AbstrakSepsis berat merupakan kondisi umum di unit perawatan intensif (UPI) dan rawat inap yang berhubungan dengan mortalitas, morbiditas, dan biaya perawatan yang tinggi. Tujuan penelitian ini ingin mendapatkan skor alternatif yang lebih sederhana, yaitu nilai mean platelet volume (MPV) sebagai prediktor mortalitas pada pasien sepsis berat selain skor APACHE II. Penelitian ini menggunakan desain cross sectional pada 76 pasien sepsis berat dewasa di Rumah Sakit H. Adam Malik Medan pada Oktober 2015-Januari 2016 yang memenuhi kriteria inklusi. Data yang diambil adalah nilai MPV dan skor APACHE II pada saat pertama sekali terdiagnosis sepsis berat, kemudian dilihat mortalitas pasien tersebut. Uji korelasi Spearman menunjukkan bahwa terdapat korelasi lemah yang signifikan (p=0,006) antara MPV dan APACHE II dengan nilai r (korelasi) = 0,314. Nilai MPV pada penelitian ini tidak memiliki kemampuan dalam memprediksi mortalitas (AUC) ROC 58,2% (IK 95%: 45,1-71,2%; p=0,223); sedangkan skor APACHE II memiliki kemampuan yang sedang dalam memprediksi mortalitas (AUC) ROC 70,4% (IK 95%: 58,6-82,2%; p= 0,002 Correlation between Mean Platelet Volume (MPV) and Apache II Score as Mortality Predictors in Severe Sepsis Patients at Haji Adam Malik General Hospital Medan AbstractSevere sepsis is a general condition in the Intensive Care Unit (ICU) and inpatient wards which correlates with mortality, morbidity, and high cost hospitalization. The main point of this study was to explore the possibility to use the mean platelet volume (MPV) as an easier alternative score for mortality predictor in addition to APACHE II score in severe sepsis patients. This study used cross-sectional design on 76 adult severe sepsis patients in Haji Adam Malik General Hospital Medan who met inclusion criteria during the periood of October 2015 to January 2016. Data collected were MPV value and APACHE II score, which were collected the first time patient was diagnosed as having severe sepsis which was then observed for their mortality The Spearman correlation tests showed that there was a weak yet significant correlation (p=0.006) between MPV and APACHE II with r (correlation) = 0.314. The MPV values in this study were unable to predict mortality (AUC) ROC 58.2% (95% CI: 45.1-71.2%, p=0.223). whereas the APACHE II score has a moderate ability to predict mortality (AUC) ROC 70,4% (95% CI: 58,6-82,2%, p= 0.002). The cut-off point of APACHE II was 19 with a sensitivity of 65.9% and a specificity of 65.7%, and a PPV of 69.2% and NPV of 62.2%. Therefore, based on this study the MPV score cannot be used as a mortality predictor in severe sepsis patients.