Common variants move the heart's recovery time by a millisecond or two — nothing on their own. They start to matter when a medicine pushes in the same direction, which several hundred medicines do.
Solid lines are connections this site curates. Dashed lines mean the two ends share a research paper — worth knowing, and not a claim that one explains the other.
The QT interval is the time it takes the lower chambers of the heart to reset between beats. It is measured on an ECG, in milliseconds, and it varies between healthy people for ordinary reasons — sex, age, heart rate, potassium levels. Longer is not automatically dangerous. Much longer, or longer in the wrong circumstances, can be.
This page is worth reading carefully, because the same gene names appear here in two roles that get confused constantly.
Rare coding changes in KCNH2 and KCNJ2 cause congenital long QT syndrome — an inherited disease diagnosed with an ECG and a family history, which carries a real risk of dangerous rhythms and is managed by cardiologists. That is not what this page is about, and no variant listed here indicates it.
Common variants in and around those same genes — the four below — shift the QT interval by roughly one to three milliseconds each. On a measurement that normally runs around 400 milliseconds, that is small enough to be invisible in any individual. Nobody has ever been diagnosed with anything because of one.
Several hundred medicines prolong the QT interval as a side effect — some antiarrhythmics, some antipsychotics, some antibiotics, some antiemetics. In most people that is absorbed without consequence, because the heart has what cardiologists call repolarisation reserve: spare capacity to reset on time even when one mechanism is impaired.
Common variants like these appear to eat into that reserve. A risk score built from common QT variants was associated with drug-induced QT prolongation and with torsade de pointes, the dangerous rhythm that long QT can produce (Strauss et al., Circulation 2017). So the finding is not "this variant is bad" but something more precise and more useful: a millisecond that means nothing by itself can mean something when a drug is pushing in the same direction.
What follows from that in practice is nothing you should do alone. It is a reason the QT-prolonging medicines page exists, and a reason that potassium, other medicines and kidney function matter more to your prescriber than any of this.
Nothing, on its own. No guideline uses common QT variants to choose or avoid a drug, and none of these results is a reason to stop or refuse a prescription. If you have fainted unexplainably, if sudden cardiac death runs in your family, or if you have been told your ECG shows a long QT, those are reasons to see a doctor — and they are reasons regardless of what any variant here says.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 55 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
Young WJ et al. 2022, Nature communications rs78078655 (DNAJB5-DT), rs196067 (near KCNJ4), rs73068914 (near ZNF579), rs74552897 (LINC01505), rs34411702 (KIAA0232), rs10490255 (SLC8A1), rs35717017 (near LINC01960), rs17774009 (POPDC1), rs10512436 (near RNU6-1134P), rs2427312 (CABLES2), rs652677 (RSKR), rs61884689 (MPPED2) and 32 more — PMID:36050321
van Duijvenboden S et al. 2021, Human molecular genetics rs1805120 (KCNH2), rs9933198 (near ATP2A1-AS1), rs2832274 (LINC00189), rs1009674 (YPEL2), rs2793409 (near PREP) — PMID:34274964
Hoffmann TJ et al. 2022, Genetics rs7865239 (AOPEP), rs8133749 (BACH1), rs78598566 (CAVIN2-AS1) — PMID:36271874
Bihlmeyer NA et al. 2018, Circulation. Genomic and precision medicine rs12053903 (SCN5A), rs1361754 (PM20D1) — PMID:29874175
Kim JW et al. 2012, American journal of human genetics rs13017846 (SLC8A1) — PMID:22726844
2026-03-01 · Genome-wide association study-identified novel genetic loci for QT interval in Taiwan: implications for population-specific arrhythmia risk. Europace. 2026. DOI:10.1093/europace/euag055
A prolonged QT interval is a well-established marker of malignant ventricular arrhythmia and sudden cardiac death risk, but most QT genetics research has been done in European cohorts, limiting how well it generalizes to East Asian populations. This first Taiwanese QT-interval GWAS, using 13,020 Taiwan Biobank participants with replication in 2,011 people from the NTU-AF Registry, found 11 significant loci total. Four were already-known QT genes (KCNQ1, KCNH2, KCNE1, NOS1AP), confirming the study's validity, while two were genuinely novel and replicated: rs55797717 near CD46/MIR29B2CHG at 1q32.2, and rs76715387 in GJA1 at 6q22.31. Colocalization analysis linked these to specific tissue expression: the CD46/MIR29B2CHG signal to gene expression in the atrial appendage and left ventricle, and the GJA1 signal to expression in the tibial artery. GJA1 encodes connexin-43, a protein directly involved in electrical coupling between heart cells -- a biologically direct candidate mechanism for arrhythmia risk, distinct from CD46's more likely inflammation-related role. This site's QT interval and drug-induced long QT page carries 53 variants; CD46, MIR29B2CHG and GJA1 are not currently among them.
The measurement. The QT interval spans ventricular depolarisation and repolarisation on the surface ECG and is rate-dependent, so it is reported corrected (QTc), most often by Bazett or Fridericia. Normal ranges differ by sex; thresholds commonly cited for concern are above roughly 450 ms in men and 460 ms in women, with risk rising materially above 500 ms. Correction formula, lead selection and measurement technique all introduce variation, which is why a single automated reading is not treated as definitive.
Common versus rare variation at the same loci. Loss-of-function coding variants in KCNH2 (LQT2) and KCNJ2 (Andersen-Tawil syndrome, sometimes classified LQT7) cause congenital long QT syndrome, with substantially prolonged QTc and a documented arrhythmic risk. The variants on this page are common non-coding or synonymous positions at or near those loci with per-allele effects of 1-3 ms. They are not diagnostic of, nor screening tests for, the congenital syndromes, and a page describing them should say so explicitly because the gene names invite the confusion.
Effect sizes. From the GWAS Catalog: rs12029454 (NOS1AP) 2.98 ms per allele, p = 3 x 10-45, risk allele frequency 0.15; rs10919071 (ATP1B1) 1.37 ms, p = 3 x 10-30, frequency 0.115, with a separate report of 2.05 ms (p = 1 x 10-15) for the alternate allele at frequency 0.87; rs2968863 (KCNH2) 1.35 ms, p = 2 x 10-15, frequency 0.29; rs17779747 (KCNJ2) 1.08 ms, p = 3 x 10-37, frequency 0.304. NOS1AP is the most consistently replicated common QT locus and has additionally been associated with drug-induced QT prolongation and with ventricular arrhythmia in separate work.
Repolarisation reserve and drug-induced long QT. Drug-induced QT prolongation is understood as unmasking of reduced repolarisation reserve: individuals with subclinical impairment tolerate normal conditions and decompensate under pharmacological challenge. Strauss et al. (Circulation 2017) reported that a risk score constructed from common QT-associated variants was associated with both drug-induced QT prolongation and torsade de pointes risk. This is population-level evidence for a mechanism, not a validated clinical test; no professional society recommends genotyping common QT variants before prescribing, and the established risk factors — hypokalaemia, hypomagnesaemia, bradycardia, renal impairment, female sex, concomitant QT-prolonging drugs and drug interactions affecting clearance — remain the ones clinicians act on.
Clinical status. None of these variants has diagnostic, screening or prescribing use. CredibleMeds maintains the reference classification of QT-prolonging medicines; that list, and the patient's electrolytes and other drugs, are what prescribing decisions rest on.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about QT Interval and Drug-Induced Long QT comes down to these specific, well-studied positions — not a diagnosis. 113 positions are linked to this page; the ones this page's own text discusses are shown first.
The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 53 of 113 linked studies with a resolved discovery ancestry.
Research-derived gene–drug associations only — not a prescription, dosing guide, or medical advice. Always follow your prescriber's guidance.
| Gene | Drug | What the research shows |
|---|---|---|
| NOS1AP | QT-prolonging medicines (certain antiarrhythmics, antipsychotics, antibiotics and antiemetics) | A mechanism with population evidence behind it and no clinical test in front of it. Several hundred medicines prolong the QT interval, and most people absorb that without consequence because the heart has spare capacity to repolarise on time. Common QT variants appear to reduce that spare capacity: a risk score built from them was associated with drug-induced QT prolongation and with torsade de pointes (Strauss et al., Circulation 2017). No society recommends genotyping before prescribing, and the factors clinicians actually act on — potassium and magnesium levels, heart rate, kidney function, other QT-prolonging drugs — matter more. Useful as an explanation of why the same drug is uneventful in most people and not in everyone. (Strauss DG et al., Common Genetic Variant Risk Score Is Associated With Drug-Induced QT Prolongation and Torsade de Pointes Risk. Circulation 2017; CredibleMeds QT drug lists) |
Databases, guidelines and references
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
QT Interval and Drug-Induced Long QT. MyGeneLog™. https://www.mygenelog.com/conditions/qt-interval-and-drug-induced-long-qt
No, and the gene names make this confusing, so it is worth being blunt. Congenital long QT syndrome is caused by rare coding changes and is diagnosed with an ECG and a family history. The variants here are common positions near the same genes that move the measurement by one to three milliseconds out of about four hundred. They are not a screening test for the syndrome and cannot rule it in or out.
Because it stops being nothing in company. Several hundred medicines prolong the QT interval, and the heart tolerates that by having spare capacity to reset on time. Common variants appear to reduce that spare capacity, which is why a score built from them tracks who gets into trouble on a QT-prolonging drug and who does not.
No. No guideline uses these variants to choose or avoid a medicine, and stopping a prescription on the basis of one would be a real risk taken to avoid a theoretical one. What your prescriber weighs — your potassium and magnesium, your heart rate, your kidney function, your other medicines — matters more and is measurable today.
If you have fainted without explanation, if sudden cardiac death has happened in your family, or if someone has told you your ECG shows a long QT. Each of those is a reason on its own, whatever any variant says.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog™. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.