Study population
In this matched case-control study, we retrospectively enrolled hospitalized cancer patients with severe ALQTS from September 2013 to April 2016 in the First Affiliated Hospital of Dalian Medical University. Study population was separated into ALQTS group and control group. The inclusion criteria for the ALQTS group were: Hospitalized cancer patients (age > 18 years) from Hematology-Oncology or other departments showing QTc (Bazett’s) ≥ 500 ms without a family history of inherited LQTS, unexplained syncope, cardiac arrest, or sudden death. The exclusion criteria included QRS duration > 120 ms, presence of complete left or right bundle branch blocks, intraventricular conduction delays or ventricular pacing. Moreover, ECG showing atrial fibrillation/flutter, second-degree and complete atrial-ventricular blocks, severely sinus, atrial and ventricular tachyarrhythmia, and acute coronary syndrome with dynamic ST-T that interfered accurate QT assessment were excluded. The inclusion criteria for the control group were same as the ALQTS group, except all had a normal QT interval (350 ms < QTc ≤ 440 ms) during hospitalization. The age, sex and the cancer diagnosis were comparable between ALQTS and control groups. This study was approved by the Ethics Committee of the First Affiliated Hospital of Dalian Medical University.
ECG evaluation
According to the inclusion and exclusion criteria, we retrieved electronic records of standard resting 12-lead ECG parameters from the Muse System (7.1.1 edition), including heart rate (HR) and QTc. ECG were recorded with a paper speed of 25 mm/s and voltage of 10 mm/1 mV in the supine position, using GE Healthcare MAC 5500 ECG diagnosis systems (India). If differences in the QTc interval exceeded 20 ms between the machine and hand measurement, the interval was corrected manually, especially in cases with complex T wave morphologies. In these cases, QT intervals were measured by an experienced QT investigator from lead II, V5 or the lead with the longest QT interval. If several ECG recordings during hospitalization were available, the longest QTc was selected.
Clinical evaluation
From the electronic medical records, we retrieved data including cancer diagnosis, classification and staging, medical history, chemotherapy drugs and presence of QT-prolonging factors such as electrolyte disorders and use of QT-prolonging drugs. Laboratory tests taken within 48 h before or after ECG recording were evaluated, including serum potassium (normal range 3.5–5.3 mmol/L) and serum calcium (2.02–2.6 mmol/L). QT-prolonging drugs used within 7 days of the ECG evaluated in this study were screened through the website (https://www.crediblemeds.org).
Follow-up
In this study, all-cause death was determined as research endpoint. Telephone survey was conducted after permission obtained from study participants or their legally authorized representatives. When death had occurred, the possible causes were investigated.
Application of the pro-QTc score system
We employed the Pro-QTc Score to explore the risk of death in patients with QTc ≥ 500 ms. QT-prolonging diagnoses and conditions include cardiomyopathy, acute coronary syndrome, prolonged QT interval, congestive heart failure, bradycardia, diabetes mellitus, stroke, hypokalemia, hypocalcemia, hypomagnesemia, female gender, old age, and several QT-prolonging drugs, which have been compiled in the QT drug list by the Arizona Center for Education and Research on Therapeutics (AZCERT). Each measure in the score system was scaled as 1 point [8]. Pro-QTc scores ≥4 were considered predictive of high mortality risk.
Statistical analyses
Continuous data were expressed as mean ± SD, and comparative analysis was performed by Student’s t-tests for independent samples. The Mann-Whitney U Test was applied when variables were non-normal distributions. Categorical variables were expressed as absolute (n) and relative proportions (%), and were analyzed using the χ2 (Chi-square) test. 2-tailed statistical significance was considered when p < 0.05. Multivariate logistic regression was performed to calculate odds ratios (OR) for various predictors of QT prolongation. For this procedure, variables were selected on the following based on clinical relevance and presence of p values < 0.1 in univariate analysis. The univariate cumulative probability of all-cause mortality during hospital stay and up to 2-year follow-up was assessed by the Kaplan-Meier survival curve using Log-rank statistics. The Cox proportional hazard survival model was used to evaluate the effect of clinical factors on the endpoint. Statistical analyses were performed using SPSS software (version 22.0, SPSS Inc., Chicago, IL).