Professional Documents
Culture Documents
net/publication/327613917
Digoxin and clinical outcomes in the Global Rheumatic Heart Disease Registry
CITATIONS READS
3 125
8 authors, including:
Some of the authors of this publication are also working on these related projects:
A systematic review of standard echocardiography versus handheld echocardiography for the detection of subclinical rheumatic heart disease View project
All content following this page was uploaded by Liesl Zuhlke on 21 October 2018.
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Digoxin and clinical outcomes in the Global
Rheumatic Heart Disease Registry
Ganesan Karthikeyan,1 Niveditha Devasenapathy,2 Liesl Zühlke,3,4
Mark Emmanuel Engel,4 Sumathy Rangarajan,5 Koon K Teo,5 Bongani M Mayosi,4
Salim Yusuf,5 on behalf of the Global Rheumatic Heart Disease Registry (REMEDY)
Investigators
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Study design, participants and outcomes We generated propensity scores for each patient using logistic
The design, baseline characteristics and 2-year outcomes of the regression with digoxin as the dependent variable. The vari-
REMEDY study have been published previously.2 3 10 Briefly, ables used in this model (listed in figure 2) were selected based
REMEDY was a prospective, multicentre, international, hospi- on their potential association with digoxin use and clinical
tal-based registry which enrolled 3343 patients with a clin- outcomes, in accordance with expert recommendations.11–14
ical diagnosis of RHD, who were seen at 25 hospitals in 14 After generating the propensity scores we tested for similarity
countries (12 African countries, Yemen and India). The key of distribution, adequate overlap of the distributions (common
outcomes of interest were death, HF, stroke, transient isch- support) (online supplementary figure S1) and ensured balance
aemic attack or non-central nervous system systemic embolism of covariates between the two groups (digoxin and no digoxin)
within blocks of propensity scores. After creating a balanced
at 2 years.
propensity score, we used a weighting method (inverse prob-
For the purposes of this analysis, we considered patients to
ability treatment weights, IPTW) and a matching method (1:2
be digoxin users if they were on the drug at baseline. The prin-
calliper matching with replacement, with calliper width set at
cipal outcome was mortality at 2 years. Other outcomes studied
0.2×SD of the logit of the propensity score) to ensure balance
were recurrent HF, all hospitalisations and the incidence of
of covariates between the two groups.14 Balance was assessed by
valve surgery or percutaneous interventions for valve disease. the standardised percentage of bias for each covariate. An overall
We also assessed the effect of digoxin use on the composite mean standardised percentage of bias <5%, or a standardised
outcomes of death or recurrent HF, and death, recurrent HF mean difference <0.1, is considered to indicate good balance.
or hospitalisation. The severity of valve lesions and other echo- We estimated average treatment effects of digoxin on outcomes
cardiographic variables were described using standard criteria.2 using the IPTW as the primary analysis and corroborated the
Patients were considered to have severe valve disease if at least results using those obtained from 1:2 calliper matching. Further,
one of the involved valves had severe disease. HF at baseline if statistically significant subgroup effects were present, we
and during the study (recurrent HF) was diagnosed if any two of generated separate propensity scores for subgroups of patients
the following criteria were present: (1) symptoms (dyspnoea on who were in AF, HF or both at baseline (variables listed in online
exertion or at rest, orthopnoea,paroxysmal nocturnal dyspnoea supplementary table S2) and followed similar procedures to
or ankle oedema) or signs (rales, increased jugular venous pres- obtain effect estimates of digoxin use on outcomes.
sure or ankle oedema) of congestive HF; (2) radiologic signs of The effect of digoxin use on outcomes was estimated using
pulmonary congestion; and (3) treatment with diuretics. Surgery ORs and their 95% CIs. A p value of <0.05 was considered
for valve disease was defined as performance of valve repair, statistically significant. All analyses were done using Stata V.14
or replacement of any affected valve with a tissue, or mechan- (StataCorp, College Station, Texas) and the propensity anal-
ical prosthesis. Percutaneous valve interventions consisted of yses were performed using the pscore, psmatch2, dr and teffects
balloon dilatation of stenosed mitral, aortic or tricuspid valves. packages.
Patients who had at least one overnight admission to hospital
were considered to have a hospitalisation event. Details and defi- Results
nitions of the baseline and outcome measures used have been Of the 3343 patients in REMEDY, 3298 (98.7%) had informa-
published previously.2 3 10 tion on baseline digoxin use, and 1144 (34.7%) of these were
on digoxin. The proportion of patients lost to follow-up among
those on and those not on digoxin was similar (figure 1). Patients
Statistical methods who were on digoxin at baseline were more likely to reside in
We compared the baseline characteristics of digoxin users and low-income countries, were older, in AF, HF and poor func-
non-users using independent t-tests and Χ2 tests as appropriate. tional class, and were more likely to be on diuretics and ACE
We performed univariable and multivariable logistic regression inhibitors (or angiotensin receptor blockers).(online supplemen-
to assess the effect of digoxin on the following outcomes: death, tary table S1) Overall, 51% (1709/3343) of patients had severe
recurrent HF, hospitalisation, surgical or percutaneous valve valve disease. The prevalence of clinical and echocardiographic
intervention and on the composites of death or recurrent HF, and markers of severe valve involvement was higher among those
death, recurrent HF or hospitalisation. The variables included prescribed digoxin (table 1). After propensity score weighting,
in the multivariable model were decided a priori based on their we achieved excellent balance between the two groups as
relevance to prognosis and were identical to the ones used previ- evidenced by the small (<5%) standardised mean difference
ously in our primary analysis of outcomes.3 They were: age, sex, between covariates (figure 2). The variables used to generate the
presence of AF or atrial flutter, NYHA class, history of HF or propensity scores and the percentage bias reduction are depicted
HF at enrolment, previous heart valve surgery or intervention, in figure 2. Balance of covariates for the 1:2 calliper matched
a history of stroke or IE, severe disease (severe valve disease analysis is shown in online supplementary figures S2 and S3.
in at least one affected valve), multivalve involvement and use
of secondary prophylaxis. We performed subgroup analyses to Effect of digoxin on clinical outcomes
assess effect modification by the presence of AF or HF at base- All patients
line, on the association of digoxin on the principal outcomes. In Among all patients, digoxin use was significantly associated
addition, we tested for interactions between gender, body mass with mortality (OR 1.63, 95% CI 1.30 to 2.04, p<0.0001)
index (BMI) and concomitant beta-blocker use with digoxin , by and recurrent HF (OR 1.48, 95% CI 1.07 to 2.04, p=0.019)
adding appropriate interaction terms in the model. Finally, we at 2 years (table 2). There was no impact of digoxin on hospi-
also assessed the effect of digoxin use in the subpopulation of talisations or the performance of percutaneous or surgical
patients who had both AF and HF at baseline when compared interventions. The composite outcome of death or HF, and
with those who had neither. death, HF or hospitalisations was also significantly increased
2 Karthikeyan G, et al. Heart 2018;0:1–7. doi:10.1136/heartjnl-2018-313614
Valvular heart disease
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Figure 1 Flow of patients contributing to the digoxin versus no digoxin analysis. Numbers are for patients using digoxin at the time of enrolment
into the study. REMEDY, Global Rheumatic Heart Disease Registry.
with digoxin use (table 2). The association of digoxin with A greater proportion of patients taking digoxin died, both
mortality was not modified by low BMI, female gender or among patients with (156, 29.3% vs 98, 18.6%) and without
the concomitant use of beta-blockers (p values for interaction HF (110, 22.9% vs 130, 9.4%). Similar to the effect in patients
0.14, 0.08 and 0.35, respectively). with AF, digoxin use was associated with a greater odds of
On propensity score weighted analyses, the effect of digoxin death among patients who did not have HF compared with
on mortality was markedly attenuated, though remaining nomi- those who had HF (table 3). On propensity adjustment, this
nally statistically significant (OR 1.05, 95% CI 1.01 to 1.09, remained significant though markedly attenuated (OR 1.08,
p=0.005). The effect on the composite outcomes was similarly 95% CI 1.04 to 1.12, p<0.0001). Propensity weight adjusted
attenuated but the association with recurrent HF was no longer analysis suggested a neutral effect of digoxin use on mortality
statistically significant (table 2). Propensity score matched anal- among patients with HF (OR 0.99, 95% CI 0.94 to 1.06,
ysis by the 2:1 calliper matching method yielded identical results p=0.83) (table 3). Similar results were seen for recurrent HF,
(online supplementary table S3). and the composite of death or HF (online supplementary table
S4 and S5).
Patients in AF
Digoxin was used by 454 (41%) patients in AF and 548 (31.5%) Patients with both AF and HF
patients without AF. More patients taking digoxin died both A significant proportion of patients (412, 14.1%) had both HF
among patients in AF (113, 24.9% vs 94, 14.4%) and among and AF at baseline. Over half of these patients (245, 59.5%)
those in sinus rhythm (148, 27% vs 133, 10.7%), compared were on digoxin. Nearly a quarter of the patients without
with those not on digoxin. In multivariable analyses, digoxin either HF or AF (265/1151, 23%) also received digoxin.
was associated with a significantly higher odds of death among Among patients with both HF and AF, a similar proportion
patients in sinus rhythm than among patients in AF (p for inter- of digoxin users died compared with non-users (65, 26.5%
action=0.03), which remained significant, although markedly vs 42, 25.2%). However, among those with neither HF nor
attenuated on propensity adjustment (OR 1.07, 95% CI 1.03 AF, the proportion of deaths among digoxin users was greater
to 1.11, p=0.001) (table 3). Propensity weight adjustment than among non-users (58, 21.9% vs 77, 8.7%). This asso-
showed no significant effect on mortality with digoxin use ciation persisted in multivariable analyses (OR for patients
among patients in AF (OR 1.01, 95% CI 0.95 to 1.07, p=0.87) not in AF or HF 2.26, 95% CI 1.5 to 3.42; OR for patients
(table 3). For recurrent HF, and the composite of death or HF, with both HF and AF 0.71, 95% CI 0.43 to 1.17, p for inter-
there was no significant difference in the effect of digoxin among action <0.0001). However, there was no difference in the
those with and without AF at baseline (online supplementary odds of developing recurrent HF. (online supplementary table
table S4 and S5). S4) Propensity score adjustment suggested persistent harm in
patients with neither AF nor HF, and a protective effect of
Patients in HF digoxin in mortality for patients who had both AF and HF (OR
About half of the patients (532, 50.2%) who were in HF, 0.88, 95% CI 0.80 to 0.98, p=0.019) (table 3). Results for the
or who had history of HF, were on digoxin compared with composite outcome of death or HF were broadly similar in
about a quarter of those who did not have HF (481, 25.8%). this group of patients (online supplementary table S5).
Karthikeyan G, et al. Heart 2018;0:1–7. doi:10.1136/heartjnl-2018-313614 3
Valvular heart disease
the three drugs, digoxin produced the least improvement in
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Table 1 Comparison of baseline characteristics of patients who
subjective symptoms and peak treadmill exercise capacity. In
completed 2-year follow-up
the Control of Rate versus Rhythm in Rheumatic Atrial Fibril-
Baseline characteristics Digoxin No digoxin P values lation Trial study, although rate control regimens were not
Participants, n (%) 1014 (34.7) 1908 (65.3) systematically evaluated, digoxin was used as add-on therapy
Mean age in years (SD) 32.3 (16) 30.2 (15.8) 0.0009 to diltiazem in 15% of patients in the rate control arm to
Female 671 (66.2) 1276 (67) 0.687 optimise treatment.15 However, neither of these studies was
Schooling beyond primary level 415 (41.1) 875 (46.2) 0.010 designed or powered to evaluate hard outcomes in relation to
Country income group <0.0001 digoxin use.
LIC 468 (46.2) 491 (25.7)
LMIC 271 (26.7) 867 (45.4) Digoxin and mortality
UMIC 275 (27.1) 550 (28.8) Numerous observational studies and a few randomised
Severe disease* 658 (64.9) 793 (41.6) <0.0001 trials have addressed the effect of digoxin use and mortality
Multivalve disease 714 (70.4) 1037 (54.4) <0.0001 in patients with non-valvular AF and HF, but have yielded
Atrial fibrillation 454 (45.3) 652 (34.5) <0.0001 contrasting results. Observational studies have in general
Heart failure 532 (52.5) 527 (27.6) <0.0001 suggested that digoxin use is associated with about a 20% higher
NYHA III/IV 315 (31.6) 353 (18.8) <0.0001
mortality.8 9 16 On the other hand, digoxin has been shown
Prior intervention or surgery 221 (21.8) 512 (26.9) 0.003
to have a neutral effect on mortality in randomised trials.7 17
Much of this difference is likely to be due to confounding
Prior stroke 77 (7.6) 134 (7) 0.557
by indication (prescription/treatment bias) in observational
Prior infective endocarditis 70 (6.9) 104 (5.5) 0.113
studies.18 In a metaregression of all-cause mortality according
Secondary prophylaxis at enrolment 454 (47) 1085 (58.5) <0.0001
to risk of bias, Ziff et al showed an increase in the risk of
Mitral valve area less than 1.5 cm2 327 (32.3) 513 (26.9) 0.002
mortality with increasing risk of bias.17 Propensity-adjusted
Dilated left atrium (≥50 mm) 562 (59.7) 555 (31.5) <0.0001 analyses may help reduce confounding due to known vari-
Dilated left ventricle† 483 (49.1) 552 (30.2) <0.0001 ables to some extent. Expectedly, studies employing propen-
Decreased ejection fraction 337 (34.5) 368 (20.2) <0.0001 sity matched analysis show a considerably attenuated risk of
Tricuspid regurgitation 449 (44.3) 416 (21.8) <0.0001 death associated with digoxin, with some reporting a neutral
Pulmonary arterial hypertension 338 (33.7) 452 (23.8) <0.0001 effect and others showing a smaller increase in risk than that
Concomitant beta-blocker use 410 (40.6) 737 (38.6) 0.3 obtained after conventional multivariable adjustment.17 In our
Calcium channel blocker use 52 (5.2) 95 (5.0) 0.819 analysis, we found an OR of 1.63 for death on multivariable
Diuretic use 915 (90.2) 1073 (56.2) <0.0001 adjustment which attenuated markedly to 1.05 in the propen-
ACE inhibitors/receptor blockers 385 (38.2) 523 (27.4) <0.0001 sity-weighted analysis, which is similar to the risk ratios (RR)
*The presence of severe disease in at least one of the involved valves. obtained by Ziff et al (RR 1.61 and 1.18).17 The large differ-
†Left ventricular end diastolic dimension >50 mm in children and >55 mm in adults. ence in risk between the propensity-adjusted and unadjusted
LIC, low-income country; LMIC, lower middle-income country; NYHA, New York analyses highlights the substantial prescription bias in digoxin
Heart Association functional class; UMIC, upper middle-income country. use in observational studies.19
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Figure 2 Standardised percentage of bias across covariates before and after inverse probability treatment weighting. Figure illustrates the balance
of covariates between treatment and comparison groups before and after inverse probability weighting. Standardised mean differences of covariates
after matching (indicated by red solid circles) are well within the prespecified ±5 margin indicating good comparability between the two groups.
NYHA, New York Heart Association; P/H/O, Past history of stroke.
who had severe disease, larger left atrial and left ventric- the benefits of digoxin are most pronounced in patients with
ular dimensions, along with diuretics and ACE inhibitors or severe HF,7 and the risk of death is highest in those with no
receptor blockers (online supplementary table S6). We found HF and the lowest baseline cardiac risk.20 The mechanisms of
a significant increase in the odds of death with digoxin use in increased mortality with digoxin among low-risk patients are
patients without either AF or HF. Digoxin appeared to confer unclear. It is plausible that among patients in AF and HF, and
a significant reduction in mortality among those with both AF the accompanying neuroendocrine derangements, the salutary
and HF which persisted in propensity-adjusted analysis. These effects of digoxin may balance its deleterious effects such as
results are consistent with previous studies which showed that proarrhythmia, while low-risk patients may only be exposed
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Table 3 Effect of digoxin on mortality by presence of AF, HF or both at baseline
Multivariable logistic regression analysis Inverse probability treatment weighted analysis
OR OR
n Digoxin No digoxin (95% CI) P values n Digoxin No digoxin (95% CI) P values
Atrial fibrillation
Patients in AF 1058 424 634 1.23 942 376 566 1.01 0.866
(0.88 to 173) (0.95 to 1.07)
Patients in sinus rhythm 1672 507 1165 2 0.03* 1549 489 1060 1.07 0.001
(1.5 to 2.7) (1.03 to 1.07)
Heart failure
Patients in HF 1009 503 506 1.26 877 448 429 0.99 0.834
(0.92 to 1.73) (0.94 to 1.06)
Patients without HF 1721 428 1293 2.09 0.024* 1641 428 1213 1.08 <0.0001
(1.53 to 2.86) (1.04 to 1.12)
Atrial fibrillation and heart failure
Patients in both AF and HF 403 237 166 0.71 348 210 138 0.88 0.019
(0.43 to 1.17) (0.80 to 0.98)
Patients in sinus rhythm 1066 241 825 2.26 <0.0001* 970 228 742 1.06 0.015
without HF (1.5 to 3.42) (1.01 to 1.12)
Overall 2730 931 1799 1.63 <0.0001 2453 842 1611 1.05 0.005
(1.3 to 2.04) (1.01 to 1.09)
*P values for interaction.
AF, atrial fibrillation; HF, heart failure.
to its deleterious effects without deriving any benefit.20 21 assist in designing large trials of the use of digoxin in patients
These data may potentially be used to guide the choice of with RHD.
patient populations for inclusion in randomised controlled
trials of digoxin in RHD. Key messages
Heart: first published as 10.1136/heartjnl-2018-313614 on 12 September 2018. Downloaded from http://heart.bmj.com/ on 29 September 2018 by guest. Protected by copyright.
Khartoum, Sudan: Tagwa Eltahir. The authors also thank Dr Shrikant Bangdiwala protocol.
(Hamilton, Canada) for reviewing the statistical methods. Dedication: The authors Provenance and peer review Not commissioned; externally peer reviewed.
wish to dedicate this manuscript to Prof Bongani Mayosi, who was a dear
colleague, friend and mentor. This work would not have been possible
without his efforts. References
1 Watkins DA, Johnson CO, Colquhoun SM, et al. Global, regional, and national burden
Collaborators The REMEDY Investigators: clinical investigators and sites: Egypt: of rheumatic heart disease, 1990-2015. N Engl J Med 2017;377:713–22.
Benha University: Azza Abul Fadl (Principal Investigator); Cairo University: Sahar S 2 Zühlke L, Engel ME, Karthikeyan G, et al. Characteristics, complications, and gaps in
Sheta (Principal Investigator). Ethiopia: University of Jimma: Abraham Haileamlak evidence-based interventions in rheumatic heart disease: the Global Rheumatic Heart
(Principal Investigator); University of Addis Ababa: Senbeta G Abdissa, Dufera Disease Registry (the REMEDY study). Eur Heart J 2015;36:1115–22.
M Begna, Wandimu Daniel, Araya G Desta, Dejuma Yadeta Goshu (Principal 3 Zühlke L, Karthikeyan G, Engel ME, et al. Clinical outcomes in 3343 children and
Investigator), Bekele A Shasho. Kenya: University of Nairobi: Bernard Gitura, Stephen adults with rheumatic heart disease from 14 low- and middle-income countries: two-
Ogendo (Principal Investigator). Malawi: University of Malawi: Neil Kennedy year follow-up of the global rheumatic heart disease registry (the REMEDY Study).
Circulation 2016;134:1456–66.
(Principal Investigator). Mozambique: Eduardo Mondlane University: Albertino
4 Nishimura RA, Otto CM, Bonow RO, et al. AHA/ACC guideline for the management
Damasceno (Principal Investigator); Instituto Nacional de Saúde: Ana Olga Mocumbi of patients with valvular heart disease: a report of the American College of
(Principal Investigator). Namibia: Windhoek Hospital: Christopher Hugo-Hamman Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation
(Principal Investigator). Nigeria: University of Abuja: Dike Ojji (Principal Investigator); 2014;2014:e521–643.
University of Jos: Ganiyu A Amusa, Fidelia Bode-Thomas (Principal Investigator), 5 Vahanian A, Alfieri O, Andreotti F, et al. Guidelines on the management of valvular
Christopher C Yilgwan, Olukemi Ige, Basil Okeahialam; Kano University: Mahmoud heart disease (version 2012). Eur Heart J 2012;33:2451–96.
U Sani (Principal Investigator); University of Ibadan: Okechukwu S Ogah (Principal 6 Ahuja RC, Sinha N, Saran RK, et al. Digoxin or verapamil or metoprolol for heart rate
Investigator); Federal Medical Centre, Abeokuta: Taiwo Olunuga; University College control in patients with mitral stenosis--a randomised cross-over study. Int J Cardiol
Hospital, Ibadan: Abiodun M Adeoye, Okechukwu Ogah (Principal Investigator). 1989;25:325–31.
Rwanda: King Faisal Hospital: Joseph Mucumbitsi (Principal Investigator). South 7 Digitalis Investigation Group. The effect of digoxin on mortality and morbidity in
patients with heart failure. N Engl J Med 1997;336:525–33.
Africa: University of Cape Town: Blanche Cupido; University of Limpopo: Phindile
8 Washam JB, Stevens SR, Lokhnygina Y, et al. Digoxin use in patients with atrial
Mntla (Principal Investigator), Christopher Sutton (Principal Investigator), Rajeev fibrillation and adverse cardiovascular outcomes: a retrospective analysis of the
Misra. Sudan: Alzaiem Alazhari University: Ahmed ElSayed (Principal Investigator), Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared with Vitamin K
Ahmed S Ibrahim; Ahmed Gasim Teaching Hospital: Huda HM Elhassan (Principal Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET
Investigator). Uganda: Makerere University: Peter Lwabi, Charles Mondo (Principal AF). Lancet 2015;385:2363–70.
Investigator), Emmy Okello. Yemen: University of Sana’a: Mohammed M Al-Kebsi 9 Vamos M, Erath JW, Hohnloser SH. Digoxin-associated mortality: a systematic review
(Principal Investigator). Zambia: University of Lusaka: John Musuku (Principal and meta-analysis of the literature. Eur Heart J 2015;36:1831–8.
Investigator). 10 Karthikeyan G, Zühlke L, Engel M, et al. Rationale and design of a Global Rheumatic
Heart Disease Registry: the REMEDY study. Am Heart J 2012;163:535–40.
Contributors Conception and design: GK, ND. Data acquisition: LZ, MEE, SR, 11 Brookhart MA, Schneeweiss S, Rothman KJ, et al. Variable selection for propensity
GK. Data analysis: ND. Data interpretation: all authors. Drafting of manuscript: GK. score models. Am J Epidemiol 2006;163:1149–56.
Critical revision for important intellectual content: all authors. Approval of final 12 Brookhart MA, Wyss R, Layton JB, et al. Propensity score methods for confounding
version: all authors. All authors agree to be accountable for all aspects of the work control in nonexperimental research. Circ Cardiovasc Qual Outcomes 2013;6:604–11.
in ensuring that questions related to the accuracy or integrity of any part of the work 13 Austin PC. An introduction to propensity score methods for reducing the effects of
are appropriately investigated and resolved. confounding in observational studies. Multivariate Behav Res 2011;46:399–424.
Funding REMEDY was funded principally by the Canadian Network and Centre for 14 Garrido MM, Kelley AS, Paris J, et al. Methods for constructing and assessing
Trials Internationally (CANNeCTIN) to GK as part of the Clinical Research Initiative propensity scores. Health Serv Res 2014;49:1701–20.
of Canadian Institutes of Health Research. The other sources of funding were the 15 Vora A, Karnad D, Goyal V, et al. Control of rate versus rhythm in rheumatic atrial
South African Medical Research Council, Lily and Ernst Hausmann Trust, the Else fibrillation: a randomized study. Indian Heart J 2004;56:65–73.
Kroner Fresenius Foundation, the University of Cape Town, the National Research 16 Ouyang AJ, Lv YN, Zhong HL, et al. Meta-analysis of digoxin use and risk of mortality
Foundation of South Africa, Harold and Ethel Pupkewitz Heart Foundation (Namibia) in patients with atrial fibrillation. Am J Cardiol 2015;115:901–6.
and the World Heart Federation. The Jos site was funded by the Jos University 17 Ziff OJ, Lane DA, Samra M, et al. Safety and efficacy of digoxin: systematic review and
Teaching Hospital, the Heart Aid Trust and Faith Alive Foundation. The Sudan sites meta-analysis of observational and controlled trial data. BMJ 2015;351:h4451.
had partial funding from Sheikan Insurance. LZ was funded in part by the Discovery 18 Bavendiek U, Aguirre Davila L, Koch A, et al. Assumption versus evidence: the case of
Foundation, a US National Institutes of Health Fogarty International Clinical digoxin in atrial fibrillation and heart failure. Eur Heart J 2017;38:ehw577.
Research Fellowship, Thrasher Research Fund Early Career Award, Wellcome Trust 19 Yusuf S, Wittes J, Bailey K, et al. Digitalis--a new controversy regarding an old drug.
Clinical Infectious Disease Research Initiative (CIDRI) Research Officer Award, the The pitfalls of inappropriate methods. Circulation 1986;73:14–18.
Hamilton Naki Clinical Scholarship and by Medtronic Foundation through support 20 Hallberg P, Lindbäck J, Lindahl B, et al. Digoxin and mortality in atrial fibrillation: a
to Rheumatic Heart Disease Action. SY is funded by the Marion Burke Chair of the prospective cohort study. Eur J Clin Pharmacol 2007;63:959–71.
Heart and Stroke Foundation of Canada. 21 Whitbeck MG, Charnigo RJ, Khairy P, et al. Increased mortality among patients taking
digoxin--analysis from the AFFIRM study. Eur Heart J 2013;34:1481–8.
Competing interests None declared. 22 Lopes RD, Rordorf R, De Ferrari GM, et al. Digoxin and mortality in patients with atrial
Patient consent Not required. fibrillation. J Am Coll Cardiol 2018;71:1063–74.