- Email: [email protected]
Therapy modifications during hospitalization in patients with chronic heart failure
EJINME-03087; No of Pages 7 European Journal of Internal Medicine xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Therapy modifications during hospitalization in patients with chronic heart failure Andreja Detiček a, Igor Locatelli a, Tina Roblek a, Aleš Mrhar a, Mitja Lainscak b,c,⁎ a b c
Faculty of Pharmacy, University of Ljubljana, Askerceva cesta 7, SI-1000 Ljubljana, Slovenia Department of Cardiology and Department of Research and Education, General Hospital Celje, Oblakova 5, SI-3000 Celje, Slovenia Faculty of Medicine, University of Ljubljana, Vrazov trg 2, SI-1104 Ljubljana, Slovenia
a r t i c l e
i n f o
Article history: Received 24 November 2015 Accepted 12 December 2015 Available online xxxx Keywords: Heart failure Drug therapy Guideline Hospitalization
a b s t r a c t Background: Guidelines on suggested pharmacological treatments for heart failure (HF) are not optimally implemented in clinical practice and whether pharmacotherapy adjustment actually happens in daily practice is largely unknown. We aimed to investigate pharmacotherapy modifications during hospitalization. Methods: This was a prospective observational survey where all admissions were screened for HF; 210 patients were included. The guideline adherence index (GAI) and modified GAI (mGAI, if ≥50% of target dose) were used to grade the pharmacotherapy. Results: Among 198 patients discharged alive (mean age 77 years, 51% male), 49% had preserved left ventricular ejection fraction (PLVEF) and 30% had left ventricular systolic dysfunction (LVSD); the echocardiography report was unavailable for 21%. Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, beta-blockers and mineralocorticoid receptor antagonists were prescribed to 78%, 58% and 20% of patients on admission and 72%, 65% and 23% at discharge, respectively. Overall, 14% of patients met GAI-3, but at discharge only 7% met mGAI-3. One of the key drugs was stopped or down-titrated in 27%. During follow-up, 21% of patients died (25% with LVSD). Patients with LVSD discharged with at least one HF drug had a lower risk of death than patients with none (HR = 0.142, 95% CI = 0.029–0.683, p = 0.015). Patients with PLVEF had better prognosis than LVSD patients when no HF drugs were prescribed at discharge (HR = 0.075, 95% CI = 0.009–0.627, p = 0.017). Conclusions: The pharmacotherapy of HF patients did not improve significantly during hospitalization, remaining suboptimal. Treatment with key drugs was terminated or reduced in a significant proportion of patients, mostly without specific written justification. © 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Long-term healthcare registries show that the implementation of the guidelines [1,2] on treatment of heart failure (HF) in clinical practice has improved over the years and that life-saving therapies are widely used in such patients [3–9]. Even when patients are hospitalized because of condition deterioration, there is ample opportunity for optimization of individual pharmacotherapy, and prescription of guidelinerecommended drugs usually increases during hospitalization [6,7, 9–11]. Different forms of guideline adherence index/indicator (GAI) have been used to assess implementation of the recommendations in
Abbreviations: ACEI, Angiotensin-converting enzyme inhibitor; ARB, Angiotensin receptor blocker; BB, Beta-blocker; CI, Confidence interval; eGFR, Bstimated glomerular filtration rate; GAI, Guideline-adherence index; HF, Heart failure; HR, Hazard ratio; LVEF, Left ventricular ejection fraction; LVSD, Left ventricular systolic dysfunction; PLVEF, Preserved left ventricular ejection fraction; MDRD, Modification of Diet in Renal Disease; MRA, Mineralocorticoid receptor antagonist; NT-proBNP, N-terminal pro-peptide of brain natriuretic peptide; NYHA, New York Heart Association. ⁎ Corresponding author. Tel.: +386 3 423 38 00; fax: +386 3 423 37 54, +386 31 379 533 (mobile). E-mail address: [email protected] (M. Lainscak).
practice, and studies have demonstrated that the GAI predicts outcome [5,7,12,13]. However, non-prescription, drug termination and dose adjustment are also reasonable in some patients and could be guided by the patient's situation and side effects [1,2,9,12,14–16]. These reasons are widely underreported and often not documented in medical records, especially reasons of a contextual nature [12,16]. They therefore remain unknown in individual patients, which could lead to overestimation of guideline non-implementation, and also to potential hazards for the patient if such therapies are restarted. The aim of this study was to evaluate modifications of HF pharmacotherapy during hospitalization in order to analyze current clinical practice. Survival analysis was used to investigate the effects of drug therapy on patient outcome.
2. Methods 2.1. Survey design, patient population and inclusion criteria The survey was part of a larger study in hospitalized patients in Slovenia. The protocol of the survey was reviewed and approved by
http://dx.doi.org/10.1016/j.ejim.2015.12.007 0953-6205/© 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
2
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
the National Ethics Committee and is available at clinicaltrials.gov (NCT01855165). We prospectively screened all admissions during a 14-week enrollment period at the University Clinic Golnik in 2013. Patients were included if they met at least one of four inclusion criteria [17]:
simplified GAI comprising only two categories: patients discharged with at least one HF drug (GAI-123) and those discharged with none (GAI-0).
- known HF diagnosis prior to admission, - echocardiography report confirming left ventricular dysfunction, - symptoms and signs of HF with elevated serum natriuretic peptide (NT-proBNP), and - treatment with a loop diuretic within 24 h after admission for reasons other than renal failure.
All statistical calculations were performed using IBM SPSS Statistics for Windows, Version 22.0. A p-value b 0.05 was considered statistically significant. McNemar's chi-square test and Wilcoxon signed-rank test were used for categorical data; t-test was used for continuous data. All tests were two-sided. Continuous variables are reported as mean (standard deviation; SD) or median (1st quartile–3rd quartile; IQR). The multivariate Cox proportional hazards model was used for survival analysis. A regression model was built to investigate the effects of the GAI on survival adjusted for potential effect of patient HF subgroup, age, sex and natriuretic peptide value (log10-transformed NT-proBNP) on survival. Interaction terms among between GAI and patient HF subgroup were also tested. The results are presented as hazard ratios (HRs) with respective 95% confidence intervals (95% CIs).
We excluded patients with any terminal chronic disease (e.g. cancer, chronic obstructive pulmonary disease, chronic kidney disease, etc.) and those who died during hospitalization. None of the authors was included in patient management. The included patients were divided into three HF subgroups according to the echocardiography report: - left ventricular systolic dysfunction (LVSD), when left ventricular ejection fraction (LVEF) was b55% - preserved left ventricular ejection fraction (PLVEF), when LVEF was ≥55% or a qualitative description of “preserved ejection fraction” or “mild left ventricular dysfunction” was given, and - no echocardiography report
We retrieved the demographic characteristics, medical histories, laboratory test results, echocardiography reports, and pharmacological treatments at admission and discharge. We also recorded the reasons for admission, the proportion of patients with anemia (serum hemoglobin level b 120 g/L in women, b 130 g/L in men), and the proportions of patients with the following findings on admission: K+ N 5.0 mmol/L, estimated glomerular filtration rate (eGFR) b30 mL/min/1.73 m2, and systolic blood pressure b 90 mm Hg. Renal function in all the patients was estimated using the simplified Modification of Diet in Renal Disease (MDRD) equation [18]. We evaluated the prescription, mean dose, and proportion of patients prescribed the target dose at admission and discharge for the following drugs: angiotensin-converting enzyme inhibitors (ACEIs)/ angiotensin receptor blockers (ARBs), beta-blockers (BBs) and mineralocorticoid receptor antagonists (MRAs) [1]. The target dose was set as the maximum daily dose recommended in the guidelines [1], with the exception of losartan where the target was set at 100 mg (instead of 150 mg) as used in the European Society of Cardiology-HF Pilot Survey [10]. Where the daily dose was stated as a range, the target dose was set at the lower limit, except in the case of lisinopril where the upper limit was used. Target doses were then considered equivalently effective within each pharmacological class, as already implemented [19]. Adherence to the guidelines was evaluated using the GAI [5]. In our survey, GAI-1, GAI-2, and GAI-3 represent the prescription of one, two, or all three neurohormonal antagonists. The indexes were also used in a modified form (mGAI) to simultaneously describe the prescription and reaching ≥50% of the target dose. In addition, we assessed therapy modification by following two main changes: the prescription or discontinuation of a drug, and up-titration to N 50% of the target dose or down-titration to b50%. When a drug was discontinued and/or downtitrated, we searched the discharge letter for the physician's explanation of the modification. We also noted the prescription of loop diuretics and digoxin. For GAI-5, the prescription of loop diuretics and digoxin was added to GAI-3. For survival analysis, we obtained all-cause mortality data from the Central Population Registry of Slovenia. Patient follow-up started at discharge and ended after a period of 6 months was exceeded for the last discharged patient. Primary categorization according to drug therapy resulted in subgroups that were too small; therefore, we used a
2.2. Statistical analysis
3. Results 3.1. Patient characteristics Of 210 eligible patients (15% of all admissions during the screening period), 198 were discharged alive after a mean (SD) of 11 (SD: 7) days and their data were analyzed. Overall, 49% of the patients had PLVEF and 30% had LVSD; the echocardiography report was not available for 21% (Fig. 1). Three inclusion criteria were met in 45% of patients, 34% reached two, 15% one, and in 6% of all the four criteria were met. The presence of symptoms and signs with elevated NT-proBNP was the most commonly met criterion (86% of patients). The mean (SD) age of the patients was 77 (SD: 8) years and 51% of the patients were male. The female patients were significantly older than the males: 78 (SD: 8) vs. 75 (SD: 8) years, p = 0.007, t-test. Patients with PLVEF were significantly older than those with LVSD: 78 (SD: 8) vs. 74 (SD: 9) years, p = 0.027, t-test. The patient characteristics are shown in Table 1. Overall, 84% of the patients were in NYHA functional classes III and IV, making HF decompensation the most common primary reason for admission (58%). Other common reasons were infections (11%), chronic
ADMITTED
1372
INCLUSION CRITERIA MET
DIED
210
12
NO ECHOCARDIOGRAPHY REPORT
DISCHARGED
198
42
ECHOCARDIOGRAPHY REPORT
156
LEFT VENTRICULAR SYSTOLIC DYSFUNCTION
PRESERVED LEFT VENTRICULAR EJECTION FRACTION
59
97 Fig. 1. Patient flow chart.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
3
Table 1 Patient characteristics on admission. All patients
LVSD
PLVEF
No report
198 (100%) 102 (51.5%) 77 (8) 11 (7)
59 (30%) 34 (58%) 74 (9) 10 (8)
97 (49%) 48 (49%) 78 (8) 11 (7)
42 (21%) 20 (48%) 79 (7) 11 (7)
1 (0.5%) 30 (15%) 86 (43%) 81 (41%)
1 (1.7%) 5 (8.5%) 24 (41%) 29 (49%)
0 (0%) 19 (20%) 41 (42%) 37 (38%)
0 (0%) 6 (14%) 21 (50%) 15 (36%)
Co-morbidities Myocardial infarction IHD and/or CAD Arterial hypertension Atrial fibrillation Aortic stenosis Valvular dysfunction Chronic kidney disease Diabetes COPD
44 (22%) 56 (28%) 161 (81%) 110 (56%) 21 (11%) 62 (31%) 50 (25%) 61 (31%) 53 (27%)
18 (31%) 22 (37%) 46 (78%) 34 (58%) 9 (15%) 29 (49%) 19 (32%) 17 (29%) 16 (27%)
13 (13%) 22 (23%) 80 (83%) 55 (57%) 12 (12%) 30 (31%) 21 (22%) 30 (31%) 31 (32%)
13 (31%) 12 (29%) 35 (83%) 21 (50%) 0 (0%) 3 (7.1%) 10 (24%) 14 (33%) 6 (14%)
Measurements LVEF (%) Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) BMI (kg/m2) S-Potassium (mmol/L) S-Sodium (mmol/L) S-NT-proBNP (ng/L) S-Creatinine (mmol/L) eGFR (mL/min per 1.73 m2)
n.a. 142 (24) 79 (14) 89 (22) 31 (7) 4.4 (0.8) 139 (4) 3127 (1040–6268) 102 (79–127) 52 (39–74)
40 (30–50) 138 (25) 81 (15) 90 (25) 30 (6) 4.3 (0.8) 139 (4) 3420 (2020–9278) 94 (68–117) 60 (39–93)
67 (60–71) 145 (23) 78 (14) 90 (23) 31 (7) 4.5 (0.8) 139 (4) 3119 (904–5615) 103 (85–129) 50 (40–70)
n.a. 138 (25) 76 (15) 85 (14) 31 (6) 4.4 (0.6) 139 (4) 2416 (1538–6581) 111 (86–129) 50 (38–61)
Frequency (%) Male Age (years): Days of hospitalization NYHA classification Class I Class II Class III Class IV
Treatment-limiting conditions Potassium N 5.0 mmol/L eGFR b 30 mL/min per 1.73 m2 Systolic BP b 90 mmHg
28 (14%) 16 (8.1%) 2 (1.0%)
6 (10%) 7 (12%) 2 (3.4%)
18 (19%) 5 (5.2%) 0 (0%)
4 (9.5%) 4 (9.5%) 0 (0%)
The values presented refer to the situation on admission. The results are presented as frequency (percentages), mean (SD), or median (1st quartile–3rd quartile; IQR). LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; IHD = Ischaemic heart disease; CAD = Coronary artery disease; NYHA = New York Heart Association; COPD = Chronic obstructive pulmonary disease; LVEF = Left ventricular ejection fraction; BP = Blood pressure; BMI = Body mass index; eGFR = Estimated glomerular filtration rate; n.a. = Non-applicable.
obstructive pulmonary disease (9%), malignancy (5%), and diagnostic procedures (3%). Hypertension (81%) and atrial fibrillation (56%) were the most frequent diseases, with no differences between the three subgroups of patients. NT-proBNP was measured in 83% of patients during hospitalization: the concentration was ≥125 ng/L in 81% and ≥300 ng/L in 78%. Anemia was present in 37%, 59% of patients had an eGFR b 60 mL/min/1.73m2, and 23% had at least one of the HF drug prescription-limiting conditions on admission (K+ N 5.0 mmol/L, eGFR b30 mL/min/1.73 m2, or systolic blood pressure b 90 mm Hg). 3.2. Pharmacotherapy The median number of all drugs prescribed was 8 (IQR: 6–10), cardiovascular drugs 5 (IQR: 3–6). At discharge, 45% of the patients were prescribed more drugs than on admission (p = 0.034). 3.2.1. Drug prescription Proportions of patients prescribed a particular drug and GAIs at admission and discharge, together with p-values, are shown in Table 2 and Fig. 2. We observed a significant increase in the prescription of BBs during hospitalization (p = 0.016), and the prescription of loop diuretics was significantly higher at discharge (p = 0.003). There were no significant differences for other drugs. In patients with LVSD, we noticed the same trends as in the two other groups, indicating lower
prescription of ACEIs/ARBs and higher prescription of other drugs at discharge (Table 2). The most commonly prescribed drugs were perindopril (ACEI), losartan (ARB), bisoprolol (BB), and spironolactone (MRA) (Table 3). The most commonly prescribed loop diuretic was furosemide. Digoxin was prescribed only to HF patients who also had atrial fibrillation. 3.2.2. Drug dose The mean doses decreased from admission to discharge in all pharmacological classes and were 60% of the target dose in ACEI, 70% in ARB, 45% in BB and 112% in MRA at discharge (Table 3). In the LVSD group of patients, the respective percentages were 57%, 62%, 43%, and 112%. The target dose was prescribed at discharge in 36% of the patients receiving ACEIs/ARBs, 15% receiving BBs and 84% receiving MRAs; ≥50% of the target dose was prescribed in 72%, 51%, and 96% of the patients, respectively (Table 3). 3.2.3. Guideline adherence indexes On admission and at discharge, 10% of the patients were not receiving any of the HF recommended drugs (GAI-0). Overall, one-third met GAI-1 and approximately 40% met GAI-2 at both time points. Only 14% met GAI-3 at discharge; 7% were also prescribed all drugs at ≥ 50% of the target dose (mGAI-3; Table 2). However, no patient reached the target dose for all the drugs. GAI-5 was met in five patients at discharge. No
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
4
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
Table 2 Therapy modifications between patient admission and discharge. All patients N = 198
p value
LVSD N = 59
p value
PLVEF N = 97
p value
No report N = 42
p value
ACEI/ARB Admission Discharge
78% 72%
0.065
76% 71%
0.549
76% 68%
0.115
83% 81%
1.000
BB Admission Discharge
58% 65%
0.016*
61% 73%
0.065
59% 63%
0.424
52% 60%
0.375
MRA Admission Discharge
20% 23%
0.405
20% 25%
0.453
21% 21%
1.000
19% 24%
0.625
LOOP DIURETIC Admission Discharge
72% 82%
0.003*
70% 86%
0.006*
73% 83%
0.093
74% 76%
1.000
DIGOXIN Admission Discharge
13% 16%
0.263
12% 15%
0.687
12% 16%
0.581
17% 19%
1.000
GAI-3 Admission Discharge
11% 14%
0.383
12% 12%
1.000
10% 11%
1.000
12% 19%
0.250
GAI-3-mod. Admission Discharge
5.1% 7.1%
0.388
5.1% 3.4%
1.000
4.1% 7.2%
0.453
7% 12%
0.500
GAI-5 Admission Discharge
0.5% 2.5%
0.219
1.7% 3.4%
1.000
0% 2.1%
n. a.
0% 2.4%
n. a.
LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; ACEI = Angiotensin-converting enzyme inhibitor; ARB = Angiotensin receptor blocker; BB = Beta-blocker; MRA = Mineralocorticoid receptor antagonist; GAI = Guideline-adherence index; n.a. = Non-applicable. ⁎ Significant if p b 0.05 (McNemar's test).
significant differences were observed between GAIs on admission and at discharge.
3.2.4. Therapy modifications Up-titration of doses to N50% of target and/or drug prescription was recorded in 27% of the patients. Therapy remained unchanged in 46%, meaning that in the remaining 27% the drugs were discontinued and/ or down-titrated to b 50% of target. In discharge letters, the reasons for these modifications were given in 45% (24 patients) and included drug intolerance or co-morbidities that limited drug initiation/up-titration.
3.3. Survival analysis During the follow-up period (7–10 months), death occurred in 25% of the patients with LVSD, 20% with PLVEF, and 19% with no echocardiography report, representing 21% of the total cohort. The probability of survival at 6 months after discharge was 86% and was lowest in the LVSD group (Table 4). An NT-proBNP measurement was available for 165 of the 198 patients in the study and was included in the survival analysis. The adjusted multivariate Cox proportional hazards analysis showed that age (p = 0.014), NT-proBNP value on admission (p = 0.004),
Fig. 2. Specific drug modifications between patient admission and discharge.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
5
Table 3 Specific drug prescription within therapeutic class at patient discharge. Pharmacological classes and drugs
Prescription (No., %)
Daily target dose in mg
Median dose in mg (IQR)
Mean percentage of target dose (%)
Patients reaching ≥50% of target dose (%)
ACEI (N = 112) Enalapril Fosinopril Captopril Lisinopril Perindopril Ramipril Trandolapril Zofenopril
112 (56.6) 11 (9.8) 1 (0.9) n.a. 150 n.a. 35 59 (52.7) 33 (29.5) 3 (2.7) 5 (4.5)
n.a. 20 20 n.a. n.a. 8 10 4 30
n.a. 10 (2.6–20) 20 (20–20) n.a. n.a. 4 (4–8) 5 (2.5–10) 4 (2–4) 11.4 (5.7–22.5)
60 59 100 n.a. n.a. 63 52 83 47
67.0 54.5 100.0
ARB (N = 30) Losartan Irbesartan Candesartan Telmisartan Valsartan
30 (15.2) 12 (40.0) n.a. 300 5 (16.7) 6 (20.0) 7 (23.3)
n.a. 100 n.a. 32 80 320
n.a. 100 (50–100) n.a. 16 (16–32) 40 (40–80) 160 (160–320)
70 76 n.a. 63 68 67
90.0 91.7
BB (N = 129) Bisoprolol Carvedilol Metoprolol tartrate Nebivolol
129 (65.2) 86 (66.7) 27 (20.9) 6 (4.7) 10 (7.8)
n.a. 10 50 200 10
n.a. 5 (2.5–5) 12.5 (12.5–25) 100 (50–100) 3.8 (2.5–5)
45 46 41 42 48
51.2 52.3 44.4 66.7 50.0
MRA (N = 45) Spironolactone Eplerenone
45 (22.7) 44 (97.8) 1 (2.2)
n.a. 25 50
n.a. 25 (25–25) 13 (13–13)
112 112 26
95.6 88.6 n.a.
78.0 51.5 100.0 40.0
80.0 85.7 100.0
ACEI = Angiotensin-converting enzyme inhibitor; ARB = Angiotensin receptor blocker; BB = Beta-blocker; MRA = Mineralocorticoid receptor antagonist; n.a. = Non-applicable.
patient HF subgroup (p = 0.047), and the GAI (p = 0.015) influenced the survival time, but no difference between male and female patients could be confirmed (p = 0.190). Variables in the subgroups of GAI (GAI-123 or GAI-0) and subgroups of HF patients (LVSD, PLVEF or no report) were found to interact, meaning that they had a combined effect on survival (Table 5). The covariates included in the Cox regression model significantly improved the prediction of survival (p = 0.004). The model demonstrated that survival of the patients without an echocardiography report and the patients with LVSD did not differ significantly in either GAI category. On the other hand, the interaction between the GAI and the HF subgroup showed that the risk of death between PLVEF and LVSD subgroups depended on whether or not at least one HF drug was prescribed at discharge. Thus, when no HF drugs were prescribed, PLVEF patients had a significantly better prognosis than the LVSD patients (HR = 0.075, 95% CI = 0.009–0.627, p = 0.017; Table 5); however, the survival of LVSD patients was significantly better when they were discharged with at least one HF drug than when they were discharged with none (HR = 0.142, 95% CI = 0.029–0.683, p = 0.015; Table 5).
Table 4 Patient 6-month survival after discharge (N = 198). Category Sex Male Female HF group LVSD PLVEF No report Number of HF drugs 0 1 2 3
Number of patients
Number of deaths
When at least one HF drug was prescribed at discharge, PLVEF patients seemed to be less at risk of death than LVSD patients (HR = 0.768, 95% CI = 0.015–40.7, p N 0.05), but among PLVEF patients better survival was observed in those with no HF drugs at discharge (HR = 1.461, 95% CI = 0.028–77.3, p N 0.05). In addition, in the HF subgroup without an echocardiography report, survival did not differ between those who received HF drugs at discharge and those who did not. Lastly, the prescription of at least one HF drug implied prescription of either ACEI/ARB or BB or both (three patients were receiving only MRAs at discharge). A combination of ACEI/ARB and BB was the most common prescription, with 82% of patients reaching GAI-2 at discharge. 4. Discussion Our survey showed that drug prescription generally increased between admission and discharge, with many patients reaching ≥50% of the target doses. However, in a significant proportion of patients, treatment with ACEIs/ARBs was terminated and MRAs were frequently not prescribed. Thus, GAI-3 was met by only a handful of patients at discharge, showing that HF pharmacotherapy was not significantly
Table 5 Multivariate Cox regression model results (N = 165). 6-month survival (SE) in %
95% CI
102 96
14 14
86.3 (3.4) 85.4 (3.6)
79.6–93.0 78.4–92.5
59 97 42
11 11 6
81.4 (5.1) 88.7 (3.2) 85.7 (5.4)
71.4–91.3 82.3–95.0 75.1–96.3
20 67 84 27
5 9 10 4
75.0 (9.7) 86.6 (4.2) 88.1 (3.5) 85.2 (6.8)
56.0–94.0 78.4–94.7 81.2–95.0 71.8–98.6
LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; HF = Heart failure; CI = Confidence interval.
Variable
HR
95% CI
p value
Male vs. female Age (years) NT-proBNP GAI-123 vs. GAI-0 PLVEF vs. LVSD No report vs. LVSD GAI-123⁎PLVEF GAI-123⁎No report
1.58 1.064 2.985 0.142 0.075 0.119 10.302 5.375
0.8–3.13 1.01–1.12 1.43–6.24 0.03–0.68 0.01–0.63 0.01–1.46 1.1–96.4 0.37–79.23
0.19 0.014 0.004 0.015 0.017 0.096 0.041 0.221
NT-proBNP = N-terminal pro-peptide of brain natriuretic peptide (log10-transformed); GAI-123 = Guideline-adherence index, at least one drug at discharge; GAI-0 = Guideline-adherence index, no drug at discharge; GAI-123*PLVEF = Interaction between GAI-123 and PLVEF; GAI-123*No report = Interaction between GAI-123 and unavailable echocardiography report; PLVEF = Preserved left ventricular ejection fraction; LVSD = Left ventricular systolic dysfunction; HR = Hazard ratio; CI = Confidence interval.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
6
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
optimized during hospitalization. In addition, loop diuretics were frequently prescribed regardless of other HF therapy modifications. Although our cohort was smaller than in most previous reports, it is comparable in terms of the co-morbidity profiles [6–8,10,11,13,20,21]. In clinical practice and in healthcare registries, as opposed to clinical trials, some patients may not have an available echocardiography report. Also in this context, our cohort resembled other studies that included all patients with HF of any etiology and any left ventricular function [6,8–10]. However, it should be emphasized that our cohort included more female patients [5–7,9–11,13,17,21,22] and more patients with PLVEF than in other studies [6,9,10]. In comparison of our findings with other studies, the proportions of patients prescribed an individual drug class were lower [5,6, 9–11,13,21,22]. Also, drug prescription during hospitalization increased significantly only for BBs, not for ACEI/ARBs or MRAs, as reported elsewhere [6,9–11]. Low prescription of MRAs has been observed previously, but it was always higher than in our cohort [5,6,9–12,19,21]. The trend of ACEI/ARB discontinuation, in particular in the patients with LVSD, has not been described previously [6, 9–11] and we have no reasonable explanation for such action. We found that the most noticeable increase in prescription of BBs and loop diuretics occurred in the LVSD patients but, together with ACEI/ARB reduction and only a slight increase in MRA prescription, these observations were the same as the general observations in the cohort as a whole. Moreover, prescription patterns were similar in patients with PLVEF and those without an echocardiography report, the proportions of patients meeting GAI-3 increasing during hospitalization in those two groups but not in LVSD patients, despite the published evidence [1]. Among all patients prescribed a particular drug, not many were at target dose, which resulted in very few patients meeting mGAI-3 at discharge. In addition, the patients with LVSD were the only ones where mGAI-3 decreased between admission and discharge. Therapy modifications toward less guideline-adherent treatment, such as drug termination or dose reduction to b50% of target dose, were recorded in 27% of all the patients and in 27% of the patients with LVSD. The reasons for those decisions were not given in 15% of the cohort and 14% of the LVSD patients, and therefore remain unknown. This could be misinterpreted as guideline non-adherence, although a valid reason might be behind such decisions. Nonetheless, what is even more important is the risk to the patient if, despite the presence of a valid contraindication, these therapies were to be reinstated. Such concerns need particular attention, because the hospital stay and the HF clinic provide key opportunities for tailoring therapy and defining goals for the management of the patient's condition. There may be some specific reasons for our findings. First, with an older population than in previous reports [4,7,11,13,15,20–22], the lower level of prescription of key pharmacological agents could be justified to some extent. Second, our survey included several patients with PLVEF and those without echocardiographically confirmed HF, which could have also contributed to lower proportions of patients with prescribed drugs; however, this appears unlikely because no major differences were found between the three HF groups. Third, it is known that HF drug use at discharge is highly associated with prescription before or on admission [11]. Nearly half of our cohort had no HF pharmacotherapy changes during hospitalization, probably reflecting the preservation of established patterns rather than true patient needs. In that context, we should consider potential relative contraindications at baseline, which may have affected the clinical decisions. However, we specifically searched for existing reasons for non-prescription, drug termination, or dose reduction in discharge letters, to add a qualitative value to the GAI assessment. The reasons given in 12% of the cohort (45% of those with drug termination and/or down-titration) included contraindications (severe renal dysfunction, hypotension, asthma, etc.) and drug intolerance, as reported elsewhere [9,12,14,22]. With regard to the complexity and comorbidities of the patients, the prevalence of contraindications to pharmacotherapy is significant;
thus, the use of guideline-recommended drugs decreases due to the “evidence gap” in treating high-risk patients. A “treatment gap” might also occur in some of these patients, who are not treated despite being eligible [22]. Other possible reasons such as polypharmacy, age, or socio-medical reasons [12,14] could have also contributed to physicians' decisions but were not documented. Similar reasons could explain why not many of the patients reached target doses and why down-titration was common. Patient needs and drug tolerability might also have played a part, as implied in the study by Maggioni et al. [9]. In our study, the exception of MRA, where the target dose was exceeded and many patients were “over-titrated”, was due to setting the target dose at the lowest effective recommended dose (25 mg spironolactone), in contrast to other studies [1,10,12,19]. Lastly, the survival analysis showed that if at least one HF drug was prescribed at discharge (ACEI/ARB, or BB, or both in most cases), the patients with LVSD had a better survival prognosis, supporting the pharmacotherapy recommendations in the guidelines and matching other published results [1,5]. At the same time, the patients with PLVEF had better survival than those with LVSD, but only when no HF drugs were prescribed at discharge (GAI-0). This finding is similar to the reported results of one meta-analysis, which also showed that when HF drugs are prescribed, better prognosis remains in favor of patients with PLVEF [23]. Nonetheless, in our study the survival of patients with PLVEF and those without an echocardiography report was not significantly different in relation to the number of HF drugs at discharge. Furthermore, the patients without a report were presumably a mixed population, with LVSD and PLVEF in unknown proportions; therefore, their survival supposedly represents the combined survival of the other two subgroups. The patients without reports did not significantly stand out from the other two subgroups in any aspect, possibly for the same reason. We can conclude that in 15% of the cohort where no reasons for nonadherence were reported, adherence to the guidelines was lower at discharge than on admission. Optimal adherence was not reached in patients where no therapy modifications were made, particularly in the LVSD group, which should have reached greatest adherence. Some of these patients might have remained undertreated or undergone omission and dosage errors that occasionally occur [9,12], which could have affected their survival. 4.1. Limitations The main limitation of our survey was the modest size of the sample. In addition, there was no definitive confirmation of HF diagnosis in the group of patients without an available echocardiography report, although it appears clinically reasonable to presume HF in these patients. The treatment guidelines in HF are specific for patients with LVSD, whereas there is limited or no evidence for other HF patients. Therapeutic patterns, however, are only marginally different, as the optimal management of risk factors in PLVEF patients would indicate a similar if not identical therapy. Lastly, the survival analysis had to be simplified in the matter of the GAI because of the small sample sizes on primary categorization. Moreover, the survival analysis was based on death from any cause, and any potential therapy adjustments or other treatment changes during the follow-up period were unknown. 4.2. Conclusions and clinical implications We have demonstrated suboptimal implementation of treatment guidelines in hospitalized patients with HF. Although this could be justified for non-prescription of target doses, withholding such therapy at discharge could translate to long-term non-prescription and potentially be fatal, especially in patients with LVSD. Increased attention
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
should be paid to defining the reasons for therapy modifications and to report them in discharge letters. The lack of such information could result in further inappropriate drug prescription and seems to imply nonadherence to the guidelines. Moreover, where pharmacological treatment could be subsequently optimized after hospital discharge, further instructions for its adjustment (for example, at the HF clinic) should be given.
[9]
[10]
[11]
Conflict of interest The authors state that they have no conflicts of interest.
[12] [13]
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ejim.2015.12.007.
[14]
[15]
References [1] McMurray JJV, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: the task force for the diagnosis and treatment of acute and chronic heart failure 2012 of the European Society of Cardiology. Developed in collaboration with the heart. Eur J Heart Fail 2012;14:803–69. [2] Jessup M, Abraham WT, Casey DE, Feldman AM, Francis GS, Ganiats TG, et al. 2009 focused update: ACCF/AHA guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines: developed in collaboration with t. Circulation 2009;119:1977–2016. [3] Cleland JGF, Cohen-Solal A, Aguilar JC, Dietz R, Eastaugh J, Follath F, et al. Management of heart failure in primary care (the IMPROVEMENT of Heart Failure Programme): an international survey. Lancet 2002;360:1631–9. [4] Komajda M, Follath F, Swedberg K, Cleland J, Aguilar JC, Cohen-Solal A, et al. The EuroHeart Failure Survey programme–a survey on the quality of care among patients with heart failure in Europe. Part 2: treatment. Eur Heart J 2003;24:464–74. [5] Komajda M, Lapuerta P, Hermans N, Gonzalez-Juanatey JR, van Veldhuisen DJ, Erdmann E, et al. Adherence to guidelines is a predictor of outcome in chronic heart failure: the MAHLER survey. Eur Heart J 2005;26:1653–9. [6] Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Follath F, Harjola V-P, et al. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population. Eur Heart J 2006;27:2725–36. [7] Frankenstein L, Remppis A, Fluegel A, Doesch A, Katus HA, Senges J, et al. The association between long-term longitudinal trends in guideline adherence and mortality in relation to age and sex. Eur J Heart Fail 2010;12:574–80. [8] Maggioni AP, Dahlström U, Filippatos G, Chioncel O, Crespo Leiro M, Drozdz J, et al. EURObservational Research Programme: regional differences and 1-year follow-up
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
7
results of the heart failure pilot survey (ESC-HF pilot). Eur J Heart Fail 2013;15: 808–17. Maggioni AP, Anker SD, Dahlström U, Filippatos G, Ponikowski P, Zannad F, et al. Are hospitalized or ambulatory patients with heart failure treated in accordance with European Society of Cardiology guidelines? Evidence from 12,440 patients of the ESC heart failure long-term registry. Eur J Heart Fail 2013;15:1173–84. Maggioni AP, Dahlström U, Filippatos G, Chioncel O, Leiro MC, Drozdz J, et al. EURObservational Research Programme: the heart failure pilot survey (ESC-HF pilot). Eur J Heart Fail 2010;12:1076–84. Krantz MJ, Ambardekar AV, Kaltenbach L, Hernandez AF, Heidenreich PA, Fonarow GC. Patterns and predictors of evidence-based medication continuation among hospitalized heart failure patients (from get with the guidelines-heart failure). Am J Cardiol 2011;107:1818–23. Oertle M, Bal R. Understanding non-adherence in chronic heart failure: a mixedmethod case study. Qual Saf Health Care 2010;19, e37. Zugck C, Franke J, Gelbrich G, Frankenstein L, Scheffold T, Pankuweit S, et al. Implementation of pharmacotherapy guidelines in heart failure: experience from the German Competence Network Heart Failure. Clin Res Cardiol 2012;101:263–72. Steinman MA, Patil S, Kamat P, Peterson C, Knight SJ. A taxonomy of reasons for not prescribing guideline-recommended medications for patients with heart failure. Am J Geriatr Pharmacother 2010;8:583–94. Steinman MA, Sudore RL, Peterson CA, Harlow JB, Fried TR. Influence of patient age and comorbid burden on clinician attitudes toward heart failure guidelines. Am J Geriatr Pharmacother 2012;10:211–8. Steinman MA, Dimaano L, Peterson CA, Heidenreich PA, Knight SJ, Fung KZ, et al. Reasons for not prescribing guideline-recommended medications to adults with heart failure. Med Care 2013;51:901–7. Cleland JGF, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, et al. The EuroHeart Failure Survey programme—a survey on the quality of care among patients with heart failure in Europe. Part 1: patient characteristics and diagnosis. Eur Heart J 2003;24:442–63. American National Kidney Foundation official webpage: http://www.kidney.org/ professionals/kdoqi/gfr_calculator.cfm. Lainscak M, Cleland JGF, Lenzen MJ, Follath F, Komajda M, Swedberg K. International variations in the treatment and co-morbidity of left ventricular systolic dysfunction: data from the EuroHeart Failure survey. Eur J Heart Fail 2007;9:292–9. Calvin JE, Shanbhag S, Avery E, Kane J, Richardson D, Powell L. Adherence to evidence-based guidelines for heart failure in physicians and their patients: lessons from the Heart Failure Adherence Retention Trial (HART). Congest. Heart Fail. 18: 73–8. Forman DE, Cannon CP, Hernandez AF, Liang L, Yancy C, Fonarow GC. Influence of age on the management of heart failure: findings from get with the guidelinesheart failure (GWTG-HF). Am Heart J 2009;157:1010–7. Peterson PN, Rumsfeld JS, Liang L, Hernandez AF, Peterson ED, Fonarow GC, et al. Treatment and risk in heart failure: gaps in evidence or quality? Circ Cardiovasc Qual Outcomes 2010;3:309–15. (MAGGIC) MGG in CHF. The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: an individual patient data meta-analysis. Eur Heart J 2011;33:1750–7.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
Contents lists available at ScienceDirect
European Journal of Internal Medicine journal homepage: www.elsevier.com/locate/ejim
Original Article
Therapy modifications during hospitalization in patients with chronic heart failure Andreja Detiček a, Igor Locatelli a, Tina Roblek a, Aleš Mrhar a, Mitja Lainscak b,c,⁎ a b c
Faculty of Pharmacy, University of Ljubljana, Askerceva cesta 7, SI-1000 Ljubljana, Slovenia Department of Cardiology and Department of Research and Education, General Hospital Celje, Oblakova 5, SI-3000 Celje, Slovenia Faculty of Medicine, University of Ljubljana, Vrazov trg 2, SI-1104 Ljubljana, Slovenia
a r t i c l e
i n f o
Article history: Received 24 November 2015 Accepted 12 December 2015 Available online xxxx Keywords: Heart failure Drug therapy Guideline Hospitalization
a b s t r a c t Background: Guidelines on suggested pharmacological treatments for heart failure (HF) are not optimally implemented in clinical practice and whether pharmacotherapy adjustment actually happens in daily practice is largely unknown. We aimed to investigate pharmacotherapy modifications during hospitalization. Methods: This was a prospective observational survey where all admissions were screened for HF; 210 patients were included. The guideline adherence index (GAI) and modified GAI (mGAI, if ≥50% of target dose) were used to grade the pharmacotherapy. Results: Among 198 patients discharged alive (mean age 77 years, 51% male), 49% had preserved left ventricular ejection fraction (PLVEF) and 30% had left ventricular systolic dysfunction (LVSD); the echocardiography report was unavailable for 21%. Angiotensin-converting enzyme inhibitors/angiotensin receptor blockers, beta-blockers and mineralocorticoid receptor antagonists were prescribed to 78%, 58% and 20% of patients on admission and 72%, 65% and 23% at discharge, respectively. Overall, 14% of patients met GAI-3, but at discharge only 7% met mGAI-3. One of the key drugs was stopped or down-titrated in 27%. During follow-up, 21% of patients died (25% with LVSD). Patients with LVSD discharged with at least one HF drug had a lower risk of death than patients with none (HR = 0.142, 95% CI = 0.029–0.683, p = 0.015). Patients with PLVEF had better prognosis than LVSD patients when no HF drugs were prescribed at discharge (HR = 0.075, 95% CI = 0.009–0.627, p = 0.017). Conclusions: The pharmacotherapy of HF patients did not improve significantly during hospitalization, remaining suboptimal. Treatment with key drugs was terminated or reduced in a significant proportion of patients, mostly without specific written justification. © 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
1. Introduction Long-term healthcare registries show that the implementation of the guidelines [1,2] on treatment of heart failure (HF) in clinical practice has improved over the years and that life-saving therapies are widely used in such patients [3–9]. Even when patients are hospitalized because of condition deterioration, there is ample opportunity for optimization of individual pharmacotherapy, and prescription of guidelinerecommended drugs usually increases during hospitalization [6,7, 9–11]. Different forms of guideline adherence index/indicator (GAI) have been used to assess implementation of the recommendations in
Abbreviations: ACEI, Angiotensin-converting enzyme inhibitor; ARB, Angiotensin receptor blocker; BB, Beta-blocker; CI, Confidence interval; eGFR, Bstimated glomerular filtration rate; GAI, Guideline-adherence index; HF, Heart failure; HR, Hazard ratio; LVEF, Left ventricular ejection fraction; LVSD, Left ventricular systolic dysfunction; PLVEF, Preserved left ventricular ejection fraction; MDRD, Modification of Diet in Renal Disease; MRA, Mineralocorticoid receptor antagonist; NT-proBNP, N-terminal pro-peptide of brain natriuretic peptide; NYHA, New York Heart Association. ⁎ Corresponding author. Tel.: +386 3 423 38 00; fax: +386 3 423 37 54, +386 31 379 533 (mobile). E-mail address: [email protected] (M. Lainscak).
practice, and studies have demonstrated that the GAI predicts outcome [5,7,12,13]. However, non-prescription, drug termination and dose adjustment are also reasonable in some patients and could be guided by the patient's situation and side effects [1,2,9,12,14–16]. These reasons are widely underreported and often not documented in medical records, especially reasons of a contextual nature [12,16]. They therefore remain unknown in individual patients, which could lead to overestimation of guideline non-implementation, and also to potential hazards for the patient if such therapies are restarted. The aim of this study was to evaluate modifications of HF pharmacotherapy during hospitalization in order to analyze current clinical practice. Survival analysis was used to investigate the effects of drug therapy on patient outcome.
2. Methods 2.1. Survey design, patient population and inclusion criteria The survey was part of a larger study in hospitalized patients in Slovenia. The protocol of the survey was reviewed and approved by
http://dx.doi.org/10.1016/j.ejim.2015.12.007 0953-6205/© 2015 European Federation of Internal Medicine. Published by Elsevier B.V. All rights reserved.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
2
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
the National Ethics Committee and is available at clinicaltrials.gov (NCT01855165). We prospectively screened all admissions during a 14-week enrollment period at the University Clinic Golnik in 2013. Patients were included if they met at least one of four inclusion criteria [17]:
simplified GAI comprising only two categories: patients discharged with at least one HF drug (GAI-123) and those discharged with none (GAI-0).
- known HF diagnosis prior to admission, - echocardiography report confirming left ventricular dysfunction, - symptoms and signs of HF with elevated serum natriuretic peptide (NT-proBNP), and - treatment with a loop diuretic within 24 h after admission for reasons other than renal failure.
All statistical calculations were performed using IBM SPSS Statistics for Windows, Version 22.0. A p-value b 0.05 was considered statistically significant. McNemar's chi-square test and Wilcoxon signed-rank test were used for categorical data; t-test was used for continuous data. All tests were two-sided. Continuous variables are reported as mean (standard deviation; SD) or median (1st quartile–3rd quartile; IQR). The multivariate Cox proportional hazards model was used for survival analysis. A regression model was built to investigate the effects of the GAI on survival adjusted for potential effect of patient HF subgroup, age, sex and natriuretic peptide value (log10-transformed NT-proBNP) on survival. Interaction terms among between GAI and patient HF subgroup were also tested. The results are presented as hazard ratios (HRs) with respective 95% confidence intervals (95% CIs).
We excluded patients with any terminal chronic disease (e.g. cancer, chronic obstructive pulmonary disease, chronic kidney disease, etc.) and those who died during hospitalization. None of the authors was included in patient management. The included patients were divided into three HF subgroups according to the echocardiography report: - left ventricular systolic dysfunction (LVSD), when left ventricular ejection fraction (LVEF) was b55% - preserved left ventricular ejection fraction (PLVEF), when LVEF was ≥55% or a qualitative description of “preserved ejection fraction” or “mild left ventricular dysfunction” was given, and - no echocardiography report
We retrieved the demographic characteristics, medical histories, laboratory test results, echocardiography reports, and pharmacological treatments at admission and discharge. We also recorded the reasons for admission, the proportion of patients with anemia (serum hemoglobin level b 120 g/L in women, b 130 g/L in men), and the proportions of patients with the following findings on admission: K+ N 5.0 mmol/L, estimated glomerular filtration rate (eGFR) b30 mL/min/1.73 m2, and systolic blood pressure b 90 mm Hg. Renal function in all the patients was estimated using the simplified Modification of Diet in Renal Disease (MDRD) equation [18]. We evaluated the prescription, mean dose, and proportion of patients prescribed the target dose at admission and discharge for the following drugs: angiotensin-converting enzyme inhibitors (ACEIs)/ angiotensin receptor blockers (ARBs), beta-blockers (BBs) and mineralocorticoid receptor antagonists (MRAs) [1]. The target dose was set as the maximum daily dose recommended in the guidelines [1], with the exception of losartan where the target was set at 100 mg (instead of 150 mg) as used in the European Society of Cardiology-HF Pilot Survey [10]. Where the daily dose was stated as a range, the target dose was set at the lower limit, except in the case of lisinopril where the upper limit was used. Target doses were then considered equivalently effective within each pharmacological class, as already implemented [19]. Adherence to the guidelines was evaluated using the GAI [5]. In our survey, GAI-1, GAI-2, and GAI-3 represent the prescription of one, two, or all three neurohormonal antagonists. The indexes were also used in a modified form (mGAI) to simultaneously describe the prescription and reaching ≥50% of the target dose. In addition, we assessed therapy modification by following two main changes: the prescription or discontinuation of a drug, and up-titration to N 50% of the target dose or down-titration to b50%. When a drug was discontinued and/or downtitrated, we searched the discharge letter for the physician's explanation of the modification. We also noted the prescription of loop diuretics and digoxin. For GAI-5, the prescription of loop diuretics and digoxin was added to GAI-3. For survival analysis, we obtained all-cause mortality data from the Central Population Registry of Slovenia. Patient follow-up started at discharge and ended after a period of 6 months was exceeded for the last discharged patient. Primary categorization according to drug therapy resulted in subgroups that were too small; therefore, we used a
2.2. Statistical analysis
3. Results 3.1. Patient characteristics Of 210 eligible patients (15% of all admissions during the screening period), 198 were discharged alive after a mean (SD) of 11 (SD: 7) days and their data were analyzed. Overall, 49% of the patients had PLVEF and 30% had LVSD; the echocardiography report was not available for 21% (Fig. 1). Three inclusion criteria were met in 45% of patients, 34% reached two, 15% one, and in 6% of all the four criteria were met. The presence of symptoms and signs with elevated NT-proBNP was the most commonly met criterion (86% of patients). The mean (SD) age of the patients was 77 (SD: 8) years and 51% of the patients were male. The female patients were significantly older than the males: 78 (SD: 8) vs. 75 (SD: 8) years, p = 0.007, t-test. Patients with PLVEF were significantly older than those with LVSD: 78 (SD: 8) vs. 74 (SD: 9) years, p = 0.027, t-test. The patient characteristics are shown in Table 1. Overall, 84% of the patients were in NYHA functional classes III and IV, making HF decompensation the most common primary reason for admission (58%). Other common reasons were infections (11%), chronic
ADMITTED
1372
INCLUSION CRITERIA MET
DIED
210
12
NO ECHOCARDIOGRAPHY REPORT
DISCHARGED
198
42
ECHOCARDIOGRAPHY REPORT
156
LEFT VENTRICULAR SYSTOLIC DYSFUNCTION
PRESERVED LEFT VENTRICULAR EJECTION FRACTION
59
97 Fig. 1. Patient flow chart.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
3
Table 1 Patient characteristics on admission. All patients
LVSD
PLVEF
No report
198 (100%) 102 (51.5%) 77 (8) 11 (7)
59 (30%) 34 (58%) 74 (9) 10 (8)
97 (49%) 48 (49%) 78 (8) 11 (7)
42 (21%) 20 (48%) 79 (7) 11 (7)
1 (0.5%) 30 (15%) 86 (43%) 81 (41%)
1 (1.7%) 5 (8.5%) 24 (41%) 29 (49%)
0 (0%) 19 (20%) 41 (42%) 37 (38%)
0 (0%) 6 (14%) 21 (50%) 15 (36%)
Co-morbidities Myocardial infarction IHD and/or CAD Arterial hypertension Atrial fibrillation Aortic stenosis Valvular dysfunction Chronic kidney disease Diabetes COPD
44 (22%) 56 (28%) 161 (81%) 110 (56%) 21 (11%) 62 (31%) 50 (25%) 61 (31%) 53 (27%)
18 (31%) 22 (37%) 46 (78%) 34 (58%) 9 (15%) 29 (49%) 19 (32%) 17 (29%) 16 (27%)
13 (13%) 22 (23%) 80 (83%) 55 (57%) 12 (12%) 30 (31%) 21 (22%) 30 (31%) 31 (32%)
13 (31%) 12 (29%) 35 (83%) 21 (50%) 0 (0%) 3 (7.1%) 10 (24%) 14 (33%) 6 (14%)
Measurements LVEF (%) Systolic BP (mm Hg) Diastolic BP (mm Hg) Heart rate (beats/min) BMI (kg/m2) S-Potassium (mmol/L) S-Sodium (mmol/L) S-NT-proBNP (ng/L) S-Creatinine (mmol/L) eGFR (mL/min per 1.73 m2)
n.a. 142 (24) 79 (14) 89 (22) 31 (7) 4.4 (0.8) 139 (4) 3127 (1040–6268) 102 (79–127) 52 (39–74)
40 (30–50) 138 (25) 81 (15) 90 (25) 30 (6) 4.3 (0.8) 139 (4) 3420 (2020–9278) 94 (68–117) 60 (39–93)
67 (60–71) 145 (23) 78 (14) 90 (23) 31 (7) 4.5 (0.8) 139 (4) 3119 (904–5615) 103 (85–129) 50 (40–70)
n.a. 138 (25) 76 (15) 85 (14) 31 (6) 4.4 (0.6) 139 (4) 2416 (1538–6581) 111 (86–129) 50 (38–61)
Frequency (%) Male Age (years): Days of hospitalization NYHA classification Class I Class II Class III Class IV
Treatment-limiting conditions Potassium N 5.0 mmol/L eGFR b 30 mL/min per 1.73 m2 Systolic BP b 90 mmHg
28 (14%) 16 (8.1%) 2 (1.0%)
6 (10%) 7 (12%) 2 (3.4%)
18 (19%) 5 (5.2%) 0 (0%)
4 (9.5%) 4 (9.5%) 0 (0%)
The values presented refer to the situation on admission. The results are presented as frequency (percentages), mean (SD), or median (1st quartile–3rd quartile; IQR). LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; IHD = Ischaemic heart disease; CAD = Coronary artery disease; NYHA = New York Heart Association; COPD = Chronic obstructive pulmonary disease; LVEF = Left ventricular ejection fraction; BP = Blood pressure; BMI = Body mass index; eGFR = Estimated glomerular filtration rate; n.a. = Non-applicable.
obstructive pulmonary disease (9%), malignancy (5%), and diagnostic procedures (3%). Hypertension (81%) and atrial fibrillation (56%) were the most frequent diseases, with no differences between the three subgroups of patients. NT-proBNP was measured in 83% of patients during hospitalization: the concentration was ≥125 ng/L in 81% and ≥300 ng/L in 78%. Anemia was present in 37%, 59% of patients had an eGFR b 60 mL/min/1.73m2, and 23% had at least one of the HF drug prescription-limiting conditions on admission (K+ N 5.0 mmol/L, eGFR b30 mL/min/1.73 m2, or systolic blood pressure b 90 mm Hg). 3.2. Pharmacotherapy The median number of all drugs prescribed was 8 (IQR: 6–10), cardiovascular drugs 5 (IQR: 3–6). At discharge, 45% of the patients were prescribed more drugs than on admission (p = 0.034). 3.2.1. Drug prescription Proportions of patients prescribed a particular drug and GAIs at admission and discharge, together with p-values, are shown in Table 2 and Fig. 2. We observed a significant increase in the prescription of BBs during hospitalization (p = 0.016), and the prescription of loop diuretics was significantly higher at discharge (p = 0.003). There were no significant differences for other drugs. In patients with LVSD, we noticed the same trends as in the two other groups, indicating lower
prescription of ACEIs/ARBs and higher prescription of other drugs at discharge (Table 2). The most commonly prescribed drugs were perindopril (ACEI), losartan (ARB), bisoprolol (BB), and spironolactone (MRA) (Table 3). The most commonly prescribed loop diuretic was furosemide. Digoxin was prescribed only to HF patients who also had atrial fibrillation. 3.2.2. Drug dose The mean doses decreased from admission to discharge in all pharmacological classes and were 60% of the target dose in ACEI, 70% in ARB, 45% in BB and 112% in MRA at discharge (Table 3). In the LVSD group of patients, the respective percentages were 57%, 62%, 43%, and 112%. The target dose was prescribed at discharge in 36% of the patients receiving ACEIs/ARBs, 15% receiving BBs and 84% receiving MRAs; ≥50% of the target dose was prescribed in 72%, 51%, and 96% of the patients, respectively (Table 3). 3.2.3. Guideline adherence indexes On admission and at discharge, 10% of the patients were not receiving any of the HF recommended drugs (GAI-0). Overall, one-third met GAI-1 and approximately 40% met GAI-2 at both time points. Only 14% met GAI-3 at discharge; 7% were also prescribed all drugs at ≥ 50% of the target dose (mGAI-3; Table 2). However, no patient reached the target dose for all the drugs. GAI-5 was met in five patients at discharge. No
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
4
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
Table 2 Therapy modifications between patient admission and discharge. All patients N = 198
p value
LVSD N = 59
p value
PLVEF N = 97
p value
No report N = 42
p value
ACEI/ARB Admission Discharge
78% 72%
0.065
76% 71%
0.549
76% 68%
0.115
83% 81%
1.000
BB Admission Discharge
58% 65%
0.016*
61% 73%
0.065
59% 63%
0.424
52% 60%
0.375
MRA Admission Discharge
20% 23%
0.405
20% 25%
0.453
21% 21%
1.000
19% 24%
0.625
LOOP DIURETIC Admission Discharge
72% 82%
0.003*
70% 86%
0.006*
73% 83%
0.093
74% 76%
1.000
DIGOXIN Admission Discharge
13% 16%
0.263
12% 15%
0.687
12% 16%
0.581
17% 19%
1.000
GAI-3 Admission Discharge
11% 14%
0.383
12% 12%
1.000
10% 11%
1.000
12% 19%
0.250
GAI-3-mod. Admission Discharge
5.1% 7.1%
0.388
5.1% 3.4%
1.000
4.1% 7.2%
0.453
7% 12%
0.500
GAI-5 Admission Discharge
0.5% 2.5%
0.219
1.7% 3.4%
1.000
0% 2.1%
n. a.
0% 2.4%
n. a.
LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; ACEI = Angiotensin-converting enzyme inhibitor; ARB = Angiotensin receptor blocker; BB = Beta-blocker; MRA = Mineralocorticoid receptor antagonist; GAI = Guideline-adherence index; n.a. = Non-applicable. ⁎ Significant if p b 0.05 (McNemar's test).
significant differences were observed between GAIs on admission and at discharge.
3.2.4. Therapy modifications Up-titration of doses to N50% of target and/or drug prescription was recorded in 27% of the patients. Therapy remained unchanged in 46%, meaning that in the remaining 27% the drugs were discontinued and/ or down-titrated to b 50% of target. In discharge letters, the reasons for these modifications were given in 45% (24 patients) and included drug intolerance or co-morbidities that limited drug initiation/up-titration.
3.3. Survival analysis During the follow-up period (7–10 months), death occurred in 25% of the patients with LVSD, 20% with PLVEF, and 19% with no echocardiography report, representing 21% of the total cohort. The probability of survival at 6 months after discharge was 86% and was lowest in the LVSD group (Table 4). An NT-proBNP measurement was available for 165 of the 198 patients in the study and was included in the survival analysis. The adjusted multivariate Cox proportional hazards analysis showed that age (p = 0.014), NT-proBNP value on admission (p = 0.004),
Fig. 2. Specific drug modifications between patient admission and discharge.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
5
Table 3 Specific drug prescription within therapeutic class at patient discharge. Pharmacological classes and drugs
Prescription (No., %)
Daily target dose in mg
Median dose in mg (IQR)
Mean percentage of target dose (%)
Patients reaching ≥50% of target dose (%)
ACEI (N = 112) Enalapril Fosinopril Captopril Lisinopril Perindopril Ramipril Trandolapril Zofenopril
112 (56.6) 11 (9.8) 1 (0.9) n.a. 150 n.a. 35 59 (52.7) 33 (29.5) 3 (2.7) 5 (4.5)
n.a. 20 20 n.a. n.a. 8 10 4 30
n.a. 10 (2.6–20) 20 (20–20) n.a. n.a. 4 (4–8) 5 (2.5–10) 4 (2–4) 11.4 (5.7–22.5)
60 59 100 n.a. n.a. 63 52 83 47
67.0 54.5 100.0
ARB (N = 30) Losartan Irbesartan Candesartan Telmisartan Valsartan
30 (15.2) 12 (40.0) n.a. 300 5 (16.7) 6 (20.0) 7 (23.3)
n.a. 100 n.a. 32 80 320
n.a. 100 (50–100) n.a. 16 (16–32) 40 (40–80) 160 (160–320)
70 76 n.a. 63 68 67
90.0 91.7
BB (N = 129) Bisoprolol Carvedilol Metoprolol tartrate Nebivolol
129 (65.2) 86 (66.7) 27 (20.9) 6 (4.7) 10 (7.8)
n.a. 10 50 200 10
n.a. 5 (2.5–5) 12.5 (12.5–25) 100 (50–100) 3.8 (2.5–5)
45 46 41 42 48
51.2 52.3 44.4 66.7 50.0
MRA (N = 45) Spironolactone Eplerenone
45 (22.7) 44 (97.8) 1 (2.2)
n.a. 25 50
n.a. 25 (25–25) 13 (13–13)
112 112 26
95.6 88.6 n.a.
78.0 51.5 100.0 40.0
80.0 85.7 100.0
ACEI = Angiotensin-converting enzyme inhibitor; ARB = Angiotensin receptor blocker; BB = Beta-blocker; MRA = Mineralocorticoid receptor antagonist; n.a. = Non-applicable.
patient HF subgroup (p = 0.047), and the GAI (p = 0.015) influenced the survival time, but no difference between male and female patients could be confirmed (p = 0.190). Variables in the subgroups of GAI (GAI-123 or GAI-0) and subgroups of HF patients (LVSD, PLVEF or no report) were found to interact, meaning that they had a combined effect on survival (Table 5). The covariates included in the Cox regression model significantly improved the prediction of survival (p = 0.004). The model demonstrated that survival of the patients without an echocardiography report and the patients with LVSD did not differ significantly in either GAI category. On the other hand, the interaction between the GAI and the HF subgroup showed that the risk of death between PLVEF and LVSD subgroups depended on whether or not at least one HF drug was prescribed at discharge. Thus, when no HF drugs were prescribed, PLVEF patients had a significantly better prognosis than the LVSD patients (HR = 0.075, 95% CI = 0.009–0.627, p = 0.017; Table 5); however, the survival of LVSD patients was significantly better when they were discharged with at least one HF drug than when they were discharged with none (HR = 0.142, 95% CI = 0.029–0.683, p = 0.015; Table 5).
Table 4 Patient 6-month survival after discharge (N = 198). Category Sex Male Female HF group LVSD PLVEF No report Number of HF drugs 0 1 2 3
Number of patients
Number of deaths
When at least one HF drug was prescribed at discharge, PLVEF patients seemed to be less at risk of death than LVSD patients (HR = 0.768, 95% CI = 0.015–40.7, p N 0.05), but among PLVEF patients better survival was observed in those with no HF drugs at discharge (HR = 1.461, 95% CI = 0.028–77.3, p N 0.05). In addition, in the HF subgroup without an echocardiography report, survival did not differ between those who received HF drugs at discharge and those who did not. Lastly, the prescription of at least one HF drug implied prescription of either ACEI/ARB or BB or both (three patients were receiving only MRAs at discharge). A combination of ACEI/ARB and BB was the most common prescription, with 82% of patients reaching GAI-2 at discharge. 4. Discussion Our survey showed that drug prescription generally increased between admission and discharge, with many patients reaching ≥50% of the target doses. However, in a significant proportion of patients, treatment with ACEIs/ARBs was terminated and MRAs were frequently not prescribed. Thus, GAI-3 was met by only a handful of patients at discharge, showing that HF pharmacotherapy was not significantly
Table 5 Multivariate Cox regression model results (N = 165). 6-month survival (SE) in %
95% CI
102 96
14 14
86.3 (3.4) 85.4 (3.6)
79.6–93.0 78.4–92.5
59 97 42
11 11 6
81.4 (5.1) 88.7 (3.2) 85.7 (5.4)
71.4–91.3 82.3–95.0 75.1–96.3
20 67 84 27
5 9 10 4
75.0 (9.7) 86.6 (4.2) 88.1 (3.5) 85.2 (6.8)
56.0–94.0 78.4–94.7 81.2–95.0 71.8–98.6
LVSD = Left ventricular systolic dysfunction; PLVEF = Preserved left ventricular ejection fraction; HF = Heart failure; CI = Confidence interval.
Variable
HR
95% CI
p value
Male vs. female Age (years) NT-proBNP GAI-123 vs. GAI-0 PLVEF vs. LVSD No report vs. LVSD GAI-123⁎PLVEF GAI-123⁎No report
1.58 1.064 2.985 0.142 0.075 0.119 10.302 5.375
0.8–3.13 1.01–1.12 1.43–6.24 0.03–0.68 0.01–0.63 0.01–1.46 1.1–96.4 0.37–79.23
0.19 0.014 0.004 0.015 0.017 0.096 0.041 0.221
NT-proBNP = N-terminal pro-peptide of brain natriuretic peptide (log10-transformed); GAI-123 = Guideline-adherence index, at least one drug at discharge; GAI-0 = Guideline-adherence index, no drug at discharge; GAI-123*PLVEF = Interaction between GAI-123 and PLVEF; GAI-123*No report = Interaction between GAI-123 and unavailable echocardiography report; PLVEF = Preserved left ventricular ejection fraction; LVSD = Left ventricular systolic dysfunction; HR = Hazard ratio; CI = Confidence interval.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
6
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
optimized during hospitalization. In addition, loop diuretics were frequently prescribed regardless of other HF therapy modifications. Although our cohort was smaller than in most previous reports, it is comparable in terms of the co-morbidity profiles [6–8,10,11,13,20,21]. In clinical practice and in healthcare registries, as opposed to clinical trials, some patients may not have an available echocardiography report. Also in this context, our cohort resembled other studies that included all patients with HF of any etiology and any left ventricular function [6,8–10]. However, it should be emphasized that our cohort included more female patients [5–7,9–11,13,17,21,22] and more patients with PLVEF than in other studies [6,9,10]. In comparison of our findings with other studies, the proportions of patients prescribed an individual drug class were lower [5,6, 9–11,13,21,22]. Also, drug prescription during hospitalization increased significantly only for BBs, not for ACEI/ARBs or MRAs, as reported elsewhere [6,9–11]. Low prescription of MRAs has been observed previously, but it was always higher than in our cohort [5,6,9–12,19,21]. The trend of ACEI/ARB discontinuation, in particular in the patients with LVSD, has not been described previously [6, 9–11] and we have no reasonable explanation for such action. We found that the most noticeable increase in prescription of BBs and loop diuretics occurred in the LVSD patients but, together with ACEI/ARB reduction and only a slight increase in MRA prescription, these observations were the same as the general observations in the cohort as a whole. Moreover, prescription patterns were similar in patients with PLVEF and those without an echocardiography report, the proportions of patients meeting GAI-3 increasing during hospitalization in those two groups but not in LVSD patients, despite the published evidence [1]. Among all patients prescribed a particular drug, not many were at target dose, which resulted in very few patients meeting mGAI-3 at discharge. In addition, the patients with LVSD were the only ones where mGAI-3 decreased between admission and discharge. Therapy modifications toward less guideline-adherent treatment, such as drug termination or dose reduction to b50% of target dose, were recorded in 27% of all the patients and in 27% of the patients with LVSD. The reasons for those decisions were not given in 15% of the cohort and 14% of the LVSD patients, and therefore remain unknown. This could be misinterpreted as guideline non-adherence, although a valid reason might be behind such decisions. Nonetheless, what is even more important is the risk to the patient if, despite the presence of a valid contraindication, these therapies were to be reinstated. Such concerns need particular attention, because the hospital stay and the HF clinic provide key opportunities for tailoring therapy and defining goals for the management of the patient's condition. There may be some specific reasons for our findings. First, with an older population than in previous reports [4,7,11,13,15,20–22], the lower level of prescription of key pharmacological agents could be justified to some extent. Second, our survey included several patients with PLVEF and those without echocardiographically confirmed HF, which could have also contributed to lower proportions of patients with prescribed drugs; however, this appears unlikely because no major differences were found between the three HF groups. Third, it is known that HF drug use at discharge is highly associated with prescription before or on admission [11]. Nearly half of our cohort had no HF pharmacotherapy changes during hospitalization, probably reflecting the preservation of established patterns rather than true patient needs. In that context, we should consider potential relative contraindications at baseline, which may have affected the clinical decisions. However, we specifically searched for existing reasons for non-prescription, drug termination, or dose reduction in discharge letters, to add a qualitative value to the GAI assessment. The reasons given in 12% of the cohort (45% of those with drug termination and/or down-titration) included contraindications (severe renal dysfunction, hypotension, asthma, etc.) and drug intolerance, as reported elsewhere [9,12,14,22]. With regard to the complexity and comorbidities of the patients, the prevalence of contraindications to pharmacotherapy is significant;
thus, the use of guideline-recommended drugs decreases due to the “evidence gap” in treating high-risk patients. A “treatment gap” might also occur in some of these patients, who are not treated despite being eligible [22]. Other possible reasons such as polypharmacy, age, or socio-medical reasons [12,14] could have also contributed to physicians' decisions but were not documented. Similar reasons could explain why not many of the patients reached target doses and why down-titration was common. Patient needs and drug tolerability might also have played a part, as implied in the study by Maggioni et al. [9]. In our study, the exception of MRA, where the target dose was exceeded and many patients were “over-titrated”, was due to setting the target dose at the lowest effective recommended dose (25 mg spironolactone), in contrast to other studies [1,10,12,19]. Lastly, the survival analysis showed that if at least one HF drug was prescribed at discharge (ACEI/ARB, or BB, or both in most cases), the patients with LVSD had a better survival prognosis, supporting the pharmacotherapy recommendations in the guidelines and matching other published results [1,5]. At the same time, the patients with PLVEF had better survival than those with LVSD, but only when no HF drugs were prescribed at discharge (GAI-0). This finding is similar to the reported results of one meta-analysis, which also showed that when HF drugs are prescribed, better prognosis remains in favor of patients with PLVEF [23]. Nonetheless, in our study the survival of patients with PLVEF and those without an echocardiography report was not significantly different in relation to the number of HF drugs at discharge. Furthermore, the patients without a report were presumably a mixed population, with LVSD and PLVEF in unknown proportions; therefore, their survival supposedly represents the combined survival of the other two subgroups. The patients without reports did not significantly stand out from the other two subgroups in any aspect, possibly for the same reason. We can conclude that in 15% of the cohort where no reasons for nonadherence were reported, adherence to the guidelines was lower at discharge than on admission. Optimal adherence was not reached in patients where no therapy modifications were made, particularly in the LVSD group, which should have reached greatest adherence. Some of these patients might have remained undertreated or undergone omission and dosage errors that occasionally occur [9,12], which could have affected their survival. 4.1. Limitations The main limitation of our survey was the modest size of the sample. In addition, there was no definitive confirmation of HF diagnosis in the group of patients without an available echocardiography report, although it appears clinically reasonable to presume HF in these patients. The treatment guidelines in HF are specific for patients with LVSD, whereas there is limited or no evidence for other HF patients. Therapeutic patterns, however, are only marginally different, as the optimal management of risk factors in PLVEF patients would indicate a similar if not identical therapy. Lastly, the survival analysis had to be simplified in the matter of the GAI because of the small sample sizes on primary categorization. Moreover, the survival analysis was based on death from any cause, and any potential therapy adjustments or other treatment changes during the follow-up period were unknown. 4.2. Conclusions and clinical implications We have demonstrated suboptimal implementation of treatment guidelines in hospitalized patients with HF. Although this could be justified for non-prescription of target doses, withholding such therapy at discharge could translate to long-term non-prescription and potentially be fatal, especially in patients with LVSD. Increased attention
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007
A. Detiček et al. / European Journal of Internal Medicine xxx (2015) xxx–xxx
should be paid to defining the reasons for therapy modifications and to report them in discharge letters. The lack of such information could result in further inappropriate drug prescription and seems to imply nonadherence to the guidelines. Moreover, where pharmacological treatment could be subsequently optimized after hospital discharge, further instructions for its adjustment (for example, at the HF clinic) should be given.
[9]
[10]
[11]
Conflict of interest The authors state that they have no conflicts of interest.
[12] [13]
Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.ejim.2015.12.007.
[14]
[15]
References [1] McMurray JJV, Adamopoulos S, Anker SD, Auricchio A, Böhm M, Dickstein K, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: the task force for the diagnosis and treatment of acute and chronic heart failure 2012 of the European Society of Cardiology. Developed in collaboration with the heart. Eur J Heart Fail 2012;14:803–69. [2] Jessup M, Abraham WT, Casey DE, Feldman AM, Francis GS, Ganiats TG, et al. 2009 focused update: ACCF/AHA guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/ American Heart Association Task Force on Practice Guidelines: developed in collaboration with t. Circulation 2009;119:1977–2016. [3] Cleland JGF, Cohen-Solal A, Aguilar JC, Dietz R, Eastaugh J, Follath F, et al. Management of heart failure in primary care (the IMPROVEMENT of Heart Failure Programme): an international survey. Lancet 2002;360:1631–9. [4] Komajda M, Follath F, Swedberg K, Cleland J, Aguilar JC, Cohen-Solal A, et al. The EuroHeart Failure Survey programme–a survey on the quality of care among patients with heart failure in Europe. Part 2: treatment. Eur Heart J 2003;24:464–74. [5] Komajda M, Lapuerta P, Hermans N, Gonzalez-Juanatey JR, van Veldhuisen DJ, Erdmann E, et al. Adherence to guidelines is a predictor of outcome in chronic heart failure: the MAHLER survey. Eur Heart J 2005;26:1653–9. [6] Nieminen MS, Brutsaert D, Dickstein K, Drexler H, Follath F, Harjola V-P, et al. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population. Eur Heart J 2006;27:2725–36. [7] Frankenstein L, Remppis A, Fluegel A, Doesch A, Katus HA, Senges J, et al. The association between long-term longitudinal trends in guideline adherence and mortality in relation to age and sex. Eur J Heart Fail 2010;12:574–80. [8] Maggioni AP, Dahlström U, Filippatos G, Chioncel O, Crespo Leiro M, Drozdz J, et al. EURObservational Research Programme: regional differences and 1-year follow-up
[16]
[17]
[18] [19]
[20]
[21]
[22]
[23]
7
results of the heart failure pilot survey (ESC-HF pilot). Eur J Heart Fail 2013;15: 808–17. Maggioni AP, Anker SD, Dahlström U, Filippatos G, Ponikowski P, Zannad F, et al. Are hospitalized or ambulatory patients with heart failure treated in accordance with European Society of Cardiology guidelines? Evidence from 12,440 patients of the ESC heart failure long-term registry. Eur J Heart Fail 2013;15:1173–84. Maggioni AP, Dahlström U, Filippatos G, Chioncel O, Leiro MC, Drozdz J, et al. EURObservational Research Programme: the heart failure pilot survey (ESC-HF pilot). Eur J Heart Fail 2010;12:1076–84. Krantz MJ, Ambardekar AV, Kaltenbach L, Hernandez AF, Heidenreich PA, Fonarow GC. Patterns and predictors of evidence-based medication continuation among hospitalized heart failure patients (from get with the guidelines-heart failure). Am J Cardiol 2011;107:1818–23. Oertle M, Bal R. Understanding non-adherence in chronic heart failure: a mixedmethod case study. Qual Saf Health Care 2010;19, e37. Zugck C, Franke J, Gelbrich G, Frankenstein L, Scheffold T, Pankuweit S, et al. Implementation of pharmacotherapy guidelines in heart failure: experience from the German Competence Network Heart Failure. Clin Res Cardiol 2012;101:263–72. Steinman MA, Patil S, Kamat P, Peterson C, Knight SJ. A taxonomy of reasons for not prescribing guideline-recommended medications for patients with heart failure. Am J Geriatr Pharmacother 2010;8:583–94. Steinman MA, Sudore RL, Peterson CA, Harlow JB, Fried TR. Influence of patient age and comorbid burden on clinician attitudes toward heart failure guidelines. Am J Geriatr Pharmacother 2012;10:211–8. Steinman MA, Dimaano L, Peterson CA, Heidenreich PA, Knight SJ, Fung KZ, et al. Reasons for not prescribing guideline-recommended medications to adults with heart failure. Med Care 2013;51:901–7. Cleland JGF, Swedberg K, Follath F, Komajda M, Cohen-Solal A, Aguilar JC, et al. The EuroHeart Failure Survey programme—a survey on the quality of care among patients with heart failure in Europe. Part 1: patient characteristics and diagnosis. Eur Heart J 2003;24:442–63. American National Kidney Foundation official webpage: http://www.kidney.org/ professionals/kdoqi/gfr_calculator.cfm. Lainscak M, Cleland JGF, Lenzen MJ, Follath F, Komajda M, Swedberg K. International variations in the treatment and co-morbidity of left ventricular systolic dysfunction: data from the EuroHeart Failure survey. Eur J Heart Fail 2007;9:292–9. Calvin JE, Shanbhag S, Avery E, Kane J, Richardson D, Powell L. Adherence to evidence-based guidelines for heart failure in physicians and their patients: lessons from the Heart Failure Adherence Retention Trial (HART). Congest. Heart Fail. 18: 73–8. Forman DE, Cannon CP, Hernandez AF, Liang L, Yancy C, Fonarow GC. Influence of age on the management of heart failure: findings from get with the guidelinesheart failure (GWTG-HF). Am Heart J 2009;157:1010–7. Peterson PN, Rumsfeld JS, Liang L, Hernandez AF, Peterson ED, Fonarow GC, et al. Treatment and risk in heart failure: gaps in evidence or quality? Circ Cardiovasc Qual Outcomes 2010;3:309–15. (MAGGIC) MGG in CHF. The survival of patients with heart failure with preserved or reduced left ventricular ejection fraction: an individual patient data meta-analysis. Eur Heart J 2011;33:1750–7.
Please cite this article as: Detiček A, et al, Therapy modifications during hospitalization in patients with chronic heart failure, Eur J Intern Med (2015), http://dx.doi.org/10.1016/j.ejim.2015.12.007