Skip to main content
. 2020 Jul 8;15(7):e0235231. doi: 10.1371/journal.pone.0235231

Table 1. Examples of past literature on predicting PMV.

Study Patient group PMV duration Predictors Predictive technique Model`s performance (AUC)
Parreco et al. (2018) [8] All ventilated level 3 ICU patients (2001–2012) >7days Oxford Acute Severity of Illness Score (OAISIS), Sequential Organ Failure Assessment, Simplified Acute Physiology Score (SAPS), Simplified Acute Physiology Score II (SAPS II), Acute Physiology Score III, and logistic organ dysfunction score (LODS), Sepsis Related Organ Failure Assessment (SOFA) Gradient-Boosted Decision Tree Algorithm Mean AUC 0.820 ± 0.016
Chang et al (2018) [16] ICU patients who survived the Sepsis/Septic shock and respiratory failure >21 days Demographics Acute Physiology, Age, Chronic Health Evaluation (APACHE II) Comorbidities Lab findings (hematology, liver function, coagulation, Urea Electrolytes, Arterial blood gases) Ventilator settings Logistic Regression AUC 0.725
Agle et al. (2006) [12] Torso trauma patients who met specific criteria for shock resuscitation and required 48 hours of mechanical ventilation >14 days Demographics, Facial trauma, chest trauma severity (abbreviated injury score AIS), ventilatory settings. Logistic Regression AUC 0.79
Clark and Lettieri (2013) [5] Adult patients requiring MV support in a medical intensive care unit (ICU) >14 days Demographics, vital signs, laboratory values (hematology, renal and liver function tests, HCO3), APACHE 2 Logistic regression AUC 0.75
Dimopoulou et al. (2003) [13] Adult patients with thoracic trauma requiring MV support in a intensive care unit (ICU) >7 days Demographics, injury characteristics, Injury severity score, AIS of the other associated injuries (head, neck, face, pelvis and extremities) and ventilatory settings Logistic regression Not declared
Figueroa- Casas et al. (2015) [14] ICU patients receiving MV >7 days Demographics, SOFA score on intubation, comorbidities, location before ICU admission, diagnosis category Logistic regression AUC 0.65–0.70