Persistent critical illness (PCI) is usually defined as a syndrome in which the acute presenting problem has been stabilized, but the patient's clinical state does not allow ICU discharge. The recovery is usually prolonged and protracted [1]. The burden associated with PCI is substantial, owing to medical complications (i.e., ventilator-associated pneumonia, surgical complication, etc) [[2], [3], [4]], increased costs [5,6], higher ICU and hospital length of stay (LOS) [3,7], increased rates of discharge to long-term health care facilities, and increased mortality [[2], [3], [4],8].

The most obvious marker of PCI is prolonged ICU LOS. Although PCI definition can ensue in 5–22 days since ICU admission [1,3,7,[9], [10], [11]], a more practical term accounts for ICU LOS longer than 10 days [2,4,8,12]. The prevalence of PCI is 5–16% of ICU admissions, accounting for 33–54% of ICU bed days [3,4,7,10].

Although case-control studies found no difference in baseline characteristics (age, sex, APACHE 2 (Acute Physiology and Chronic Health Evaluation), Charlson score, clinical frailty score, or activities of daily living dependence) between patients who suffered from PCI and those who did not [2,4], other studies demonstrated such differences. Those who suffer from PCI are more likely to include a greater proportion of males, have a higher APACHE 2 score, be admitted to the ICU due to respiratory disease, be older, and have a higher SOFA (Sequential Organ Failure Assessment) score at admission [3,8,[12], [13], [14]]. Harrison et al. & Zhang et al. demonstrated a greater impact of acute illness parameters (compared to chronic illness parameters) during the first days following ICU admission in the prediction of PCI [3,9]. Differences in patient characteristics between different admission reason categories to the ICU (e.g. trauma, sepsis, etc.), might be associated with PCI development [3].

The proposed pathophysiology underlying PCI is persistent inflammation-immunosuppression and catabolism syndrome (PICS). In brief, after an inflammatory trigger (whether infection, trauma or major surgery), a chemokine reaction favors bone marrow myeloid cell proliferation; However, maturation of these cells in PICS patients is blocked. These immature cells have both immunosuppressive and pro-inflammatory properties, resulting in prolonged inflammation and protein catabolism [13,15]. Several clinically applicable markers for PCI were suggested for early recognition of PCI, such as neutrophil-to-lymphocyte ratio (NLR) monocyte-to-lymphocyte ratio (MLR), and urea-to-creatinine ratio (UCr) [9,12,16]. The UCr was found significantly higher in trauma patients who developed PCI compared to those who didn't, starting in the early days of ICU admission [9,12]. Interestingly, a machine learning model demonstrated neither urea nor creatinine at admission were associated with PICS development in orthopedic trauma patients [17].

UCr is a marker of protein catabolism [18]. it is increased when urea increases in a greater proportion than creatinine, or when urea decreases to a lesser extent than creatinine. Increased urea at the time of ICU admission is associated with increased mortality. Urea increment during ICU stay is also a recognized finding, attributable to several reasons such as dehydration (or excessive negative fluid balance), gastrointestinal bleeding, steroid treatment, protein catabolism and more. Creatinine changes during ICU stay are usually attributed to kidney function, however, creatinine decrement during hospitalization is a marker of muscle wasting and ICU-acquired weakness. Therefore, UCr has been demonstrated to increase during ICU stay [[18], [19], [20]]. Haines et al. demonstrated elevated UCr in polytrauma patients who suffer from PICS starting from the 5th day after admission [12], and showed a correlation to ongoing muscle catabolism. Early recognition of these patients, might allow adequate nutritional support, taking into account the increased protein catabolism [21].

This work primarily aimed to characterize the patients who suffer from PCI, validate the previously suggested UCr as an early marker of PCI in trauma patients, and explore the application of this for prediction of PCI in other admissions reasons to the ICU. Secondly, we wished to explore whether UCr might serve as a mortality marker in this population.