Digoxin in advanced heart failure patients: a question of rhythm

Re Po Ca diol. 2013;32(4):303---310 Re is a Po uguesa de Ca diologia Po uguese Jou nal o Ca diology www. e po ca diol.o g ORIGINAL ARTICLE Digoxin in ad anced hea ailu e pa ien s: A ques ion o hy hm Elisabe e Jo ge∗, Rui Bap is a, Hélia Ma ins, Fá ima Sa ai a, Susana Cos a, Hen ique Viei a, Lou enc¸o Coelho, Ped o Mon ei o, Fá ima F anco, Luís A. P o idência Hospi ais da Uni e sidade de Coimb a e Clínica Uni e si á ia de Ca diologia da Faculdade de Medicina da Uni e sidade de Coimb a, Coimb a, Po ugal Recei ed 6 Oc obe 2011; accep ed 5 No embe 2012 A ailable online 23 Ma ch 2013 KEYWORDS Ad anced hea ailu e; Digoxin; A ial fib illa ion; P ognosis Abs ac Backg ound: The impac o digoxin on ou comes o pa ien s wi h ad anced hea ailu e (HF) emains unce ain and i s e ec may be di e en o pa ien s in a ial fib illa ion (AF) o sinus hy hm (SR). Objec i es: To de e mine he impac o digoxin on ou comes o ad anced HF pa ien s and o assess whe he p ognosis di e s in pa ien s in AF and SR. Me hods: A o al o 268 consecu i e pa ien s admi ed o an in ensi e ca e uni wi h decom- pensa ed HF we e e alua ed. Pa ien s we e di ided in o wo g oups: A --- pa ien s wi h AF (n=89), and B --- pa ien s in SR (n=179). Fo each g oup we compa ed pa ien s medica ed and no medica ed wi h digoxin. A mean ollow-up o 3.3 yea s was pe o med. Resul s: Addi ion o digoxin o con empo a y s anda d HF he apy showed no impac on mo al- i y o pa ien s in g oup B (all-cause mo ali y in ollow-up: 19.1% s. 22.5%, p=0.788). Rega ding g oup A, we obse ed significan ly lowe medium- e m mo ali y o pa ien s on digoxin he apy (18.6% s. 46.6%, p=0.048). Digoxin he apy did no influence eadmissions o decompensa ed HF. Among AF pa ien s, no di e ences we e ound ega ding demog aphic, clinical, echo- ca diog aphic and labo a o y a iables be ween pa ien s medica ed and no medica ed wi h digoxin. Conclusions: Digoxin he apy may imp o e he p ognosis o ad anced HF pa ien s wi h AF unde op imal medical he apy. Howe e , no benefi o digoxin was demons a ed o pa ien s in SR. These esul s may help o imp o e pa ien selec ion o digoxin he apy. © 2011 Sociedade Po uguesa de Ca diologia Published by Else ie España, S.L. All igh s ese ed. ∗Co esponding au ho . E-mail add ess: [email p o ec ed] (E. Jo ge). 0870-2551/$ – see on ma e © 2011 Sociedade Po uguesa de Ca diologia Published by Else ie España, S.L. All igh s ese ed. h p://dx.doi.o g/10.1016/j. epc.2012.11.007 304 E. Jo ge e al. PALAVRAS-CHAVE Insuficiência ca díaca a anc¸ada; Digoxina; Fib ilhac¸ão au icula ; P ognós ico Digoxina na insuficiência ca díaca a anc¸ada: uma ques ão de i mo Resumo In oduc¸ão: O impac o p ognós ico da digoxina na insuficiência ca díaca (IC) a anc¸ada pe - manece mal escla ecido. A ele ância da e apêu ica digi álica pode se di e en e na fib ilhac¸ão au icula (FA) ela i amen e ao i mo sinusal (RS). Objec i os: De e mina o impac o p ognós ico da digoxina na IC e e ifica se es e é di e en e consoan e se encon em em FA ou em RS. Mé odos: Es uda am-se 268 doen es in e nados numa unidade de cuidados in ensi os po IC descompensada. Di idiu-se a populac¸ão em dois g upos: A --- 89 doen es em FA; g upo B --- 179 doen es em RS. Pa a cada g upo compa a am-se os doen es medicados com os não medicados com digoxina. Realizou-se um seguimen o clínico com a du ac¸ão mediana de 3,3 anos. Resul ados: A digoxina não e e impac o na mo alidade dos doen es com IC a anc¸ada que se encon a am em RS (mo alidade a 3,3 anos: 19,1% e sus 22,5%, p = 0,788), adicionada à e apêu ica médica o imizada. Nos doen es em FA obse ou-se uma educ¸ão significa i a da mo alidade nos doen es medicados com digoxina (18,6% e sus 46,6%, p = 0,048). A digoxina não al e ou a axa de ein e namen os po IC descompensada. No g upo A não se e ifica am di e enc¸as en e os doen es medicados e não medicados com digoxina ela i amen e aos pa âme os demog áficos, clínicos, ecoca diog áficos ou labo a o iais. Conclusão: Es e es udo suge e que a digoxina pode melho a o p ognós ico dos doen es com IC a anc¸ada e FA, sob e apêu ica médica op imizada. Con udo, es a não demons ou bene ício nos doen es em RS. Es es esul ados podem con ibui pa a uma melho selec¸ão dos doen es a medica com digoxina, o imizando a elac¸ão isco/bene ício des a e apêu ica. © 2011 Sociedade Po uguesa de Ca diologia. Publicado po Else ie España, S.L. Todos os di ei os ese ados. In oduc ion Despi e mo e han 200 yea s o esea ch, he ole o digoxin in hea ailu e he apy emains con o e sial. The Ame i- can College o Ca diology and Ame ican Hea Associa ion (ACC/AHA) guidelines o he managemen o pa ien s wi h hea ailu e (HF) ecommend he use o digoxin, unless con- aindica ed, o pe sis en symp oms a e op imal medical he apy.1Digoxin use is also endo sed by he Eu opean Soci- e y o Ca diology guidelines.2Al hough widely accep ed, hese ecommenda ions a e de i ed mainly om sho - e m s udies assessing non-mo ali y ou comes.3 --- 8 The Digi alis In es iga o G oup (DIG) s udy,9a la ge andomized ial, showed no impac on all-cause o ca dio ascula mo al- i y wi h digoxin in oligosymp oma ic male HF pa ien s in sinus hy hm (SR). Howe e , a educ ion in hospi aliza ions due o wo sening HF and an imp o emen in quali y o li e we e epo ed.10 Two o he ials showed ha wi hd awal o digoxin esul ed in wo sening HF symp oms.6,7 Al hough mo e han 90% o pa ien s we e ecei ing angio ensin- con e ing enzyme inhibi o s (ACEIs) in he DIG ial,9 he use o be a-blocke s and aldos e one an agonis s was no epo ed and likely was e y limi ed. Simila ly, imp o ed ou comes a e also seen wi h ca diac de ices, which a e now widely used in HF pa ien s. A ecen ly published s udy sugges s no benefi o digoxin in ad anced HF pa ien s on con empo a y medical he apy.11 In his epo , HF pa ien s in SR ea ed wi h digoxin had a wo se p ognosis han hose who we e no . Howe e , his did no appea o be ue o pa ien s in a ial fib illa ion (AF). In pa ien s wi h AF, he beneficial ole o digoxin is widely accep ed, mainly o op imize en icula a e con ol, and he Ca edilol A ial Fib illa ion E alua ion (CAFE) s udy sugges s ha he e may be syne gis ic benefi s be ween digoxin and ca edilol.11 In ou s udy, we sough o de e mine he impac o digoxin he apy in ad anced HF pa ien s on op imal con- empo a y medical he apy, and o assess whe he p ognosis di e s acco ding o he p esence o AF. Me hods Pa ien popula ion and da a collec ion This single-cen e s udy included 268 consecu i e pa ien s admi ed o an in ensi e ca e uni be ween Janua y 2003 and June 2006 due o decompensa ed HF in New Yo k Hea Associa ion (NYHA) class III o IV. S anda dized eco ds we e used o desc ibe he s udy popula ion in e ms o clinical and demog aphic cha ac e is ics, ca dio ascula isk ac o s and como bidi ies. The eco ds also showed HF e iology, hemodynamic s a us on admission, labo a o y pa ame- e s, elec oca diog aphic and echoca diog aphic da a, and ea men . Renal dys unc ion was defined as c ea inine clea ance o <60 ml/min calcula ed by he Cockc o ---Gaul o mula. Ischemic ca diomyopa hy was defined as le en- icula dys unc ion associa ed wi h significan co ona y disease (>70% s enosis in a leas one majo epica dial co o- na y a e y and/o >50% le main s enosis). In he mino i y o pa ien s wi hou co ona y angiog aphy (4%) he diagnos- ic c i e ion o ischemic hea disease was p io myoca dial in a c ion (his o y o in a c ion o in a c sca on ECG). Non- ischemic ca diomyopa hy was defined as le en icula dys unc ion in he absence o he abo e al e a ions. Co o- na y angiog aphy was pe o med in 70% o his g oup; an Digoxin in ad anced hea ailu e pa ien s 305 Table 1 Gene al cha ac e is ics o he s udy popula ion. (%) Sinus hy hm (n=179) A ial fib illa ion (n=89) SR s. AF No digoxin Digoxin To al p No digoxin Digoxin To al p p Male 62.8 76.5 73.2 0.078 95.7 69.7 76.4 0.012 0.600 Anemia 28.1 16.5 19.3 0.145 15.8 19.1 18.2 0.749 0.855 Dyslipidemia 54.8 35.5 39.7 0.051 52.4 28.0 35.2 0.500 0.524 COPD 18.2 13.5 14.6 0.503 26.3 31.3 68.8 0.736 0.114 Diabe es 36.4 29.1 30.8 0.427 26.3 36.4 33.8 0.425 0.651 S oke 10.0 6.0 6.9 0.449 5.3 6.7 6.3 0.832 0.860 PAD 13.8 15.5 15.1 0.825 5.6 15.9 12.9 0.270 0.689 Dialysis 2.4 2.9 2.8 0.666 0.0 4.5 3.4 0.141 0.968 Thy oid d. 6.5 11.4 10.3 0.423 0.0 29.4 22.1 0.011 0.023 Li e d. 0.0 5.9 4.6 0.179 5.6 0.0 1.6 0.111 0.291 AF: a ial fib illa ion; COPD: ch onic obs uc i e pulmona y disease; d: disease; PAD: pe iphe al a e ial disease; SR: sinus hy hm. ischemia s ess es (s ess echoca diog aphy, myoca dial pe usion scanning o pe usion magne ic esonance imag- ing) was a ailable in he emainde , which e ealed no sig- nifican al e a ions. A mean clinical ollow-up o 40 mon hs was pe o med. Clinical da a we e collec ed du ing pa ien s’ isi s o he ou pa ien clinic o by elephone in e iew. The occu ence o eadmission due o decompensa ed HF and dea h was eco ded. Medical eco ds we e e iewed o con- fi m and o ob ain addi ional in o ma ion. The popula ion was di ided in o wo g oups, acco ding o he p esence o AF: g oup A --- 89 pa ien s wi h AF and g oup B --- 179 pa ien s in sinus hy hm. A p e iously defined subg oup analysis was pe o med in pa ien s in SR s. AF. Fo each g oup pa ien s medica ed and no medica ed wi h digoxin we e compa ed. Endpoin s The p ima y endpoin was long- e m all-cause mo ali y and he seconda y endpoin was a combined ou come o mo al- i y and ehospi aliza ion due o decompensa ed HF. S a is ical analysis The s a is ical analysis was pe o med wi h SPSS 13.0 o Windows (SPSS Inc., Chicago, Ill.). The esul s we e exp essed as means ± s anda d de ia ion o con inuous a iables and as equencies and pe cen ages o ca ego ical a iables. The equencies o he ca ego ical a iables in he wo g oups we e compa ed by he chi-squa e es o Fishe ’s exac es . Con inuous a iables we e compa ed using he S uden ’s unpai ed es (no mal a iables) and he Mann---Whi ney U es (non-no mal a iables). Kaplan---Meie su i al analysis and he B eslow es we e used o compa e he g oups in e ms o cumula i e su i al. A alue o p<0.05 was conside ed s a is ically significan . Resul s Gene al cha ac e is ics o he s udy g oups The baseline cha ac e is ics o he s udy popula ion a e lis ed in Table 1. O he o al popula ion hospi alized due o ad anced HF du ing he s udy pe iod, 33.2% (89/268) p esen ed wi h AF. These pa ien s we e olde (62.4±12.8 s. 55.1±15.4 yea s; p<0.001), wi h highe le en icula ejec ion ac ion (LVEF) (31.0% s. 26.0%, p=0.003) and a highe p e alence o al ula disease as HF e iology (3.4% s. 13.6%; p<0.001). The p e alence o hy oid disease was highe in AF pa ien s (22.1% s. 10.3%; p=0.023). Among he AF pa ien s, he e we e no significan di e - ences be ween pa ien s wi h o wi hou digoxin ega ding demog aphic pa ame e s such as age o gende . We obse ed a highe p e alence o dyslipidemia in pa ien s no med- ica ed wi h digoxin (64.7% s. 31.7%, p=0.039), bu no di e ences we e ound be ween he g oups ega ding o he ca dio ascula isk ac o s o como bidi ies. Analy- sis o hemodynamic, labo a o y and echoca diog aphic da a on admission showed no significan di e ences in blood p essu e, hea a e, diu esis a e, LVEF, se um sodium, po assium, hemoglobin and B- ype na iu e ic pep ide (BNP) le els, ca diac ou pu , peak VO2o pulmona y a e y sys- olic p essu e (Table 2). Dila ed ca diomyopa hy was mo e equen in pa ien s on digoxin he apy (86.8% s. 13.2%, p=0.032) and ischemic ca diomyopa hy was mo e p e alen in pa ien s wi hou his he apy (37.5% s. 62.5%, p<0.001), as shown in Table 3. Conce ning g oup B, no di e ences we e ound be ween pa ien s wi h o wi hou digoxin ega ding demog aphic pa ame e s, ca dio ascula isk ac o s o como bidi ies. LVEF was significan ly lowe in pa ien s on digoxin he apy (25.0% s. 30.0%, p=0.003) and admission c ea inine was also lowe (1.4±0.6 s. 1.7±1.1 mg/dl; p=0.028) (Table 2). Dila ed ca diomyopa hy was mo e equen in pa ien s on digoxin he apy (57.4% s. 35.7%, p=0.014) (Table 3). The apy Medica ions on admission, du ing hospi al s ay and a dis- cha ge a e desc ibed in Table 4. AF pa ien s we e mo e o en medica ed wi h amioda one (42.5% s. 21.0%, p=0.001) and wa a in (50.7% s. 23.1%, p=0.001) on admission and a discha ge (61.8% s. 43.3%; p=0.004 and 65.2% s. 34.1%; p=0.001, espec i ely, o amioda one and wa a in) and he p esc ip ion o be a-blocke s was significan ly lowe 306 E. Jo ge e al. Table 2 Hemodynamic, echoca diog aphic and labo a o y da a. Sinus hy hm (n=179) A ial fib illa ion (n=89) SR s. AF To al No digoxin Digoxin p To al No digoxin Digoxin p p Age, yea s 55.08 ± 15.43 55.35 ± 16.04 55 ± 15.29 0.898 62.36 ± 12.81 64.09 ± 12.99 61.76 ± 12.8 0.456 <0.001 LVEF, % 0.26 ± 0.1 0.30 ± 0.12 0.25 ± 0.08 0.003 0.31 ± 0.12 0.31 ± 0.13 0.30 ± 0.12 0.746 0.003 SBP, mmHg 113.95 ± 23.64 119.15 ± 27.49 112.43 ± 22.28 0.146 117.08 ± 23.74 120.43 ± 26.6 116.29 ± 23.18 0.561 0.540 DBP, mmHg 68.93 ± 14.74 72.53 ± 16.35 67.87 ± 14.14 0.105 70.37 ± 16.36 73.93 ± 13.6 69.53 ± 16.94 0.369 0.727 HR, bpm 84.72 ± 20.3 84.79 ± 19.32 84.7 ± 20.65 0.981 89.19 ± 21.38 86.21 ± 19.03 89.9 ± 22 0.566 0.011 Admission C , mg/dl 1.42 ± 0.74 1.68 ± 1.13 1.35 ± 0.58 0.028 1.59 ± 0.97 1.66 ± 0.67 1.57 ± 1.04 0.749 0.353 Discha ge C , mg/dl 1.42 ± 0.68 1.59 ± 0.88 1.37 ± 0.61 0.106 1.46 ± 0.84 1.61 ± 0.69 1.42 ± 0.88 0.438 0.704 Na+, mmol/l 136.89 ± 5.32 138.34 ± 3.68 136.5 ± 5.63 0.082 135.97 ± 5.64 136.33 ± 6.99 135.88 ± 5.31 0.783 0.047 K+, mmol/l 4.38 ± 0.72 4.43 ± 0.7 4.36 ± 0.73 0.600 4.28 ± 0.64 4.03 ± 0.77 4.35 ± 0.59 0.090 0.403 Hb, g/dl 13.19 ± 2.14 12.79 ± 1.89 13.29 ± 2.2 0.267 13.4 ± 2.22 12.96 ± 2.46 13.5 ± 2.17 0.435 0.489 BNP, pg/ml 1354.25 ± 1251.43 1408.58 ± 1242.94 1339.22 ± 1259.98 0.804 1000.83 ± 909.93 1004.75 ± 1186.02 999.74 ± 834.34 0.987 0.140 PASP, mmHg 50.67 ± 16.47 40.55 ± 17.17 48.77 ± 18.65 0.187 47.59 ± 11.56 65 ± 18.91 53.24 ± 19.2 0.232 0.153 VO2, ml/kg/min 17.25 ± 4.38 18.28 ± 4.26 17.04 ± 4.4 0.290 15.59 ± 4.37 17.07 ± 2.84 15.25 ± 4.63 0.366 0.081 CI, l/min/m22.04 ± 0.51 2.23 ± 0.48 1.99 ± 0.51 0.164 2.15 ± 1.12 2.85 ± 2.3 1.96 ± 0.41 0.082 0.672 Hospi al s ay, days 5.94 ± 4.37 7.02 ± 6.23 5.60 ± 3.55 0.062 6.84 ± 4.67 6.09 ± 4.28 7.11 ± 4.8 0.371 0.032 Blood glucose, mg/dl 130.01 ± 60.01 143.74 ± 70.41 125.70 ± 55.96 0.089 139.74 ± 55.45 135.77 ± 49.72 141.08 ± 57.55 0.701 0.282 BNP: B- ype na iu e ic pep ide; CI: ca diac index by he Fick me hod; C : c ea inine; DBP: dias olic blood p essu e; Hb: hemoglobin; HR: hea a e; LVEF: le en icula ejec ion ac ion; PASP: pulmona y a e y sys olic p essu e; SBP: sys olic blood p essu e; VO2: peak oxygen up ake. Digoxin in ad anced hea ailu e pa ien s 307 Table 3 E iology o hea ailu e. (%) Sinus hy hm (n=179) A ial fib illa ion (n=89) SR s. AF No digoxin Digoxin To al p No digoxin Digoxin To al p p Dila ed 35.7 57.4 52.2 0.014 21.7 60.0 50.0 0.002 0.730 Res ic i e 2.4 0.0 0.6 0.072 8.7 1.5 3.4 0.104 0.074 Hype ophic 0.0 1.5 1.1 0.428 0.0 3.1 2.3 0.395 0.500 Ischemic 47.6 34.6 37.6 0.127 52.2 15.4 25.0 0.001 0.400 Val ulopa hy 0.0 4.4 3.4 0.165 8.7 15.4 13.6 0.422 <0.001 AF: a ial fib illa ion; SR: sinus hy hm. (57.3% s. 83.2%; p=0.001) compa ed wi h pa ien s in SR. In AF pa ien s, spi onolac one p esc ip ion on admission was highe in pa ien s on digoxin he apy (58.5% s. 25.0%; p=0.011), bu no di e ences we e ound ega ding o he p io medica ions o discha ge he apy. Rega ding SR pa ien s, we obse ed ha pa ien s on digoxin he apy we e mo e o en medica ed wi h spi ono- lac one, u osemide and wa a in p io o hospi al admission (67.0% s. 38.2%; p=0.03; 88.0% s. 67.6%; p=0.006; 27.5% s. 8.8%; p=0.024, espec i ely) and a discha ge (91.9% s. 62.8%; p=0.001; 99.3% s. 93.0%; p=0.016; 39.0% s. 18.6%; p=0.014, espec i ely). Long- e m p ognosis Abou one-fi h (21.1%) o pa ien s unde wen hea ans- plan a ion du ing he ollow-up pe iod. Addi ion o digoxin o con empo a y s anda d HF he apy showed no impac on long- e m mo ali y in pa ien s wi h ad anced HF and SR (19.1% s. 22.5%, p=0.788) (Figu e 1). Rega ding AF pa ien s, we obse ed significan ly lowe long- e m mo ali y in pa ien s on digoxin he apy (18.6% s. 46.6%, p=0.048) (Figu e 2). Digoxin he apy did no influence he composi e endpoin o all-cause mo ali y and ead- missions o decompensa ed HF, in he o al popula ion o s a ified by AF/SR. Table 4 Medica ion on admission, du ing hospi al s ay and a discha ge. Medica ion Sinus hy hm (n=179) A ial fib illa ion (n=89) SR s. AF No digoxin (%) Digoxin (%) To al (%) p No digoxin (%) Digoxin (%) To al (%) p p Admission ACEI 67.6 63.3 64.3 0.644 55.0 71.7 67.1 0.176 0.684 ARB 5.9 14.7 12.6 0.177 20.0 17.0 17.8 0.764 0.391 Be a-blocke 44.1 57.8 54.5 0.162 40.0 47.2 45.2 0.583 0.194 Spi onolac one 38.2 67.0 60.1 0.003 25.0 58.5 49.3 0.011 0.129 Digoxin 11.8 52.3 42.7 <0.001 10.0 67.9 52.1 <0.001 0.190 Fu osemide 67.6 88.0 83.1 0.006 85.0 88.7 87.7 0.670 0.378 Amioda one 17.6 22.0 21.0 0.585 45.0 41.5 42.5 0.788 <0.001 S a in 32.4 21.1 23.8 0.178 25.0 17.0 19.2 0.438 0.442 Wa a in 8.8 27.5 23.1 0.024 40.0 54.7 50.7 0.262 <0.001 In-hospi al Dobu amine 14.3 18.5 17.5 0.528 8.7 21.2 18.0 0.178 0.926 Dopamine 9.5 11.1 10.7 0.772 17.4 9.1 11.2 0.278 0.901 No ad enaline 2.4 0.7 1.1 0.380 0.0 3.0 2.2 0.398 0.480 IV amioda one 23.1 21.2 21.7 0.890 25.0 25.0 25.0 1.000 0.758 Le osimendan 67.4 81.5 78.1 0.053 56.5 80.0 73.9 0.028 0.443 Discha ge ACEI 79.1 82.2 81.5 0.643 65.2 78.5 75.0 0.207 0.221 ARB 14.0 14.9 14.7 0.876 17.4 21.2 20.2 0.694 0.252 Be a-blocke 81.4 83.8 83.2 0.710 52.2 59.1 57.3 0.564 <0.001 Spi onolac one 62.8 91.9 84.9 <0.001 82.6 86.4 85.4 0.661 0.918 Fu osemide 93.0 99.3 97.8 0.016 100.0 98.5 98.9 0.553 0.527 Amioda one 38.1 44.9 43.3 0.440 47.8 66.7 61.8 0.109 0.004 S a in 46.5 31.6 31.7 0.075 26.1 22.7 23.6 0.744 0.054 Wa a in 18.6 39.0 34.1 0.014 65.2 65.2 65.2 0.995 <0.001 ACEI: angio ensin-con e ing enzyme inhibi o ; AF: a ial fib illa ion; ARB: angio ensin ecep o blocke ; IV: in a enous; SR: sinus hy hm. 308 E. Jo ge e al. 1.0 0.8 0.6 Cumula i e su i al 0.4 0.2 0.0 0.00 500.00 1000.00 1500.00 Time o dea h, days 2000.00 2500.00 Digoxin No Yes No-censo ed Yes-censo ed Figu e 1 Cumula i e 40-mon h su i al in pa ien s in sinus hy hm s a ified acco ding o digoxin medica ion. Discussion Despi e ecen ad ances in he apy, HF emains associa ed wi h high mo ali y and hospi aliza ion a es, as we epo in ou con empo a y coho . Digoxin is he oldes HF d ug and one o he leas expensi e. The ACC/AHA HF guide- lines ecommend conside ing adding digoxin in pa ien s wi h pe sis ing HF symp oms al eady ea ed wi h diu e ics, an ACEI (o angio ensin ecep o blocke ) and a be a-blocke .1 1.0 Cumula i e su i al 0.8 0.6 0.4 0.2 0.0 0.00 500.00 1000.00 1500.00 Time o dea h, days 2000.00 2500.00 Digoxin No Yes No-censo ed Yes-censo ed Figu e 2 Cumula i e 40-mon h su i al in pa ien s wi h a ial fib illa ion s a ified acco ding o digoxin medica ion. Howe e , he e is limi ed e idence on he ole o digoxin in pa ien s on backg ound he apy wi h be a-blocke s and ACEIs. The landma k DIG s udy, a andomized con olled ial o digoxin in pa ien s wi h ch onic HF al eady medica ed wi h diu e ics, showed ha digoxin had no impac on o e all mo ali y, bu did educe he a e o hospi aliza ions due o HF.9Howe e , he p esc ip ion a e o be a-blocke s was no epo ed and, a he ime he ial was conduc ed, be a- blocke s we e no accep ed as s anda d he apy o pa ien s wi h HF. Pa icipan s in he DIG ial we e male ou pa ien s wi h ch onic s able HF ( he majo i y in NYHA class I o II) in SR, and hus did no eflec eal-wo ld HF popula ions.9 Simila ly o he DIG ial, we did no see a su i al benefi in ad anced HF pa ien s in sinus hy hm. Howe e , we ailed o demons a e a educ ion in HF hospi aliza ions, p obably due o lack o powe . Ou popula ion is significan ly di e en om ha o he DIG ial, wi h mo e se e e pa ien s hospi alized o decompensa ed HF, all in NYHA class III o IV wi h se e ely dep essed LVEF, and a high p e alence o como bidi ies. The se e i y o his popula ion is highligh ed by he signi - ican p opo ion o pa ien s who ul ima ely p oceeded o hea ansplan a ion. Fu he mo e, he use o diu e ics, ACEIs, be a-blocke s and de ices in ou s udy popula ion was highe han in o he s udies examining he ole o digoxin. This may ha e educed he independen beneficial e ec o digoxin on HF eadmissions. The same is epo ed in a subs udy o he Valsa an in Hea Failu e T ial (Val- HeFT).12 In a coho o 6800 pa ien s, he in es iga o s also ailed o show any hospi aliza ion benefi wi h he use o digoxin he apy, u he suppo ing ou esul s. One ela- i ely ecen e ospec i e s udy sugges s ha he e was no benefi wi h digoxin in a hea ansplan e e al popula ion o 455 pa ien s wi h ad anced HF ecei ing con empo a y medical he apy.13 In his s udy, digoxin use was associa ed wi h inc eased isk o he p ima y ou come o dea h, u gen ansplan a ion o en icula assis de ice implan a ion, and he isk was highe in pa ien s in SR compa ed wi h AF. Ou s udy sugges s ha digoxin he apy may imp o e he i al p ognosis o pa ien s wi h ad anced HF and AF unde op imal medical he apy. In AF pa ien s, one po en ial mech- anism o benefi o digoxin is simply o educe en icula a e, he eby limi ing myoca dial oxygen consump ion, a concep in which in e es was ecen ly eawakened wi h he publica ion o he landma k SHIFT ial.14 The idea ha be a-blocke s migh elimina e he need o digoxin in pa ien s wi h AF and HF was he a ionale o he CAFE s udy.11 Howe e , a he han showing ha digoxin is edun- dan , he s udy sugges ed ha he e could be syne gis ic benefi s be ween digoxin and ca edilol. The e a e easons o he han a e con ol why digoxin may be mo e e ec i e in he p esence o be a blockade. Be a-blocke s, especially nonselec i e ones, ha a enua e s ess-induced educ ions in se um po assium15 migh neu- alize he inc ease in sudden dea h associa ed wi h he use o digoxin. The ino opic e ec o digoxin may imp o e ol- e abili y o be a-blocke s, enabling highe dose i a ion. Aldos e one an agonis s also inc ease se um po assium, educing he incidence o hypokalemia; i is well known ha sudden dea h educ ion was one o he p ima y mech- anisms o mo ali y educ ion in he Randomized Aldac one Digoxin in ad anced hea ailu e pa ien s 309 E alua ion S udy (RALES)16 and he EPHESUS ial.17 Spi ono- lac one ended o exe a g ea e educ ion in mo ali y among pa ien s ea ed wi h digoxin. Pos -hoc analyses o he DIG ial sugges ha digoxin educed mo ali y a low (0.5---0.9 ng/ml) se um digoxin con- cen a ions (SDC), bu had no e ec a highe (≥1 ng/ml) SDC.18 A close examina ion o he Kaplan---Meie su i al plo s in he DIG ial e eals an ea ly mo ali y educ- ion in he digoxin g oup, ollowed by a la e i ual o e lap o he plo s, sugges ing ha , al hough he ea ly su i al benefi was elimina ed in la e yea s, he e was no inc ease in mo ali y.19 This lack o a long- e m e ec o digoxin on mo ali y may be due o use o open-labeled digoxin in he placebo g oup, and a cumula i e e ec o he use o high-dose digoxin; mo e han 80% o he pa icipan s in he DIG ial we e ecei ing ≥0.25 mg o digoxin daily o ma ching placebo, which was highe han he cu en ly ec- ommended daily dosage.18 In ou popula ion, he s anda d daily dose was 0.125 mg, wi h a majo i y o pa ien s paus- ing on weekends; he maximum dose p esc ibed du ing da a collec ion was 0.25 mg/day, fi e imes a week. The con- inued use o high-dose digoxin in elde ly pa ien s, wi h de e io a ing kidney unc ion, may ha e esul ed in highe SDC du ing he la e yea s o he DIG ial. The benefi- cial e ec s o digoxin a low SDC a e p ima ily due o i s e ec on he neu oho monal sys em.18 By inhibi ing he sodium-po assium adenosine iphospha e pump in enal ubules and agal a e en fibe s, digoxin supp esses bo h he enin-angio ensin-aldos e one and sympa he ic ne ous sys ems.18 Low SDC also educes he isk o digoxin oxic- i y and associa ed mo bidi y and mo ali y.18 Concomi an he apy wi h be a-blocke s and aldos e one an agonis s and use o lowe doses o digoxin could adically imp o e he isk/benefi a io o digoxin.20 The explana ion o di e en ou comes in AF and SR emains elusi e. In ou popula ion, we obse ed significan ly lowe LVEF in pa ien s in SR (26.0% s. 31.0%, p=0.003), bu no o he significan di e ences we e ound be ween pa ien s in SR and AF. Ou s udy has se e al limi a ions. Fi s , his obse a ional e ospec i e s udy eflec s a single-cen e expe ience and includes a ela i ely small numbe o pa ien s. Second, we did no ha e da a o SDC; howe e , he s anda d dose p esc ibed was 0.125 mg/day fi e imes a week and he maximum dose p esc ibed du ing da a collec ion was 0.25 mg/day, and we know ha digoxin dose is a s ong p e- dic o o SDC. Gi en he na u e o he da a se , we we e unable o asce ain he leng h o he apy on digoxin be o e he index HF admission. Thi d, da a ega ding discon inu- ing o s a ing he apy, o op imiza ion o d ug doses du ing ollow-up, we e no a ailable. P e ious s udies on digoxin in pa ien s wi h ch onic HF epo ed imp o emen s in LVEF, HF symp oms and exe cise pe o mance. In ou s udy, hese we e no measu ed, and hus we canno exclude po en ial beneficial e ec s on hese so endpoin s. One issue wi h e ospec i e ou comes analysis is always whe he o no he mo e se e e pa ien s p e e en ially ecei ed a gi en he - apy. Howe e , in ou popula ion he e we e no significan di e ences be ween AF pa ien s wi h o wi hou digoxin ega ding demog aphic pa ame e s, significan como bidi- ies, o hemodynamic, labo a o y, elec oca diog aphic and echoca diog aphic da a on admission, making selec ion bias unlikely. Conclusions The p esen s udy sugges s ha digoxin may imp o e he i al p ognosis o AF pa ien s wi h ad anced HF unde op i- mal medical he apy. Digoxin did no demons a e benefi in pa ien s in SR. These esul s may help o imp o e pa ien selec ion o ea men wi h digoxin, in o de o achie e a be e isk/benefi a io in ad anced HF pa ien s. Polypha - macy is known o be associa ed wi h an inc eased isk o ad e se d ug eac ions, d ug in e ac ions and poo compli- ance. Thus, o mul iple easons o bo h e ficacy and sa e y, whe he o no digoxin he apy has any ole in he cu en managemen o ad anced HF needs o be e alua ed. Fu - he andomized con olled ials a e equi ed o find ou whe he digoxin has a ole in con empo a y managemen o HF. Conflic s o in e es The au ho s ha e no conflic s o in e es o decla e. Re e ences 1. Hun SA. ACC/AHA 2005 guideline upda e o he diagnosis and managemen o ch onic hea ailu e in he adul : a epo o he Ame ican College o Ca diology/Ame ican Hea Associa ion Task Fo ce on P ac ice Guidelines (W i ing Commi ee o Upda e he 2001 Guidelines o he E alua ion and Managemen o Hea Failu e). J Am Coll Ca diol. 2005;46:e1---82. 2. Remme WJ, Swedbe g K. Guidelines o he diagnosis and ea - men o ch onic hea ailu e. Eu Hea J. 2001;22:1527---60. 3. Dobbs SM, Kenyon WI, Dobbs RJ. Main enance digoxin a e an episode o hea ailu e: placebo-con olled ial in ou pa ien s. B Med J. 1977;1:749---52. 4. Guya GH, Sulli an MJ, Fallen EL, e al. A con olled ial o digoxin in conges i e hea ailu e. Am J Ca diol. 1988;61:371---5. 5. Lee DC, Johnson RA, Bingham JB, e al. Hea ailu e in ou pa- ien s: a andomized ial o digoxin e sus placebo. N Engl J Med. 1982;306:699---705. 6. U e sky BF, Young JB, Shahidi FE, e al. Randomized s udy assessing he e ec o digoxin wi hd awal in pa ien s wi h mild o mode a e ch onic conges i e hea ailu e: esul s o he PROVED ial. PROVED In es iga i e G oup. J Am Coll Ca diol. 1993;22:955---62. 7. Packe M, Gheo ghiade M, Young JB, e al. Wi hd awal o digoxin om pa ien s wi h ch onic hea ailu e ea ed wi h angio ensin-con e ing-enzyme inhibi o s. RADIANCE S udy. N Engl J Med. 1993;329:1---7. 8. The Cap op il-Digoxin Mul icen e Resea ch G oup. Compa a- i e e ec s o he apy wi h cap op il and digoxin in pa ien s wi h mild o mode a e hea ailu e. JAMA. 1988;259:539---44. 9. The Digi alis In es iga ion G oup. The e ec o digoxin on mo - ali y and mo bidi y in pa ien s wi h hea ailu e. N Engl J Med. 1997;336:525---33. 10. Lade E, Egan D, Hunsbe ge S, e al. The e ec o digoxin on he quali y o li e in pa ien s wi h hea ailu e. J Ca d Fail. 2003;9:4---12. 11. Khand AU, Rankin AC, Ma in W, e al. Ca edilol alone o in combina ion wi h digoxin o he managemen o a ial 310 E. Jo ge e al. fib illa ion in pa ien s wi h hea ailu e? J Am Coll Ca diol. 2003;42:1944. 12. Bu le J, Anand I, Kuskowski M, e al. Highe mo ali y and ad e se ca dio ascula ou comes isk wi h digoxin use in pa ien s wi h hea ailu e: a Val-HeFT subs udy. Ci cula ion. 2005; Suppl. II:641. 13. Geo giopoulou V, Kaloge opoulos A, Giamouzis G, e al. Digoxin he apy does no imp o e ou comes in pa ien s wi h ad anced hea ailu e on con empo a y medical he apy. Ci c Hea Fail. 2009;2:90---7. 14. Swedbe g K, Komajda M, Bohm M, e al. I ab adine and ou - comes in ch onic hea ailu e (SHIFT): a andomised con olled s udy. Lance . 2010;376:875---85. 15. Poole-Wilson PA, Swedbe g K, Cleland JGF, e al. Compa ison o ca edilol and me op olol on clinical ou comes in pa ien s wi h ch onic hea ailu e in he Ca edilol o Me op olol Eu opean T ial (COMET): andomised con olled ial. Lance . 2003;362:7---13. 16. Pi B, Zannad F, Remme WJ, e al. The e ec o spi onolac one on mo bidi y and mo ali y in pa ien s wi h se e e hea ailu e. N Engl J Med. 1999;341:709---17. 17. Pi B, Bak is G, Ruilope LM, e al. Se um po assium and clinical ou comes in he Eple enone Pos -Acu e Myoca dial In a c ion Hea Failu e E ficacy and Su i al S udy (EPHESUS). Ci cula- ion. 2008;118:1643---50. 18. Ahmed A, Pi B, Rahim oola S, e al. E ec s o digoxin a low se um concen a ions on mo ali y and hospi aliza ion in hea ailu e: a p opensi y-ma ched s udy o he DIG ial. In J Ca diol. 2008:138---46. 19. Ahmed A, Waags ein F, Pi B, e al. E ec i eness o digoxin in educing one-yea mo ali y in ch onic hea ailu e in he Digi alis In es iga ion G oup T ial. Am J Ca diol. 2009;103: 82---7. 20. Cleland J, Culling on D. Digoxin, Quo Vadis? Ci c Hea Fail. 2009;2:81---5.