Tesis

Model Skoring Diagnostik Atresia Bilier Berdasarkan USG Abdomen 2 fase, Elastografi, Klinis, dan Biomarker = Diagnostic Scoring Model for Biliary Atresia Based on Two-Phase Abdominal Ultrasonography, Elastography, Clinical Findings, and Biomarkers.

Pendahuluan: Atresia bilier merupakan penyebab utama kolestasis neonatal yang dapat berkembang cepat menjadi sirosis dan gagal hati bila tidak didiagnosis dini. Modalitas diagnosis baku saat ini seperti kolangiografi intraoperatif bersifat invasif dan tidak ideal untuk skrining awal. Penelitian ini bertujuan mengembangkan model skoring diagnostik non-invasif berbasis integrasi klinis, ultrasonografi abdomen dua fase, elastografi, serta biomarker MMP-7 dan GGT untuk membedakan kolestasis atresia bilier dari kolestasis bukan atresia bilier. Metode: Studi potong lintang dilakukan terhadap neonatus dengan kolestasis. Subjek menjalani pemeriksaan klinis, laboratorium, shear wave elastografi, dan USG abdomen dua fase. Parameter yang dianalisis meliputi triangular cord sign, diameter arteri hepatika, rasio hepatika-vena porta, clay stool, kadar serum GGT dan MMP- 7. Analisis ROC digunakan untuk menentukan nilai potong, dan regresi logistik multivariat membentuk sistem skoring akhir. Hasil: Model skoring terdiri dari empat parameter utama: clay stool, diameter arteri hepatika >1,05 mm, kadar MMP-7 >7,8 ng/mL, dan GGT >318,5 U/L, masingmasing dengan skor 0–1 (total skor 0–4). Rentang skor 3–4 menunjukkan risiko tinggi atresia bilier. Pada nilai potong probabilitas 6,2%, sistem skoring menunjukkan sensitivitas 68,4%, spesifisitas 100%, nilai prediktif positif (PPV) 100%, nilai prediktif negatif (NPV) 76%, dan akurasi 84,2%. Uji chi-square menunjukkan signifikansi statistik (p=0,000) dan nilai Kappa 0,684 menunjukkan tingkat kesesuaian yang baik antara model dan diagnosis akhir. Kesimpulan: Model skoring diagnostik ini menunjukkan spesifisitas dan akurasi yang baik dalam membedakan kolestasis atresia bilier dari kolestasis bukan atresia. Meskipun sensitivitasnya masih terbatas, sistem ini menawarkan alternatif non-invasif yang menjanjikan sebagai alat bantu diagnosis awal, terutama kolestasis akibat bukan atresia bilier untuk menghindari prosedur invasif yang tidak perlu.
Kata Kunci: Atresia bilier, kolestasis neonatal, skoring diagnostik, USG abdomen dua fase, elastografi, MMP-7, GGT, clay stool


Introduction: Biliary atresia is the leading cause of neonatal cholestasis and can rapidly progress to cirrhosis and liver failure if not diagnosed early. The current diagnostic gold standard, intraoperative cholangiography, is invasive and not ideal for early screening. This study aims to develop a non-invasive diagnostic scoring model based on the integration of clinical findings, two-phase abdominal ultrasonography, elastography, and biomarkers MMP-7 and GGT to differentiate biliary atresia from non-biliary atresia cholestasis. Methods: A cross-sectional study was conducted on neonates with cholestasis. Subjects underwent clinical examination, laboratory testing, shear wave elastography, and two-phase abdominal ultrasound. Analyzed parameters included triangular cord sign, hepatic artery diameter, hepatic-to-portal vein ratio, presence of clay-colored stool, and serum levels of GGT and MMP-7. ROC analysis was used to determine diagnostic cut-off values, and multivariate logistic regression was applied to construct the final scoring system. Results: The diagnostic model comprised four key parameters: clay-colored stool, hepatic artery diameter >1.05 mm, MMP-7 >7.8 ng/mL, and GGT >318.5 U/L, each assigned a score of 0 or 1 (total score range: 0–4). A score of 3–4 indicated high risk for biliary atresia. At a probability cut-off of 6.2%, the model showed a sensitivity of 68.4%, specificity of 100%, positive predictive value (PPV) of 100%, negative predictive value (NPV) of 76%, and overall accuracy of 84.2%. Chi-square analysis showed statistical significance (p = 0.000), and a Kappa coefficient of 0.684 indicated good agreement between the model and final diagnosis. Conclusion: This diagnostic scoring model demonstrated good specificity and accuracy in distinguishing biliary atresia from other causes of neonatal cholestasis. Although its sensitivity remains limited, the system offers a promising non-invasive alternative for early diagnosis, particularly in ruling out non-biliary atresia to avoid unnecessary invasive procedures.
Keyword: Biliary atresia, neonatal cholestasis, diagnostic scoring, two-phase abdominal ultrasound, elastography, MMP-7, GGT, clay stool.