論文ID: CJ-19-0158
Background: In Japan, tolvaptan is indicated for patients with heart failure and volume overload who have inadequate response to other diuretics. In contrast to the USA and Europe, tolvaptan can be used in Japan in patients with normal sodium levels.
Methods and Results: In this multicenter, non-interventional, post-marketing surveillance study, prospective data from 3,349 patients treated with tolvaptan over a 5-year period were analyzed to identify benefits and risks. By Day 2 of treatment, 76.9% of evaluable patients had an increase in baseline 24-h urine volume (tolvaptan responders). Mean change in body weight was similar between 7.5 mg and 15 mg dosage groups (−3.6±3.9 kg and −3.7±4.0 kg, respectively). Improvement or disappearance rates for congestive symptoms from baseline to Day 14 ranged from 77.7% for lower limb edema to 51.1% for 3rd sound. Adverse drug reactions were reported in 18.1% of patients, most frequently thirst (8.4%). No case of central pontine myelinolysis was reported. All-cause mortality was significantly lower in patients with improved sodium concentration and increased 24-h urine volume.
Conclusions: The effectiveness and safety of tolvaptan in real-world clinical settings was confirmed in this large-scale analysis. The 7.5-mg dose was equally as effective as the 15-mg dose and had a better safety profile. Improvements in all-cause mortality were suggested in tolvaptan responders.
The prevalence of heart failure (HF) is known to increase with age. In Japan, where the proportion of the population aged ≥65 years has been increasing rapidly in recent years, the number of patients with chronic HF is expected to increase in parallel, representing an enormous healthcare burden.1 The Acute Decompensated Heart Failure Syndromes (ATTEND) registry, the largest study of hospitalized HF patients conducted to date in Japan, found that approximately 70% of patients had fluid retention and 76.3% of patients had received in-hospital management with diuretics.2 Frequent use of diuretics in these patients is a concern, given that loop diuretics are known to adversely affect renal function when administered at higher doses.3 The Japanese Cardiac Registry of Heart Failure in Cardiology (JCARE-CARD) reported that 27% of patients hospitalized with worsening HF who were discharged following symptom remission were rehospitalized within 6 months.4
Tolvaptan, a vasopressin receptor 2 antagonist,5 is approved for the indications of hyponatremia and syndrome of inappropriate antidiuretic hormone (SIADH) in the USA and for SIADH in Europe. In contrast, in Japan, tolvaptan is indicated for ‘volume overload in HF patients when adequate response was not obtained with other diuretics (e.g., loop diuretics)’.6 The pivotal phase 3 Qualification of Efficacy and Safety in the study of Tolvaptan (QUEST) in cardiac edema study in support of this indication involved a small sample of 53 patients treated with tolvaptan for 7 days.7 This randomized controlled trial applied strict inclusion and exclusion criteria for enrollment. To confirm whether the efficacy and safety profile of tolvaptan in clinical trials extends to the general population of HF patients, its effectiveness and tolerability must also be investigated during use in real-world clinical settings. The Samsca post-Marketing surveillance In heart faiLurE (SMILE) study was conducted over 5 years with the aim of observing 3,000 patients treated with tolvaptan in usual clinical practice; interim results have been reported.8–10 Here, we report results of the final analysis of the SMILE study.
The prospective, multicenter, observational SMILE study was conducted in compliance with Good Post-Marketing Study Practice, an ordinance issued by the Ministry of Health, Labour and Welfare for implementation of post-marketing surveillance of new drugs approved in Japan. The study took place between 2011 and 2015, and the target number of patients was 3,000. As this naturalistic, non-interventional survey study involved the collection of anonymous data from clinical settings, obtaining informed consent from patients and approval by respective institutional review boards was not mandatory, although both were obtained according to the regulations of participating medical institutions. This approach was fully compliant with Japanese regulations for post-marketing surveillance.
The study enrolled HF patients with volume overload who were resistant to loop diuretics. Patients with anuria, consciousness disturbance, difficulties with water intake or hypernatremia, or who were pregnant, were excluded. Although the standard tolvaptan treatment period was specified as 2 weeks, some patients were treated for longer than 2 weeks at the discretion of attending physicians.
Study AssessmentsEffectiveness of tolvaptan was evaluated by measuring body weight and 24-h urine volume. Baseline 24-h urine volume was defined as the urine volume over 24 h up to the morning of the starting day of tolvaptan treatment.
Congestive symptoms were recorded from baseline through the consecutive treatment periods. Lower limb edema and pulmonary congestion were evaluated by symptomatic grading (none, mild, moderate, and severe). Dyspnea, jugular venous distention, hepatomegaly, 3rd sound and rales were evaluated according to their presence or absence.
Safety of tolvaptan was evaluated by monitoring laboratory test values and vital signs. Main laboratory parameters examined were serum sodium and serum potassium concentrations, alanine aminotransferase, aspartate aminotransferase, total bilirubin, blood urea nitrogen (BUN), creatinine, and estimated glomerular filtration rate (eGFR). Adverse events (AEs) and adverse drug reactions (ADRs) were summarized according to attending physicians’ reports and listed by frequency. An ADR was defined as an AE for which a causal relationship to tolvaptan could not be ruled out. A serious AE as assessed by attending physicians was defined as any untoward medical occurrence that was life-threatening or required inpatient hospitalization.
Statistical AnalysesChanges in 24-h urine volume from baseline to Day 2, the peak period of aquaresis, were analyzed and described as mean±standard deviation (SD); Day 1 data were used for patients without available Day 2 data. For statistical analysis of changes in body weight and urine volume by given dose of tolvaptan, patients who had received tolvaptan at a specific dose (3.75, 7.5 or 15 mg) without adjustment were extracted and assigned into groups. Changes in weight and urine volume at each observation timepoint from baseline to Day 14 were analyzed per dose group and expressed as mean±SD.
For statistical analysis of congestive symptoms, the proportions of patients with each grade of symptom (lower limb edema and pulmonary congestion) or with the presence/absence of a symptom (dyspnea, jugular venous distention, hepatomegaly, 3rd sound, and rales) were analyzed for each observation timepoint from baseline to Day 14.
The incidence of ADRs was calculated according to time of onset.
For analyses of initial tolvaptan dose, baseline loop diuretic dose, and baseline eGFR, the mean, median or proportion of each variable was calculated according to the year of initiation of tolvaptan treatment.
For analyses of survival, all-cause mortality was estimated using the Kaplan-Meier method to create survival curves. Survival in patient subgroups was compared by the log-rank test.
All statistical analyses were performed using SAS version 9.2 software (SAS Institute, Cary, NC, USA).
Of 3,350 enrolled patients, 3,349 were included in the analysis population; 1 patient who did not receive tolvaptan was excluded from analyses. Baseline demographic and clinical characteristics of SMILE study patients are presented in Table 1 and are compared below with those in the ATTEND registry.
| Characteristic | |
| n | 3,349 |
| Age, years | 77.3±12.4 |
| Median, range | 80.0 (3–106) |
| Sex, male | 56.1 |
| Body weight, kg | 57.6±13.9 |
| Urine volume, 24 h, mL | 1,954±1,224 |
| NYHA class I/II | 27.6 |
| NYHA class III/IV | 61.8 |
| Systolic blood pressure, mmHg | 119±22 |
| Diastolic blood pressure, mmHg | 66±14 |
| Underlying disease | |
| Ischemic heart disease | NR |
| Hypertensive heart disease | 22.4 |
| Cardiomyopathy/Myocarditis | 17.1 |
| Valvular heart disease | 33.0 |
| Medications | |
| Loop furosemide equivalent, mg | 63.0±81 |
| Angiotensin-converting enzyme inhibitor | 16.5 |
| Angiotensin II receptor blocker | 26.2 |
| β-blocker | 40.0 |
| Thiazide | 7.5 |
| Aldosterone antagonist | 42.6 |
| Carperitide | 16.2 |
| Inotropes | 17.2 |
| Laboratory data | |
| Serum albumin, g/dL | 3.2±0.6 |
| Blood urea nitrogen, mg/dL | 33.5±20.8 |
| Serum creatinine, mg/dL | 1.56±1.08 |
| eGFR, mL/min/1.73 m2 | 43.5±26.8 |
| Serum sodium, mEq/L | 136.9±6.2 |
| Serum potassium, mEq/L | 4.14±0.65 |
| Bilirubin total, mg/dL | 1.0±1.2 |
| Congestive symptoms | |
| Lower limb edema | 83.2 (2,671/3,209)a |
| Pulmonary congestion | 84.5 (2,568/3,040)a |
| Dyspnea | 70.0 (2,230/3,187)a |
| Jugular venous distension | 58.2 (1,386/2,382)a |
| Hepatomegaly | 31.9 (589/1,846)a |
| 3rd sound | 36.1 (843/2,333)a |
| Rales | 50.4 (1,435/2,846)a |
| B-type natriuretic peptide, pg/mL | 663 (8–31,761) |
Values are mean±SD, median with lower and upper quartiles, or percent. aDenominator indicates the number of patients with at least 2 measurements for a symptom: 1 at baseline and 1 post-baseline. eGFR, estimated glomerular filtration rate; NYHA, New York Heart Association; SD, standard deviation.
The mean age of patients was 77.3±12.4 years (median age, 80 years) and the population was 56.1% male, similar to corresponding values in ATTEND. Etiology of HF was valvular heart disease in 33.0% of SMILE patients, compared with 19.5% of ATTEND patients. Mean systolic blood pressure (SBP) was 119±22 mmHg in SMILE patients vs. 145.5±36.7 mmHg in ATTEND patients.
In the SMILE study, mean 24-h urine volume before the start of tolvaptan treatment was 1,954±1,224 mL. Mean serum sodium concentration was lower (136.9±6.2 vs. 139.3±4.4 mEq/L), and mean serum creatinine concentration was higher (1.56±1.08 vs. 1.43±1.57 mg/dL), in SMILE vs. ATTEND patients. Use of β-blockers was recorded in 40.0% of SMILE patients, and in 33.9% of ATTEND patients. Median B-type natriuretic peptide concentration was similar in both studies, at 663 pg/mL and 707 pg/mL, respectively.
Dosages and Administration PeriodThe mean tolvaptan dosage in the overall study period was 9.4±3.8 mg/day. In total, 56.0% of patients completed treatment within 14 days, and 44.0% of patients continued treatment for ≥15 days. Although it was not possible to capture accurate data on administration periods for individual patients, at least 12% of the analysis population received tolvaptan for ≥1 month.
EffectivenessThe distribution of changes in 24-h urine volume on Day 2 of tolvaptan treatment is shown in Figure 1. Of 1,231 patients with evaluable data (i.e., those with at least two 24-h urine volume measurements), 947 (76.9%) patients had increased 24-h urine volume compared with baseline and were regarded as tolvaptan responders.
Distribution of increase in 24-h urine volume from baseline to Day 2.
Figure 2 shows changes in body weight and 24-h urine volume for patients who initiated tolvaptan at doses of 3.75 mg, 7.5 mg, or 15 mg and continued to receive tolvaptan at these doses without adjustment. Mean changes in body weight from baseline to Day 14 were −3.0±3.3 kg in the 3.75 mg group, −3.6±3.9 kg in the 7.5 mg group, and −3.7±4.0 kg in the 15 mg group (Figure 2A). Mean increases in 24-h urine volume from baseline to Day 2 were 418±885 mL in the 3.75 mg group, 772±1,204 mL in the 7.5 mg group, and 888±1,209 mL in the 15 mg group (Figure 2B).
Mean change in (A) body weight and (B) 24-h urine volume (UV) from baseline to Day 14 according to fixed dose of tolvaptan.
At baseline, the proportion of patients with congestive symptoms ranged from 84.5% for pulmonary congestion to 31.9% for hepatomegaly (Table 1). Changes over time in symptomatic grades or presence/absence of congestive symptoms are shown in Table 2. Improvement or disappearance was observed at Day 14 for all seven symptoms. Rates ranged from 77.7% (improvement) for patients with lower limb edema at baseline to 51.5% (disappearance) for patients with 3rd sound at baseline.
| Symptom | Improvementa/ disappearanceb rates |
|---|---|
| Lower limb edemaa | 77.7 (2,024/2,606) |
| Pulmonary congestiona | 72.5 (1,753/2,419) |
| Dyspneab | 69.5 (1,517/2,182) |
| Jugular venous distensionb | 62.6 (828/1,322) |
| Hepatomegalyb | 52.4 (289/552) |
| 3rd soundb | 51.1 (409/801) |
| Ralesb | 67.8 (938/1,383) |
Values are percent. aImprovement rate. bDisappearance rate.
Incidences of ADRs reported by attending physicians according to time of onset are summarized in Table 3. The overall incidence of ADRs was 18.1%. The most common ADR was thirst (8.4%), followed by hypernatremia (4.4%). Of reported cases of hypernatremia, 0.4% were assessed as serious by the attending physicians. No case of central pontine myelinolysis (CPM) was reported. Rapid correction of serum sodium concentration, defined as a 12 mEq/L increase within 24 h, was reported for 8 (0.2%) patients.
| Parameter | Total | 1–3 days | 4–7 days | 8–14 days | 15–30 days | ≥31 days |
|---|---|---|---|---|---|---|
| Incidence | 18.1 (607/3,349) | 11.7 (393/3,349) | 4.3 (121/2,850) | 3.3 (70/2,106) | 1.7 (18/1,087) | 0.8 (3/392) |
| Thirst | 8.4 | 6.8 | 1.2 | 0.7 | 0.3 | 0.5 |
| Hypernatremia (seriousa) | 4.4 (0.4) | 2.9 (0.1) | 1.4 (0.3) | 0.4 (0.1) | 0.3 (0.0) | 0.0 (0.0) |
| Renal dysfunctionb | 1.2 | 0.3 | 0.4 | 0.7 | 0.3 | 0.5 |
| Liver dysfunctionc | 0.9 | 0.2 | 0.3 | 0.5 | 0.2 | 0.3 |
Values are percent. aSeriousness was assessed by the attending physician. bIncludes renal failure, renal function disorder, chronic renal disease, acute renal dysfunction, prerenal failure, increased blood creatinine. cIncludes cirrhosis, hepatic function abnormality, fulminant hepatitis, ischemic hepatitis, liver damage, drug-induced liver injury, aspartate aminotransferase increased, alanine aminotransferase increased, bilirubin increased, γ-GTP increased, liver enzymes increased.
Serious renal dysfunctional was reported for 11 patients, all of whom had renal dysfunction caused by loop diuretics or worsening HF as a complication before the start of tolvaptan treatment. Serious hepatic dysfunction was reported for 6 patients, of which 4 cases were considered to be related to tolvaptan treatment; 2 of the 6 patients died (liver failure; disseminated intravascular coagulation) and the remaining 4 patients recovered with or without a dose reduction or discontinuation of tolvaptan.
Table 4 shows the incidence of ADRs by tolvaptan fixed dose group. ADRs were most frequent in the group treated with tolvaptan 15 mg. This group also had the highest incidence of thirst and hypernatremia. Dose proportionality was not observed for renal or hepatic dysfunction.
| Parameter | 3.75 mg | 7.5 mg | 15 mg | P valuea |
|---|---|---|---|---|
| Adverse drug reaction | 12.6 (40/317) | 15.4 (271/1,760) | 18.1 (140/772) | 0.0162 |
| Thirst | 5.1 | 6.8 | 9.2 | 0.0080 |
| Hypernatremia | 2.8 | 3.7 | 6.4 | 0.0017 |
| Renal dysfunctionb | 1.0 | 1.0 | 1.2 | 0.7366 |
| Hepatic dysfunctionc | 1.3 | 0.8 | 0.7 | 0.5820 |
Values are percent. aCochran-Armitage trend test. bIncludes renal failure, renal function disorder, chronic renal disease, acute renal dysfunction, prerenal failure, blood creatinine increased. cIncludes cirrhosis, hepatic function abnormality, fulminant hepatitis, ischemic hepatitis, liver damage, drug-induced liver injury, AST increased, ALT increased, bilirubin increased, γ-GTP increased, liver enzyme increased.
Kaplan-Meier estimates of all-cause death are shown in Figure 3. The overall cumulative 6-month mortality rate was 14% (Figure 3A). Subgroup analyses indicated that mortality rates were significantly higher (P<0.0001) in patients with a baseline sodium concentration <135 mEq/L vs. ≥135 mEq/L (Figure 3B). In the subgroup with a baseline sodium concentration <135 mEq/L, mortality rates were significantly higher (P=0.0221) in patients without a correction vs. patients with a correction to ≥135 mEq/L by Day 14 (Figure 3C). Mortality rates were significantly higher (P=0.0275) in patients without an increase (non-responders) vs. patients with an increase (responders) in 24-h urine volume up to Day 2 of tolvaptan treatment (Figure 3D).
Kaplan-Meier estimates of (A) all-cause death; (B) all-cause death by baseline serum sodium concentration; (C) all-cause death in patients with a baseline serum sodium concentration <135 mEq/L who did or did not have a correction to ≥135 mEq/L by Day 14 of tolvaptan treatment; (D) all-cause death in patients with an increase (responder) or without an increase (non-responder) in 24-h urine volume up to Day 2 of tolvaptan treatment.
A comparison of baseline characteristics in patients with or without a correction to a serum sodium concentration ≥135 mEq/L by Day 14 of tolvaptan treatment (Supplementary Table 1) indicated important differences. Patients with a lower serum sodium concentration at baseline were not prone to reach normonatremia. Baseline renal function was better in patients whose sodium concentration had corrected to ≥135 mEq/L by Day 14, reflecting the dependency of urinary excretion on renal function. Baseline hyponatremia was associated with greater disease severity, as indicated by lower BP and less frequent use of carperitide. The lower tolvaptan dose in the hyponatremic group suggested that increasing the dose may have facilitated a correction to normonatremia.
The same comparison in patients with (responder) or without (non-responder) an increase in 24-h urine volume up to Day 2 of tolvaptan treatment (Supplementary Table 2) indicated that 24-h urine volume at baseline was higher in patients whose urine volume had increased by Day 2. Older age was associated with a poorer response in urine volume.
Treatment Evolution 2011–2015Evolution in the starting doses of tolvaptan and patients’ baseline characteristics from the start (2011) to the end (2015) of the study period are shown in Table 5. The mean initial tolvaptan dose decreased from 9.8 mg to 8.2 mg. During the study timeframe, the proportion of patients who started tolvaptan at a dose of 3.75 mg increased from 8.6% to 20.0%, while the proportion of patients who began tolvaptan treatment at a dose of 15 mg decreased from 35.1% to 19.6%. From 2011 to 2015, the dose of other baseline loop diuretics converted to furosemide equivalents decreased from 69.3 mg to 54.9 mg, and patients’ baseline eGFR increased from 40.4 mL/min/1.73 m2 to 45.6 mL/min/1.73 m2.
| Parameter | Total | 2011 | 2012 | 2013 | 2014 | 2015 |
|---|---|---|---|---|---|---|
| Tolvaptan initial dose, mg (mean) | 9.0 | 9.8 | 9.6 | 8.8 | 8.3 | 8.2 |
| Tolvaptan initial dose, mg (median) | 7.5 | 7.5 | 7.5 | 7.5 | 7.5 | 7.5 |
| Tolvaptan 3.75 mg (%) | 12.2 | 8.6 | 9.0 | 10.1 | 16.8 | 20.0 |
| Tolvaptan 7.5 mg (%) | 62.0 | 56.3 | 58.7 | 67.1 | 63.5 | 60.1 |
| Tolvaptan 15 mg (%) | 25.5 | 35.1 | 32.1 | 22.4 | 19.5 | 19.6 |
| Loop diuretics, mg (mean) | 63.0 | 69.3 | 69.0 | 59.7 | 60.4 | 54.9 |
| Loop diuretics, mg (median) | 40 | 40 | 40 | 40 | 40 | 40 |
| eGFR, mL/min/1.73 m2 (mean) | 43.5 | 40.4 | 41.8 | 43.5 | 46.4 | 45.6 |
eGFR, estimated glomerular filtration rate.
The SMILE study, which investigated use of tolvaptan in 3,349 HF patients with volume overload who were resistant to loop diuretics, complements and extends the real-world evidence for tolvaptan previously reported in the ATTEND registry study.2 The broad similarities between SMILE and ATTEND patient populations in terms of age and sex ratio confirm that most patients hospitalized with worsening HF in Japan are elderly and male. Conversely, the differences observed in certain baseline clinical characteristics between the respective patient populations may reflect additional years of experience with tolvaptan since the time frame of the ATTEND registry (2006–2011).
Patients in ATTEND were in the acute phase of HF, whereas those enrolled in SMILE were resistant to loop diuretics and thus can be assumed to have had more severe HF. This assumption is supported by the lower mean SBP in SMILE vs. ATTEND patients (119 vs. 145.5 mmHg). The higher death rate among ATTEND patients with SBP <120 mmHg11 emphasizes that, in the SMILE study, tolvaptan was being used in patients with extremely severe HF. Higher baseline serum creatinine and BUN concentrations in SMILE vs. ATTEND patients were also indicative of more advanced disease.
In the SMILE study, more than three-quarters (76.9%) of evaluable patients treated with tolvaptan in real-world clinical settings had increased 24-h urine volume on Day 2 and were thus considered to be tolvaptan responders. A better prognosis was expected in responders and survival analyses confirmed this expectation. With the exception of older age, however, predictors of response in terms of baseline characteristics were unclear. Several predictors of response to tolvaptan have been proposed based on urine osmolality or aquaporin levels.12,13 Urine osmolality is a simple method for identifying responders. If identification of probable tolvaptan responders before the start of treatment proves possible with either of these methods, use of tolvaptan can be reserved for patients most likely to benefit from treatment.
Evaluation of changes in body weight and urine volume according to tolvaptan dose showed minimal differences in mean values between the 7.5 mg and 15 mg groups. Although tolvaptan was considered to be less effective at a dose of 3.75 mg, the decrease in mean body weight and increase in mean urine volume observed in this group nonetheless indicate a diuretic effect at very low doses. With respect to a dose-response relationship, a clinical pharmacology study of tolvaptan in healthy volunteers reported a dose-dependent response at doses between 15 mg and 120 mg.14 Elsewhere, a randomized, double-blind, phase III study of tolvaptan in Japanese patients with HF and volume overload refractory to furosemide reported lower responses in the 7.5 mg vs. 15 mg dose group, although the generalizability of these results is limited by the small sample size (10 patients per dose group).15 An important difference between clinical trials and naturalistic studies is that, in clinical practice, dosages can be adjusted according to patient response. Based on our results, tolvaptan might be expected to have a substantial diuretic effect in HF patients with volume overload at dosages <15 mg/day.
An interim analysis of the SMILE study in 1,053 evaluable patients reported significant improvement or absence of lower limb edema, pulmonary congestion, dyspnea, jugular venous distention, and hepatomegaly at Days 7 and 14 (P<0.0001 for all symptoms).16 This final analysis confirms the interim results and indicates clinically meaningful symptom-free rates also for 3rd sound and rales.
Although the potent aquaretic effects of tolvaptan may raise safety concerns about the potential for CPM, no cases occurred among 3,349 patients over the 5-year study period. Another potential concern with tolvaptan is hypernatremia. Risk factors for tolvaptan-induced hypernatremic events have previously been analyzed and reported in detail.16 A risk score nomogram based on the final SMILE dataset has been developed to identify at-risk patients who should begin tolvaptan treatment at a dosage of 3.75 mg/day. Although cases of elevated liver enzymes related to tolvaptan were reported in clinical trials of patients with autosomal dominant polycystic kidney disease,17 the tolvaptan dosage was much higher (60–120 mg/day) than that indicated for use in patients with HF. In our study, tolvaptan-related serious hepatic dysfunction was reported in 4 patients and a trend was apparent for association with the 15 mg dose. These results suggest that a lower tolvaptan dose may be safe in patients with hepatic dysfunction, although treatment discontinuation should always be considered in the event that liver function abnormalities arise.
Kaplan-Meier estimates indicated a low cumulative 6-month all-cause mortality rate with tolvaptan treatment. Death was inversely associated with baseline serum sodium concentration. Correction of serum sodium to normal range values was associated with lower mortality rates among patients with less severe baseline hyponatremia. An increase in 24-h urine volume by Day 2 was also associated with a lower mortality rate. A comparison of characteristics between patient subgroups with baseline sodium concentrations of <135 mEq/L or ≥135 mEq/L indicated that a lower sodium concentration may be related to thiazide treatment; switching such patients to tolvaptan might improve hyponatremia. To the best of our knowledge there are few reports to show a relationship between correction of serum sodium concentration and death. A retrospective chart review study found that improving serum sodium concentration after hospital discharge in patients with decompensated HF was a strong predictor of long-term survival;18 our findings are consistent with that report. Konishi and colleagues reported that progression to hyponatremia during hospitalization in patients with acute decompensated HF who were normonatremic at admission was a strong predictor of poor cardiac prognosis.19 In a retrospective analysis of 4,067 Japanese patients hospitalized for HF, a serum sodium concentration ≤135 mEq/L at discharge was associated with a significantly lower 10-year survival rate compared with normonatremia.20 It is noteworthy that these studies were performed prior to approval of tolvaptan for use in clinical practice, meaning that serum sodium concentrations were being managed without a vasopressin receptor 2 antagonist. Correcting the serum sodium concentration with tolvaptan may improve long-term prognosis. Our study is the first to show the contribution of tolvaptan to improved survival in real-world settings. Further studies are needed to confirm the results.
In patients hospitalized for acute decompensated HF, removal of excess fluid is a priority to alleviate patient discomfort (especially dyspnea) and improve prognosis.21 Recent studies of tolvaptan suggest a time lag of 2–3 days between early diuresis (net fluid and weight loss) and patient perceived improvement of dyspnea,22–24 underlying the importance of identifying appropriate outcome measures to assess response. In the current study, our interest was the effect of tolvaptan on decongestive symptoms at the end of scheduled treatment (Day 14), as residual congestion is a risk factor for readmission and worse prognosis.21 Inomata and colleagues compared additive tolvaptan with an increased dose of furosemide in patients with residual congestion despite optimal medical therapy.25 Over 7 days’ treatment, the significant increase in daily urine output with tolvaptan vs. furosemide, in the absence of any detrimental effect on renal function, suggested a protective effect.
Consistent with our findings, a meta-analysis of tolvaptan studies in patients with acute HF (6 randomized controlled trials in 746 patients) reported improvements compared with control (placebo or conventional therapy) in body weight, serum sodium concentration and congestive symptoms, but without effect on short-term (≤30 days) or long-term (>30 days) all-cause death.26 Our subgroup analyses showed that certain disease/patient factors affect the risk of death. Specifically, hyponatremia severity at baseline, the extent of tolvaptan-induced improvement in hyponatremia, and patients’ ability to respond to tolvaptan with increased urine output are associated with lower 6-month all-cause death.
Study LimitationsThe main strengths of the SMILE study are that it provides robust real-world evidence from a large population of hospitalized HF patients treated with tolvaptan under usual clinical practice conditions, and demonstrates evolution in the use of tolvaptan in this indication in line with accumulating clinical experience. A limitation is that the study does not necessarily demonstrate the efficacy and safety of tolvaptan alone, as clinical results were based on actual clinical practice where use of concomitant medications was not restricted.
The effectiveness and tolerability of low-dose tolvaptan (≤15 mg/day) in real-world clinical settings was confirmed by analyzing data from 3,349 patients with HF and volume overload who were resistant to loop diuretics. The dose of 7.5 mg dose was almost equally as effective as the 15 mg dose and had a more favorable safety profile. No cases of CPM were reported. More severe hyponatremia at the start of tolvaptan treatment may be associated with a poorer outcome.
The authors thank all investigators for their participation in the study. Medical writing assistance was provided by Content Ed Net (Osaka, Japan) with funding from Otsuka Pharmaceuticals Co., Ltd.
The study was funded by Otsuka Pharmaceuticals Co., Ltd., Japan. Otsuka Pharmaceuticals Co., Ltd. provided oversight on conduct of the study, including design, collection, compilation, and data analysis.
K.K., N.S., T.I., and M.Y. contributed to analysis and interpretation of the data. T.S. contributed to study design. All authors contributed to drafting and revising the manuscript and provided final approval of the publication version.
K.K., N.S. and T.I. have received lecture fees from Otsuka Pharmaceutical Co., Ltd. T.S. and Y.F. are employees of Otsuka Pharmaceutical. M.Y. was an employee of Otsuka Pharmaceutical Co., Ltd. at the time the study was conducted.
Please find supplementary file(s);
http://dx.doi.org/10.1253/circj.CJ-19-0158
