S174 Journal of Cardiac Failure Vol. 20 No. 10S October 2014

4h, 8h, and 24h after tolvaptan administration. Patients were divided into 3 groupsaccording to the change of sodium level (D-group: decrease of sodium level vs.M-group: mild increase (1-4Meq/L) vs. I-group: increase (5-13Meq/L). Results: Hy-pernatremia (O149Meq/L) was observed in 3 patients (4.2%). Timing of reachingmaximum sodium level was 4h in 2 (2.7%) and 24h in 35 (48.6%) patients, whilesodium level did not increase after tolvaptan treatment in 35 (48.6%) patients. Thelevel of baseline serum sodium was significantly different among the three groups(D-group: 139.263.6 vs. M-group: 135.966.7 vs. I-group:134.565.5Meq/L, respec-tively, p50.03). The age, potassium level, creatinine level, decrease of body weight,or tolvaptan dose was not associated with change of sodium level.Extreme sodium elevation (O12Meq/L) was observed in 1 (1.4%) patient, and theelevation timing was at 24h after tolvaptan administration. Conclusions: Rapid so-dium elevation was rare, although prevalence of mild sodium elevation within 24hwas about 50%. In future, we must check how frequently the serum sodium levelshould be measured according to the patient characteristics.

O-172Baseline Urine Sodium Concentration Best Predicts the Response to Short-termTolvaptan Treatment in Patients with Heart FailureYUICHI SATO1, KAORU DOHI1, TETSUSHIRO TAKEUCHI1, MUNEYOSHITANIMURA1, EMIYO SUGIURA1, NAOTO KUMAGAI1, SHIRO NAKAMORI1,NAOKI FUJIMOTO2, MASHIO NAKAMURA3, MASAAKI ITO1

1Department of Cardiology and Nephrology, Mie University Graduate School ofMedicine, 2Department of Molecular and Laboratory Medicine, Mie UniversityGraduate School of Medicine, 3Department of Clinical Cardiovascular Research,Mie University Graduate School of Medicine

Purpose: We identified predictor of response to tolvapan (TLV) in patients with heartfailure (HF). Methods: We enrolled 48 patients with HF with excess fluid retentiondespite receiving oral diuretics. All patients received low-dose TLV (7.5 mg/day) for7 days, and underwent right heart catheterization at baseline and after 7-day treatmentperiod. Patients were defined as non-responder only if they require specific interventionincluding intravenous vasodilator and inotropic therapy beyond simply holding diureticafter TLV treatment. Results: Responders (77%) had higher urine sodium concentra-tion, lower plasma BNP levels, lower mean pulmonary capillary wedge pressure(mPCWP) and mean right atrial pressure (mRAP) than non-responders at baseline.However, serum sodium levels and urine osmolality at baseline were similar betweenthe two groups. Responders had greater reduction in mPCWP and mRAP than non-re-sponders after 7-day treatment period. Multivariable logistic regression analysisconfirmed that plasma BNP levels and urinary sodium concentration were the indepen-dent predictors of responders to TLV (BNP: OR 0.943, 95% CI 0.900-0.988, P50.010,urine sodium concentration: OR 1.006, 95% CI 1.001-1.010, P50.013). ROC curveanalysis showed that urine sodium concentration was the best predictor of respondersto TLV with cut-off point of O46.5 mEq/gCre (AUC 0.883, 95% CI; 0.768-0.999,sensitivity 79%, specificity 91%, PO0.001). Conclusion: Urinary sodium concentra-tion is the most important predictor of responders to TLV in patients with HF.

O-173Serum Creatinine and Estimated GFR Predict Long-term Efficacy ofVasopressin Receptor Antagonist in Patients with Decompensated Heart FailureSHIGETO TSUKAMOTO, TETSUO SAKAI, TERUO SEKIMOTO, TAROADACHI, YUJI HAMAZAKI, YOUICHI KOBAYASHIDivision of Caldiology, Department of Medicine, Showa University School ofMedicine

Background: To prevent re-hospitalization due to recurring decompensated heartfailure, long-term administration of the vasopressin type 2 receptor antagonist

Variable

Not re-hospitalized

(n58)

Re-hospitalized

(n59)P-

value

Age, yrs 65.5 6 17.8 73.9 6 8.2 0.251Male, n (%) 5 (62.5) 7 (77.8) 0.490Tolvaptan dosage (mg) 7.97 6 4.4 10.0 6 3.5 0.336Left ventricular ejection

fraction (%)39 6 15.2 38 6 11.7 0.888

Clinical scenarioclassification (2/3/5)

6/2/0 6/2/1

Serum sodium (mEq/L) 137.8 6 2.6 135.2 6 4.9 0.220Serum creatinine

(mg/dL)1.16 6 0.53 1.79 6 0.59 0.048

Estimated GFR (mL/min/1.73m2)

56.9 6 26.2 33.2 6 15.4 0.047

B-type natriureticpeptide (pg/mL)

906.3 6 750.9 1226.8 6 816.5 0.463

Loop diuretics dosage(mg)

77.5 6 55.2 128.9 6 63.2 0.121

(Tolvaptan) is often provided. However, their efficacy and indications are not to bedetermined. We retrospectively examined the patients treated with Tolvaptan forlong term. Methods: Tolvaptan was administrated to 102 patients during hospitaliza-tion from October 2010 to May 2014. Long-term treatment of Tolvaptan afterdischarge was provided to 22 patients. Among 22 patients, 17 patients were fol-lowed-up for six months. We divided them into two groups whether re-hospitalizedwithin six months after discharge or not, and compared clinical characteristics atthe time of Tolvaptan administration. Results: Results are showed in Table. Conclu-sion: Serum creatinine and estimated GFR at the time of administration may predictlong-term efficacy of Tolvaptan.

O-174Increased Urine Aquaporin-2 Levels Relative to Plasma Arginine Vasopressin isa Novel Marker of Response to TolvaptanTERUHIKO IMAMURA1, KOICHIRO KINUGAWA1, TAKEO FUJINO2,TOSHIRO INABA2, HISATAKA MAKI2, MASARU HATANO2, ATSUSHI YAO2,ISSEI KOMURO2

1Department of Therapeutic Strategy for Heart Failure, Graduate School of Medicine,University of Tokyo, Tokyo, Japan, 2Department of Cardiovascular Medicine,Graduate School of Medicine, University of Tokyo, Tokyo, Japan

Background: Preserved function of collecting duct may be essential for the responseto vasopressin V2 receptor antagonist, tolvaptan (TLV), but the relationship with re-sponses to TLV has been unknown. Methods and Results: Consecutive 60 patientswith stage D decompensated heart failure (HF) who had received TLV on a de novobasis were retrospectively enrolled [TLV(+) group]. Among them, 41 patients were re-sponders defined by urine volume (UV) increase after TLV initiation. In the UV-definedresponders, plasma levels of arginine vasopressin (P-AVP) had a close correlation(r50.843, p!0.001) with urine aquaporin-2 (U-AQP2) levels (5.42 6 3.54 ng/mL).In contrast, 19 were the UV-defined non-responders, and they had extremely low U-AQP2 levels (0.766 0.59 ng/mL, p!0.001 vs. responders) regardless of P-AVP levels.Calculated by the ROC analysis, U-AQP2/P-AVPO0.5 �103 completely separated theUV-defined responders from the non-responders. Then, we determined the AQP-defined responders as having U-AQP2/P-AVP O0.5 �103. Propensity score-matched60 HF patients without TLV treatment were examined, and exactly the same numbersof patients (N541) were selected as the AQP-defined responders. These patients had apoorer survival without TLV than the TLV-treated responders over 2-year observationperiod (73.8% vs. 94.8%, p50.034). Conclusions: U-AQP2/P-AVP is a novel predictorfor the response to TLV in patients with decompensated HF. The AQP-defined re-sponders may enjoy a better prognosis by TLV treatment.

O-175BNP Level Predicts Clinical Efficacy of Tolvaptan in Patients with Heart FailureSEIICHIROU YOSHIMURA1, YOSHITAKA SUGAWARA1, YASUYUKIMARUYAMA2, TAKAAKI SHIONO3, TOSHIKAZU FUNAZAKI4, YOSHIHIKOSAKAI5, NORIHUMI KUBO6, HIROSHI WADA1, TAKESHI MITSUHASHI1,SHIN-ICHI MOMOMURA1

1Cardiovascular Division Saitama Medical Center, Jichi Medical University, Saitamacity, Saitama, Japan, 2Iwatsuki Minami Hospital, 3Kitazato Medical Center,4Kawaguchi Saiseikai Hospital, 5Koshigaya Hospital, Dokkyo Medical University,6Saitama Medical Center

Objective: We examined the effectiveness and safety of tolvaptan in patients hospi-talized for heart failure in multi center. Method: Clinical parameters including urinevolume, body weight, serum sodium level and renal function, were obtained from inhospital patients with heart failure who had received tolvaptan at 3.75 to 15 mg daily.Results: Tolvaptan was administered in 66 patients (male 48/female18, age 70.3+/14.1, NYHA2:3:457:42:15). During hospitalization, urinary volume increase(pre12526684 ml, post18426877ml, p! 0.0001) and body weight decreased signif-icantly(pre 64.1617.6Kg, post 59.8616.3Kg, p ! 0.0001), without significantchanges in blood pressure, heart rate, serum Na level and serum creatine level.BNP level had correlations with change of urine volume. Conclusions: PatientsBNP level predicts changes of urine volume.

O-176RV Dysfunction Plays an Important Role in Predicting Non-response toTolvaptan in Patients with Heart Failure with Reduced Ejection FractionNORIAKI IWAHASHI, TOSHIAKI EBINA, KAZUO KIMURADivision of Cardiology, Yokohama City University Medical Center, Yokohama,Japan

We explored the role of RV function for the prediction of Tolvaptan (TLV) in patientswith heart failure with reduced EF(HFrEF).Methods: 55 patients admitted for acute decompensated HF were enrolled (Age73years, 35male). EF were uniformly reduced (LVEF532%). They were administeredTLV at 7.5-15 mg/day for one week. Non-response was defined as no evidence of a 1kg decrease in body weight (BW) during a week. Blood sampling and echocardiog-raphy, including strain imaging, were performed just prior to starting TLV. Vivid-q