Tesis

Performa Seattle Heart Failure Model (SHFM) dalam Memprediksi Major Adverse Cardiac Events (MACE) pada Pasien Gagal Jantung Kronik Pasca Acute Decompensated Heart Failure (ADHF) dalam Satu Tahun = Performance of the Seattle Heart Failure Model (SHFM) in Predicting Major Adverse Cardiac Events (MACE) in Chronic Heart Failure Patients Post-Acute Decompensated Heart Failure (ADHF) within One Year.

Latar Belakang: Pasien gagal jantung kronik pasca Acute Decompensated Heart Failure (ADHF) di Indonesia memiliki risiko tinggi mengalami Major Adverse Cardiac Events (MACE). Seattle Heart Failure Model (SHFM) merupakan sistem prediksi risiko mortalitas yang telah divalidasi secara luas, namun performanya belum pernah dievaluasi untuk memprediksi MACE pada populasi Indonesia. Tujuan: Penelitian ini bertujuan untuk mengevaluasi performa prediktif mortalitas SHFM, yang meliputi aspek kalibrasi dan diskriminasi, dalam mengestimasi kejadian MACE pada pasien gagal jantung kronik pasca ADHF di Indonesia. Metode: Studi kohort retrospektif ini mengikutsertakan 137 pasien di RSUPN Dr. Cipto Mangunkusumo. Performa SHFM dievaluasi menggunakan uji kalibrasi (Hosmer-Lemeshow) dan diskriminasi (Area Under the Curve, AUC). Analisis regresi logistik multivariat digunakan untuk memodifikasi model SHFM. Hasil: SHFM menunjukkan kalibrasi yang baik (p = 0,16) namun kemampuan diskriminasi yang buruk (AUC 0,63; 95% CI: 0,53-0,73). Analisis tambahan berhasil memodifikasi model SHFM menjadi lebih superior. Modifikasi ini memasukkan kelas fungsional NYHA, dosis diuretik, kadar asam urat, dan interaksinya sebagai prediktor utama menunjukkan kemampuan diskriminasi yang jauh lebih baik (AUC 0,77; 95% CI: 0,68-0,85). Kesimpulan: Pada pasien gagal jantung kronik pasca ADHF, nilai prediksi mortalitas SHFM terbukti terkalibrasi dengan baik (p=0,16) untuk prediksi MACE satu tahun. Akan tetapi, kemampuan diskriminasinya tergolong lemah (AUC=0,63), menunjukkan keterbatasan model dalam mengidentifikasi individu berisiko secara akurat.
Kata Kunci: Gagal jantung, MACE, Seattle Heart Failure Model


Background: Chronic heart failure patients post-Acute Decompensated Heart Failure (ADHF) in Indonesia have a high risk of experiencing Major Adverse Cardiac Events (MACE). The Seattle Heart Failure Model (SHFM) is a widely validated risk prediction system, but its performance has never been evaluated for predicting MACE in the Indonesian population. Objective: This study aims to evaluate the predictive performance of SHFM, including calibration and discrimination aspects, in estimating the incidence of MACE in chronic heart failure patients post-ADHF in Indonesia. Methods: This retrospective cohort study included 137 patients at Dr. Cipto Mangunkusumo National Central Public Hospital. The performance of the SHFM was evaluated using the Hosmer-Lemeshow test for calibration and the Area Under the Curve (AUC) for discrimination. Multivariate logistic regression analysis was used to modify the SHFM model. Results: SHFM showed good calibration (p = 0.16) but poor discrimination (AUC 0.63; 95% CI: 0.53-0.73). Further analysis successfully modified the SHFM, resulting in a superior model. This modification, which incorporated NYHA functional class, diuretic dose, uric acid levels, and their interactions as primary predictors, demonstrated significantly better discrimination (AUC 0.77; 95% CI: 0.68-0.85). Conclusion: In patients with chronic heart failure post ADHF, the SHFM mortality prediction score was found to be well-calibrated (p=0.16) for one-year MACE prediction. However, its discriminatory ability was poor (AUC=0.63), indicating the model's limitation in accurately identifying at-risk individuals.
Keywords: Heart failure, MACE, Seattle Heart Failure Model