Latar belakang: Penentuan kebutuhan nutrisi secara tepat pada anak sakit kritis perludilakukan untuk menghindari underfeeding dan overfeeding. Rumus estimasi menjadidasar perkiraan kebutuhan energi jika kalorimetri indirek sebagai baku emas tidaktersedia. Akurasi rumus pada studi terdahulu sangat bervariasi dan dipengaruhi olehkarakteristik populasi setempat, sehingga akurasinya perlu diuji pada populasi Indonesia.Tujuan: Mengevaluasi akurasi rumus Schofield WH, Schofield W, dan WHOdibandingkan kalorimetri indirek, serta mengevaluasi dampak penambahan faktor stresterhadap akurasi.Metode: Penelitian deskriptif analitik potong lintang ini mengikutsertakan pasien anakyang menggunakan ventilasi mekanik Mei sampai Juli 2019. Analisis kesesuaiandilakukan dengan membandingkan perhitungan rumus Schofield WH, Schofield W, danWHO, dengan dan tanpa faktor stres terhadap pengukuran kalorimetri indirek.Hasil: Penelitian mengikutsertakan 52 subjek pada hari perawatan 1-5 di PICU denganmedian usia 5 tahun (1 bulan 10 hari hingga 17 tahun 9 bulan). Kebutuhan energi yangdiukur kalorimetri indirek adalah 60,7 ± 23,5 Kkal/kg/hari. Estimasi rumus SchofieldWH, Schofield W, dan WHO lebih rendah dari hasil pengukuran tersebut dengan %biasberturut-turut -13 ± 19, -15 ± 20, dan -16 ± 21. Nilai estimasi dan hasil pengukurankalorimetri indirek berkorelasi kuat (intraclass correlation coefficient r > 0,9) namuninterval kesesuaian (limit of agreement) dari %bias sangat lebar. Hanya 12 (23%) subjekyang memiliki nilai estimasi akurat sesuai dengan kalorimetri indirek. Pada populasipenelitian ini faktor stres meningkatkan akurasi rumus estimasi.Simpulan: Rumus Schofield WH, Schofield W, dan WHO tidak akurat sebagai estimasikebutuhan energi anak sakit kritis. Hasil prediksi rumus tersebut lebih rendah darikebutuhan aktual jika faktor stres tidak digunakan.

---

Background: Accurate estimation of energy expenditure in critically ill children isimportant to avoid underfeeding and overfeeding. Prediction formula helps to estimateenergy expenditure when the gold standard indirect calorimetry is not available. Previousstudy on estimation accuracy yielded variable result in different populationcharacteristics, therefore the accuracy of prediction formula in Indonesian populationneeds to be evaluated.Objective: To assess the accuracy of Schofield WH, Schofield W, and WHO formulacompared to indirect calorimetry. To evaluate the impact of additional stress factor on theaccuracy of prediction formula.Methods: This is a descriptive analytic cross-sectional study on mechanically ventilatedcritically ill children held in May-July 2019. We analyze the agreement of measuredenergy expenditure using indirect calorimetry and estimated energy expenditurecalculated by Schofield WH, Schofield W, and WHO formula, with and withoutadditional stress factor.Results: This study included 52 subjects with median age 5 years old (1 month 10 days -17 years 9 months) on day 0-5 after they were admitted to PICU. Mean measuredenergy expenditure was 60,7 ± 23,5 Kcal/kg/day. All estimated energy expenditure bySchofield WH, Schofield W, and WHO were lower than measured energy expenditurewith % bias of -13 ± 19, -15 ± 20, and -16 ± 21, respectively. Estimated and measuredvalue have strong correlation (intraclass correlation coefficient r > 0.9) but the limit ofagreement interval is too wide. Only 12 (23%) subjects have accurate estimation ofenergy expenditure. In this population stress factor improves the accuracy of predictionformulas.Conclusion: Schofield WH, Schofield W, and WHO formula have poor accuracy inestimating energy expenditure in critically ill children. Without additional stress factor,the estimated value were lower than actual/measured energy expenditure