Abstrak
Acute Respiratory Distress Syndrome (ARDS) meningkatkan kebutuhan perawatan intensif pada pasien COVID-19. Diperlukan upaya deteksi cepat untuk memilah prioritas pasien COVID-19 yang berisiko ARDS. Penelitian bertujuan memprediksi derajat ARDS menggunakan kuantifikasi luas opasitas/konsolidasi foto dan mengidentifikasi faktor prediktor ARDS. Metode: Desain studi crossectional dilakukan di Rumah Sakit Umum Pusat Fatmawati Jakarta pada Juni-Desember 2022. Uji diagnostik terhadap hasil skoring foto toraks dengan komparasi 3 metode yaitu Brixia, Reeves dan modifikasi skoring foto dengan metode skoring Universal Thorax ARDS Measurement Index (UTAMI) dibandingkan dengan diagnosis ARDS menggunakan kriteria Berlin sebagai gold standard. Hasil: Sebanyak 318 pasien rawat inap COVID-19 dengan pneumonia dianalisis. Faktor laboratorium seperti kadar neutrofil, CRP, D-dimer, saturasi dan respiratory rate merupakan faktor prediktor ARDS dengan metode Berlin. Pada metode skoring UTAMI, diketahui komorbid CAD, CRP dan saturasi oksigen dapat memprediksi kejadian ARDS. CRP merupakan faktor prediktor terhadap ARDS pada kedua metode Berlin (PR 1,28; 95% CI 0,97 -1,70) dan UTAMI (PR 1,71; 95% CI 1,19– 2,46). Pada uji AUROC diketahui bahwa nilai PaO2/FiO2 dengan metode berlin bisa memisahkan pasien ARDS ICU dan non-ARDS ICU dengan akurasi 81,2% sedang Metode skoring UTAMI sebesar 79,8 %. sehingga sensitifitas dan spesifisitas pada metode skoring UTAMI terhitung kategori baik. Kesimpulan: Metode UTAMI dapat digunakan untuk memprediksi level ARDS pasien yang membutuhkan ICU . Saran: Klinisi dapat mengaplikasikan model prediktif ARDS ini untuk meningkatkan pelayanan ICU di rumahsakit
Background: Acute Respiratory Distress Syndrome (ARDS) has an impact to increase the need for intensive care among COVID-19 patients. Early detection is needed to prioritize COVID-19 patients at risk of ARDS. The research aims to predict the level of ARDS using quantification of the extent of photo opacity/consolidation and identifying factors of ARDS. Method: A cross-sectional study design was carried out at the Fatmawati Central General Hospital, Jakarta in June-December 2022. Diagnostic tests on the results of thorax x-ray scoring using a comparison of 3 methods, consist of Brixia, Reeves and modified photo scoring using the UTAMI method were compared with ARDS diagnosis using the Berlin criteria as gold standards. Results: A total of 318 hospitalized COVID-19 patients with pneumonia were analyzed. Laboratory factors such as neutrophil levels, CRP, D-dimer, saturation and respiratory rate are predictor factors for ARDS using the Berlin method. Meanwhile, using UTAMI scoring, it is known that comorbid CAD, CRP and Oxygen Saturation are predictor the incidence of ARDS. CRP was a predictor factor for ARDS in both the Berlin (PR 1.28; 95% CI 0.97 –1.70) and UTAMI (PR 1.71; 95% CI 1.19–2.46) methods. In the AUROC test, it was found that the PaO2/FiO2 using the Berlin method could separate ARDS ICU patients from non-ARDS ICU patients with an accuracy of 81.2%. Meanwhile, the UTAMI scoring method was 79.8%. so that the sensitivity and specificity of the UTAMI scoring method are in fair discrimination. Conclusion: The UTAMI method can be used to predict a patient's ARDS level. Recommendation: Clinicians could use UTAMI method as predictive model score to estimate the need of Intensive Care Unit in ARDS.