OMI ECG findings

T wave in V1 taller than V6 ➜ NTTV1

• A T wave in V1 taller than V6, especially when it's new (a New Upright T wave, or TV1>TV6) ➜ suspect CAD or acute ischemia. The mechanism is a mirror image of posterior (posterior/inferobasal) ischemia or injury, which is why it's tied to the LCx, not the LAD.
• How to judge it: compare the T waves in V1 and V6 on the same ECG, not against a memorized absolute height (in mm). If there's an old ECG, always pull it up and compare. The "N" in NTTV1 stands for New; without an old ECG all you have is TTV1, and the evidence is a notch weaker.
• NTTV1 isn't there to make a diagnosis; it's there to wake your eyes up. When you see it, go back and look for STD in V2–V3 and for taller R waves, dig out the old ECG, and get another ECG, rather than calling the CV man right away.

T wave too fat and too symmetric relative to the QRS ➜ HATW. Remember: it's "the R wave gets shorter," not "the T wave gets taller."

• A hyperacute T wave isn't "tall," it's "fat." The criterion is area under the curve relative to the QRS, plus symmetry.
• HATW isn't "the mountain got taller," it's "the mountain got fatter." The area gets bigger, not the height.

New pathologic Q waves (especially with STE or hyperacute T waves) ➜ Pathologic Q wave

• Pathologic Q wave ≠ old MI. Q waves may appear within the first hour of the acute phase, a Q wave doesn't exclude acute OMI, and abnormal Q waves on the initial ECG did not eliminate the effect of thrombolytic therapy in reducing infarct size.
• The logic of the criteria is asymmetric: meeting either of the two = pathologic (pathologic Q wave, or); both must be met to count as normal (septal q wave, and). To be called normal, the bar is actually higher.
• An isolated QS in V1 can be normal; but in an OMI context, any Q wave in V2–V3 is abnormal (unless it's lead misplacement, LBBB, or a paced rhythm). Also, Qr in V1 is a fairly specific ECG sign of PE: when you see Qr in V1, both lights should go on at once, anterior OMI and PE.

STE present but not reaching STEMI criteria ➜ Subtle STE

• Subtle STE = ST elevation that doesn't reach the STEMI criteria threshold. The sensitivity of STEMI criteria themselves isn't high: Hillinger 2019 (prospective, multicenter, software-automated measurement of the first ECG on arrival, compared against the final adjudicated STEMI diagnosis) and Martin 2007 (with ceMRI-diagnosed AMI as the gold standard) both show that relying on STEMI criteria alone misses a large group of MIs.
• Don't look only at absolute height (in mm); look at proportion: in a lead with small QRS voltage, an occluded artery may not be able to squeeze out even 1 mm. Look instead at ST-segment morphology (is it straightening or turning convex? normal early repol is upwardly concave), whether there's reciprocal change (often easier to spot than the STE itself), whether the T wave is getting fatter along with it, and compare with the old ECG and with a repeat ECG.
• For subtle STE to count, you need clinical context + morphology + teammates all present together; it's not about picking up the phone over any tiny bit of STE under 1 mm. That's acting like you just found a loaded gun.

Upside-down U wave ➜ Inverted U wave. Very rare, but every one you catch counts.

• Definition: in a lead with an upright T wave, a downward U wave after the T wave ends (Neg U under T(+)). A normal U wave goes in the same direction as the T wave and gets more prominent as the heart rate slows.
• The key to recognizing it: use the QT interval as calipers. Find a lead where the QT interval is clear cut and carry that QT length over to the suspicious lead: if it falls outside the QT interval, it's an inverted U wave; if it falls inside, it's the second half of a biphasic T wave.
• It can appear earlier than STE, and it can appear when there's no chest pain. That's what makes it as precious as HATW: an early signal.

New bifascicular block (RBBB + fascicular block, most commonly LAFB) + chest pain ➜ New-onset bifascicular block. It means the septal perforator has been cut off: an ominous sign of proximal LAD.

• Mechanism: the right bundle branch and the left anterior fascicle are supplied mainly by the LAD's septal perforators; the first septal perforator (S1) comes off the proximal LAD and supplies the basal septum, bundle branches included. Proximal LAD or left main occlusion ➜ no blood to the septal perforator ➜ new RBBB+LAFB.
• Catching "new" is the one key move: in Widimsky 2012, overall in-hospital mortality for RBBB and LBBB was similar; but new (or presumed new) RBBB 18.8% vs old RBBB 6.4%, a nearly threefold difference. When you see RBBB, the first thing isn't deciding left vs right; it's digging up the old ECG.
• You'll often see downsloping STE: it doesn't look like a typical STEMI, so it's the easiest thing to call "secondary change from the BBB" and turn the page. Pendell Meyers put it even more bluntly on Dr. Smith's ECG Blog: patients with acute OMI plus RBBB+LAFB are either about to arrest, just resuscitated, or in cardiogenic shock.

Maximal STD in V1–V4 ➜ STDmaxV1-4. Posterior OMI until proven otherwise. Get posterior leads.

• STDmaxV1-4: ischemic STD in the precordial leads (V1–V6), with the maximal STD in one of V1, V2, V3, or V4 (relative to V5 and V6). The point isn't "how deep," it's "which lead is deepest."
• The mechanism is vectors: posterior OMI is the mirror image of transmural STE on the opposite (posterior) wall, projecting as STD in V1–V4; the summed STD of diffuse subendocardial ischemia points toward the apex (V5, V6). When you see STDmaxV1-4, treat it as posterior OMI until proven otherwise.
• The precordial QRS/T waves have four bonus clues: horizontal (not upsloping) STD, a dominant R wave in V2 (R/S ratio > 1), a broad R wave (> 30 ms), and upright T waves in V1–V4; these are all what the posterior "mirror image" looks like flipped over. You can use Ken Grauer's "Mirror Test": flip V1–V3 upside down and see whether it turns into an anterior STEMI you'd recognize.

STE, and it's in aVR (+ multiple leads STD) ➜ this is that wastebasket. Everything is in there, but for a sick patient, call CV and talk it through together.

• aVR STE is itself a mirror: there's no ventricular myocardium under aVR (Smith's version), or it faces the basal septum (the Mattu/LITFL version). The two camps differ on the anatomy but reach the same conclusion: that STE isn't the artery under it being occluded; it's the mirror image of a bunch of STD elsewhere. You can think of aVR as (−)aVR: with STD in I and II, (−)aVR mathematically has to have STD, so (+)aVR has to have STE.
• The conclusion deserves to be said loudly: aVR STE + diffuse STD is not left main occlusion, and it's not OMI; it's diffuse subendocardial ischemia (supply/demand mismatch). Patients with a true 100% left main occlusion usually don't make it to you: the patient who walks into the ED, alert, holding this ECG is there precisely because their left main "isn't completely occluded yet."
• This pattern has high mortality, but the mechanism isn't a single occluded artery; it's "the whole heart globally short on oxygen." It may be ACS (left main insufficiency, all three vessels bad, proximal LAD), or not ACS at all (sepsis, severe anemia, GI bleeding, hypoxia, SVT/AF RVR, hypovolemia, aortic stenosis, hypertensive emergency). The first step isn't picking up the phone; it's finding and correcting the supply/demand cause. After correcting it, get another ECG; if the STD is still there and the patient is still in pain (or the chest pain is refractory, or the blood pressure is dropping), talk with CV.

Inferior or lateral STD/TWI, mirroring an STE somewhere ➜ Reciprocal STD / TWI. Reciprocal change is often more obvious than the STE itself.

• The STD in this finding isn't the answer itself; it's a witness for "a subtle STE in another lead": a subtle STE plus reciprocal STD/TWI in the opposite lead goes from "something's off" to "diagnostic."
• Remember two things first (the words on my own slide): reciprocal change may appear before the STE, and it may be more obvious than the STE. One mechanism is low voltage: in someone whose QRS is small to begin with, STE that's proportional to the QRS will be less obvious too, but the STD of reciprocal change isn't limited that way and may actually stand out more. So in these patients you often see the STD first and the STE after; that STD is pointing the way for you.
• Memorize the opposite-wall relationships: STD in V1–V4 → think posterior (see STDmaxV1-4); STD in I/aVL → think inferior (II/III/aVF); STD in II/III/aVF → think high lateral (I/aVL); STD in V5–V6 with STE in V1 → think septal (see Precordial Swirl). Lead III and aVL have a "magical" mirror-image relationship: proximal LAD occlusion almost always has reciprocal STD in the inferior leads; mid/distal LAD occlusion doesn't necessarily.

STE or a disproportionately upright T in V1 (±V2) + reciprocal STD/TWI in V5–V6 ➜ Precordial Swirl. Think occlusion of the proximal LAD, before the septal branch.

• The definition formally published in 2025 (Goss et al.): STE and/or hyperacute T waves in V1-V2, with reciprocal STD and/or TWI in V5-V6, forming a "clockwise swirl" across the precordial leads. Mechanism: if the LAD is occluded before the first septal perforator (S1), even the whole septum becomes transmurally ischemic; the septum sits right and anterior, so the injury vector projects onto V1 (and aVR) as STE, while the apex and lateral wall sit at the other end of the vector, so V5 and V6 get reciprocal STD.
• How to look: first V1. Does the ST-T "not look the way it should" (disproportionate STE/disproportionately upright T/coved shape)? The paper's quantitative version is a V2 T-wave amplitude ÷ S-wave amplitude > 0.40. Then V6: is there STD? Is that STD a "flat shelf" rather than the "downsloping" LVH strain kind? Is the T wave still upright (classic swirl is flat STD with an upright T)?
• Telling it apart from aVR STE + multiple leads STD (diffuse subendocardial ischemia) is just a matter of which is taller: aVR STE > V1 STE → think subendocardial ischemia/left main/3-vessel; V1 STE > aVR STE → think precordial swirl/proximal LAD occluded before S1. This is a reading trick, not a criterion with sens/spec numbers; use it as a clue, not as the only reason.

If V2 or V3 has no S wave and no J wave at the same time ➜ treat it as OMI outright, no calculation needed.

• TQRSD = in either V2 or V3, the S wave and the J wave are both gone. Note it's "or," not "and": if either V2 or V3 has no S wave and no J wave at the same time, it counts.
• TQRSD is a rule-in tool, not a rule-out tool. No TQRSD does not mean, in any way, that it isn't OMI.

Biphasic or deep symmetric TWI in V2–V4, and the patient is now pain-free ➜ Wellens' syndrome. Don't relax just because there's no pain.

• Wellens waves ≠ Wellens' syndrome: Wellens waves are just a T-wave shape (biphasic or deep symmetric TWI); Wellens' syndrome is the T-wave shape + the full clinical context. What's the difference? Whether the patient is in pain right now.
• The mechanism is a reperfusion T wave: the LAD was just occluded and has now opened on its own. That's why the criteria require pain-free, troponin not elevated, no Q wave, and preserved R wave.

J-point STD + upright, tall, symmetric T wave in V1–V6 ➜ de Winter's T wave. This is the LAD's anterior STEMI equivalent.

• de Winter = Mount Fuji, but the foot of the mountain sinks 1-3 mm first: upsloping ST depression at the J point runs straight into a tall, symmetric T wave. This is an acute, ongoing proximal LAD occlusion.
• de Winter and Wellens both show up in V2-V4, but one is "present progressive" (acute proximal LAD occlusion) and the other is "past perfect" (sub-acute occlusion: occluded before, open now).
• The 2022 ACC expert consensus has listed de Winter's sign as a STEMI equivalent.

STE in all three leads I, aVL, and V2 (the STE in V2 that "skips V1 and isn't contiguous" is the soul of it), reciprocal STD in lead III (often extending to II, aVF) ➜ South African Flag sign.

• STE in leads I, aVL, V2 + reciprocal STD in lead III (II and aVF often go along, but III is the most obvious). On paper, aVL and V2 sit in two different zones, so under the traditional grouping they aren't contiguous.
• Ken Grauer's added third criterion: no STE in any chest lead other than V2. This is the key to distinguishing it from an LAD occlusion that's still evolving, where the STE hasn't yet spread to chest leads beyond V2.
• Why it's easy to miss: aVL and V2 aren't contiguous in the traditional grouping, so the pattern a 1st or 2nd diagonal occlusion produces, "aVL + V2 STE, nothing in the other chest leads," may by definition fail to meet the two-contiguous-leads STEMI criteria.

STE only in lead III (not II/aVF) + STD in V4–V6, but no STD in V2 ➜ Aslanger's pattern. An inferior OMI layered on multivessel disease.

• Three criteria (Aslanger 2020 original paper): ① STE in lead III but not in the other inferior leads (II, aVF) ② STD in any of V4–V6 with a positive or terminally positive T wave, and no STD in V2 ③ ST in lead V1 higher than in V2. The mechanism: the ST vector of the inferior MI plus the ST vector of subendocardial ischemia from multivessel disease add up and point rightward; II and aVF get canceled out, leaving STE visible only in lead III ("two fans blowing at each other, and II and aVF in the middle get canceled out").
• Ken Grauer's add-on: the only leads with obvious STE may be just III, aVR, V1; reciprocal STD in lead I may be more obvious than in aVL (because the ST vector points rightward).
• The two incidence figures have different denominators; don't mix them up: 6.3% of the NSTEMI population; the paper also says 13.3% of inferior MIs may present with this pattern. In-hospital and one-year mortality in this group are similar to inferior STEMI (no statistically significant difference), yet they may miss out on emergent revascularization because they get labeled NSTEMI: that's the real clinical significance of Aslanger's pattern.

So far there's only one single-center, retrospective study on this pattern; for criteria and level of evidence, go straight to Part 3 (B). I'm not giving my own definition here, so you don't end up memorizing the wrong thing.

• Northern OMI means the ST vector points straight up (north): STE in aVR and aVL together, with reciprocal STD inferiorly. Dr. Smith's ECG Blog puts it this way: it's a STEMI pattern associated with subepicardial (transmural) ischemia of the heart's base (the very top of the heart, just below the atria, possibly especially the base of the septum). In the same post, Amandeep Singh, who contributed the case, wrote that it may come from left main occlusion, proximal LAD occlusion, or multivessel coronary artery disease.
• ⚠️ The evidence is thin, and we have to be honest about it: Dr. Smith's ECG Blog itself calls it "a vaguely defined pattern." The only original study we found is Ni 2026, a small retrospective case analysis (the authors are all from the same hospital), and the authors also say the patients they enrolled did not fully meet the current Northern OMI diagnostic criteria, which is why they propose a modified version. Sources differ: Dr. Smith's ECG Blog (2025-12-18) lists aVR, aVL STE (often with negative T-waves) + inferior (II, III, aVF) and lateral (V4–V6) STD (T wave positive or biphasic); Ni 2026's modified version only says aVR, aVL STE + reciprocal STD in at least II, III, aVF. Recognize it, but don't rely on it.
• Not the same as aVR STE + multiple leads STD: when Smith compiled cases of complete left main occlusion in 2019 (which he himself called anecdotal cases), the Northern OMI pattern was among them, and he states plainly, "LM occlusion does not present with isolated STE in aVR." So when you see STE in aVR, the first thing to do is look at aVL: if aVL is elevated too and there's inferior reciprocal STD, think Northern OMI.

In II/III/aVF or I/aVL, a terminal QRS notch/deflection and a wider QRS in that lead ➜ N wave sign. Full criteria in Part 3 (B).

• N wave sign = an extra notch or deflection at the end of the QRS. The original Niu 2013 study defines it with four criteria: ① a notch/deflection at the terminal QRS ② a height cutoff of 2 mm (referenced to the PR segment) ③ continuous change of the notch within 24 hours (even disappearing or turning into an S wave) ④ prolonged QRS duration in these leads. The mechanism is a delayed activation wave from the left ventricular basal region (LCx territory): "the whole class hands in their exams together, but one student writes too slowly and hands it in last."
• To tell it apart from a benign J wave (early repolarization), don't just look at which lead it's in: lead location can only tell you "if it is an N wave, where the culprit might be," not "whether it is an N wave." Look at what the criteria themselves require: ① height (benign ones often aren't this big) ② dynamics (an N wave changes within 24 hours; benign early repol is static, the old ECGs looked like this too) ③ whether the QRS is widened (an N wave is delayed activation and drags out the QRS); then add the clinical context (ACS symptoms or not) and comparison with old ECGs.
• Picking up the LCx: N wave in II, III, aVF, sensitivity 77%, specificity 89%. ⚠️ The target condition for these numbers is whether the culprit is the LCx, in patients with confirmed NSTEMI who underwent coronary angiography, not "picking out OMI among chest pain patients," so they can't be lined up side by side with the numbers for the earlier findings. A study from another group (Rostoff 2020) came out more conservative: among patients with an LCx/OM culprit, the most common finding was T-wave precordial instability, not the N wave; an N wave in aVL was an independent predictor of an LCx/OM culprit, but patients with an LAD/diagonal culprit also showed inferior N waves. The N wave is a clue pointing to the LCx, not a high-sensitivity screening tool.

STE, with well-formed Q waves / QS waves sitting in front of it in V1–V4 ➜ T/QRS ratio. If the single-lead and sum methods disagree ➜ ED echo.

• Prerequisite: well-formed Q waves, or at least one QS wave, sitting in front of the STE in V1–V4. First confirm you're in LVA (LV aneurysm) morphology territory, then use this rule to sort "old scar" from "burning right now." For the exact definition of a QS wave, see Smith's original blog post. This rule can't be used to separate out early repol; that's the 4-variable formula's job, and plugging into the wrong socket gives the wrong answer.
• Mnemonic: the T wave is the fire, the QRS is the wall. In acute MI the T wave is large relative to the QRS; in old MI/LV aneurysm the T wave is small: see whether the fire has burned over the top of the wall.
• How to calculate: measure T-wave amplitude and QRS amplitude in each of V1–V4. Compute T÷QRS once for each lead and take the largest; then sum the four T's and the four QRS's and divide, for one sum ratio. Each ratio goes with its own cutoff; don't mix them.

STE, no Q waves, and V2–V4 looks like early repol ➜ check for TQRSD first; no TQRSD and you've also passed the other gates ➜ 4-variable formula. Remember: the formula can only push suspicion up.

• Prerequisites (you must pass the gate first; if you don't, don't calculate): ≥1 mm STE in at least one lead of V2–V4, and none of the following: LVH, Q waves/QS in V2–V4, inferior or precordial STD, TQRSD, precordial TWI, straight or convex ST, STE >5 mm. Failing the gate sends you different places: TQRSD or STD ➜ already obvious OMI, Call CV directly; Q waves/QS ➜ switch to T/QRS ratio; LVH ➜ don't calculate, look at the baseline ECG and get an echo; precordial TWI ➜ think Wellens; straight/convex ST or STE >5 mm ➜ already beyond subtle, no need to calculate.
• How the two formulas relate: the 3-variable formula (Smith 2012) uses three variables, STE60V3, QTc, and RAV4; the 4-variable formula (Driver 2017) adds QRSV2 (the total QRS amplitude in V2) on top of those three. The derivation study and the Turkish external validation both show the 4-variable beats the 3-variable, and MDCalc has already replaced the old 3-variable version with the 4-variable one. Don't change a single coefficient.
• The single most important line: this formula can only "escalate suspicion," not "cancel suspicion." A result below the cutoff doesn't mean you can comfortably send the patient home. If it already looked like an LAD occlusion to you, a low formula result shouldn't overturn that.

Wide QRS (LBBB / ventricular paced) ➜ go straight to MSC. Scan the three criteria lead by lead; if paced, also scan V4–V6 for concordant STD.

• In the world of wide QRS, MSC is nearly the only road with literature behind it: in LBBB or ventricular paced patients, STEMI criteria, the 4-variable formula (its premise is distinguishing early repol from LAD occlusion), the HATW score (its derivation study itself excluded patients with a wider QRS), and Precordial Swirl (the first item of its definition is narrow QRS without LVH) are all unusable. Four out, one left: only MSC is backed by three studies, derivation, LBBB validation, and paced validation.
• Rule of Appropriate Discordance (the seesaw): in LBBB/paced rhythm the ventricles are depolarized through an abnormal pathway, so the ST-T is supposed to point opposite the main QRS direction; that's normal. When you see STE in LBBB, your first reaction shouldn't be "STE ➜ STEMI." Ask first: is this STE "supposed to tip up," or has it "tipped up too far"?
• What Smith 2012 did, in one line: replaced the absolute 5 mm in the original Sgarbossa third criterion with 25% of the preceding S-wave depth. Seesaws come big and small, so an absolute value isn't fair: LBBB patients without coronary occlusion may also have ≥5 mm discordant STE; a proportion is fair.