. 2022 Nov 7;28(1):e13015. doi: 10.1111/anec.13015

TABLE 3.

Summary of results for acquired long QT syndrome.

Citation	Medication(s) and drug monitoring	ECG recording, QTc correction and analysis	Outcome measure	Results
Sugrue, Noseworthy, et al. (2017)	N/A	12 lead ECG Bazett formula Automated QT‐alert system (QTc ≥500 ms) Cardiologist review	Novel proprietary TW program: TW left/right slope, TW amplitude, TW area, TW COG x/y coordinates, COG first 25% TW, COG last 25% of the TW, Tpeak–Tend interval Diagnostic accuracy: Sn, Sp, NPV, PPV Sensitivity analysis (age, sex)	aLQTS TWM biomarkers: Lead V₅ Shallower TW right slope (−2322 vs. −3593 mV/s; p < .001) Greater T‐peak‐Tend (109 vs. 92 ms; p < .001) Smaller TW COG (290 ms vs. 310 ms; p < .001) cLQTS vs. aLQTS: Lead V₅ distinguishes cLQTS from aLQTS in 77% cases Sn 90%, Sp 58%, PPV 83%, NPV 71% Sensitivity analysis: Age 78.3% No sex difference (M: 76%, F: 78%)
Sugrue et al. (2015)	Sotalol Dofetilide	12 lead ECG Bazett formula Automated TW analysis software (MATLAB)	Novel proprietary TW program QT interval and QTc Pre and post TdP ECG Risk prediction: Sn, Sp, NPV, PPV	Discrimination accuracy Lead V₃: QTc discrimination (p < .001, r = .72) Drug: 480 ms Control: 420 ms Lead I: TW right slope (p = .002, r = .45) Torsadogenic risk prediction: ECG biomarkers TWRS: 88% QTc alone: 79% Risk prediction accuracy Sn: 92.1% (combined), 88.1% (QTc), 79.7% (TWRS) Sp: 81.4% (combined), 72% (QTc), 46% (TWRS) PPV: 90.3% (combined), 85% (QTc), 58.9% (TWRS) NPV: 84.6% (combined), 76.9% (QTc), 70% (TWRS)
Vicente et al. (2015)	Dofetilide (500 mcg) Quinidine (400 mg) Ranolazine (1500 mg) Verapamil (120 mg) 24 hr treatment period 7 day washout period Plasma drug level paired with ECG	12 lead ECG (THEW): Baseline 15 time‐points postdose (0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7, 8, 12, 14, 24 hr) Fridericia formula QTGuard+ automated TWM and vectorcardiographic biomarker analysis	ECG biomarkers: QT interval, QRS, J‐Tpeak interval, Tpeak–Tend interval TWM architectural patterns: Flatness, asymmetry, notching Vectorcardiographic biomarkers: ERD_30%, LRD_30%, QRS‐T angle, ventricular gradient, maximum magnitude of T vector, TCRT Patch clamp experiments: Ion channel block (hERG, late sodium, calcium)	Ion channel block: Dofetilide: 55% hERG Quinidine: 71% hERG, 8% calcium, 3% sodium Ranolazine: 26% hERG, 21% sodium Verapamil: 17% calcium, 7% hERG ΔQTc & TWM metrics: Dofetilide: ΔQTc 73.6 ms, flatness 0.16, asymmetry 0.25, notching 55% (p < .001) Quinidine: ΔQTc 78.9 ms, flatness 0.21, asymmetry 0.34 (p < .01), notching 69.7% (p < .001) Ranolazine: ΔQTc 12 ms, flatness 0.06 (p < .01), asymmetry 0.10, notching 1.4% (p < .001) Vectorcardiographic biomarkers: ERD_30%, LRD_30% increase for dofetilide, quinidine, ranolazine QRS‐T angle decrease for dofetilide, quinidine TCRT decrease for dofetilide
Johannessen et al. (2014)	Dofetilide (500 mcg) Quinidine (400 mg) Ranolazine (1500 mg) Verapamil (120 mg) 24‐hr treatment period 7 ‐day washout period Plasma drug level paired with ECG	12 lead ECG (THEW) Fridericia formula QTGuard+	ECG biomarkers (lead II): P‐onset, PR, QRS, QRS‐onset, QRS‐offset, QT interval, J‐Tpeak interval, Tpeak–Tend interval Concentration‐dependent analysis	QTc & TWM metric prolongation: Dofetilide: QTc 79.3 ms, J‐Tpeak 39.5 ms, Tpeak–Tend 40 ms, CDA; (p < .001) Quinidine: QTc 78.1 ms, J‐Tpeak 29.1 ms, Tpeak–Tend 49.8 ms, CDA; (p < .001) Ranolazine: QTc 12.6 ms (p < .001), Tpeak–Tend 8.8 ms (p < .013), CDA (p < .01)
Couderc et al. (2011)	Moxifloxacin ECG (2 hr post)	12 lead ECG Bazett and Fridericia formulae ECGScan software for ECG digitization Automated ECG interval assessment with COMPAS software	Cardiac ventricular repolarization: Lead II: QT apex, QT, Tpeak‐Tend, TW amplitude, left and right TW slope Eigen lead: T‐loop (roundness) Vectorcardiographic biomarkers: ERD, LRD (30–50% threshold) Cardiac events	Moxifloxacin model: ERD positive with drug (p = .0001) LQT2 model: Left slope detected KCNH2 mutation (p = .0002) Cardiac event prediction: T‐roundness (p = .007) Equivalent to QTc
Graff et al. (2010)	Trial 1 Placebo: 7 days Moxifloxacin Days 1–6: placebo Day 7: 400 mg Trial 2 Sotalol Day 1: no drug Day 2: 160 mg Day 3: 320 mg	Trial 1 12 lead ECG postdose Trial 2 Digital Holter ECG Fridericia formula MUSE/Interval Editor software	MCS = asymmetry + notch + (1.6 x flatness) QTc	Sotalol MCS > QTc (z score 2.5, CI 1.26 to 3.66) 320 mg equivalent to 160 mg (p = .25) 160 mg: MCS 7x greater, QTc 4x greater (vs. moxifloxacin) 320 mg: MCS 15x greater, QTc 6x greater (vs. moxifloxacin) Covariate effects on QTc and TWM: Parabolic regression>linear regression for sotalol (p < .01) No difference for moxifloxacin (p = .45) ΔMCS: Sotalol 320 mg > 160 mg (pooled fit, p < .01) Sotalol 160 mg > moxifloxacin and placebo Moxifloxacin sex differences QTc and TWMΔ: F > M
Graff et al. (2009)	Sotalol Day 1: 0 mg Day 2: 160 mg Day 3: 320 mg Plasma drug level	12 lead ECG Control LQT2 12 lead Holter ECG Sotalol Fridericia formula 12SL algorithm	MCS = asymmetry + notch + (1.6 x flatness) QTc Test accuracy (AUC)	Sotalol: MCS > QTc (p < .001) 160 mg (AUC 84% vs. 72%) 320 mg (AUC 94% vs. 87%) Effect size: MCS > QTc (50%) at maximal AUC and ECGΔ (p < .001) MCS > QTc (3‐fold) at peak ECGΔ (p < .001) MCS discrimination accuracy (normal QTc) Similar for sotalol and LQT2 (p = .9)

Citation

Medication(s) and drug monitoring

ECG recording, QTc correction and analysis

Outcome measure

Results

Sugrue, Noseworthy, et al. (2017)

N/A

12 lead ECG

Bazett formula

Automated QT‐alert system (QTc ≥500 ms)

Cardiologist review

Novel proprietary TW program:

TW left/right slope, TW amplitude, TW area, TW COG x/y coordinates, COG first 25% TW, COG last 25% of the TW, Tpeak–Tend interval

Diagnostic accuracy:

Sn, Sp, NPV, PPV

Sensitivity analysis (age, sex)

aLQTS TWM biomarkers:

Lead V₅

Shallower TW right slope (−2322 vs. −3593 mV/s; p < .001)

Greater T‐peak‐Tend (109 vs. 92 ms; p < .001)

Smaller TW COG (290 ms vs. 310 ms; p < .001)

cLQTS vs. aLQTS:

Lead V₅ distinguishes cLQTS from aLQTS in 77% cases

Sn 90%, Sp 58%, PPV 83%, NPV 71%

Sensitivity analysis:

Age 78.3%

No sex difference (M: 76%, F: 78%)

Sugrue et al. (2015)

Sotalol

Dofetilide

12 lead ECG

Bazett formula

Automated TW analysis software (MATLAB)

Novel proprietary TW program

QT interval and QTc

Pre and post TdP ECG

Risk prediction:

Sn, Sp, NPV, PPV

Discrimination accuracy

Lead V₃:

QTc discrimination (p < .001, r = .72)

Drug: 480 ms

Control: 420 ms

Lead I:

TW right slope (p = .002, r = .45)

Torsadogenic risk prediction:

ECG biomarkers

TWRS: 88%

QTc alone: 79%

Risk prediction accuracy

Sn: 92.1% (combined), 88.1% (QTc), 79.7% (TWRS)

Sp: 81.4% (combined), 72% (QTc), 46% (TWRS)

PPV: 90.3% (combined), 85% (QTc), 58.9% (TWRS)

NPV: 84.6% (combined), 76.9% (QTc), 70% (TWRS)

Vicente et al. (2015)

Dofetilide (500 mcg)

Quinidine (400 mg)

Ranolazine (1500 mg)

Verapamil (120 mg)

24 hr treatment period

7 day washout period

Plasma drug level paired with ECG

12 lead ECG (THEW):

Baseline

15 time‐points postdose (0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 5, 6, 7, 8, 12, 14, 24 hr)

Fridericia formula

QTGuard+ automated

TWM and vectorcardiographic biomarker analysis

ECG biomarkers:

QT interval, QRS, J‐Tpeak interval, Tpeak–Tend interval

TWM architectural patterns:

Flatness, asymmetry, notching

Vectorcardiographic biomarkers:

ERD_30%, LRD_30%, QRS‐T angle, ventricular gradient, maximum magnitude of T vector, TCRT

Patch clamp experiments:

Ion channel block (hERG, late sodium, calcium)

Ion channel block:

Dofetilide: 55% hERG

Quinidine: 71% hERG, 8% calcium, 3% sodium

Ranolazine: 26% hERG, 21% sodium

Verapamil: 17% calcium, 7% hERG

ΔQTc & TWM metrics:

Dofetilide: ΔQTc 73.6 ms, flatness 0.16, asymmetry 0.25, notching 55% (p < .001)

Quinidine: ΔQTc 78.9 ms, flatness 0.21, asymmetry 0.34 (p < .01), notching 69.7% (p < .001)

Ranolazine: ΔQTc 12 ms, flatness 0.06 (p < .01), asymmetry 0.10, notching 1.4% (p < .001)

Vectorcardiographic biomarkers:

ERD_30%, LRD_30% increase for dofetilide, quinidine, ranolazine

QRS‐T angle decrease for dofetilide, quinidine

TCRT decrease for dofetilide

Johannessen et al. (2014)

Dofetilide (500 mcg)

Quinidine (400 mg)

Ranolazine (1500 mg)

Verapamil (120 mg)

24‐hr treatment period

7 ‐day washout period

Plasma drug level paired with ECG

12 lead ECG (THEW)

Fridericia formula

QTGuard+

ECG biomarkers (lead II):

P‐onset, PR, QRS, QRS‐onset, QRS‐offset, QT interval, J‐Tpeak interval, Tpeak–Tend interval

Concentration‐dependent analysis

QTc & TWM metric prolongation:

Dofetilide: QTc 79.3 ms, J‐Tpeak 39.5 ms, Tpeak–Tend 40 ms, CDA; (p < .001)

Quinidine: QTc 78.1 ms, J‐Tpeak 29.1 ms, Tpeak–Tend 49.8 ms, CDA; (p < .001)

Ranolazine: QTc 12.6 ms (p < .001), Tpeak–Tend 8.8 ms (p < .013), CDA (p < .01)

Couderc et al. (2011)

Moxifloxacin

ECG (2 hr post)

12 lead ECG

Bazett and Fridericia formulae

ECGScan software for ECG digitization

Automated ECG interval assessment with COMPAS software

Cardiac ventricular repolarization:

Lead II: QT apex, QT, Tpeak‐Tend, TW amplitude, left and right TW slope

Eigen lead: T‐loop (roundness)

Vectorcardiographic biomarkers:

ERD, LRD (30–50% threshold)

Cardiac events

Moxifloxacin model:

ERD positive with drug (p = .0001)

LQT2 model:

Left slope detected KCNH2 mutation (p = .0002)

Cardiac event prediction:

T‐roundness (p = .007)

Equivalent to QTc

Graff et al. (2010)

Trial 1

Placebo: 7 days

Moxifloxacin

Days 1–6: placebo

Day 7: 400 mg

Trial 2

Sotalol

Day 1: no drug

Day 2: 160 mg

Day 3: 320 mg

Trial 1

12 lead ECG postdose

Trial 2

Digital Holter ECG

Fridericia formula

MUSE/Interval Editor software

MCS = asymmetry + notch + (1.6 x flatness)

QTc

Sotalol

MCS > QTc (z score 2.5, CI 1.26 to 3.66)

320 mg equivalent to 160 mg (p = .25)

160 mg: MCS 7x greater, QTc 4x greater (vs. moxifloxacin)

320 mg: MCS 15x greater, QTc 6x greater (vs. moxifloxacin)

Covariate effects on QTc and TWM:

Parabolic regression>linear regression for sotalol (p < .01)

No difference for moxifloxacin (p = .45)

ΔMCS:

Sotalol 320 mg > 160 mg (pooled fit, p < .01)

Sotalol 160 mg > moxifloxacin and placebo

Moxifloxacin sex differences

QTc and TWMΔ: F > M

Graff et al. (2009)

Sotalol

Day 1: 0 mg

Day 2: 160 mg

Day 3: 320 mg

Plasma drug level

12 lead ECG

Control

LQT2

12 lead Holter ECG

Sotalol

Fridericia formula

12SL algorithm

MCS = asymmetry + notch + (1.6 x flatness)

QTc

Test accuracy (AUC)

Sotalol:

MCS > QTc (p < .001)

160 mg (AUC 84% vs. 72%)

320 mg (AUC 94% vs. 87%)

Effect size:

MCS > QTc (50%) at maximal AUC and ECGΔ (p < .001)

MCS > QTc (3‐fold) at peak ECGΔ (p < .001)

MCS discrimination accuracy (normal QTc)

Similar for sotalol and LQT2 (p = .9)

Abbreviations: Δ, delta change; AUC, area under the curve; CI, confidence interval; COG, center of gravity; COMPAS, COMPrehensive Analysis of the repolarization Segment; ECG, electrocardiogram; ERD, early repolarisation; F, female; hr, hour; LRD, late repolarization; LQTS, long QT syndrome; M, male; MCS, morphology combination score; NPV, negative predictive value; PPV, positive predictive value; Sp, specificity; Sn, sensitivity; TCRT, total cosine R‐to‐T; THEW, Telemetric and Holter ECG Warehouse database; TW, T wave; TWM, T wave morphology; TWRS, T wave right slope.