation ties it to prolonged ventricular diastole, where sustained stretch of the ventricular wall delays the activation of stretch-sensitive channels and produces a delayed post-potential of reversed polarity (Duque-González et al., 2021).

Clinical significance also splits along that same line. A prominent U wave is frequently benign — the expected companion of a slow heart rate — but when it grows large enough to rival or exceed the T wave, it is a marker of severe hypokalemia and the electrical substrate for torsades de pointes, particularly once a flattening T wave fuses with the enlarging U wave (LITFL U Wave, 2024; LITFL Hypokalaemia, 2024). An inverted U wave carries a different weight: seen in a lead where the T wave itself remains upright, it is described as a highly specific marker of underlying heart disease, and in a chest-pain presentation as “a very specific sign of myocardial ischaemia” that can appear before any ST-segment or T-wave change and can point toward critical LAD or left main disease (LITFL U Wave, 2024; Duque-González et al., 2021). A 2023 literature review found that patients with negative U waves have a higher prevalence of hypertension, faster heart rates, more cardiac disease, and more left ventricular hypertrophy than patients with normal U waves, and that negative U waves were associated with increased all-cause mortality, cardiac death, and cardiac hospitalization in men; the same review notes that a T wave that is also negative alongside the U wave is especially linked to cardiac disease (Kihlgren et al., 2023). A 2023 JAMA Internal Medicine teaching case makes the same point at the bedside: a patient with intermittent exertional chest pain, normal initial troponin, and normal electrolytes showed an inverted U wave on the ECG captured during a symptomatic episode that was not present on the resting tracing — the U wave itself was the red flag (Shu, Liao, and Peng, JAMA Internal Medicine, 2023).

Neither form of U wave abnormality produces symptoms of its own; it is a tracing finding, not a sensation. Whatever the patient feels comes from the underlying process — palpitations, weakness, or cramping from the electrolyte disturbance behind a prominent U wave, or chest pain, dyspnea, and diaphoresis from the ischemia behind an inverted one. Many strips carrying this label come from an asymptomatic patient.

Causes of a prominent U wave include bradycardia (the most common benign cause), severe hypokalemia (the most common pathological cause), hypocalcemia, hypomagnesemia, hypothermia, raised intracranial pressure, left ventricular hypertrophy, and hypertrophic cardiomyopathy; recognized drug causes include digoxin, phenothiazines, and Class Ia and Class III antiarrhythmics such as quinidine (LITFL U Wave, 2024). Causes of an inverted U wave are concentrated in structural and ischemic heart disease: coronary artery disease with active or exercise-induced ischemia (classically in the left anterior descending or left main territory), hypertension, and valvular disease (LITFL U Wave, 2024; Duque-González et al., 2021).

Interpretation Guide

Key Features:

• Rate: not a defining feature of either form, but a direct input to the prominent form — the U wave becomes more visible as the rate slows, and is seen in around 90% of tracings at 65 bpm or below (Duque-González et al., 2021; LITFL U Wave, 2024). A prominent U wave that shrinks as the rate rises is behaving like the ordinary rate-dependent wave, not like a pathological finding
• Rhythm: not a defining feature. Both forms are read against whatever rhythm is present and say nothing about impulse origin
• P waves: not part of either U wave criterion
• PR interval: within normal limits unless a separate, coexisting abnormality is present
• QRS complex: within normal limits for an isolated U wave finding, though left ventricular hypertrophy is itself a recognized cause of a prominent U wave — check the voltage criteria before attributing prominence to another cause (LITFL U Wave, 2024)
• ST segment: worth reading alongside the U wave rather than separately. In hypokalemia, ST depression, T wave flattening, and a prominent U wave form a recognized triad (LITFL Hypokalaemia, 2024). In ischemia, an inverted U wave can appear before any ST-segment change develops, making it an earlier clue than the ST segment itself (LITFL U Wave, 2024)
• T waves: read together with the U wave, not independently. A normal U wave shares the T wave’s polarity and stays under roughly one-quarter of its height (LITFL U Wave, 2024), though the practical threshold that matters most is whether the U wave has grown clearly larger than the T wave, which is the point at which hypokalemia-associated risk rises (LITFL Hypokalaemia, 2024). An inverted U wave is only meaningful in a lead where the T wave itself is upright (LITFL U Wave, 2024). As hypokalemia worsens, the T wave flattens while the U wave grows, and once potassium falls below about 2.7 mmol/L the U wave becomes clearly prominent (LITFL Hypokalaemia, 2024); a 2025 cohort study found U waves on 80% of ECGs in patients with potassium under 2.5 mEq/L, versus 35.9% at moderate and 7.3% at mild hypokalemia (p<0.001) (Ramadurai and Varadarajan, Cureus, 2025)
• QT interval: the classic pitfall. When a flattened T wave fuses with an enlarging U wave, the interval that gets measured is really a QU interval, not a true QT, and reporting it as QT prolongation is a documented source of false alarm (LITFL Hypokalaemia, 2024). A 2025 case report demonstrated this directly: an apparent QT of 540 ms measured in the standard leads (II and V5, where the T and U waves stayed fused) turned out to be a true QT of 400 ms once measured in V2, where the two waves separated cleanly; a premature atrial complex helped confirm the diagnosis, because the U wave’s amplitude visibly dropped on the post-PAC beat while a true bifid T wave would not behave that way (Kerkouri et al., European Heart Journal Case Reports, 2025)
• Other findings: amplitude and direction relative to the T wave are the two things to record explicitly — whether the U wave is larger than roughly one-quarter of the T wave (prominent) or opposite in polarity to an upright T wave (inverted), since the two forms carry different differentials and different urgency (LITFL U Wave, 2024)

The single most important distinction on this page is direction, not just size. A large upright U wave in a bradycardic or hypokalemic patient is frequently benign; a U wave that inverts in a lead where the T wave stays upright is abnormal by definition and warrants an ischemia workup, even when the initial troponin and ST segments look unremarkable (LITFL U Wave, 2024; Shu, Liao, and Peng, 2023).

Key Leads

• Lead V2 – Along with V3, the lead where the U wave is normally best seen, and the lead a 2025 case report used to resolve a T-U fusion problem that made the QT appear falsely prolonged in the standard leads (LITFL U Wave, 2024; Kerkouri et al., 2025)
• Lead V3 – The other standard lead for U wave visualization, and where the amplitude comparison against the T wave is usually made (LITFL U Wave, 2024)
• Leads II and V5 – The leads standard guidance names for QT measurement, but also where T-U fusion is most likely to obscure the true QT endpoint; if the T and U waves will not separate here, switch to V2 rather than measure a fused complex (Kerkouri et al., 2025)
• Leads V1-V4 – Where an exercise-induced or ischemic inverted U wave is classically described, and where it has been associated with critical left anterior descending or left main coronary disease (LITFL U Wave, 2024; Duque-González et al., 2021)
• Precordial leads generally, over the limb leads – The U wave is a low-amplitude deflection that shows up more reliably in the precordial (semidirect) leads than in the limb leads, so a limb-lead-only tracing is a weaker basis for excluding either form (Merck Manual Electrocardiography, 2026)

Differential Diagnosis

• QT Interval Extension (QTIE) — the finding a prominent U wave most often mimics. Distinguishing clue: find the true end of the T wave rather than the end of the last visible deflection. When the T and U waves fuse in the standard measurement leads, switch to a lead where they separate (V2 is the leading choice), and use a premature atrial complex if one is available — the U wave’s amplitude drops on the post-PAC beat, while a true bifid T wave does not (Kerkouri et al., 2025).
• T Wave Change (TWC) — the finding that a bifid or notched T wave can be confused with. Distinguishing clue: a true second hump on the T wave itself behaves differently from a separate U wave under a change in cycle length — a genuine U wave’s amplitude is rate- and cycle-sensitive (shrinking after a PAC, growing as the rate slows), while a notched or bifid T wave’s morphology is comparatively fixed (Kerkouri et al., 2025).
• Sinus Bradycardia (SB) — the rate explanation for a prominent upright U wave. Distinguishing clue: the ordinary U wave becomes more visible simply because the rate is slow, reaching around 90% visibility at 65 bpm or below. A U wave that is proportionate to the T wave and tracks the rate (more visible when slow, less visible as the rate rises) is the expected rate effect, not a separate pathological finding (Duque-González et al., 2021; LITFL U Wave, 2024).
• Left Ventricle Hypertrophy (LVH) — a structural cause of a prominent U wave in its own right. Distinguishing clue: check the voltage criteria (Sokolow-Lyon, Cornell, or Romhilt-Estes) before attributing a prominent U wave to an electrolyte or drug cause — if LVH voltage criteria are already met, the hypertrophy itself is a recognized, sufficient explanation (LITFL U Wave, 2024).

Treatment Brief

The U wave itself is not treated. What a monitoring technician or nursing student owns here is telling the two forms apart, reporting the measurement defensibly, and recognizing which pattern is the one that cannot wait.

• Record both amplitude and direction, not just “U wave present.” A U wave larger than roughly one-quarter of the T wave is prominent; a U wave opposite in polarity to an upright T wave is inverted — and the two point in different clinical directions (LITFL U Wave, 2024).
• Before reporting a long QT, check whether the T and U waves have fused. If the standard leads (II, V5) show fusion, remeasure in V2, and use a premature atrial complex if one appears on the strip — a true U wave’s amplitude drops on the post-PAC beat (Kerkouri et al., 2025).
• For a prominent U wave, send electrolytes urgently, especially potassium. U wave prevalence rises sharply with hypokalemia severity — from roughly 7% at mild levels to 80% once potassium falls under 2.5 mEq/L — and the same patients are the ones at risk for torsades de pointes (Ramadurai and Varadarajan, 2025; LITFL Hypokalaemia, 2024).
• Review the medication list for digoxin and Class Ia or III antiarrhythmics (quinidine and related agents) when a prominent U wave has no obvious electrolyte or rate explanation (LITFL U Wave, 2024).
• Treat a new inverted U wave — especially one appearing with or during chest pain and absent on a prior resting tracing — as a possible ischemic red flag, even if the troponin and ST segments look unremarkable. It has been reported as the only visible ECG finding preceding classic ischemic changes (LITFL U Wave, 2024; Shu, Liao, and Peng, 2023).
• Compare against a prior ECG whenever one exists. A stable, longstanding prominent U wave at a stable slow rate is a very different report from a new inverted U wave or one that has grown disproportionately since the last tracing.
• Check for left ventricular hypertrophy before treating a prominent U wave as unexplained; voltage criteria alone can account for it (LITFL U Wave, 2024).