Prolongation of the QT interval in palliative care patients

Vol. 26 No. 3 September 2003

Journal of Pain and Symptom Management

855

Original Article

Prolongation of the QT Interval in Palliative Care Patients Georgina Walker, MRCP, Andrew Wilcock, DM, FRCP, Ann Marie Carey, MRCP, Cathann Manderson, RN, Rebecca Weller, RN, and Vincent Crosby, MRCGP Hayward House Macmillan Specialist Palliative Care Unit, Nottingham City Hospital NHS Trust, Nottingham, United Kingdom

Abstract Prolonged QT interval on the electrocardiogram (ECG) is associated with an increased risk of cardiac arrhythmia and sudden death. Many drugs used in palliative medicine increase the QT interval and several have had their licenses withdrawn or severely restricted. The relative importance of prolonged QT interval will increase for palliative medicine physicians when dealing with patients with longer prognoses and especially cardiac disease. Given these safety concerns, the aim of this study was to determine the prevalence of a prolonged QT interval in palliative care patients who were not in the terminal stage and were referred to a specialist service. Of 300 patients, 47 (16%) had prolonged QTc but only two had QT ⬎500ms. The presence of coexistent cardiac disease or high levels of serum alkaline phosphatase appear to be the clinical features most robustly associated with a prolonged QTc. Although prolonged QTc is relatively common in patients referred to a specialist palliative care service, severely prolonged QT is rare. J Pain Symptom Manage 2003;26:855–859. 쑖 2003 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved. Key Words QT interval, cardiac arrhythmia, adverse drug reaction, palliative care

Introduction The QT interval lies between the beginning of the QRS, marking the start of ventricular depolarization, and the end of the T wave, marking the end of ventricular repolarization. A prolonged QT interval on the electrocardiograph (ECG) is associated with an increased risk

Address reprint requests to: Andrew Wilcock, DM, FRCP, Hayward House Macmillan Specialist Palliative Care Unit, Nottingham City Hospital NHS Trust, Hucknall Road, Nottingham, NG5 1PB, United Kingdom. Accepted for publication: February 4, 2003.

쑖 2003 U.S. Cancer Pain Relief Committee Published by Elsevier Inc. All rights reserved.

of life-threatening ventricular tachyarrhythmia, especially torsades de pointes, and of sudden death in cardiac and other diseases.1 Growing safety concerns in relation to druginduced prolonged QT has seen several drugs withdrawn, e.g., terodiline, astemizole, cisapride, and droperidol, or their licensed indications severely restricted, e.g., thioridazine. Some other commonly used antipsychotics, e.g., haloperidol, and tricyclic antidepressants, e.g. amitriptyline, also prolong the QT interval. Palliative care patients may be at particular risk of prolonged QT given the high prevalence of multiple drug use, malnutrition and metabolic disturbance. 0885-3924/03/$–see front matter doi:10.1016/S0885-3924(03)00313-0

856

Walker et al.

For any drug treatment, balancing the potential benefit against the risk of harm is part of the prescribers’ responsibilities and is aided by an understanding of the magnitude of the risk. Achieving the right balance in relation to the risk of drug-induced prolonged QT interval will require more careful consideration as palliative care practitioners become increasingly involved with patients with longer prognoses and cardiac disease. Given these safety concerns and the lack of information regarding the incidence of prolonged QT in palliative care patients, it was decided to routinely carry out an ECG on all new referrals to Hayward House, except those referred for terminal care. We report our findings in the first 300 patients.

Methods All patients referred for day care, outpatient review or inpatient admission that were not deemed to be in the terminal phase, i.e., their last few days of life, as part of their medical assessment had a single 12 lead ECG carried out using a Esaotebiomedica P80 electrocardiograph, Seca, Birmingham, UK. This automatically calculates the normalized QTc interval (QT interval corrected for heart rate) using the formula, QTc ⫽ QT *冪1000/RR, where RR is the average time interval between two consecutive ventricular complexes, computed on the basis of average heart rate. In accordance with the European Agency for the Evaluation of Medicinal Products Agency, we defined a prolonged QTc interval as ⬎450 milliseconds(ms) for males and ⬎470ms for females, and identified those at particular risk with an uncorrected QT of ⬎500ms.3 Performance status, history of cardiac disease, prolonged QT or family history of prolonged QT, drug use, serum potassium, magnesium and calcium, survival and mode of death were noted. A death was defined as ‘unexpected’ if it occurred suddenly and without warning in a patient not considered to be in their terminal phase. We used published data of drugs known to prolong the QT interval to identify at risk drugs (Appendix 1).2,4,5 Patients with normal QTc were compared to those with prolonged QTc for age, and for serum potassium, calcium and magnesium using Student’s t-test; sex, inpatient to outpatient ratio, co-existent cardiac disease and

Vol. 26 No. 3 September 2003

number who had died using Chi-square; use of tricyclics, phenothiazines, other at-risk drugs and number who had died unexpectedly by Fisher’s exact test. For performance status and survival, the Mann-Whitney U test and Kaplan Meier survival plots were applied, respectively. Correlation between QTc and age, performance status, electrolytes, renal and hepatic function were assessed using Pearson’s correlation coefficient. For all calculations, the Statistical Package for Social Scientists Software for Windows (SPSS version 9) was used and P ⱕ 0.05 was regarded as statistically significant.

Results All patients had a diagnosis of cancer. No patient reported a past history or family history of prolonged QTc. Full biochemistry was carried out in all but 90 and 13 patients who did not have a serum magnesium or calcium analyzed, respectively. Forty-seven patients (16%) had prolonged QTc and two had a QT ⬎500ms. The distribution of QTc, the demographics of patients with a normal or prolonged QTc and the patients with a QT ⬎500ms are shown in Figure 1 and Table 1, respectively. Compared to those with a normal QTc, those with a prolonged QTc had a significantly greater male:female ratio (P ⫽ 0.005); an increased number with cardiac disease (P ⫽ 0.05); lower serum potassium (P ⫽ 0.005), calcium (P ⫽ 0.002) creatinine (P ⫽ 0.01); and higher serum alkaline phosphatase (P ⫽ 0.003) and bilirubin (P ⫽ 0.009). There were weak but statistically significant correlations between QTc and serum potassium (r ⫽ ⫺0.22, P ⬍ 0.01), serum calcium (r ⫽ ⫺0.16, P ⬍ 0.01), urea (r ⫽ ⫺0.15, P ⬍ 0.05), creatinine (r ⫽ ⫺0.19, P ⬍ 0.01) and bilirubin (r ⫽ 0.18, P ⬍ 0.01). There was no significant correlation between QTc and age, performance status, number of at risk drugs used and other serum biochemistry. The only apparent risk factors in the patients with a QT ⬎500ms were ischemic heart disease in both and in one, the use of tamoxifen.

Discussion QT prolongation may be primary, e.g., hereditary, or more commonly, secondary to cardiac disease, drugs, metabolic disturbance, malnutrition or toxins. Classes of drugs commonly

Vol. 26 No. 3 September 2003

QT Interval Prolongation in Palliative Care Patients

857

Fig. 1. Distribution of QTc interval.

implicated include cardiac antiarrhythmics, antipsychotics, antihistamines and macrolides. The basic mechanism by which most drugs prolong the QT interval appears to be through potassium channel blockade, interfering with

potassium currents in (enhanced) and out (reduced) of the cardiac myocytes; these changes modify the action potential duration.2 Using the EMEA definition, 16% of patients had prolonged QTc, as calculated by the

Table 1 Demographics of Patients with Normal or Prolonged QTc and Details of the 2 Patients with QT ⬎500ms

Normal QTc Number (%) of patients Male:female ratioa Mean age (range) Mean performance status Inpatient:outpatient ratio Number (%) using a suspect drug: Tricyclic antidepressants Phenothiazines Other suspect drug Total (%) using a suspect drug Mean number of suspect drugs used per patient Number (%) with cardiac diseasea Mean (range) potassium (mmol/L)a Mean (range) calcium (mmol/L)a Mean (range) magnesium (mmol/L) (for 161 patients) Mean (range) urea (mmol/L) Mean (range) creatinine (µmol/L)a Mean (range) bilirubin (mmol/L)a Mean (range) alkaline phosphatase (mmol/L)a Number (%) died Mean (median) survival, days Number (%) of deaths ‘unexpected’

253 (84.3) 128:126 67 (23–94) 2.6 180:73 34 49 28 82

(13) (19) (11) (32) 0.5

Prolonged QTc ⬎450ms (males) ⬎470ms (females) 47 (15.7) 34:13 68 (36–92) 2.7 34:13 5 9 7 18

(11) (19) (15) (38) 0.6

58 (23) 4.3 (2.2–6.1) 2.60 (2.17–3.19) 0.8 (0.5–1.1)

17 4.0 2.44 0.8

(36) (2.0–6.1) (1.74–2.86) (0.6–0.9)

7.9 104 14 572

6.7 84 16 1035

(2.0–26.3) (44–251) (3–320) (100–6182)

(1.7–45.5) (34–780) (1–442) (79–7613)

209 (83) 88 (32) 12 (5)

ms ⫽ millisecond a Indicates significant difference (for all P ⱕ 0.05, see text).

42 (89) 72 (40) 2 (4)

Prolonged QT ⬎500ms 2 (0.7) Female 65 (51–78) 1.5 Outpatients – – 1 1 0.5 2 (100) 4.6 (4.3–4.8) 2.5 –

Alive

858

Walker et al.

ECG machine. There are no previous reports in palliative care patients for comparison but it is within the range reported in psychiatric patients (8–23%), higher than reported in nonmalnourished hospital patients (4%), and less than reported in malnourished hospital patients (23%).6–8 Direct comparison is difficult due to the variation in methodology and the upper limit of normal used, sometimes as low as 440ms. We relied solely on automatic calculation of the QTc and this may lead to a degree of overestimation. Although rapid and in most cases accurate, automatically calculated values of QTc may be overestimated when the end of the T wave is obscured by the beginning of the U wave, particularly at slow heart rates.9 To avoid this, some also analyze QTc intervals manually.6 Identifying at risk patients appears difficult without carrying out an ECG. Although recommended, enquiring about a personal or family history of prolonged QT appears unhelpful in practice. Of the features that differed significantly between patients with prolonged and normal QTc, the presence of cardiac disease and level of alkaline phosphatase appeared to be most robust. Ischemic heart disease was also present in the two patients with severely prolonged QT ⬎500ms. The differences in serum potassium, calcium, bilirubin and creatinine between the two groups were relatively small and the excess of male patients in the prolonged QT group, which is contrary to what is expected, likely reflects the uneven male:female ratio in the group as a whole. Nevertheless, correlations suggest that the QT interval increases as serum levels of bilirubin increase and potassium, calcium and creatinine fall. The latter is perhaps contrary to expectations, but the upper range of values for creatinine in the two groups are very different and a patient in the normal QTc group with severe renal failure may have skewed the results (Table 1). Thus, the presence of cardiac disease and/or abnormal liver function in particular, together with low levels of potassium or calcium could act as a useful reminder to clinicians of the need for greater consideration when prescribing drugs known to prolong QT. Given the link between prolonged QT and malnutrition, a more detailed assessment of nutrition may be of use in future work. Although a crude endpoint without post mortem examination, it

Vol. 26 No. 3 September 2003

is of some reassurance that there were no significant differences in median survival or incidence of sudden death in patients with prolonged QTc. Many drugs prolong the QT interval, although the incidence of torsades de pointes appears greatest with the use of cardiac antiarrhythmics, e.g., quinidine (2–9%), compared to other classes of drug (e.g., 1 in 120,000 patients treated with cisapride.)2 For some drugs, the risk is only present with plasma levels seen with high doses, IV administration, or as a result of a drug interaction. For example, imidazole antifungals, by inhibiting the P450 isoenzyme CYP 3A4, prevent the metabolism of cisapride, leading to dangerously high plasma levels. Conversely, the degree of QT prolongation is not always dose related. Further, many of the at risk drugs are substrates for CYP 2D6 and the 5–10% of the European population that is CYP 2D6 poor metabolizer phenotype may be exposed to dangerously high plasma levels even with normal dosages.10 There is a risk of torsades de pointes with any prolongation of the QT, although the risk grows as the QT increases, particularly ⬎500ms. Alternatively, in an individual, a drug which increases the QTc by 30-60ms should raise concern, and if by ⬎60ms, serious concern about the risk of arrhythmia.3 The risk of torsades de pointes is also increased in those with congenital long QT syndrome, females, a history of symptomatic arrhythmias, bradycardia ⬍50bpm, hypertension and other cardiac disease, electrolyte imbalance (especially hypokalemia, hypomagnesemia), impaired liver and renal function (relating to reduced drug metabolism and excretion), hypothyroidism, alcoholism and malnutrition. In conclusion, palliative medicine physicians should be aware that prolonged QT is relatively common in their patients, although severely prolonged QT appears relatively rare. These patients may be exposed to many risk factors for prolonged QT and torsades de pointes, including some of the drugs commonly used in palliative care, but the magnitude of this risk remains unknown. Future work should involve patients with a broader range of diagnoses, continue to identify those most at risk and how QTc changes with time or following the introduction of an at risk drug.

Vol. 26 No. 3 September 2003

QT Interval Prolongation in Palliative Care Patients

Acknowledgments The authors would like to thank Dr Ray Corcoran, Hayward House, Dr. Robert Twycross, Sir Michael Sobell House, Oxford, Dr. Satish Singh, Medicines Control Agency, London and Dr. Sara Lewis, Nottingham City Hospital for their respective help with the manuscript, data interpretation and statistics.

References 1. De Ponti F, Poluzzi E, Montanaro N. QT-interval prolongation by non-cardiac drugs: lessons to be learned from recent experience. Euro J Clin Pharmacol 2000;56:1–18. 2. Haverkamp W, Breithardt G, Camm AJ, et al. The potential for QT prolongation and proarrhythmia by non-antiarrhythmic drugs: clinical and regulatory implications. Euro Heart J 2000;21:1216–1231. 3. Anonymous. Points to consider: the assessment of the potential for QT interval prolongation by

859

noncardiovascular medicinal products. European Agency for the Evaluation of Medicinal Products (EMEA); Committee for Proprietary Medicinal Products: CPMP/986/96. 4. Viskin S. Long QT syndromes and torsade de pointes. The Lancet 1999;354:1625–1633. 5. www.torsades.org. 6. Reilly JG, Ayis SA, Ferrier IN, Jones SJ, Thomas SHL. QTc-interval abnormalities and psychotropic drug therapy in psychiatric patients. The Lancet 2000;355:1048–1052. 7. Warner JP, Barnes TRE, Henry JA. Electrocardiographic changes in patients receiving neuroleptic medication. Acta Psychiatr Scand 1996;93:311–313. 8. Da Cunha DF, De Carvalho Da Cunha F, Ferreira TPS et al. Prolonged QTc intervals on the electrocardiograms of hospitalized malnourished adults. Nutrition 2001;17:370–372. 9. Sheridan DJ. Drug-induced proarrhythmic effects: assessment of changes in QT interval. Br J Clin Pharmacol 2000;50:297–302. 10. Idle JR. The heart of psychotropic drug therapy. The Lancet 2000;355:1824.

Appendix 1 Commonly Used Drugs Associated with Prolonged QT Interval Class Antiarrhythmic drugs

Drug

Class

Amiodarone Bretylium Disopyramide N-acetyl-procainamide

Antimicrobial and antimalarial drugs

Amantadine Chloroquine Clarithromycin Clindamycin Co-trimoxazole Erythromycin Ketoconazole Pentamidine Quinine Spiramycin

Antihistamines

Astemizolea Diphenhydramine Hydroxyzine Terfenadinea Fexofenadine

Miscellaneous

Cisapridea Fos-Phenytoin

Procainamide Propafenone Quinidine Sotalol, d-sotalol Psychiatric drugs

a

Amitriptyline Clomipramine Cloral hydrate Chlorpromazine Citalopram Desipramine Doxepin Droperidola Fluphenazine Haloperidol Imipramine Lithium Maprotiline Nortriptyline Paroxetine Pimozidea Prochlorperazine Risperidone Sertindolea Thioridazinea Trifluoperazine Venlafaxine

Drugs withdrawn, suspended or restricted because of safety concerns regarding prolonged QT. Adapted from references.2,4,5

Drug

Nicardipine Octreotide Probucol Salmeterol Sumatriptan Tamoxifen Terodilinea Vasopressin