Cardiovasc Toxicol DOI 10.1007/s12012-013-9236-9
QT Dispersion and Prognostication of the Outcome in Acute Cardiotoxicities: A Comparison with SAPS II and APACHE II Scoring Systems Hossein Hassanian-Moghaddam • Hassan Amiri • Nasim Zamani • Mitra Rahimi • Shahin Shadnia • Maryam Taherkhani
Ó Springer Science+Business Media New York 2013
Abstract We aimed to evaluate the efficacy of QT dispersion (QTD) in determining the outcome of the patients poisoned by cardiotoxic medications and toxins. Patients who referred to our emergency department (ED) due to acute toxicity with any cardiotoxic medication or toxin and were admitted to medical toxicology intensive care unit (MTICU) were enrolled into the study. A questionnaire containing the demographic characteristics, vital signs, laboratory tests, electrocardiographic (ECG) parameters of the first ECG taken on MTICU or ED admission, simplified acute physiology score (SAPS), and acute physiology and chronic health evaluation (APACHE) score was filled for every single patient. QTD was manually calculated. The patients were divided into two groups of survivors and nonsurvivors and compared. Although QTD was not significantly different between the survivors and non-survivors (P = 0.8), SAPS II and APACHE II score were so. SAPS and APACHE had the highest sensitivity and specificity in determining the patients’ mortality, respectively. SAPS had the highest sensitivity, and QTD had the highest specificity in predicting the later development of the complications. SAPS II and APACHE II scoring systems are the best systems for prognostication of death in patients with acute cardiotoxic medication-induced poisonings. QTD can be successfully used for the prediction of complications.
H. Hassanian-Moghaddam H. Amiri N. Zamani (&) M. Rahimi S. Shadnia Toxicological Research Center, Department of Clinical Toxicology, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Karegar Street, Tehran, Iran e-mail: [email protected] M. Taherkhani Department of Cardiology, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran
Keywords SAPS II APACHE II QT dispersion Poisoning Outcome Prognosis Complications
Introduction Determination of the prognosis is always a challenge for the treating physicians in all fields of medicine. In poisoned patients, determination of the prognosis on admission seems to be even more difficult because many of these patients refer with decreased level of consciousness and rather unstable vital signs. Different scoring systems have been developed to prognosticate mortality and later development of complications in intensive care unit (ICU) patients. Two of the most efficient scoring systems are simplified acute physiology score (SAPS) II and acute physiology and chronic health evaluation (APACHE) II score that have more or less been evaluated in different poisoning settings including tricyclic antidepressant (TCA), aluminum phosphide (ALP), and carbon monoxide (CO) poisonings [1–3]. APACHE II is a severity-of-disease classification system applied within 24 h of admission of a patient to ICU. An integer score from 0 to 71 is computed based on several measurements; higher scores correspond to more severe disease and a higher risk of death [4]. SAPS II is another scoring system used for the same purpose calculated to be between 0 and 163 [5]. These scoring systems, however, are not definite and, even if applied, need different laboratory tests and calculations by expert staff. QT dispersion (QTD) was first defined by Day and colleagues as the difference between the QTc Max-QTc Min in 12-lead electrocardiogram (ECG) [6]. This interlead difference was suggested to reflect the disparity in myocardial recovery times. It has been proposed that QTD is a noninvasive measure of the degree of homogeneity in
123
Cardiovasc Toxicol
myocardial repolarization and a significant predictor of serious arrhythmias and cardiac mortality in human beings [7]. Its normal range has been reported to be between 10.5 ± 10 and 71 ± 7 ms with a mean of 33.4 ± 20.3 ms [5]. The upper normal limit of this measure has been defined to be 75 ms [5]. A QTD [ 40 ms can predict the development of sustained VTach with a sensitivity of 88 % and specificity of 57 % in patients with cardiovascular disorders [8]. In general, prolonged QTD has been accompanied by a higher risk of cardiac arrest and mortality [9, 10]. A QTD [ 80 ms has prognosticated postmyocardial infarction VTach with a sensitivity and specificity of 70 and 86 %, respectively, and is definitely a valuable factor in determining the outcome [11]. Although few studies have evaluated QTD in poisoned patients mainly in CO toxicity [12–15], no study has been performed on patients poisoned by cardiotoxic medications or toxins to elucidate its efficacy in prognosticating the mortality and development of further complications. We aimed to calculate QTD in patients poisoned by cardiotoxic medications and toxins, evaluate its efficacy in determining the patients’ outcome, and compare it with two well-known ICU scoring systems (SAPS II and APACHE II) to see whether it can be a good substitute for them in acute cardiotoxicities.
Materials and Methods In this prospective study, a minimum sample size of 59 patients was determined using the sample size formula for descriptive studies [n = (Z1-a ? Zb)2(S1 ? S2)2/(l1-l2)2] [16]. All patients referred to the emergency department (ED) of our center due to acute toxicity with any cardiotoxic medication or toxin [17] and were admitted to our medical toxicology ICU (MTICU) were enrolled into the study. The total number of the patients who were recruited in the study was 64 patients. The type of medication or toxin was determined according to the history given by the patient him/herself if he/she was conscious on admission. If the patient was unconscious, the history was taken from the relatives. In case they did not give the history, the history was taken from the patient after regaining consciousness. In cases with methadone, tramadol, and methamphetamine toxicity, the diagnosis was confirmed by positive urine tests. In organophosphate-poisoned patients, the low level of choline esterase confirmed the diagnosis. Since the confirmatory serum tests are limited in our center, the history was relied on in other cases. The patients with underlying cardiologic diseases were excluded. Also, those who had ingested different types of medications, some of which were not cardiotoxic, were excluded. On admission, a questionnaire containing the demographic characteristics
123
of each patient (including gender, age, the type and amount of the medication or toxin consumed, and the time elapsed between the toxin consumption and admission) was filled. All patients had been directly admitted to the MTICU from ED. The first vital signs and laboratory tests as well as the ECG parameters withdrawn from the first ECG taken on MTICU or ED admission were also recorded. SAPS II and APACHE II score were calculated using the online calculators [18, 19] and recorded, as well. QT interval was marked between the beginning of the QRS complex and end of the T wave. If a U wave interrupted the returning T to baseline, QT interval was measured to the nadir of the curve between the T and U waves. Corrected QT was measured using the Bazett’s formula (QTc = measured QT/HRR interval) if the heart rate was between 60 and 100 bpm and the Framingham linear regression analysis (QTLC = QT ? 0.154[1 - RR]) if the heart rate was [100 bpm [11]. QTD was determined as the difference between the longest and the shortest QTc intervals in 12-lead ECG and defined to be prolonged if it was [75 ms (Fig. 1) [7]. All measures were double-extracted and manually calculated by a single researcher and re-checked by an attending cardiologist. The final outcome (death versus survival), complications developed during MTICU stay, and the duration of hospital stay were also recorded. The patients were then divided into two groups of survivors and non-survivors, and a comparison was performed between them regarding all characteristics of the questionnaire. Data were entered into the Statistical Package for Social Sciences (SPSS software version 17, IBM Co., Chicago, Ill, USA) and analyzed using t test, Mann– Whitney U test, Fisher’s exact test, and Chi-square test. ROC curves were used to determine the power of each parameter in determining the patients’ mortality and development of later complications. A P value \0.05 was considered to be statistically significant. The study was in accordance with Helsinki declaration and approved by our local ethics committee.
Fig. 1 QT interval is marked between the beginning of the QRS complex and the end of the T wave (left green mark 0.34 s in this example). RR is defined between the two consecutive R waves (right pink mark 0.52 s in this example). Corrected QT is measured using the Bazett’s formula (QTc = measured QT/HRR interval) if the heart rate is between 60 and 100 bpm and the Framingham linear regression analysis (QTLC = QT ? 0.154[1 - RR]) if the heart rate is[100 bpm. In the example shown, since the heart rate is[100 bpm, Framingham formula should be used and the QTc will be 0.41 s. QTD is determined as the difference between the longest and the shortest QTc intervals in 12-lead ECG (Color figure online)
Cardiovasc Toxicol
Results A total of 64 patients were prospectively evaluated. Thirtysix (56.2 %) were male. Mean age was 31 ± 12 years (range; 7 to 65 years). Forty-eight patients (66.6 %) survived and 16 (33.3 %) died. No statistically significant difference was noted between those who survived and those who died in terms of age and sex. The most common overdosed cardiotoxic drugs were methadone (13 patients; 20.3 %), followed by TCA and ALP (each 9 patients; 14.1 %), methamphetamine (crystal) and tramadol HCl (each 5 patients; 7.8 %), and carbon monoxide and organophosphate (each 3 patients; 4.7 %; Table 1). Six patients had consumed a combination of medications and toxins, all of which were cardiotoxic. Mean time between consumption and hospital admission was 10.9 ± 16.4 h (range 0.5–96 h) in the patients without a significant difference between the survivors and non-survivors (P = 0.3). On admission to the MTICU, no significant difference was detected between the temperature, respiratory rate, pulse rate, systolic and diastolic blood pressure, and mean arterial pressure of the groups (all P values [0.05). Also, mean Glasgow coma scale (GCS), blood urea nitrogen and creatinine, serum sodium and potassium, and bicarbonate and pH were not significantly different between these two groups. Mean QTD, SAPS II, and APACHE II score were 101 ± 52 ms, 40 ± 16 (range 8–76), and 15 ± 6 (range 4–33), respectively. Although QTD was insignificantly different between the survivors and non-survivors (P = 0.8), SAPS (P = 0.03) and APACHE score (P = 0.009) were so. Corrected QT measures in all 12 ECG leads were not significantly different between the two groups, either. Table 1 Overdosed medications and toxins and number of the patients who consumed them Medication/toxin overdosed
Number of the patients (%)
Methadone
13 (20.3)
Tricyclic antidepressants
9 (14.1)
Aluminum phosphide
9 (14.1)
Crystal
5 (7.8)
Tramadol HCl
5 (7.8)
Carbon monoxide
3 (4.7)
Organophosphates
3 (4.7)
Lidocaine
2 (3.1)
Carbamazepine
2 (3.1)
Colchicine
2 (3.1)
Calcium channel blockers
2 (3.1)
Beta blockers
2 (3.1)
Chloroquine Multi-drug overdose
1 (1.5) 6 (9.3)
Except for GCS (P = 0.02), none of the vital signs and laboratory tests on MTICU admission were different between those who survived and those who died. Among the corrected QTs in 12 leads of ECG, QTD, SAPS II, and APACHE II score, QTD (P = 0.04), SAPS II (P = 0.004), and APACHE II score (P = 0.04) could successfully predict the later development of the complications. Our complications included refractory hypotension in 5, right bundle branch block in 3, and abnormal liver function tests, re-intubation, and refractory seizures each in 2 patients. Brugada pattern, T-40 ms frontal plane axis deviation, premature ventricular contractions, complete heart block, atrial fibrillation, and acute respiratory distress syndrome were each detected in one patient. Mean hospital stay was 5.0 ± 4.7 days without a significant difference between the two groups. SAPS II (P = 0.009) was the only score that could predict the duration of hospital stay. QTC in leads avR and avF (in comparison with other leads) had a better sensitivity and specificity for
Fig. 2 The efficacy of QTC in leads avR and avF, QTD, SAPS II, and APACHE II score is compared in determining the patients’ death. Sensitivity of a test = number of true positives/number of all positives. Specificity of a test = number of true negatives/number of all negatives
Fig. 3 The efficacy of QTC in leads avR and avF, QTD, SAPS II, and APACHE II score is compared in determining the patients’ complications
123
Cardiovasc Toxicol
prognostication of the patients’ mortality (Fig. 2). SAPS II had the highest sensitivity, and APACHE II had the highest specificity in determining the patients’ mortality (Fig. 2). However, SAPS II was not so specific in determination of the mortality. Also, SAPS II had the highest sensitivity, and QTD had the highest specificity in predicting later development of the complications (Fig. 3).
Discussion According to our results, mean QTD was 101 ± 52 ms which is significantly higher than the previously reported mean QTD in normal people (mean 33.4 ± 20.3 ms; max 75 ms) [7, 8]. This finding shows that in medication- and toxin-induced cardiotoxicities, mean QTD is increased. This is probably due to the abnormalities in the homogeneity of the myocardial repolarization, as the previous authors have stated [7]. Our study is novel in that it has evaluated QTD in the prediction of the prognosis in acutely poisoned patients. Also, to our best knowledge, no study has compared the efficacy of such potential prognostic factor (QTD) with previously developed scoring systems (SAPS II and APACHE II) in the prediction of mortality and complications in critically ill poisoned patients admitted to MTICU. However, our results show that QTD is not a good predictor of the mortality in these patients. Based on the obtained P values, SAPS II is the most sensitive and APACHE II is the most specific method for the prediction of mortality in medication-induced cardiotoxicities. Although SAPS II had the highest sensitivity in determination of the mortality rates in these patients, it was not so specific for this purpose. QTD, with a cutoff point of 85 ms, can predict the mortality with a sensitivity of almost 55 % and specificity of 53 % which is not precise enough to be applied in the MTICU setting. On the other hand, SAPS II was the most sensitive and QTD was the most specific tool that could successfully predict development of the complications. All factors had lower specificities in comparison with the sensitivities for determination of further complications. Considering both sensitivity and specificity, SAPS II was the best factor in determination of the later complications. SAPS II was the only scoring system that could successfully predict the duration of hospital stay. Wang and associates reported that QTD could predict the risk of atherosclerotic events in chronically poisoned patients with arsenic. They concluded that this measure could be a significant and strong prognostic factor of cardiovascular complications in those patients [20]. In another study, cardiac enzymes, arrhythmias, and death rate increased with QTD increasing in tetramine poisoning [21]. The only study we found in the literature that had evaluated QTD in determination of the prognosis in critically ill
123
patients was the study performed by Zhou and colleagues that had concluded that QTD had predictable value in prognosis of the serious illnesses; however, they had mentioned that the value of its clinical use was limited [22]. Although Manini and colleagues suggested that QTD correlated with the adverse cardiovascular events in suspected poisonings [23], they did not evaluate it as a prognostic factor in such setting. Regarding our results, it seems that although SAPS II and APACHE II scoring systems are yet the best systems for prognostication of death in acutely poisoned patients with cardiotoxic medications and toxins, QTD can be successfully used for the prediction of further complications, almost all of which are cardiovascular. However, more prospective studies with greater sample sizes are warranted to confirm such a relation.
Limitations QTD evaluation is not widely used in the setting of acute toxicity. Although the process of determination of QTD is operator-dependent and this is a potential limitation of the current study, we tried to decrease such probable discrepancy by re-evaluation of the calculated measurements by an attending cardiologist. Also, since QRS width may affect the QTD measurement, we meant not to evaluate those ECGs with QRS widening to eliminate such a problem. However, none of our patients had QRS widening. In our center, the confirmatory serum levels are not available for all medications and toxins. We had to rely on the patients’ history in some cases (for example in cases with colchicine toxicity). This is another limitation of our study that we tried to obviate by taking accurate history from the patients or their relatives.
Conclusion SAPS II and APACHE II scoring systems are the best systems for prognostication of death in acutely poisoned patients with cardiotoxic medications and toxins. QTD can be successfully used for the prediction of further complications in these patients. Acknowledgments The authors would like to thank Professor Abdolkarim Pajoumand for his comments on designing and performance of the study.
References 1. Cekmen, N., & Erdemli, O. (2009). Urgent hyperbaric oxygen therapy (HBO2) for acute carbon monoxide poisoning. Bratislavske Lekarske Listy, 110, 716–718.
Cardiovasc Toxicol 2. Louriz, M., Dendane, T., Abidi, K., Madani, N., Abouqal, R., & Zeggwagh, A. A. (2009). Prognostic factors of acute aluminum phosphide poisoning. Indian Journal of Medical Sciences, 63, 227–234. 3. Arranto, C. A., Mueller, C., Hunziker, P. R., Marsch, S. C., & Eriksson, U. (2003). Adverse cardiac events in ICU patients with presumptive antidepressant overdose. Swiss Medical Weekly, 133, 479–483. 4. Knaus, W. A., Draper, E. A., Wagner, D. P., & Zimmerman, J. E. (1985). APACHE II: A severity of disease classification system. Critical Care Medicine, 13, 818–829. 5. Le Gall, J. R., Lemeshow, S., & Saulnier, F. (1993). A new simplified acute physiology score (SAPS II) based on a European/ North American multicenter study. JAMA, 270, 2957–2963. 6. Day, C. P., McComb, J. M., & Campbell, R. W. (1990). QT dispersion: An indication of arrhythmia risk in patients with long QT intervals. British Heart Journal, 63, 342–344. 7. Atescelik, M., Bozdemir, M. N., Yildiz, M., Gurbuz, S., Ayranci, M., Goktekin, M. C., et al. (2012). QT dispersion in carbon monoxide poisoning. European Review for Medical and Pharmacological Sciences, 1, 25–29. 8. Malik, M., & Batchvarov, V. N. (2000). Measurement, interpretation, and clinical potential of QT dispersion. Journal of the American College of Cardiology, 6, 1749–1766. 9. de Bruyne, M. C., Hoes, A. W., Kors, J. A., Hofman, A., van Bemmel, J. H., & Grobbee, D. E. (1998). QTc dispersion predicts cardiac mortality in the elderly: The Rotterdam Study. Circulation, 97(5), 467–472. 10. Okin, P. M., Devereux, R. B., Howard, B. V., Fabsitz, R. R., Lee, E. T., & Welty, T. K. (2000). Assessment of QT interval and QT dispersion for prediction of all-cause and cardiovascular mortality in American Indians: The Strong Heart Study. Circulation, 101(1), 61–66. 11. Clancy, C. (2011). Electrophysiologic and electrocardiographic principles. In L. S. Nelson, N. A. Lewin, M. A. Howland, R. S. Hoffman, L. R. Goldfrank, & N. E. Flomenbaum (Eds.), Goldfrank’s Toxicologic Emergencies (pp. 314–329). NewYork: McGraw-Hill Press. 12. Huang, C. Y. C. H. (2005). Carbon monoxide Intoxication. Annals of Disaster Medicine, 4, S8–S16.
13. Gu¨rkan, Y., Canatay, H., Toprak, A., Ural, E., & Toker, K. (2002). Carbon monoxide poisoning—a cause of increased QT dispersion. Acta Anaesthesiologica Scandinavica, 46(2), 180–183. 14. Yelken, B., Tanriverdi, B., Cetinbas, F., Memis, D., & Sut, N. (2009). The assessment of QT intervals in acute carbon monoxide poisoning. Anadolu Kardiyoloji Dergisi, 9, 397–400. 15. Hancı, V., Ayog˘lu, H., Yurtlu, S., Yıldırım, N., Okyay, D., Erdog˘an, G., et al. (2011). Effects of acute carbon monoxide poisoning on the P-wave and QT interval dispersions. Anadolu Kardiyoloji Dergisi, 1, 48–52. 16. Fox, N., Hunn, A., & Mathers, N. (2013). Sampling and sample size calculation. National Institute for Health Research. Available form http://rds-eastmidlands.nihr.ac.uk/resources/doc_download/ 9-sampling-and-sample-size-calculation.html. Accessed July 9, 2013. 17. Dan, M., & Roden, M. D. (2004). Drug-induced prolongation of the QT interval. New England Journal of Medicine, 350, 1013–1022. 18. SAPS II calculator. (2013). OPUS 12 Foundation, Inc. Available from opus12.org/SAPS_II.html. Accessed July 9, 2013. 19. APACHE II calculator-GlobalRPH. (2013). Available from: http://www.globalrph.com/apacheii.html. Accessed July 9, 2013. 20. Wang, C. H., Chen, C. L., Hsiao, C. K., Yıldırım, N., Okyay, D., Erdog˘an, G., et al. (2010). Arsenic-induced QT dispersion is associated with atherosclerotic diseases and predicts long-term cardiovascular mortality in subjects with previous exposure to arsenic: A 17-year follow-up study. Cardiovascular Toxicology, 10(1), 17–26. 21. Liu, H. K. (2004). Clinical study of QT dispersion in tetramine poisoning. Journal of North Sichuan Medical College, 4. 22. Zhou, C. Y., Liang, Z. Q., Liu, R. S. (2004). Significance of QT dispersion for prognosis in the critical illness. Chinese Journal of Critical Care Medicine, 12. 23. Manini, A. F., Nelson, L. S., Skolnick, A. H., Slater, W., & Hoffman, R. S. (2010). Electrocardiographic predictors of adverse cardiovascular events in suspected poisoning. Journal of Medical Toxicology, 6(2), 106–115.
123
QT Dispersion and Prognostication of the Outcome in Acute Cardiotoxicities: A Comparison with SAPS II and APACHE II Scoring Systems Hossein Hassanian-Moghaddam • Hassan Amiri • Nasim Zamani • Mitra Rahimi • Shahin Shadnia • Maryam Taherkhani
Ó Springer Science+Business Media New York 2013
Abstract We aimed to evaluate the efficacy of QT dispersion (QTD) in determining the outcome of the patients poisoned by cardiotoxic medications and toxins. Patients who referred to our emergency department (ED) due to acute toxicity with any cardiotoxic medication or toxin and were admitted to medical toxicology intensive care unit (MTICU) were enrolled into the study. A questionnaire containing the demographic characteristics, vital signs, laboratory tests, electrocardiographic (ECG) parameters of the first ECG taken on MTICU or ED admission, simplified acute physiology score (SAPS), and acute physiology and chronic health evaluation (APACHE) score was filled for every single patient. QTD was manually calculated. The patients were divided into two groups of survivors and nonsurvivors and compared. Although QTD was not significantly different between the survivors and non-survivors (P = 0.8), SAPS II and APACHE II score were so. SAPS and APACHE had the highest sensitivity and specificity in determining the patients’ mortality, respectively. SAPS had the highest sensitivity, and QTD had the highest specificity in predicting the later development of the complications. SAPS II and APACHE II scoring systems are the best systems for prognostication of death in patients with acute cardiotoxic medication-induced poisonings. QTD can be successfully used for the prediction of complications.
H. Hassanian-Moghaddam H. Amiri N. Zamani (&) M. Rahimi S. Shadnia Toxicological Research Center, Department of Clinical Toxicology, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Karegar Street, Tehran, Iran e-mail: [email protected] M. Taherkhani Department of Cardiology, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences, Tehran, Iran
Keywords SAPS II APACHE II QT dispersion Poisoning Outcome Prognosis Complications
Introduction Determination of the prognosis is always a challenge for the treating physicians in all fields of medicine. In poisoned patients, determination of the prognosis on admission seems to be even more difficult because many of these patients refer with decreased level of consciousness and rather unstable vital signs. Different scoring systems have been developed to prognosticate mortality and later development of complications in intensive care unit (ICU) patients. Two of the most efficient scoring systems are simplified acute physiology score (SAPS) II and acute physiology and chronic health evaluation (APACHE) II score that have more or less been evaluated in different poisoning settings including tricyclic antidepressant (TCA), aluminum phosphide (ALP), and carbon monoxide (CO) poisonings [1–3]. APACHE II is a severity-of-disease classification system applied within 24 h of admission of a patient to ICU. An integer score from 0 to 71 is computed based on several measurements; higher scores correspond to more severe disease and a higher risk of death [4]. SAPS II is another scoring system used for the same purpose calculated to be between 0 and 163 [5]. These scoring systems, however, are not definite and, even if applied, need different laboratory tests and calculations by expert staff. QT dispersion (QTD) was first defined by Day and colleagues as the difference between the QTc Max-QTc Min in 12-lead electrocardiogram (ECG) [6]. This interlead difference was suggested to reflect the disparity in myocardial recovery times. It has been proposed that QTD is a noninvasive measure of the degree of homogeneity in
123
Cardiovasc Toxicol
myocardial repolarization and a significant predictor of serious arrhythmias and cardiac mortality in human beings [7]. Its normal range has been reported to be between 10.5 ± 10 and 71 ± 7 ms with a mean of 33.4 ± 20.3 ms [5]. The upper normal limit of this measure has been defined to be 75 ms [5]. A QTD [ 40 ms can predict the development of sustained VTach with a sensitivity of 88 % and specificity of 57 % in patients with cardiovascular disorders [8]. In general, prolonged QTD has been accompanied by a higher risk of cardiac arrest and mortality [9, 10]. A QTD [ 80 ms has prognosticated postmyocardial infarction VTach with a sensitivity and specificity of 70 and 86 %, respectively, and is definitely a valuable factor in determining the outcome [11]. Although few studies have evaluated QTD in poisoned patients mainly in CO toxicity [12–15], no study has been performed on patients poisoned by cardiotoxic medications or toxins to elucidate its efficacy in prognosticating the mortality and development of further complications. We aimed to calculate QTD in patients poisoned by cardiotoxic medications and toxins, evaluate its efficacy in determining the patients’ outcome, and compare it with two well-known ICU scoring systems (SAPS II and APACHE II) to see whether it can be a good substitute for them in acute cardiotoxicities.
Materials and Methods In this prospective study, a minimum sample size of 59 patients was determined using the sample size formula for descriptive studies [n = (Z1-a ? Zb)2(S1 ? S2)2/(l1-l2)2] [16]. All patients referred to the emergency department (ED) of our center due to acute toxicity with any cardiotoxic medication or toxin [17] and were admitted to our medical toxicology ICU (MTICU) were enrolled into the study. The total number of the patients who were recruited in the study was 64 patients. The type of medication or toxin was determined according to the history given by the patient him/herself if he/she was conscious on admission. If the patient was unconscious, the history was taken from the relatives. In case they did not give the history, the history was taken from the patient after regaining consciousness. In cases with methadone, tramadol, and methamphetamine toxicity, the diagnosis was confirmed by positive urine tests. In organophosphate-poisoned patients, the low level of choline esterase confirmed the diagnosis. Since the confirmatory serum tests are limited in our center, the history was relied on in other cases. The patients with underlying cardiologic diseases were excluded. Also, those who had ingested different types of medications, some of which were not cardiotoxic, were excluded. On admission, a questionnaire containing the demographic characteristics
123
of each patient (including gender, age, the type and amount of the medication or toxin consumed, and the time elapsed between the toxin consumption and admission) was filled. All patients had been directly admitted to the MTICU from ED. The first vital signs and laboratory tests as well as the ECG parameters withdrawn from the first ECG taken on MTICU or ED admission were also recorded. SAPS II and APACHE II score were calculated using the online calculators [18, 19] and recorded, as well. QT interval was marked between the beginning of the QRS complex and end of the T wave. If a U wave interrupted the returning T to baseline, QT interval was measured to the nadir of the curve between the T and U waves. Corrected QT was measured using the Bazett’s formula (QTc = measured QT/HRR interval) if the heart rate was between 60 and 100 bpm and the Framingham linear regression analysis (QTLC = QT ? 0.154[1 - RR]) if the heart rate was [100 bpm [11]. QTD was determined as the difference between the longest and the shortest QTc intervals in 12-lead ECG and defined to be prolonged if it was [75 ms (Fig. 1) [7]. All measures were double-extracted and manually calculated by a single researcher and re-checked by an attending cardiologist. The final outcome (death versus survival), complications developed during MTICU stay, and the duration of hospital stay were also recorded. The patients were then divided into two groups of survivors and non-survivors, and a comparison was performed between them regarding all characteristics of the questionnaire. Data were entered into the Statistical Package for Social Sciences (SPSS software version 17, IBM Co., Chicago, Ill, USA) and analyzed using t test, Mann– Whitney U test, Fisher’s exact test, and Chi-square test. ROC curves were used to determine the power of each parameter in determining the patients’ mortality and development of later complications. A P value \0.05 was considered to be statistically significant. The study was in accordance with Helsinki declaration and approved by our local ethics committee.
Fig. 1 QT interval is marked between the beginning of the QRS complex and the end of the T wave (left green mark 0.34 s in this example). RR is defined between the two consecutive R waves (right pink mark 0.52 s in this example). Corrected QT is measured using the Bazett’s formula (QTc = measured QT/HRR interval) if the heart rate is between 60 and 100 bpm and the Framingham linear regression analysis (QTLC = QT ? 0.154[1 - RR]) if the heart rate is[100 bpm. In the example shown, since the heart rate is[100 bpm, Framingham formula should be used and the QTc will be 0.41 s. QTD is determined as the difference between the longest and the shortest QTc intervals in 12-lead ECG (Color figure online)
Cardiovasc Toxicol
Results A total of 64 patients were prospectively evaluated. Thirtysix (56.2 %) were male. Mean age was 31 ± 12 years (range; 7 to 65 years). Forty-eight patients (66.6 %) survived and 16 (33.3 %) died. No statistically significant difference was noted between those who survived and those who died in terms of age and sex. The most common overdosed cardiotoxic drugs were methadone (13 patients; 20.3 %), followed by TCA and ALP (each 9 patients; 14.1 %), methamphetamine (crystal) and tramadol HCl (each 5 patients; 7.8 %), and carbon monoxide and organophosphate (each 3 patients; 4.7 %; Table 1). Six patients had consumed a combination of medications and toxins, all of which were cardiotoxic. Mean time between consumption and hospital admission was 10.9 ± 16.4 h (range 0.5–96 h) in the patients without a significant difference between the survivors and non-survivors (P = 0.3). On admission to the MTICU, no significant difference was detected between the temperature, respiratory rate, pulse rate, systolic and diastolic blood pressure, and mean arterial pressure of the groups (all P values [0.05). Also, mean Glasgow coma scale (GCS), blood urea nitrogen and creatinine, serum sodium and potassium, and bicarbonate and pH were not significantly different between these two groups. Mean QTD, SAPS II, and APACHE II score were 101 ± 52 ms, 40 ± 16 (range 8–76), and 15 ± 6 (range 4–33), respectively. Although QTD was insignificantly different between the survivors and non-survivors (P = 0.8), SAPS (P = 0.03) and APACHE score (P = 0.009) were so. Corrected QT measures in all 12 ECG leads were not significantly different between the two groups, either. Table 1 Overdosed medications and toxins and number of the patients who consumed them Medication/toxin overdosed
Number of the patients (%)
Methadone
13 (20.3)
Tricyclic antidepressants
9 (14.1)
Aluminum phosphide
9 (14.1)
Crystal
5 (7.8)
Tramadol HCl
5 (7.8)
Carbon monoxide
3 (4.7)
Organophosphates
3 (4.7)
Lidocaine
2 (3.1)
Carbamazepine
2 (3.1)
Colchicine
2 (3.1)
Calcium channel blockers
2 (3.1)
Beta blockers
2 (3.1)
Chloroquine Multi-drug overdose
1 (1.5) 6 (9.3)
Except for GCS (P = 0.02), none of the vital signs and laboratory tests on MTICU admission were different between those who survived and those who died. Among the corrected QTs in 12 leads of ECG, QTD, SAPS II, and APACHE II score, QTD (P = 0.04), SAPS II (P = 0.004), and APACHE II score (P = 0.04) could successfully predict the later development of the complications. Our complications included refractory hypotension in 5, right bundle branch block in 3, and abnormal liver function tests, re-intubation, and refractory seizures each in 2 patients. Brugada pattern, T-40 ms frontal plane axis deviation, premature ventricular contractions, complete heart block, atrial fibrillation, and acute respiratory distress syndrome were each detected in one patient. Mean hospital stay was 5.0 ± 4.7 days without a significant difference between the two groups. SAPS II (P = 0.009) was the only score that could predict the duration of hospital stay. QTC in leads avR and avF (in comparison with other leads) had a better sensitivity and specificity for
Fig. 2 The efficacy of QTC in leads avR and avF, QTD, SAPS II, and APACHE II score is compared in determining the patients’ death. Sensitivity of a test = number of true positives/number of all positives. Specificity of a test = number of true negatives/number of all negatives
Fig. 3 The efficacy of QTC in leads avR and avF, QTD, SAPS II, and APACHE II score is compared in determining the patients’ complications
123
Cardiovasc Toxicol
prognostication of the patients’ mortality (Fig. 2). SAPS II had the highest sensitivity, and APACHE II had the highest specificity in determining the patients’ mortality (Fig. 2). However, SAPS II was not so specific in determination of the mortality. Also, SAPS II had the highest sensitivity, and QTD had the highest specificity in predicting later development of the complications (Fig. 3).
Discussion According to our results, mean QTD was 101 ± 52 ms which is significantly higher than the previously reported mean QTD in normal people (mean 33.4 ± 20.3 ms; max 75 ms) [7, 8]. This finding shows that in medication- and toxin-induced cardiotoxicities, mean QTD is increased. This is probably due to the abnormalities in the homogeneity of the myocardial repolarization, as the previous authors have stated [7]. Our study is novel in that it has evaluated QTD in the prediction of the prognosis in acutely poisoned patients. Also, to our best knowledge, no study has compared the efficacy of such potential prognostic factor (QTD) with previously developed scoring systems (SAPS II and APACHE II) in the prediction of mortality and complications in critically ill poisoned patients admitted to MTICU. However, our results show that QTD is not a good predictor of the mortality in these patients. Based on the obtained P values, SAPS II is the most sensitive and APACHE II is the most specific method for the prediction of mortality in medication-induced cardiotoxicities. Although SAPS II had the highest sensitivity in determination of the mortality rates in these patients, it was not so specific for this purpose. QTD, with a cutoff point of 85 ms, can predict the mortality with a sensitivity of almost 55 % and specificity of 53 % which is not precise enough to be applied in the MTICU setting. On the other hand, SAPS II was the most sensitive and QTD was the most specific tool that could successfully predict development of the complications. All factors had lower specificities in comparison with the sensitivities for determination of further complications. Considering both sensitivity and specificity, SAPS II was the best factor in determination of the later complications. SAPS II was the only scoring system that could successfully predict the duration of hospital stay. Wang and associates reported that QTD could predict the risk of atherosclerotic events in chronically poisoned patients with arsenic. They concluded that this measure could be a significant and strong prognostic factor of cardiovascular complications in those patients [20]. In another study, cardiac enzymes, arrhythmias, and death rate increased with QTD increasing in tetramine poisoning [21]. The only study we found in the literature that had evaluated QTD in determination of the prognosis in critically ill
123
patients was the study performed by Zhou and colleagues that had concluded that QTD had predictable value in prognosis of the serious illnesses; however, they had mentioned that the value of its clinical use was limited [22]. Although Manini and colleagues suggested that QTD correlated with the adverse cardiovascular events in suspected poisonings [23], they did not evaluate it as a prognostic factor in such setting. Regarding our results, it seems that although SAPS II and APACHE II scoring systems are yet the best systems for prognostication of death in acutely poisoned patients with cardiotoxic medications and toxins, QTD can be successfully used for the prediction of further complications, almost all of which are cardiovascular. However, more prospective studies with greater sample sizes are warranted to confirm such a relation.
Limitations QTD evaluation is not widely used in the setting of acute toxicity. Although the process of determination of QTD is operator-dependent and this is a potential limitation of the current study, we tried to decrease such probable discrepancy by re-evaluation of the calculated measurements by an attending cardiologist. Also, since QRS width may affect the QTD measurement, we meant not to evaluate those ECGs with QRS widening to eliminate such a problem. However, none of our patients had QRS widening. In our center, the confirmatory serum levels are not available for all medications and toxins. We had to rely on the patients’ history in some cases (for example in cases with colchicine toxicity). This is another limitation of our study that we tried to obviate by taking accurate history from the patients or their relatives.
Conclusion SAPS II and APACHE II scoring systems are the best systems for prognostication of death in acutely poisoned patients with cardiotoxic medications and toxins. QTD can be successfully used for the prediction of further complications in these patients. Acknowledgments The authors would like to thank Professor Abdolkarim Pajoumand for his comments on designing and performance of the study.
References 1. Cekmen, N., & Erdemli, O. (2009). Urgent hyperbaric oxygen therapy (HBO2) for acute carbon monoxide poisoning. Bratislavske Lekarske Listy, 110, 716–718.
Cardiovasc Toxicol 2. Louriz, M., Dendane, T., Abidi, K., Madani, N., Abouqal, R., & Zeggwagh, A. A. (2009). Prognostic factors of acute aluminum phosphide poisoning. Indian Journal of Medical Sciences, 63, 227–234. 3. Arranto, C. A., Mueller, C., Hunziker, P. R., Marsch, S. C., & Eriksson, U. (2003). Adverse cardiac events in ICU patients with presumptive antidepressant overdose. Swiss Medical Weekly, 133, 479–483. 4. Knaus, W. A., Draper, E. A., Wagner, D. P., & Zimmerman, J. E. (1985). APACHE II: A severity of disease classification system. Critical Care Medicine, 13, 818–829. 5. Le Gall, J. R., Lemeshow, S., & Saulnier, F. (1993). A new simplified acute physiology score (SAPS II) based on a European/ North American multicenter study. JAMA, 270, 2957–2963. 6. Day, C. P., McComb, J. M., & Campbell, R. W. (1990). QT dispersion: An indication of arrhythmia risk in patients with long QT intervals. British Heart Journal, 63, 342–344. 7. Atescelik, M., Bozdemir, M. N., Yildiz, M., Gurbuz, S., Ayranci, M., Goktekin, M. C., et al. (2012). QT dispersion in carbon monoxide poisoning. European Review for Medical and Pharmacological Sciences, 1, 25–29. 8. Malik, M., & Batchvarov, V. N. (2000). Measurement, interpretation, and clinical potential of QT dispersion. Journal of the American College of Cardiology, 6, 1749–1766. 9. de Bruyne, M. C., Hoes, A. W., Kors, J. A., Hofman, A., van Bemmel, J. H., & Grobbee, D. E. (1998). QTc dispersion predicts cardiac mortality in the elderly: The Rotterdam Study. Circulation, 97(5), 467–472. 10. Okin, P. M., Devereux, R. B., Howard, B. V., Fabsitz, R. R., Lee, E. T., & Welty, T. K. (2000). Assessment of QT interval and QT dispersion for prediction of all-cause and cardiovascular mortality in American Indians: The Strong Heart Study. Circulation, 101(1), 61–66. 11. Clancy, C. (2011). Electrophysiologic and electrocardiographic principles. In L. S. Nelson, N. A. Lewin, M. A. Howland, R. S. Hoffman, L. R. Goldfrank, & N. E. Flomenbaum (Eds.), Goldfrank’s Toxicologic Emergencies (pp. 314–329). NewYork: McGraw-Hill Press. 12. Huang, C. Y. C. H. (2005). Carbon monoxide Intoxication. Annals of Disaster Medicine, 4, S8–S16.
13. Gu¨rkan, Y., Canatay, H., Toprak, A., Ural, E., & Toker, K. (2002). Carbon monoxide poisoning—a cause of increased QT dispersion. Acta Anaesthesiologica Scandinavica, 46(2), 180–183. 14. Yelken, B., Tanriverdi, B., Cetinbas, F., Memis, D., & Sut, N. (2009). The assessment of QT intervals in acute carbon monoxide poisoning. Anadolu Kardiyoloji Dergisi, 9, 397–400. 15. Hancı, V., Ayog˘lu, H., Yurtlu, S., Yıldırım, N., Okyay, D., Erdog˘an, G., et al. (2011). Effects of acute carbon monoxide poisoning on the P-wave and QT interval dispersions. Anadolu Kardiyoloji Dergisi, 1, 48–52. 16. Fox, N., Hunn, A., & Mathers, N. (2013). Sampling and sample size calculation. National Institute for Health Research. Available form http://rds-eastmidlands.nihr.ac.uk/resources/doc_download/ 9-sampling-and-sample-size-calculation.html. Accessed July 9, 2013. 17. Dan, M., & Roden, M. D. (2004). Drug-induced prolongation of the QT interval. New England Journal of Medicine, 350, 1013–1022. 18. SAPS II calculator. (2013). OPUS 12 Foundation, Inc. Available from opus12.org/SAPS_II.html. Accessed July 9, 2013. 19. APACHE II calculator-GlobalRPH. (2013). Available from: http://www.globalrph.com/apacheii.html. Accessed July 9, 2013. 20. Wang, C. H., Chen, C. L., Hsiao, C. K., Yıldırım, N., Okyay, D., Erdog˘an, G., et al. (2010). Arsenic-induced QT dispersion is associated with atherosclerotic diseases and predicts long-term cardiovascular mortality in subjects with previous exposure to arsenic: A 17-year follow-up study. Cardiovascular Toxicology, 10(1), 17–26. 21. Liu, H. K. (2004). Clinical study of QT dispersion in tetramine poisoning. Journal of North Sichuan Medical College, 4. 22. Zhou, C. Y., Liang, Z. Q., Liu, R. S. (2004). Significance of QT dispersion for prognosis in the critical illness. Chinese Journal of Critical Care Medicine, 12. 23. Manini, A. F., Nelson, L. S., Skolnick, A. H., Slater, W., & Hoffman, R. S. (2010). Electrocardiographic predictors of adverse cardiovascular events in suspected poisoning. Journal of Medical Toxicology, 6(2), 106–115.
123
