Agung Pradnyana Suwirya, I Gusti Bagus Putu Suwarjana Kaler, Bagus Ari Pradnyana Dwi Sutanegara, Renny Anggreni Rena

Perbandingan skor Padua, Khorana, Protecht dan Modified Padua sebagai prediktor kejadian deep vein thrombosis (DVT) pada pasien kanker yang menjalani kemoterapi di Rumah Sakit Umum Pusat Prof. Dr. I. G. N. G. Ngoerah, Bali, Indonesia

Introduction

Perbandingan skor padua, khorana, protecht dan modified padua sebagai prediktor kejadian deep vein thrombosis (dvt) pada pasien kanker yang menjalani kemoterapi di rumah sakit umum pusat prof. Dr. I. G. N. G. Ngoerah, bali, indonesia. Studi ini membandingkan skor Padua, Khorana, Protecht, dan Modified Padua sebagai prediktor DVT pada pasien kanker kemoterapi. Skor Modified Padua ≥4 sangat efektif.

Abstract

Introduction: Venous thromboembolism is a very common complication of cancer. Due to the high risk and potential complications, a risk assessment model (RAM) is needed to predict the likelihood of thromboembolism in cancer patients. Methods: This study used a case-control design, utilizing a duplex ultrasound (DUS) registry and medical record searches in cancer patients undergoing chemotherapy and undergoing comprehensive duplex ultrasound (CDUS). A comparison was made between the Padua, Khorana, Protecht, and Modified Padua scoring systems as predictors of deep vein thrombosis. Independent t-tests, chi-square tests, logistic regression, and receiver operator regression (ROC) analyses were used. Results: The Padua, Khorana, and Protecht scores had an area under the curve (AUC) of 0.714, 0.757, and 0.801, respectively, while the Modified Padua score had an AUC of 0.915. Significant variables found to be predictors of deep vein thrombosis were decreased mobility (OR: 6.662), history of trauma and/or surgery within one month (OR: 16.972), pre-chemotherapy leukocyte count >11x109/L (OR: 15.964), and respiratory failure (OR 8.474). Conclusion: A modified Padua score ≥4 (high risk) can be considered as a guideline for performing CDUS to confirm the possibility of DVT in patients. Pendahuluan: Tromboemboli vena merupakan komplikasi dari kanker yang sangat sering terjadi. Akibar tingginya risiko dan kemungkinan komplikasi yang terjadi diperlukan suatu risk assessment model (RAM) guna memprediksi kemungkinan terjadinya tromboemboli pada pasien kanker. Metode: penelitian ini menggunakan rancangan kasus kontrol, menggunakan register duplex ultrasound (DUS) dan penelusuran rekam medis pada pasien kanker yang menjalani kemoterapi dan dilakukan comprehensive duplex ultrasound (CDUS). Dilakukan perbandingan terhadap system skoring Padua, Khorana, Protecht dan Modified Padua sebagai prediktor dari thrombosis vena dalam. Analisis uji T independent, uji chi-square, regresi logistik, dan kurva receiver operator (ROC) digunakan dalam pebelitian ini. Hasil: Skor Padua, Khorana dan Protecht memili area under the curve (AUC) (0,714; 0,757 dan 0,801), sedangkan skor Modified Padua AUC (0,915). Variabel signifikan yang ditemukan dapat menjadi prediktor dari kejadian thrombosis vena dalam adalah penurunan mobilitas (OR: 6,662), riwayat trauma dan/atau pembedahan dalam satu bulan (OR: 16,972), kadar leukosit pre kemoterapi >11x109/L (OR: 15,964), gagal napas (OR 8,474). Simpulan: Skor modified Padua ≥4 (resiko tinggi) dapat dipertimbangkan sebagai panduan dalam melakukan CDUS untuk mengkonfirmasi kemungkinan terjadinya DVT pada pasien.

Review

This study addresses a critical clinical need for effective risk assessment models to predict deep vein thrombosis (DVT) in cancer patients undergoing chemotherapy, a population known for its high incidence of venous thromboembolism (VTE). The authors undertook a comparative analysis of four prominent scoring systems—Padua, Khorana, Protecht, and a Modified Padua score—within a specific patient cohort at Prof. Dr. I. G. N. G. Ngoerah Hospital in Bali, Indonesia. By evaluating these models in a local context, the research aims to identify the most accurate predictor to guide diagnostic procedures and potentially improve patient outcomes. The chosen case-control design, utilizing existing duplex ultrasound registries and medical records, provides a pragmatic approach to assessing these predictive tools. Methodologically, the study employed standard statistical analyses including independent t-tests, chi-square tests, logistic regression, and receiver operator characteristic (ROC) curves to robustly compare the predictive capabilities of each scoring system. The results unequivocally demonstrate the superior performance of the Modified Padua score, which achieved a remarkable Area Under the Curve (AUC) of 0.915. This significantly outcompetes the other scores (Padua: 0.714, Khorana: 0.757, Protecht: 0.801), indicating its strong discriminatory power in identifying patients at high risk of DVT. Furthermore, the study identified several significant independent predictors of DVT, including decreased mobility, a history of recent trauma or surgery, elevated pre-chemotherapy leukocyte count, and respiratory failure, each carrying substantial odds ratios and highlighting specific clinical vulnerabilities in this patient group. The findings carry significant clinical implications, particularly for the local setting. The conclusion that a Modified Padua score ≥4 should be considered a guideline for performing comprehensive duplex ultrasound (CDUS) offers a clear, actionable recommendation for clinicians. This locally validated tool can enhance targeted DVT screening, potentially reducing unnecessary diagnostic procedures while ensuring that high-risk patients receive timely confirmation and intervention. While the single-center, retrospective nature of this case-control study may limit its immediate generalizability to broader populations, it provides a crucial foundation for future prospective validation studies. The identification of specific local risk factors also suggests avenues for further refinement of VTE risk assessment models tailored to regional patient characteristics and healthcare contexts.

Full Text

Comments

You need to be logged in to post a comment.