ARDS onset is often rapid and progressive, resulting from either septic or non-septic insults [20]. Sepsis as the most commonly encountered cause of ARDS is generally associated with higher mortality than other risk factors [3, 21–23]. Our study demonstrated that sepsis- and non-sepsis- related ARDS present differently in terms of clinical characteristics, with higher severity of illness and worse clinical outcomes in sepsis-related ARDS. Response to treatment and clinical outcomes differ on the basis of sub-phenotype [24]. Early recognition of individual patients with sepsis-related ARDS is recommended during ICU treatment in order to establish targeted therapies.
However, the diagnosis and stratification of ARDS are based on clinical definitions that lack both sensitivity and specificity. Clinical symptoms and signs can be misleading, especially in patients with variable clinical characteristics or several comorbidities; thus, early diagnosis of sepsis-related ARDS is not straightforward. Recent interest has focused on blood biomarkers, which might capture aspects of pathophysiology that are not otherwise well captured in clinical data, and generally contribute more prominently to the sub-phenotype determination in clinical work [25].
Presepsin, a 13-kDa fragment of sCD14, is released in the plasma as a consequence of cellular phagocytosis after bacterial infection and appears to be a novel biomarker of sepsis [26]. It has been used for early diagnosis, risk stratification and prognostic evaluation of sepsis in recent years [8, 9, 11, 14]. However, no studies have tested the possible value of presepsin in patients with sepsis who have developed ARDS. Thus, we designed a multicenter prospective study to validate the diagnostic and prognostic role of presepsin in sepsis-related ARDS; and compare it with the clinical value of PCT.
As our results show, compared to that in healthy controls, presepsin increases considerably in patients with ARDS. Subpopulation analysis showed a significant correlation between presepsin levels and the different populations in the ARDS group (sepsis-related ARDS and non-sepsis-related ARDS). Presepsin was invariably elevated in patients with sepsis-related and non-sepsis-related ARDS. Patients with sepsis-related ARDS had notably higher plasma presepsin levels than patients with non-sepsis-related ARDS. The difference was present in the early phase of evolving ARDS, thereby allowing the discrimination between sepsis-related and non-sepsis-related ARDS before the results of microbiological testing are generally available. Presepsin seems to be a diagnostic tool for differentiating between septic and non-septic underlying disease in early ARDS.
Although serum PCT has been used as a biomarker in the diagnosis of sepsis, PCT is also increased in other risk conditions of ARDS, such as multiple trauma, extensive burns, pancreatitis, organ transplantation, and major surgery [13]. In our setting, PCT was elevated in both patient groups with a substantial overlap between patients with sepsis-related and non-sepsis-related ARDS patients. The median levels of PCT were significantly more elevated in sepsis-related ARDS, while ROC curve analysis showed that PCT concentrations had less valuable diagnostic capacity for sepsis-related ARDS than presepsin. Therefore, PCT shows limited value for distinguishing sepsis from other etiologies in patients with ARDS.
Higher levels of plasma presepsin were associated with worse clinical outcomes in patients with sepsis [8, 27, 28]. In our patients with sepsis-related ARDS, presepsin concentrations were higher in decedents than in survivors. Analysis of mortality in sepsis-related ARDS showed that a significant correlation between presepsin levels and mortality at early stages. Measurements of presepsin levels revealed valuable prognostic capacity to predict all-cause in-hospital mortality. After adjustments for other clinical variables, presepsin retained acceptable prognostic value for in-hospital mortality. However, we are not suggesting the application of presepsin as the sole marker for predicting mortality, but merely emphasizing its association with mortality and the associated implications for potentially contributing to risk stratification in combination with other clinical tools (such as risk-prediction scores).
The SOFA score, a complex score evaluaing six different organ systems and addressing diverse clinical parameters, predicts the severity and mortality of critically ill patients with high accuracy [17]. A significant correlation was found between presepsin levels and the SOFA score in patients with sepsis-related ARDS. Additionally, the prognostic accuracy of presepsin appeared to be similar to that of the SOFA score for early mortality in patients with sepsis-related ARDS. It is enticing that the power of presepsin as a robust circulating biomarker for patient prognosis is comparable to that of the complex SOFA score in sepsis-related ARDS.
To the best of our knowledge, the results presented herein are the first with regard to the discriminative value and prognostic capacity of presepsin for sepsis-related ARDS. Presepsin may add to diagnostic accuracy and facilitate early recognition of patients with sepsis-related ARDS who are likely to benefit from promptly appropriate broad-spectrum antibiotics. Our results also pointed out the possible prognostic role of presepsin in promptly identifying high-risk patients with sepsis-related ARDS and predicting in-hospital mortality. The application of presepsin in ARDS sub-phenotypes (sepsis-related and non-sepsis-related) might be a useful tool to stratify patients in future clinical research trials, which might be advantageous for future differential treatments.
A major strength of this study is that it was conducted in two large, well-defined centers and four ICUs. All data were collected prospectively to avoid recall biases. Nevertheless, several limitations should be acknowledged. First, the number of patients was relatively small; subsequent studies with large-scale and independent cohorts should confirm and validate the clinical indications for presepsin in sepsis-related ARDS. Second, the study design limited patients to those who required IMV, thereby not generalising patients who met the ARDS criteria but only received non-IMV. Third, only one measurement of presepsin was available. Dynamic monitoring of circulating biomarkers is more vital and rewarding during the management of disease. Finally, we were not able to perform the commonly accepted Cox regression because of limitations in our data.