Acute respiratory distress syndrome (ARDS) and acute kidney injury (AKI) are two common contributing causes of admission to the intensive care unit (ICU), with important implications for patient morbidity, mortality and health resource utilization. We will briefly examine recent evidence from studies focused on how novel organ damage biomarkers could be used to drive clinical strategies in acutely ill patients with AKI or ARDS (including the weaning phase after ARDS resolution) (Table 1).

Table 1 Examples of biomarkers that could drive clinical strategies during acute respiratory distress syndrome and acute kidney injury
Full size table

Acute respiratory distress syndrome and weaning

Many biomarkers have been explored for the diagnosis and prognosis of ARDS and recent studies highlight their potential to drive clinical strategies. One important area is the differentiation of the degree of lung epithelial versus endothelial injury, which may be imprecise when based only on clinical presentation. Currently, the most promising plasma biomarker for assessing endothelial injury during ARDS is angiopoietin-2 [1], whereas the receptor for advanced glycation end products (RAGE) may best depict the extent of lung epithelial injury [2]. The development of rapid point-of-care (POC) tests for circulating levels of these biomarkers could lead to clinical trials of biomarker-guided, cell-specific ARDS therapy testing epithelial-targeted therapies such as epithelial growth factors [3] or endothelial barrier permeability modifiers [4].

Biomarkers also have the potential to guide ventilator management. In the ARMA trial, lower tidal volumes led to lower levels of plasma IL-6, IL-8, and TNFR1 over the subsequent 1–3 days [5]. More recent studies also showed fluctuations in circulating inflammatory cytokines within 6 h of a change in mechanical ventilator strategy [6] and variations in plasma RAGE levels within 1 h of a recruitment maneuver [7]. Given their rapid change with ventilator settings modifications, serial bedside measurement of circulating inflammatory cytokines and/or RAGE could be used to assess for ongoing ventilator-induced lung injury and to guide ventilator management in order to minimize lung strain. In latter stages of ARDS, the determination of the N-terminal-peptide type III procollagen on bronchoalveolar lavage (BAL NT-PC-III) done at day 7 was recently shown to identify patients with fibroproliferation; ARDS patients with a BAL NT-PC-III above 9 µg/L could be selected for inclusion in a trial of corticosteroid therapy [8].

The usefulness of cardiac biomarkers like natriuretic peptides in differentiating ARDS from hemodynamic pulmonary edema is limited because of the frequent rise in left-sided filling pressures during ARDS and the influence of right ventricle failure and other confounding factors in this context. However, in patients recovering from ARDS or other forms of respiratory failure, natriuretic peptides may be useful to optimize weaning from mechanical ventilation, which is frequently jeopardized by acute heart failure. Studies on natriuretic peptides during weaning have suggested that elevated baseline value [9] or increases during failed spontaneous breathing trial (SBT) [10] may help predict difficult weaning and identify its cardiac origin, respectively. In a randomized trial, a simple POC natriuretic peptide-guided fluid management strategy was associated with increased diuretic use, a more negative fluid balance, and a shorter duration of mechanical ventilation, especially in patients with left ventricular systolic dysfunction [11]. An increase in plasma protein or hemoglobin concentration during SBT, reflecting the blood volume contraction induced by pulmonary edema formation, has also been demonstrated to diagnose cardiogenic weaning failure [12].

Acute kidney injury

The diagnosis of AKI in critical care settings has essentially relied on measurement of serum creatinine and urine output for many decades with very little progress to inform clinical decision-making [13]. Serum creatinine only represents a surrogate biomarker for glomerular filtration, not acute injury to the kidney per se, and has numerous limitations including wide variations in normal range (due to age, sex, diet, muscle mass, drug exposures, fluid status), more than 24 h delay to reach steady state after insults, and the need for more than 50 % of kidney function loss prior to significant increases being detected. While numerous novel kidney damage biomarkers for AKI have been characterized [e.g., cystatin c, neutrophil gelatinase-associated lipocalin (NGAL), kidney injury molecule-1 (KIM-1), and L-type fatty acid binding protein (L-FABP)] across a range of clinical settings, few have proven to have clear utility for integration at the bedside to inform clinical decision-making [e.g., when to start renal replacement therapy (RRT)] [14].

Recently, two promising novel biomarkers for AKI were discovered and characterized from approximately 300 candidate biomarkers: tissue inhibitor of metalloproteinase-2 (TIMP-2) and insulin-like growth factor binding protein 7 (IGFBP7) [15]. Both TIMP-2 and IGFBP7 are inducers of G1 cell cycle arrest during early phases of injury, implying that they may have an important role in the adaptive response, preservation, and repair processes associated with acute kidney insults (i.e., sepsis, ischemia). In this derivation and validation study (SAPPHIRE) of 522 critically ill adults, the combined urine biomarker product of [TIMP-2*IGFBP7] was found to predict the development of AKI (KDIGO stage 2) within 12 h [AUC 0.80 (95 % CI 0.75–0.84); as compared to 0.78 (0.72–0.83) and 0.79 (0.75–0.84) for either alone], was stable across clinical syndromes, and was shown to be superior to all previously described AKI biomarkers. Urinary [TIMP-2*IGFBP7] >0.3 and >2 ng/ml measured with a POC test were further validated for predicting AKI as high sensitivity and high specificity cutoffs, respectively [16, 17]. The addition of urinary [TIMP-2*IGFBP7] to the risk reference multivariable model resulted in an integrated discrimination improvement (IDI) of 0.098 (95 % CI 0.032–0.265) and a net reclassification improvement (NRI) of 70 % (95 % CI 19–96), with an improvement model AUC of 0.06 (95 % CI 0.04–0.36) [15, 18]. Further data have shown urinary [TIMP-2*IGFBP7] to be a sensitive and specific predictor of AKI, along with showing association with long-term mortality, and preliminary data suggesting it may have value for informing about recovery after major cardiac and non-cardiac surgery [16, 19, 20]. In 2014, the US Food and Drug Administration permitted marketing of NephroCheck® (Astute Medical, Inc., USA) as the first point of care device using a fluorescence immunoassay to measure urinary [TIMP-2*IGFBP7]. Among patients with perceived risks for AKI, applying the high sensitivity threshold could focus patient care towards strategies of intensified monitoring (e.g., short-term transfer to a high-dependency unit or ICU from the emergency department, avoidance of episodes of hemodynamic instability, limitation of exposure to kidney toxins, adaptation of drug dosing, and reduction of unnecessary fluid accumulation), prior to any increase in serum creatinine. Alternatively, the high specificity threshold, where there is greater confidence in fewer false positives, could be utilized and integrated with additional clinical information to help inform more complex and invasive management decisions, such as if and when to start RRT.

In conclusion, although significant research efforts have been deployed in this era, future prospective studies are still needed to validate biomarkers to be used in clinical strategies in ICU patients with ARDS or AKI.