Recent Patient Characteristics and Medications at Admission and Discharge in Hospitalized Patients With Heart Failure

Background To improve the clinical outcome of heart failure (HF), it is important to evaluate the etiology and comorbidities of HF. We previously reported the baseline clinical characteristics and medications in hospitalized patients with HF in years 2000 - 2002 (group 2000) and 2007 - 2009 (group 2008). Methods We conducted a retrospective study of 158 patients who were hospitalized due to HF between 2012 and 2014 (group 2013) in the Department of Cardiology, Fukuoka University Hospital. We analyzed the clinical characteristics and medications at admission and discharge, and compared the findings in group 2013 to those in group 2000 and group 2008. Results The major causes of HF were ischemic heart disease, hypertensive cardiomyopathy, valvular heart disease, and dilated cardiomyopathy. The New York Heart Association classification in group 2013 was significantly higher than those in group 2000 and group 2008. There was no difference in the level of brain natriuretic peptide at admission between group 2008 and group 2013. Tolvaptan began to be administered in group 2013. The median dose of furosemide just before the use of tolvaptan was 40 mg/day. At discharge, group 2013 showed higher rates of β-blocker and aldosterone antagonist. There was no difference in the frequency of loop diuretics. The dose of carvedilol at discharge was only 6.2 ± 4.0 mg/day. Antiarrhythmic drugs and β-blocker were used more frequently in HF with reduced ejection fraction (EF) than in HF with preserved EF. Conclusions We may be able to improve the clinical outcome of HF by examining the differences in the clinical characteristics and medications at admission and discharge in hospitalized patients with HF.


Introduction
Heart failure (HF) has a poor prognosis. However, previous studies have shown that several medications can improve the prognosis of HF [1][2][3][4][5][6][7][8][9][10][11][12][13][14]. It is important that patients are provided the optimal medications according to guidelines for HF [15,16]. Unfortunately, not all patients with HF receive appropriate medical treatment, and there are differences in the compliance with guidelines between hospitals [17]. In addition, the prognosis is known to worsen if the compliance with guidelines is poor [18]. Previous registration studies include ADHERE (characteristics and outcomes of patients hospitalized for heart failure in the United States) [19], ATTEND (acute decompensated heart failure syndromes registry) [20] and EHFSII (Eu-roHeart Failure Survey II: a survey of hospitalized acute heart failure patients) [21]. Although these studies included patients with similar ages and genders, there were differences in complications and the duration of hospitalization. The use of medications also differed. We may be able to provide appropriate medical treatment by knowing the present patient characteristics and medications in our hospital. We previously reported the baseline clinical characteristics and medications of hospitalized patients with HF in years 2000 -2002 (group 2000) and 2007 -2009 (group 2008) [22]. Various clinical trials have recently been performed in the field of HF [23][24][25][26][27][28][29], and new medications are now available [30]. Therefore, we analyzed the patient characteristics and medications at admission and discharge, and compared the findings in 2012 -2014 (group 2013) to those in group 2000 and group 2008.

Clinical parameters
The blood pressure and heart rate were determined, and echocardiography was performed at admission. The echocardiographic parameters examined were the left atrial dimension (LAd), left ventricular end diastolic dimension (LVEDd) and LV ejection fraction (LVEF). Information regarding medications was collected at three time points (at admission, in the hospital and at discharge). Data on the body mass index (BMI), systolic blood pressure (SBP), diastolic blood pressure (DBP), blood levels of brain natriuretic peptide (BNP), amino-terminal pro-brain natriuretic peptide (NT-proBNP), creatinine (Cr), estimated glomerular filtration rate (eGFR), creatinine clearance (CCr), uric acid (UA), sodium (Na), potassium (K), hemoglobin (Hb), C-reactive protein (CRP), total cholesterol (TC), triglyceride (TG), high-density lipoprotein-cholesterol (HDL-c) and low-density lipoprotein-cholesterol (LDL-c) were also collected at admission. HFrEF (HF with reduced EF) was defined as EF equal to or less than 40%. HFpEF (HF with preserved EF) was defined as EF equal to or more than 50%, and borderline was defined as EF 41-49% [15].

Dose of furosemide just before the use of tolvaptan
Tolvaptan began to be used in 2010, and its % use in group 2013 was 24%. The dose of tolvaptan was 1.875 mg/day (n = 1), 3.75 mg/day (n = 12), 7.5 mg/day (n = 23) or 15 mg/day (n = 2). As shown in Figure 2, the median dose of furosemide just before the use of tolvaptan was 40 mg/day (IQR 20 -55 mg/day), when we converted azosemide 30 mg and torasemide 4 mg to furosemide 20 mg. Seventy-nine percent of patients were given tolvaptan with carperitide, and 21% of patients were only given tolvaptan.

Cardiac rehabilitation
In our hospital, cardiac rehabilitation was started in 2011. The induction rate of cardiac rehabilitation during hospitalization in group 2013 was 85% (n = 135). On the other hand, only 5% (n = 8) of patients received cardiac rehabilitation as outpatients.

Discussion
In this study, the major causes of HF were IHD, HTCM, valvular heart disease and DCM. Group 2013 at discharge showed higher rates of β-blocker and aldosterone antagonist than group 2000, whereas the dose of carvedilol at discharge was only 6.2 ± 4.0 mg/day. There were no differences in the use of ACE-I or ARB between group 2000 and group 2013. Tolvaptan became available for use in group 2013, and was administered in 24% of patients. In addition, antiarrhythmic drugs and β-blocker were used more often in HFrEF than in HFpEF. In recent studies in the United States and Europe, including ADHERE [19], OPTIMIZE-HF [31] and EHFS II [21], the major causes of HF were IHD (about 30-40%), HTCM (about 20%), and valvular heart disease (about 20%). The causes of HF in registry studies in Japan, including JCARE-CARD [32], CHART [33] and ATTEND [20], were similar to those in our group 2013. Nonetheless, the patients in our study showed various causes of HF, including unknown causes. The etiology of HF needs to be evaluated more precisely.
Previous studies have shown that β-blocker improved the clinical outcome of patients with HFrEF [10][11][12][13]. Furthermore, other clinical studies showed that ACE-I [1][2][3], ARB [4][5][6][7] and aldosterone antagonist [8,9,34] improved the survival rate of patients with HFrEF. The guidelines for HF refer to these studies and recommend the administration of as many of these medications as possible, based on the condition of patients with HF. β-blocker and aldosterone antagonist were used more often in group 2013 than in group 2000. On the other hand, the dose of carvedilol in this study (6.2 ± 4.0 mg/day) was lower than that in a previous clinical study (45 ± 27 mg/day) [13]. Aldosterone antagonists are recommended in patients with HF (NYHA classification II-IV) who have LVEF of 35% or less, unless contraindicated, to reduce morbidity and mortality [15]. Therefore, we should combine aldosterone antagonists with other effective drugs for HF therapy, although aldosterone antagonists were used more often in group 2013 than in group 2000. In this case, serum K and Cr need to be carefully monitored to avoid hyperkalemia and renal insufficiency. There was no difference in %ACE-I + ARB between group 2000 and group 2013, whereas there was a difference in %ACE-I or ARB between the groups. The stage of HF in this study was C or D (ACCF/AGA stages of HF). In stage C or D, ACE-I is recommended in patients with HFrEF and current or prior symptoms, unless contraindicated, to reduce morbidity and mortality [15]. ARBs are recommended in patients with HFrEF with current or prior symptoms who do not tolerate ACE-I. In the ADHERE trial [19], a clinical trial that was performed in the United States in 2004, the usage rates of ACE-I and ARB were 41% and 12%. In EHFS II [21], which was performed in Europe in 2004 -2005, the usage rates of ACE-I and ARB were 71.1% and 10.4%. In JCARE-CARD, which was performed in Japan in 2004 -2005, the usage rates of ACE-I and ARB were 36.7% and 46.1% [32]. In Japan, ARBs are used more often than in the United States and Europe, probably because of adverse effects (e.g., cough, angioedema, rash and taste disturbances).
Tolvaptan became available in Japan for the treatment of HF in 2010. Gheorghiade et al [35] reported that tolvaptan induced an increase in urine volume, a decrease in body weight and normalization of serum sodium. In the EVEREST study [30], tolvaptan showed short-term effectiveness, but did not improve the long-term prognosis. Tolvaptan has only been used for the treatment of HF in Japan. When sufficient diuresis is not achieved with diuretics such as loop diuretics, tolvaptan is recommended. On the other hand, there is currently no clear evidence regarding whether tolvaptan should be used together with loop diuretics. Based on the results of post-marketing surveillance, the median dose of furosemide just before the administration of tolvaptan was 40 mg/day [36]. The median dose of tolvaptan was also 40 mg/day in this study. Since the IQR is from 20 to 55 mg/day, we should start tolvaptan with furosemide ≥ 40 mg/day.
There is considerable evidence regarding the pharmacological treatment of HFrEF. Beneficial drugs include diuretics, ARB, ACE-I, β-blocker, aldosterone antagonist, hydralazine, isosorbide dinitrate, digoxin, anticoagulants, and omega-3 fatty acids. Based on the data in group 2013, we should increase the dose of carvedilol to improve the clinical outcome of HFrEF. On the other hand, statins, nutritional supplements, hormonal therapies, long-term infusion of a positive inotropic drug, and CCBs appeared to be no effect or harmful in patients with HFrEF. No treatment has been shown to be beneficial in HFpEF. Since β-blockers were not effective in HFpEF, they were not used as much as in HFrEF (HFrEF (83%) vs. HFpEF (61%), P < 0.05)). In addition, we have to consider the higher incidence of HFpEF in Japan (HFpEF (25.4%) vs. HFrEF (58.2%)) [32]. Although %HFpEF was equivalent to %HFrEF in group 2013, the only treatments with evidence class I were diuretics and blood pressure management [15].
Torasemide (30% at discharge) was used in addition to furosemide in group 2013. Furosemide and azosemide are short-and long-acting loop diuretics, respectively. Azosemide, compared with furosemide, reduced the unplanned admission to a hospital for congestive HF [29]. In addition, azosemide suppresses activation of the sympathetic nervous system compared with furosemide in patients with HF [37]. Since a few reports have reported that azosemide is superior to furosemide, further studies will be needed to resolve this issue.
In this study, BNP and NT-proBNP were measured in 80% and 70% of the patients. Although the half-life of BNP in blood is short (20 min), that of NT-proBNP is relatively long (120 min). Since NT-proBNP shows high renal clearance, the correlation between NT-proBNP and BNP is worsened when eGFR is < 30 mL/min/1.73 m 2 . NT-proBNP is more stable than BNP after centrifugal separation and freeze-thaw. Since there is considerable evidence regarding the use of BNP [38][39][40][41] and NT-proBNP [42][43][44][45][46] as markers for diagnosis, their measurement is useful for screening and determining the prognosis in HF. It is still unclear whether the measurement of NT-proBNP offers any advantages over the measurement of BNP.

Conclusions
We may be able to improve the clinical outcome of HF by examining the differences in the clinical characteristics and medications at admission and discharge in hospitalized patients with HF. The present findings suggest that we should increase the doses of β-blocker and administer aldosterone antagonist more frequently in our hospital.