High prevalence of risk factors in elderly patients using drugs associated with acquired torsades de pointes chronically in Colombia

Abstract

Aims

Medication is one of the main causes of long QT syndrome (LQTS) and torsades de pointes (TdP), and the older adult population is at particularly high risk. The aim of the present study was to describe the prescription patterns of drugs with a risk of TdP in the Colombian older adult population.

Methods

Patients older than 65 years who received medication with a risk of TdP during three consecutive months were selected. The medication was obtained and classified according to the QT Drug List from Crediblemeds.org. The data were analysed using SPSS‐22.

Results

A total of 55 932 patients were chronically receiving QT‐prolonging drugs; 61.9% (n = 34 ,632) were women and the mean age of the sample was 75.6 years. Drugs with a conditional risk were consumed by 95.2% of patients, 5.3% received drugs with a known risk and 2.9% received drugs with a possible risk. Two or more QT‐prolonging drugs were consumed by 10.3% of the patients (n = 5786). Most of the sample (96.8%, n = 54 170) had at least one additional risk factor for LQTS, with a mean of 3.1 ± 0.9 risk factors. Patients receiving QT‐prolonging drugs for psychiatric and neurological disease were at a higher risk of major polypharmacy [odds ratio (OR) 3.0; 95% confidence interval (CI) 2.80, 3.22) and of receiving high doses of QT‐prolonging drugs (OR 3.8; 95% CI 3.52, 4.05).

Conclusions

The widespread use of medication that causes TdP and the high prevalence of additional risks in the older adult population raise the need for accurate prediction of risk and constant patient monitoring. Patients taking psychiatric drugs are at a higher risk of TdP.

What is Already Known about this Subject

• Some drug classes are considered to be one of the main causes of long QT syndrome (LQTS) and torsades de pointes (TdP), and this serious adverse drug reaction has gained attention over time.

• The older adult population using medication related to LQTS and TdP has multiple risk factors that may increase the probability of these adverse outcomes.

• Studies have identified risk patterns for TdP in acute care but data on their chronic use are scarce.

What this Study Adds

• 10.6% of the older adult population in Columbia was found to consume chronically drugs with QT interval‐prolonging potential. Most of these patients were consuming drugs with a conditional risk, but 98% of them had at least one risk factor for LQTS and TdP.

• Use of QT‐prolonging drugs for psychiatric and neurological disorders was higher among women, and increased with age. In addition, the use of this type of drug was associated with an increase in the risk of major polypharmacy, and of receiving concurrent use of more than one QT drug and high doses.

• Clinicians should keep in mind that common cardiovascular and noncardiovascular drugs are associated with LQTS and TdP, and should routinely assess the associated risk factors, including an electrocardiographic study before and after prescribing these drugs.

Introduction

The long QT syndrome (LQTS) is an anomaly in myocardial repolarization that may cause death by triggering torsades de pointes (TdP), a type of polymorphic ventricular tachycardia 1. Some classes of drugs are considered to be one of the major causes of acquired LQTS, although TdP is experienced by only a small portion of patients that develop drug‐induced QT interval prolongation 2, 3. In fact, the first cause of withdrawal or restriction of the use of drugs on the market in the 1990s was their relationship with the occurrence of LQTS and TdP 4, 5, 6, and regulatory agencies have paid increasing attention to this serious adverse drug reaction in recent years 7.

Drugs that prolong the QT interval include medications for cardiovascular disease, antimicrobial agents, psychotropic drugs and other pharmacological groups 2. Use of these medications is common in clinical practice; 22.8% of all prescriptions claimed by 4.8 million patients in a US national database in 1999 included at least one QT‐prolonging drug, of which 9.4% had an overlapping prescription for either another QT‐prolonging drug or a drug inhibiting the clearance of the QT‐prolonging drug 8.

Drugs usually trigger TdP in individuals with one or more pre‐existing genetic or environmental risk factors, such as structural heart disease, age, gender, capacity of metabolic pathways, electrolyte alterations and overt or subclinical ion channel mutations. Therefore, drug withdrawals and restrictions often occur in the marketing period and are based on data from spontaneous reporting systems, but it remains difficult to predict the risk of TdP in routine clinical practice 3, 9, 10, 11, 12. The prevalence of these and other risk factors increases in the elderly, making this population more prone to develop drug‐induced LQTS and TdP 13.

In Colombia, there are no national programmes regarding TdP risk, and this type of serious adverse event has not been previously reported in the pharmacovigilance programme of the database used for the present study 14. Therefore, our aim was to describe the chronic use patterns of drugs associated with LQTS and TdP and the prevalence of coexistent risk factors in an older adult population from Colombia.

Methods

Descriptive study

Drug dispensing records were obtained from a private company, Audifarma S.A., which delivers medication to approximately 6.5 million people insured by the Colombian Health Care System (approximately 13.8% of the Colombian population). This database includes records of all medication prescribed by five health insurance companies and dispensed by Audifarma S.A. to the patient. Information about medications purchased outside of the insurance system and over‐the‐counter medications are not included.

Patients between 65 and 100 years of age were included in the study if they received drugs with a risk of TdP continuously between September and November 2015, to confirm that patients were receiving stable treatment and indirectly reflect adherence to therapy 15. Drugs related to acquired LQTS were included and classified as having a known, possible or conditional risk of TdP according to the QTDrugs List (available at Crediblemeds.org; accessed on 17 November 2015) (Table 1). Drugs that should be avoided in congenital LQTS, according to this list, include the medications for cardiovascular disease, anti‐infective agents, psychotropic drugs listed above plus drugs that pose a unique risk to patients with this anomaly by mechanisms other than QT prolongation 16. This last group was not included in the present study as the prevalence of congenital LQTS is unknown in the population of study and recommendations regarding these drugs may not be applicable to the population of interest.

Table 1.

Drugs risk categories for torsade de pointes (TdP)

Risk category for Tdp
Conditional	Substantial evidence supports the conclusion that these drugs prolong the QT interval and are associated with a risk of developing TdP, but only under certain known conditions (e.g. excessive dose or overdose, or being the index or interacting agent in a drug–drug interaction)
Possible	Substantial evidence supports the conclusion that these drugs can cause QT prolongation but there is insufficient evidence that the drugs, when used as directed in the labelling, have a risk of causing TdP
Known	Substantial evidence supports the conclusion that these drugs prolong the QT interval and are associated with a risk of TdP when used as directed in the labelling

Source: www.crediblemeds.org

The following variables were obtained from the dispensing records and exported to a database using Microsoft Excel 2013 (Seattle, USA): (i) sociodemographic factors: gender, age; (ii) medications: mean number of defined daily doses (DDD) dispensed during the 3‐month period; (iii) risk factors for LQTS:

1. Cardiovascular disease 9, 17 (consumption of antihypertensive, inotropic, diuretic and antianginal medication).

2. Use of diuretics (risk of electrolyte imbalances) 17, 18;

3. Use of digitalis 6.

4. End‐stage renal disease 9, 17 (dispensing of erythropoietin).

5. Thyroid disease 9 (use of thyroid hormone or antithyroid medication).

6. Polypharmacy 9 (minor polypharmacy 2–4 drugs; major polypharmacy ≥5 drugs 15).

7. Use of high doses of medication related to LQTS 17 (as there is no consensus for the definition of this parameter, a cut‐off point of >1 DDD for an individual drug was defined).

8. Use of medication that causes significant inhibition of the cytochrome P450 (CYP) 3A4 isoenzyme 9, 19.

9. Use of more than one drug with a risk of TdP 20.

An analysis was performed using IBM SPSS Statistics 22, Armonk, New York, USA. The Mann–Whitney U test was used to compare the number of drugs among users and non‐users of psychiatric and neurological QT‐prolonging drugs. A P‐value of <0.05 was considered to be significant. The chi‐square test was used to compare the presence of risk factors among users and non‐users of QT‐prolonging drugs for psychiatric and neurological disorders. Separate multivariate logistic regression analyses were used to assess the risk for use of QT‐prolonging medication for cardiovascular diseases and for psychiatric and neurological disorders, including as independent variables gender (using male gender as a reference) and age groups (using the 65–74‐year group as a reference). The results were expressed as odds ratios (ORs) and 95% confidence intervals (CIs). The Bioethics Committee of the Universidad Tecnológica of Pereira (Pereira, Colombia) reviewed and approved the research as ‘research without risk’ and guaranteed the anonymity of the patients, following the Declaration of Helsinki.

Results

From the 6.5 million patients included in the database, 8.1% were ≥ 65 years of age (n = 525 498). Of these, 55 932 patients (10.6%) consumed 66 out of the 98 QT‐prolonging drugs available in Colombia from the Crediblemeds.org list continuously during the observation period (3 months) (Appendix S1). Women comprised 61.9% of all patients (n = 34 632). The mean age was 75.6 ± 7.2 years; 36.7% of them (n = 20 511) were 65–74 years of age, 51.3% (n = 28 676) were 75–84 years of age and 12.1% (n = 6745) were ≥85 years of age, corresponding to 9.2%, 12.7% and 12.8% of the total population in the database for each age group, respectively.

Drugs that cause LQTS or TdP

Most patients (89.7%, n = 50 146) were taking one, 9.1% (n = 5105) were taking two and 1.1% (n = 613) were taking three QT‐prolonging drugs; 63 patients were taking four of these drugs and five were taking five. Patients received a mean of 2.94 ± 1.04 drugs in total (range 1–9) and 1.12 ± 0.36 from the QTDrugs List (range 1–5).

Drugs with a conditional risk of TdP were taken by 95.2% of the patients (n = 53,258), of which hydrochlorothiazide was the most frequently taken (60.1%, n = 32 030), followed by furosemide (30.3%, n = 16 151). Drugs with a known risk were prescribed to 5.3% of the patients (n = 2980) and drugs with a possible risk were prescribed to 2.9% (n = 1650), of which 46.7% (n = 1393) and 31.9% (n = 526) were receiving at least one drug with a conditional risk concomitantly, they are known risk drugs and drugs with the possible risk and of these cases, it was found that 46.7% and 31.9% received at the same time also a drug with conditional risk respectively. The prescription patterns are shown in Table 2.

Table 2.

Prescription patterns of QT‐prolonging drugs in the older adult population from Colombia, 2015

QT‐prolonging drugs (%)	Total (n = 55 932)		Aged 65–74 years (n = 28 676)		Aged 75–84 years (n = 20 511)		Aged ≥ 85 years (n = 6745)		DDD per 1000 inhabitants per day
	n	%	n	%	n	%	n	%
Known risk of TdP *	2980	5.3	1284	4.5	1176	5.7	520	7.7
Amiodarone	1062	35.6	404	31.5	465	39.5	193	37.1	1.99
Cilostazol	443	14.9	157	12.2	180	15.3	106	20.4	0.57
Levomepromazine	371	12.5	164	12.8	143	12.2	64	12.3	0.09
Chloroquine	354	11.9	229	17.8	107	9.1	18	3.5	0.20
Escitalopram	316	10.6	167	13.0	114	9.7	35	6.7	0.95
Haloperidol	216	7.3	68	5.3	88	7.5	60	11.5	0.13
Fluconazole	122	4.1	72	5.6	40	3.4	10	1.9	0.0693
Possible risk of TdP *	1650	3.0	876	3.1	574	2.8	200	3.0
Imipramine	461	27.9	229	26.1	168	29.3	64	32.0	0.26
Tolterodine	294	17.8	139	15.9	122	21.3	33	16.5	0.58
Clozapine	232	14.1	126	14.4	78	13.6	28	14.0	0.14
Tamoxifen	121	7.3	73	8.3	33	5.7	15	7.5	0.24
Risperidone	110	6.7	52	5.9	42	7.3	16	8.0	0.11
Mirtazapine	109	6.6	50	5.7	46	8.0	13	6.5	0.22
Olanzapine	89	5.4	47	5.4	28	4.9	14	7.0	0.20
Venlafaxine	68	4.1	36	4.1	25	4.4	7	3.5	0.18
Conditional risk of TdP *	53 258	95.2	27 302	95.2	19 560	95.4	6396	94.8
Hydrochlorothiazide	32 030	60.1	18 839	69.0	10 884	55.6	2307	36.1	58.91
Furosemide	16 151	30.3	6182	22.6	6792	34.7	3177	49.7	31.37
Trazodone	3119	5.9	1218	4.5	1240	6.3	661	10.3	1.18
Fluoxetine	1952	3.7	990	3.6	686	3.5	276	4.3	4.18
Sertraline	1460	2.7	606	2.2	586	3.0	268	4.2	3.63
Amitriptyline	864	1.6	442	1.6	288	1.5	134	2.1	0.61
Quetiapine	803	1.5	228	0.8	330	1.7	245	3.8	0.60

DDD, defined daily dose; TdP, torsades de pointes,.

^* Number of patients with at least one drug in the category.

QT‐prolonging drugs for cardiovascular conditions were consumed by 87.1% of the patients , and for psychiatric and neurological conditions were consumed by 16.7% of the patients. The use of cardiovascular QT‐prolonging drugs seemed to be lower in women and to decrease with age, whereas use of psychotropic and neurological QT‐prolonging drugs was higher among women compared with men and increased with age (Table 3). QT‐prolonging drugs with actions on the digestive system were used by 0.6% of patients (n = 358), antimicrobial agents were used chronically by 0.3% (n = 174) and antineoplastic medication by 20 patients.

Table 3.

Use of QT‐prolonging drugs for cardiovascular and psychological and neurological disorders by gender and age groups in an elderly population from Colombia, 2015

	Cardiovascular drugs ( n = 48 739)		Psychotropic drugs (n = 9342)
	n (%)	OR (95% CI)	n (%)	OR (95% CI)
Gender
Male	18 776 (88.2)	Ref	3066 (14.4)	Ref
Female	29 963 (86.5)	0.868 (0.824, 0.915)	6276 (18.1)	1.304 (1.244, 1.367)
Age category
65–74 years	25 258 (88.1)	Ref	4064 (14.2)	Ref
75–84 years	17 897 (87,3)	0.929 (0.880, 0.981)	3544 (17.3)	1.259 (1.198, 1.322)
≥85 years	5584 (82.8)	0.654 (0.609, 0.703)	1734 (25.7)	2.070 (1.943, 2.207)

CI, confidence interval; OR, odds ratio; Ref. reference.

Risk factors

Excluding age and female gender, 96.8% of the population (n = 54 170) included had at least one additional risk factor for LQTS, with a mean of 3.1 ± 0.9 risk factors (range 0–7). One to three risk factors were found in 71.6% of the total sample (n = 40 023), 20.7% (n = 11 590) had four risk factors and 4.6% (n = 2557) had between five and seven risk factors.

At least one risk factor was identified in 98.0% (n = 53 215) of the population consuming drugs with a conditional risk, in 91.3% (n = 2721) taking drugs with a known risk and in 72.1% (n = 1190) taking drugs with a possible risk. The prevalences of risk factors according to the risk category of the QT‐prolonging drug, gender and age group are shown in Figure 1 and Table 4.

Figure 1.

Prevalence of risk factors for torsades de pointes in patients using QT‐prolonging drugs, Colombia, 2015. Drug risk category: known, possible, conditional

Table 4.

Number of risks factors for torsades de pointes (TdP) and use of QT‐prolonging drugs in an elderly population from Colombia, 2015

Number of risk factors	QT‐prolonging drug risk category						Total
	Known		Possible		Conditional
	n	%	n	%	n	%	n	%
0	259	8.7	460	27.9	1043	2.0	1762	3.2
1	294	9.9	100	6.1	211	0.4	605	1.1
2	679	22.8	405	24.5	2858	5.4	3820	6.8
3	463	15.5	270	16.4	35 108	65.9	35 598	63.6
4	766	25.7	267	16.2	11 489	21.6	11 590	20.7
5	444	14.9	134	8.1	2278	4.3	2286	4.1
6	70	2.3	13	0.8	253	0.5	253	0.5
7	5	0.2	1	0.1	18	0.0	18	0.0

Note: patients are included in each category for which they were receiving at least one drug.

Almost all patients were consuming drugs for cardiovascular disease (94.4%, n = 52 810), while 86.1% (n = 48 178) were taking diuretics. Minor polypharmacy (2–4 drugs) was the second most common risk factor and occurred in 87.4% of patients (n = 48 866), while major polypharmacy (five or more drugs) occurred in 7.0% (n = 3918). Medication for thyroid disease was consumed by 18.1% of patients (n = 10 135) and this consumption was 1.811 times higher in patients using amiodarone (n = 1 062 29.1%, 95% CI 16.45, 21.51). High doses of QT‐prolonging drugs were given to 6.7% of patients (n = 3751) and 10.3% (n = 5786) received more than one QT‐prolonging drug. There were 553 patients (1.0%) with end‐stage renal disease, 467 (0.8%) using digoxin and nine using CYP3A4 inhibitors.

The risk of receiving more than one QT‐prolonging drug was highest among patients using such drugs for psychiatric and neurological disease (OR 27.667; 95% CI 25.886, 29.571). Almost half of these patients (46.6%, n = 4359) received more than one QT‐prolonging drug, of which 6.9% (n = 647) used between three and five such drugs. These patients with drugs for psyquiatric and neurological diseases received a mean of 1.54 QT‐prolonging drugs and 3.01 drugs of any kind – significantly higher than patients not consuming these types of drugs (P < 0.001).

Furthermore, the use of QT‐prolonging drugs for psychiatric and neurological disease increased the prevalence of major polypharmacy to 14.7% (n = 1379; OR 3.003; 95% CI 2.801, 3.220), patients in this group had a 3.777 times higher risk of receiving high doses of QT‐prolonging drugs, with a prevalence of 16.1% (n = 1502; OR 3.775; 95% CI 3.522, 4.048), and had higher rates of thyroid disease (25.9%, n = 2422; OR 1.763; 95% CI 1.673, 1.858) compared with those not using this group of drugs.

Discussion

The present study showed that 10.6% of the older adult population in a database of drug dispensing in Colombia were chronically consuming drugs associated with LQTS and TdP. Almost all of these patients were using drugs classified as having a conditional risk of LQTS, but 96.8% of patients had at least one additional risk factor, including use of more than one drug in the QTDrugs List. Cardiovascular disease, use of diuretics and polypharmacy were present in most of the patients, and nearly two‐thirds of the population were women, which is itself a risk factor 21.

Worldwide, studies focusing on the chronic use of QT‐prolonging medications are scarce. In a large study that measured the incidence of use of QT‐prolonging drugs in more than 1 million adults in the USA in 1999, clarithromycin and erythromycin made up 47.4% of a total of QT‐prolonging drugs 8, in contrast to our study, in which drugs to treat cardiovascular disease were the most commonly taken QT‐prolonging medications. Accordingly, chronic and acute use patterns of these drugs should be considered separately for the design of preventive measures. Nonetheless, both studies found a high rate of coprescriptions of antidepressants and antipsychotic medications that may prolong the QT interval, together with other QT‐prolonging drugs (26.9% and 42.3%, respectively) 8; in our case, this included patients taking between three and five of these drugs.

Use of QT ‐prolonging drugs for psychiatric and neurological disorders was higher among women and increased with age, and was related to major polypharmacy and high doses of QT‐prolonging drugs. As we age, concentrations of antipsychotic drugs need to be higher and they accumulate in tissues, increasing the body reservoir. High doses of such drugs are especially dangerous for women because they possess a relatively greater lipid : lean ratio and therefore more space for drug accumulation 22. A recent case‐control study of Swedish elderly patients consuming antidepressant and antipsychotic drugs showed that patients using drugs with a known or possible risk of TdP are at higher risk for mortality compared with drugs with a conditional risk or without a TdP classification 23.

Drugs with a known or possible risk of TdP were consumed by a small fraction of the population included, which may indicate that physicians are aware that they should avoid this medication in older patients when possible. Nonetheless, more than one‐third of the people consuming these drugs were also receiving at least one more medication in the QTDrugs list. The combination of more than one QT‐prolonging drug may increase this risk owing to the additive effect on inward‐rectifier potassium ion channels and drug–drug interactions that alter the metabolism of these drugs5.

Chronic use of diuretics exposes several disadvantages in older adults, which may outweigh the benefit for the prevention of cardiovascular events, with some studies even showing an increase in the risk of sudden cardiac death 24, 25. Diuretics are the most frequent cause of hypokalaemia 26, found in 44.8% of TdP and 32.1% of LQTS cases retrieved from the French Pharmacovigilance Database over a 10‐year period 27. In our study, furosemide and hydrochlorothiazide were the most prescribed QT‐prolonging drugs, and use of diuretics was one of the main risk factors detected in all three risk groups; therefore, interventions aimed at promoting the rational use of this medication group is diuretics can have a big impact on risk minimization.

Use of inappropriate medication in the older adult population contributes to polypharmacy, adverse drug events and drug‐related healthcare costs 18. Fifteen drugs that should be avoided in the older adult population according to Beers criteria 28 were being used by 7.97% (n = 4457) of the patients. Some of these drugs have strong anticholinergic properties that make older adults more susceptible to their undesired effects, including severe central nervous system adverse events that may often go unrecognized and be attributed to age‐related impairment 29. Similarly, use of antibiotics for more than 4 weeks should be avoided in the elderly, except when treating osteomyelitis, prostatitis, tuberculosis or endocarditis 30.

Amiodarone represents a small percentage of the QT‐prolonging drugs used, but it was the most frequent medication used within the known risk group. This medication is likely to be a major concern regarding risk of TdP because, despite the described low incidence (<1.0%) 5, several studies have shown it to be the main cause of TdP alone 27, 31 or in combination with other substances (especially antiarrhythmic or antipsychotic agents) 1, 32, 33. Furthermore, amiodarone is classified as a potentially inappropriate medication for chronic use in older adults according to Beers criteria because of its QT prolongation risk, toxicity and the related thyroid disorders 34; the latter association was evidenced in our data.

Even though the use of high doses (>1 DDD) of QT‐prolonging drugs was found in only 6.7% of patients, using low doses of these drugs should not be adopted as a strategy to minimize risk because for some medications, LQTS and TdP can appear in a dose‐dependent manner (psychiatric drugs, methadone and class III antiarrhythmic agents such as amiodarone), while for other drugs, such as class Ia antiarrhythmic agents, TdP can occur at low or subtherapeutic doses 5, 26.

A recent study found that 60% of patients who developed LQTS were treated on an outpatient basis 31. However, the estimation of real TdP risk is extremely difficult to achieve, both for individuals and populations 9, 13, which increases the importance of identifying the risk factors before prescribing these drugs and balancing them by using the findings of an electrocardiographic study before and after starting the medication 4, 18, 33, a consideration that would apply to our population.

For example, a retrospective study showed that only 50.0% of intensive care unit patients who developed LQTS were receiving a medication with an identified risk, with no other risk factors significantly predictive 32, while Kallergis et al. showed that patients who develop this outcome (LQTS and TdP) have at least one identifiable risk factor 5. Another recent study found that only a prolonged uncorrected QT was a predictor of TdP in older adult patients (in this population, female gender and drug therapy were not associated with the outcome) 27. Furthermore, a retrospective study of LQTS and TdP cases found that neither the number of QT‐prolonging drugs nor female gender were predictors of arrhythmia development 27, contradicting previous evidence 2, 6.

Evidence regarding the risk of TdP for each drug changes constantly; this makes it almost impossible for physicians and pharmacists to keep updated. Therefore, pharmacovigilance programmes should complement the assessment of risk by using various strategies, such as the incorporation of alert systems in prescription interfaces or programmes for detecting patients at risk according to dispensing records. The latter was used in a study conducted in Spain which also included sending information about patients at risk, together with a survey, to primary care physicians in order to detect additional risk factors; this showed positive impacts on the prescription of QT‐prolonging drugs, such as withdrawal, dose reduction or electrocardiogram monitoring, although only 25.4% of the physicians returned the questionnaire 35.

Only prescriptions of drugs with a risk of TdP or that were used for the treatment of additional risk factors were included in the present study as access to the clinical records was not possible, so there were no data on electrolyte levels, QT interval length and the total number of drugs received. Thus, the present study showed that, although there is a large population of elderly patients chronically exposed to these drugs, the real risk status of this population, as well as the clinical outcomes related to this exposure, is unknown.

Elderly patients are being exposed to the risk of sudden cardiac death through the use of common medications with a risk of TdP in the presence of multiple chronic diseases that result in the prescription of numerous drugs; this in itself increases the risk of negative health consequences 18, 36. Considering the challenge that most older adults will have at least one additional risk factor, the inability to predict the risk for a given individual raises the need for adequate tools to help in the decision‐making process when prescribing QT‐prolonging medication for long‐term use in older patients 5.

Considering that most elderly patients have numerous risk factors for TdP at baseline, the assessment of these, including electrocardiographic screening, should be carried out routinely by physicians before and after starting any drug that could prolong the QT interval 18, 23. In the present study, diuretics (furosemide and hydrochlorothiazide) were the most commonly prescribed QT‐prolonging drugs and therefore their use requires close monitoring, especially because of their hypokalaemic effect and associated cardiovascular comorbidity. In addition, patients using psychotropic and neurological drugs could be at higher risk of TdP owing to an increased rate of major polypharmacy, use of multiple QT‐prolonging drugs and prescription of high doses of such drugs, which could be related to low awareness about the QT‐prolonging potential of these noncardiac drugs 37. More studies about the occurrence of drug‐induced TdP and its relationship with the associated risk factors are required to be able to measure adequately the risk in individual patients.

Competing Interests

All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.

The authors would like to thank Viviana Orozco, Andrea Orozco and Soffy López for their contribution to our work database, and Katarina Selling for her kind advice. This study did not received funding sources.

Supporting information

Appendix S1 List of all the drugs that prolong the QT interval used chronically by elderly patients in Colombia, 2015

Supporting info item

Moreno‐Gutiérrez, P. A. , Gaviria‐Mendoza, A. , Cañón, M. M. , and Machado‐Alba, J. E. (2016) High prevalence of risk factors in elderly patients using drugs associated with acquired torsades de pointes chronically in Colombia. Br J Clin Pharmacol, 82: 504–511. doi: 10.1111/bcp.12969.