European Journal of Heart Failure (2011) 13, 1238–1243 doi:10.1093/eurjhf/hfr091
Exposure to sennoside–digoxin interaction and risk of digoxin toxicity: a population-based nested case–control study† Meng-Ting Wang 1*, I-Hsun Li 1,2, Wan-Ju Lee 1, Tien-Yu Huang 3, Hsin-Bang Leu 4, and Agnes L.F. Chan 5* 1 School of Pharmacy, National Defense Medical Center, Taipei, Taiwan, Republic of China; 2Department of Pharmacy Practice, Tri-Service General Hospital, Taipei, Taiwan, Republic of China; 3Division of Gastroenterology, Department of Internal Medicine, Tri-Service General Hospital, Taipei, Taiwan, Republic of China; 4Division of Cardiology, Healthcare and Management Center, Taipei Veterans General Hospital, Taipei, Taiwan, Republic of China; and 5Department of Pharmacy, Chi Mei Medical Center, Tainan, Taiwan, Republic of China
Received 27 December 2010; revised 22 March 2011; accepted 22 April 2011; online publish-ahead-of-print 28 July 2011
Aims
Digoxin is an important medication for heart failure (HF) patients and sennosides are widely used to treat constipation. Recently, safety concerns have been raised about a possible interaction between sennosides and digoxin, an issue that has not been studied empirically. This study therefore aimed to evaluate whether exposure to sennoside– digoxin interaction is associated with an increased risk of digoxin toxicity. ..................................................................................................................................................................................... Methods This was a population-based nested case–control study that analysed data obtained from the Taiwan National Health and results Insurance Research Database between 1 January 2001 and 31 December 2004. All HF patients treated with digoxin for the first time were included as the study cohort. Of these, cases were identified as subjects hospitalized for digoxin toxicity (International Classification of Diseases, Ninth Revision, Clinical Modification, ICD-9-CM 972.1), and matched to randomly selected controls. Use of sennosides was compared between the two groups. Odds ratios (ORs) were employed to quantify the risk associated with exposure to sennoside– digoxin interaction by conditional logistic regression. The study cohort comprised 222 527 HF patients, of whom 524 were identified as cases and 2 502 as matched controls. Use of sennosides during the 14 days preceding the index date was found to be associated with a 1.61-fold increased risk of digoxin toxicity [95% confidence interval (CI) ¼ 1.15, 2.25]. Additionally, a greater risk was observed for sennosides prescribed at an average daily dose ≥24 mg (adjusted OR ¼ 1.93; 95% CI ¼ 1.27, 2.94). ..................................................................................................................................................................................... Conclusion The combined use of sennosides and digoxin was found to be associated with a modest increased risk of digoxin toxicity in HF patients.
----------------------------------------------------------------------------------------------------------------------------------------------------------Keywords
Digoxin intoxication † Drug safety † Sennoside–digoxin interaction † Nested case–control study
Introduction Digoxin still plays an important role in the treatment of symptomatic heart failure (HF).1 Studies have demonstrated that patients treated with low doses of digoxin and who have low-serum digoxin concentrations have reduced mortality and HF-specific hospitalizations.2 – 5 Currently, the use of digoxin is recommended in HF patients with impaired left ventricular function.1,6 However, the well-known toxic effects of digoxin are not uncommon, and †
include confusion, anorexia, visual disturbances, as well as rhythm disturbances such as atrioventricular block, atrial arrhythmias, and ventricular tachycardia.7 Observational population-based studies have found that the risk of digoxin toxicity can be significantly increased by the concomitant use of drugs such as clarithromycin, diuretics, and sertraline.8 – 10 It has also been suggested that increased potassium loss caused by herbal laxatives might potentiate the toxicity of digoxin;11,12 however, empirical evidence is not currently available.
Part of this manuscript had been presented at the 4th Asian Conference on Pharmacoepidemiology, Tainan, Taiwan, 23 – 25October 2009.
* Corresponding author. 9 F, No.161, Section 6, MinChuan East Road, Taipei 114, Taiwan, Tel: +886 2 8792 3100 ext. 18870, Fax: +886 2 8792 3169, Email:
[email protected].
edu.tw (M-T.W.), No. 901, Chung-Hwa Road, Young Kang City, Tainan 710, Taiwan, Tel: +886 6 2812811 ext. 53109, Email:
[email protected] (A.L.F.C.) Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2011. For permissions please email:
[email protected].
1239
Sennoside– digoxin interactions
Sennosides are widely used as stimulant herbal laxatives due to their relatively low cost, natural origin, and ready availability without the need for a prescription.13 Sennosides are hydroxyanthracene glycosides and are contained in the leaves and fruit of Cassia acutifolia and Cassia angustifolia. They are hydrolysed by the bacterial flora in the large intestine and converted to the active metabolite rhein anthrone, which induces a laxative effect through modification of intestinal motility and accumulation of intraluminal fluid. Sennosides are generally considered safe drugs;14 however, safety concerns have recently been raised about an interaction between sennosides and digoxin.11 In theory, sennosides may increase the risk of digoxin toxicity during digoxin therapy. Direct potassium loss in the faeces and sodium loss causing a secondary renal effect are possible underlying mechanisms which may account for sennoside-induced hypokalaemia.15 In addition, it has been reported that hypokalaemia is a significant factor in digoxin toxicity; therefore, an interaction between sennosides and digoxin is believed to exist and is mentioned in many drug-interaction compendiums11,12 and computerized drug information databases such as Micromedex.16 However, to the best of our knowledge, there has been no previous research to evaluate the adverse clinical consequences of sennoside– digoxin interaction. We therefore conducted a population-based nested case –control study to examine the risk of digoxin toxicity associated with the use of sennosides among HF patients treated continuously with digoxin.
Methods Data source This study employed a population-based nested case – control study design, utilizing data from the National Health Insurance Research Database (NHIRD). The NHIRD contains inpatient, outpatient, and pharmacy claims made by over 99% of the 23 million inhabitants of Taiwan under the National Health Insurance (NHI) programme. Specifically, detailed information obtained from the database included encrypted identification number, medications prescribed, gender, date of visit, length of hospitalization, and diagnoses based on the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM). The NHIRD has been widely used to study drug safety.17 A five-year dataset of NHIRD that comprised claims data for all patients diagnosed with HF between 1 January 2000 and 31 December 2004 was retrieved and analysed for this study.
Identification of study cohort The study cohort comprised patients with a diagnosis of HF who started to use digoxin continuously during the five-year study period. Specifically, diagnoses of HF in both outpatient and inpatient settings were identified from the validated ICD-9-CM codes 428.xx, 398.91, 402.01, 402.11, 402.91, 404.00, 404.01, 404.03, 404.10, 404.11, 404.13, 404.90, 404.91, and 404.93.18 New use of digoxin in 2001– 2004 was defined as no previous prescription for digoxin during the previous year, and the date of new use of digoxin following HF diagnosis served as the date for cohort entry. We excluded patients whose first prescription date for digoxin preceded the first diagnosis date of HF and patients who had received any previous prescriptions for digoxin during the 1 year before the cohort entry date, to define new users of digoxin. The study cohort was then followed
until hospitalization for digoxin toxicity (ICD-9-CM 972.1), discontinuation of digoxin, or the end of the study period, whichever came first. Patients were considered to have discontinued digoxin therapy if the number of days between the calculated end date of a digoxin prescription and the start date of the following digoxin prescription exceeded 30 days. For patients who discontinued digoxin, the observation period was extended to one month after the last prescription for digoxin, because digoxin toxicity may occur even after cessation of digoxin therapy. A similar cohort has been analysed and reported previously for evaluation of the adverse clinical consequences of exposure to digoxin– clarithromycin interactions8 and digoxin– diuretic interactions.9
Case definition and selection of controls Case patients were identified as those hospitalized for digoxin intoxication (ICD-9-CM 972.1), and the index date was taken as the date of hospital admission for digoxin intoxication. Only the first hospitalization was considered for analysis in patients who had more than one episode of digoxin toxicity. Using an incidence density sampling approach, each case patient was matched with up to five controls nested within the study cohort by age (+5 years), sex, cohort entry date (+6 months), occurrence of chronic renal disease (ICD-9-CM code 585.xx) in the six months prior to the index date, use of potassium-depleting diuretics, and use of P-glycoprotein inhibitors in the 3 months before the index date. The specific drugs defined as potassium-depleting diuretics and P-glycoprotein inhibitors are shown in Supplementary material online, Table S1. Each control was assigned the same index date as the corresponding case patient.
Measurement of sennoside use Any use of sennosides during 1 – 14, 15 – 30, 31 –90, and 91 – 365 days before the index date was evaluated. Study subjects who did not use sennosides in the year prior to the index date were defined as non-users. For those patients prescribed sennosides in the 14 days prior to the index date, the average daily sennoside dose prescribed during the 1 month prior to the index date was calculated and categorized into three levels for evaluating the dose– response relationship: ≤12, 12.1– 23.9, and ≥24 mg. We categorized the dose of sennosides using 12 mg as the cut-off point because in .60% of the sennoside prescriptions evaluated in this study each tablet contained 12 mg of sennoside. The use of cephalexin was also examined to test the specificity of the main analysis, because cephalexin is known not to interact with digoxin.
Statistical analysis The association between sennoside use and risk of digoxin toxicity requiring hospitalization was evaluated by conditional logistic regression, which provided estimates of odds ratios (ORs). We made adjustments for previous hospitalization for HF, and any diagnosis of ventricular arrhythmias (ICD-9-CM 410.xx, 412.xx)19 measured during the 1-year period prior to cohort entry. Prescriptions of medications that might affect potassium levels20 or the metabolism of digoxin, were recorded in the 3 months before the index date and served as covariates in the multivariate models; these medications comprised potassium supplements, angiotensin-converting enzyme inhibitors (ACEIs), b-adrenergic antagonists, non-steroidal antiinflammatory drugs, medications that increase renal potassium loss and medications that might cause transcellular potassium shift (see Supplementary material online, Table S1). In addition, the average daily dose of digoxin was calculated as the ratio of prescribed daily dose to defined daily dose in the 3 months before the index date.
1240 We also adjusted for comorbidities present in the 6 months prior to the index date, including myocardial infarction (ICD-9-CM 410.xx, 412,xx), chronic renal disease (ICD-9-CM 585.xx) and diabetes mellitus (ICD-9-CM 250.xx),21 which may alter the digoxin response, leading to digoxin toxicity. The software programmes SAS version 9.1 (SAS Institute, Cary, NC, USA) and STATA version 10.0 (STATA, College Station, TX, USA) were used for data management and statistical analyses, respectively.
M.-T. Wang et al.
Table 1 Demographic and clinical characteristics of cases and matched controls Cases (n 5 516)
Controls (n 5 2502)
P-value
................................................................................ Agea, mean + SD
76.7 + 11.9
76.7 + 10.5
Malea Comorbidity
186 (36.1)
898 (35.9)
0.967 NAb
Sensitivity analysis
Chronic renal diseasea
69 (13.4)
267 (10.7)
0.053
Three additional analyses were conducted to evaluate the robustness of the main results under the conditions examined. First, we limited cases and controls to those with no previous episodes of HF-related hospitalization. Second, we further adjusted for the use of three other types of laxative (bulk-forming, osmotic, and stimulant laxatives) during multivariate analyses. Third, we specified cases and matched controls to those with use of potassium-sparing diuretics in the 3 months prior to the index date.
Myocardial infarction Diabetes mellitus
44 (8.5) 156 (30.2)
143 (5.7) 658 (26.3)
0.012 0.061
Ventricular arrhythmias
106 (20.5)
500 (20.0)
0.659
83 (16.1)
238 (9.5)
,0.001
0.9 + 0.8
0.6 + 0.4
,0.001
450 (87.2)
2182 (87.2)
141 (27.3)
557 (22.3)
278 (53.9)
1182 (47.2)
0.008
99 (19.2)
354 (14.2)
0.004
P-glycoprotein inhibitorsa
200 (38.8)
968 (38.7)
Increased renal potassium loss
91 (17.6)
459 (18.4)
0.665
Transcellular potassium shift
219 (42.4)
1047 (41.9)
0.961
NSAIDs
255 (49.4)
1134 (45.3)
0.096
Results We identified 222 527 patients with a diagnosis of HF and at least one digoxin prescription from the NHIRD during the study period. Of these, 49 402 patients were excluded due to fulfilment of the exclusion criteria. The study cohort therefore consisted of 173 125 HF patients continuously treated with digoxin. From these patients a total of 524 cases were identified; however, eight cases could not be matched with any control. Accordingly, there were 516 case patients and 2502 matched controls. Table 1 shows the characteristics of the cases and matched controls. The matching variables and the examined comorbidities were similar in the two groups, except for a higher prevalence of myocardial infarction and previous HF-related hospitalization observed in the cases. A greater proportion of cases received a prescription for potassium supplements (P ¼ 0.009), ACEIs (P ¼ 0.008), and b-adrenergic blockers (P ¼ 0.004). Cases were also found to receive a higher dose of digoxin than controls (P , 0.001). The risk of digoxin toxicity associated with use of sennosides over various periods of time before the index date is presented in Table 2. Use of sennosides in the 14 days preceding the index date was found to be statistically associated with a 1.61-fold (95% CI ¼ 1.15, 2.25; P ¼ 0.005) increase in the risk of digoxin toxicity requiring hospitalization after adjustment for the covariates. However, the risk diminished and was found to be nonsignificant when sennosides were taken between 15 and 30 days, 31 and 90 days, or 91 and 365 days before the index date. In addition, the use of cephalexin was not associated with an increased risk of digoxin toxicity requiring hospitalization across the four different periods of time. The use of sennosides in HF patients treated with digoxin is associated in a dose-dependent manner with risk of hospitalization for digoxin toxicity (Figure 1). An average daily dose of sennosides ≤12 mg or between 12.1 and 23.9 mg was not found to be associated with an increased risk of digoxin toxicity requiring hospitalization. However, an increased risk was observed for sennosides prescribed at an average daily dose .24 mg (adjusted OR ¼ 1.93; 95% CI ¼ 1.27, 2.94). Trend analysis also indicated a gradually
Previous HF-related hospitalization Medication history Doses of digoxin, PDD:DDD, mean + SD Potassium-depleting diureticsa Potassium supplements ACEIs b-adrenergic blockers
NAb 0.009
NAb
Values are expressed as n (%). HF, heart failure; PDD, prescribed daily dose; DDD, defined daily dose; ACEIs, angiotensin-converting enzyme inhibitors; NSAIDs, non-steroidal anti-inflammatory drugs; NA, not applicable; SD, standard deviation. a The matching variables. b P-values are not able to be provided due to no within-group variance.
increasing trend in the adjusted ORs across the three strata of average daily dose of sennosides (P , 0.001). The main results were robust in the sensitivity analyses for specification of subjects to: those with no previous HF hospitalization, those who were prescribed potassium-sparing diuretics in the 3 months before the index date, and inclusion of the three types of laxatives as three additional covariates in the multivariate analyses (Figure 2).
Discussion In this large population-based nested case– control study, we found that exposure to sennoside–digoxin interaction in HF patients leads to an adverse clinical outcome in the present clinical setting. Specifically, the increased risk of digoxin toxicity associated with the use of sennosides is concentrated in the first 2 weeks of treatment among HF patients receiving digoxin. Dose– response
1241
Sennoside– digoxin interactions
Table 2 Risk of digoxin toxicity associated with sennosides use in heart failure patients receiving digoxin Time interval before index date
Controls (n 5 2502)
Univariate OR (95% CI)
Multivariate ORa (95% CI)
59 (11.4) 20 (3.9)
187 (7.5) 79 (3.2)
1.63 (1.19, 2.24)b 1.31 (0.79, 2.18)
1.61 (1.15, 2.25)b 1.30 (0.76, 2.22)
19 (3.7) 11 (2.1)
104 (4.2) 41 (1.6)
0.93 (0.56, 1.54) 1.39 (0.70, 2.76)
0.97 (0.58, 1.65) 1.43 (0.70, 2.93)
18 (3.5) 65 (12.6)
109 (4.4) 264 (10.6)
0.80 (0.48, 1.35) 1.24 (0.92, 1.69)
0.83 (0.49, 1.41) 1.20 (0.86, 1.69)
18 (3.5) 105 (20.4)
82 (3.3) 539 (21.5)
1.11 (0.66, 1.89) 0.97 (0.76, 1.24)
1.03 (0.59, 1.78) 0.93 (0.71, 1.21)
Cases (n 5 516)
............................................................................................................................................................................... 1 –14 days, n (%) Sennosides Cephalexin 15– 30 days, n (%) Sennosides Cephalexin 31– 90 days, n (%) Sennosides Cephalexin 91– 365 days, n (%) Sennosides Cephalexin
CI, confidence interval; OR, odds ratio. a Adjusted for any diagnosis of chronic renal disease, myocardial infarction, diabetes mellitus, ventricular arrhythmias, previous hospitalization for heart failure, doses of digoxin, any prescription of potassium supplements, angiotensin-converting enzyme inhibitors, b-adrenergic antagonists, non-steroidal anti-inflammatory drugs, and medications that increase renal potassium loss or cause transcelluar potassium shift. b P , 0.05.
Figure 1 Risk of digoxin toxicity by average daily doses of sennosides.
analysis also indicated that the risk of digoxin toxicity increased with an increasing daily dose of sennosides, which lends further support to the causal relationship. This investigation provides empirical data about the safety of combined use of sennosides and digoxin, which is of great importance because hypothetically, sennosides are thought to interact with digoxin.16 Although the cause of the sennoside–digoxin interaction is currently unknown, an imbalance in electrolytes, particularly hypokalaemia, might be the plausible mechanism accounting for the observed increased risk of digoxin toxicity caused by sennosides.11,12 Specifically, use of sennosides is associated with hypokalaemia partly due to direct potassium loss in the faeces and also due to secondary renal effects resulting from sodium loss.12 On the other hand, digoxin reversely inhibits sodium –potassium ATPase (Na+/K+ pump) which is the same binding site for extracellular potassium to compete with.22,23 Thus, hypokalaemia
induced by sennosides may lead to a further inhibitory effect of digoxin on the Na+/K+ pump, resulting in an increased risk of digoxin toxicity. We identified a significantly increased risk of digoxin toxicity requiring hospitalization only when sennosides were administered during the 14 days prior to the index date. Kohvakka et al. 24 reported that 46% of 134 digitalized HF patients who had been treated previously with hydrochlorothiazide for 6 weeks had a serum potassium level ,3.6 mmol/L, indicating that HF patients are prone to hypokalaemia while receiving digoxin and diuretics. Similarly, in our study the examined cases and controls may have been prone to hypokalaemia because around 85% of them received potassium-depleting diuretics. Additionally, patients taking sennosides for .1– 2 weeks may experience hypokalemia.15 Consequently, the characteristics of digitalized patients and the onset of hypokalaemia caused by sennosides after their use for 1 –2 weeks, might subsequently have led to digoxin toxicity in our study subjects. Those patients who tolerate exposure to sennosides within the 2 weeks of treatment prior to the index date may be less susceptible to digoxin toxicity caused by the sennoside –digoxin interaction. Due to the rapid onset of adverse clinical consequences, physicians should be cautious about prescribing sennosides during the early stages of treatment in HF patients taking digoxin. Combined therapy with inappropriately-used sennosides and digoxin might threaten patient safety. Sennosides are widely available not only on prescription but also as an over-the-counter (OTC) laxative.25 Even though sennosides are not recommended to be used for .8–10 days when used as a self-prescribed medication,26 inappropriate use of sennosides has been reported.25,27 The´ophile et al. 27 reported that 31.4% of laxative users including sennoside users take self-prescribed laxatives daily for .1 year, indicating that chronic use of OTC laxatives is still a major issue.
1242
M.-T. Wang et al.
Figure 2 Risk of digoxin toxicity in the sensitivity analyses.
Therefore, prolonged use of sennosides with concomitant use of digoxin raises safety concerns, including inappropriate use of selfprescribed sennosides. Our findings deserve public notice since exposure to sennoside –digoxin interaction is not rare in HF patients. Constipation is a common gastrointestinal disorder, ranging in prevalence from 1.9 to 27%, with most estimates in North America being in the range of 12– 19%.28 The use of laxatives increases with increasing age, particularly in those aged over 65 years.29 In this study, 24% of the study cohort were prescribed sennosides at least once during the entire follow-up period. In addition, digoxin is still one of the most frequently prescribed medications for HF in the elderly.30 Therefore, HF patients might be more likely to encounter sennoside–digoxin interaction than those with other diseases, and attention should be paid to these patients to avoid exposure to combined sennoside and digoxin therapy.
Strengths and limitations This study has several unique attributes. To our knowledge, it is the first to empirically evaluate the clinical adverse consequences of interaction between digoxin and sennosides (laxatives) in a population-based manner. Specifically, all HF patients treated continuously with digoxin for the first time during the study period were identified as the study cohort, from which all incident cases of digoxin toxicity were evaluated for exposure to sennoside– digoxin interaction. In addition, the use of a nested case –control study design not only attenuated selection bias, but also enabled us to analyse a large number of cases of digoxin toxicity requiring hospitalization. Furthermore, the causal inference is strengthened based on the observed dose–response relationship between sennoside use and risk of digoxin toxicity requiring hospitalization. Several limitations of this study need to be addressed. First, sennosides are available as OTC drugs in Taiwan, and misclassification was possible for both the cases and controls. However, misclassification should be random, and is believed to be non-differential between the cases and controls, which tends to attenuate our findings; in other words, the magnitude of the observed risk could have
been underestimated. Second, the dose–response relationship could not be comprehensively examined due to the limited variation in the doses of sennosides used. Third, despite the fact that the cases and controls were identified from the same cohort, the existence of selection bias cannot be ruled out. Fourth, information on medication compliance, serum digoxin concentrations, potassium levels, and laboratory data regarding renal function and HF severity could not be obtained from the database. Finally, digoxin is also indicated for the treatment of atrial fibrillation; however, since this study was aimed at HF patients the findings are not generalizable to all digoxin users.
Conclusions and implications This large observational study confirms that the combined use of sennosides with digoxin is associated with a modestly increased risk of digoxin toxicity requiring hospitalization in HF patients. Health care professionals should therefore be vigilant when assessing the presence of any symptoms of digoxin toxicity, monitor serum digoxin concentrations closely, and titrate doses of digoxin as necessary for the concomitant use of sennosides.
Supplementary material Supplementary material is available at European Journal of Heart Failure online.
Acknowledgements The interpretation and conclusions contained herein do not represent those of Bureau of National Health Insurance, Department of Health or National Health Research Institutes, Taiwan. We also thank Dr Yu-Ching Chou for verification of the statistics we employed and data we presented.
Funding This study was supported in part by a grant funded by Department of Health, Taiwan (DOH96 – TD– 113 – 028), and the data were retrieved
Sennoside– digoxin interactions
from the National Health Insurance Research Database provided by the Bureau of National Health Insurance. Conflict of interest: All authors declare no conflicts of interests relevant to this study.
References 1. Dickstein K, Cohen–Solal A, Filippatos G, McMurray JJ, Ponikowski P, Poole-Wilson PA, Stromberg A, van Veldhuisen DJ, Atar D, Hoes AW, Keren A, Mebazaa A, Nieminen M, Priori SG, Swedberg K. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2008. Eur J Heart Fail 2008;10:933 –989. 2. Ahmed A, Rich MW, Love TE, Lloyd-Jones DM, Aban IB, Colucci WS, Adams KF, Gheorghiade M. Digoxin and reduction in mortality and hospitalization in heart failure: a comprehensive post hoc analysis of the DIG trial. Eur Heart J 2006;27: 178 –186. 3. Ahmed A. Digoxin and reduction in mortality and hospitalization in geriatric heart failure: importance of low doses and low serum concentrations. J Gerontol A Biol Sci Med Sci 2007;62:323 – 329. 4. Ahmed A, Pitt B, Rahimtoola SH, Waagstein F, White M, Love TE, Braunwald E. Effects of digoxin at low serum concentrations on mortality and hospitalization in heart failure: a propensity-matched study of the DIG trial. Int J Cardiol 2008;123: 138 –146. 5. Rathore SS, Curtis JP, Wang Y, Bristow MR, Krumholz HM. Association of serum digoxin concentration and outcomes in patients with heart failure. JAMA 2003; 289:871 –878. 6. Vaz Pe´rez A, Turan RG, Rauchhaus M. Digoxin in chronic heart failure: possibility of a second chance? QJM 2011;104:369 – 371. 7. Williamson KM, Thrasher KA, Fulton KB, LaPointe NM, Dunham GD, Cooper AA, Barrett PS, Patterson JH. Digoxin toxicity: an evaluation in current clinical practice. Arch Intern Med 1998;158:2444 –2449. 8. Chan ALF, Wang MT, Su CY, Tsai FH. Risk of digoxin intoxication caused by clarithromycin–digoxin interactions in heart failure patients: a population-based study. Eur J Clin Pharmacol 2009;65:1237 –1243. 9. Wang MT, Su CY, Chan ALF, Lian PW, Leu HB, Hsu YJ. Risk of digoxin intoxication in heart failure patients exposed to digoxin –diuretic interactions: a population-based study. Br J Clin Pharmacol 2010;70:258 –267. 10. Juurlink DN, Mamdani MM, Kopp A, Herrmann N, Laupacis A. A populationbased assessment of the potential interaction between serotonin-specific reuptake inhibitors and digoxin. Br J Clin Pharmacol 2005;59:102 – 107. 11. Valli G, Giardina EGV. Benefits, adverse effects and drug interactions of herbal therapies with cardiovascular effects. J Am Coll Cardiol 2002;39:1083 –1095. 12. Blumenthal M, Busse WR, Goldberg A, Gruenwald J, Hall T, Klein S, Riggins CW, Rister RS. The Complete German Commission E Monographs. 1st ed. Austin, USA: American Botanical Council; 1998. 13. Bouras EP, Tangalos EG. Chronic constipation in the elderly. Gastroenterol Clin North Am 2009;38:463 –480.
1243 14. Mascolo N, Capasso R, Capasso F. Senna. A safe and effective drug. Phytother Res 1998;12(Suppl 1):S143 –S145. 15. Westendorff J. Anthranoid derivatives-general discussion. In: De Smet PAGM, Keller K, Hansel R, Chandler RF, eds. Adverse Effects of Herbal Drugs. Berlin, Germany: Springer-Verlag; 1993. p105 –118. 16. Micromedexw Healthcare Series, (electronic version). Thomson Micromedex, Greenwood Village, CO, USA. http://www.thomsonhc.com (8 July 2010). 17. Lee CH, Wang JD, Chen PC. Increased risk of hospitalization for acute hepatitis in patients with previous exposure to NSAIDs. Pharmacoepidemiol Drug Saf 2010;19: 708 –714. 18. McCullough PA, Philbin EF, Spertus JA, Kaatz S, Sandberg KR, Weaver WD. Confirmation of a heart failure epidemic: findings from the Resource Utilization Among Congestive Heart Failure (REACH) study. J Am Coll Cardiol 2002;39: 60 – 69. 19. De Bruin ML, van Hemel NM, Leufkens HG, Hoes AW. Hospital discharge diagnoses of ventricular arrhythmias and cardiac arrest were useful for epidemiologic research. J Clin Epidemiol 2005;58:1325 –1329. 20. Gennari FJ. Hypokalemia. N Engl J Med 1998;339:451 –458. 21. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saunders LD, Beck CA, Feasby TE, Ghali WA. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005;43:1130 –1139. 22. Kjeldsen K, Norgaard A, Gheorghiade M. Myocardial Na, K-ATPase: the molecular basis for the hemodynamic effect of digoxin therapy in congestive heart failure. Cardiovasc Res 2002;55:710 –713. 23. Katzung BG, Parmley WW. Drugs used in heart failure. In Katzung BG. Basic & Clinical Pharmacology. 9th ed. New York, USA: McGraw Hill; 2004. p. 201 –215. 24. Kohvakka A, Heinonen L, Pietinen P, Salo H, Eisalo A. Potassium and magnesium balance in thiazide-treated cardiac patients with special reference to diet. Acta Med Scand Suppl 1982;668:102 –109. 25. Motola G, Mazzeo F, Rinaldi B, Capuano A, Rossi S, Russo F, Vitelli MR, Rossi F, Filippelli A. Self –prescribed laxative use: a drug-utilization review. Adv Ther 2002; 19:203–208. 26. European Medicines Agency. Overview of comments received on community herbal monograph. EMEA/HMPC/342482/2006, EMEA/HMPC/237716/ 2006, EMEA/HMPC/342513/2006. London; 2007. http://www.emea.europa. eu/pdfs/human/hmpc/aloe/34248206en.pdf, http://www.emea.europa.eu/pdfs/ human/hmpc/sennae_fructus/23771606en.pdf, http://www.emea.europa.eu/pdfs/ human/hmpc/frangulae_cortex/34251306en.pdf (8 July 2010) 27. The´ophile H, Miremont– Salame´ G, Abouelfath A, Be´gaud B, Haramburu F. Patterns of laxative use in self-medication. Ann Pharmacother 2009;43:2122 –2123. 28. Higgins PDR, Johanson JF. Epidemiology of constipation in North America: a systematic review. Am J Gastroenterol 2004;99:750 –759. 29. Bosshard W, Dreher R, Schnegg JF, Bu¨la CJ. The treatment of chronic constipation in elderly people: an update. Drugs Aging 2004;21:911 –930. 30. Pervaiz MH, Dickinson MG, Yamani M. Is digoxin a drug of the past?. Cleve Clin J Med 2006;73:821 –834.
