Serum heme oxygenase-1 measurement is useful for evaluating disease activity and outcomes in patients with acute respiratory distress syndrome and acute exacerbation of interstitial lung disease

Acute respiratory distress syndrome (ARDS) is one of the major manifestations of multiple organ failure syndrome and is a leading cause of death in intensive care units [1]. Within the clinical course of interstitial lung disease (ILD), an acute exacerbation (AE) can occur at any time and is associated with significant morbidity and mortality [2]. Diffuse alveolar damage (DAD) is considered the histological hallmark of the acute phase of ARDS and AE-ILDs, while alternative histological appearances comprise organizing pneumonia, alveolar haemorrhage, and unspecific inflammatory changes [3, 4]. The clinical course and rate of progression of ARDS and AE-ILDs are extremely variable among patients. Therefore, biomarkers including symptoms, blood, physiological, radiological, and pathological findings and these combinations may be useful in characterizing disease severity and predicting the rate of progression and response to therapies [5, 6].

Oxidative/nitrosative stress results from an imbalance between cellular production of reactive oxygen species (ROS)/reactive nitrogen species (RNS) and the endogenous antioxidants such as stress response protein (heme oxygenase-1 (HO-1)), classic antioxidant enzymes (superoxide dismutases (SODs), catalase, glutathione peroxidase (GPx)) [7]. Nuclear factor-erythroid 2 p45 subunit-related factor 2 (Nrf2) is essential for activating response in the lung due to induction of the expression of antioxidant and these enzymes are expressed in bronchial and alveolar epithelial cells and macrophages of the lung [8]. Several clinical evidences suggested that increased oxidative/nitrosative stress might play a major role in the progression of various lung diseases such as idiopathic pulmonary fibrosis (IPF), chronic obstructive pulmonary disease, and ARDS [9‐14]. HO-1 catalyzes heme degradation to biliverdin-IXα, carbon monoxide, and iron. These metabolites mediate the antiapoptotic, anti-inflammatory, vasodilatory, anticoagulant, antioxidant, and antiproliferative properties of HO-1 under the control of the microsomal nicotinamide adenine dinucleotide phosphate-cytochrome p450 reductase [15, 16]. HO-1 expression is induced by various stimuli such as exposure of cigarette smoke extract (CSE), heme, hypoxia, endotoxin, and pro-inflammatory cytokines. Suzuki et al. reported that in human alveolar macrophages, acute CSE exposure increases HO-1 mRNA for 2 h [17]. Mumby et al. reported that HO-1 protein concentrations are significantly elevated in lung tissue and bronchoalveolar lavage fluid taken from ARDS patients compared with controls, and HO-1 expression contributes to changes in iron mobilization, signalling, and regulation seen in this condition [18]. Also, we have demonstrated the usefulness of measuring serum HO-1 in the diagnosis and prognosis of patients with ARDS and AE-ILDs [19, 20]. Therefore, we speculate that HO-1 characterized as rapid stress response protein with various physiological activities caused by HO-1 metabolites could reflect pulmonary cellular damage induced by ROS and RNS more closely and directly [7, 15‐17].

In the present study, we investigated whether the degree of oxidative stress measured by serum HO-1 levels at the diagnosis could be useful for predicting prognosis and these levels could decrease during the clinical courses among patients with ARDS and AE-ILDs.

Increased oxidative/nitrosative stress might play a major role in the progression of various lung diseases including AE-ILDs and ARDS [9‐14]. HO-1, a rate-limiting enzyme in heme catabolism, has antioxidative activities in patients with these diseases [27‐29]. We previously investigated whether evaluating the degree of oxidative stress by measuring serum HO-1 using the sandwich ELISA method is useful for assessing disease activities and predicting prognosis in patients with ARDS and AE-ILDs [19, 20]. In the present study which was an integrated analysis of these previous studies, we investigated whether composite parameters including serum HO-1 and other clinical parameters could predict the prognosis of ARDS and AE-ILDs more accurately than serum HO-1 alone and oxidative stress measured by serum HO-1 levels could be reduced in the clinical course.

ARDS and AE-ILDs are a life-threatening event and the mortality rate is high [30, 31]. A retrospective analysis in patients with ARDS showed a significant increase in hospital mortality in patients with DAD compared to those without DAD (71.9% vs. 45.5%) [30]. Other retrospective cohort study in patients with AE of chronic fibrosing IP showed the overall survival after admission was 67% at 1 month and 40% at 3 months [31]. In both patients with ARDS and AE-ILDs, our data showed better prognosis than these cohort studies, however, the patients with ARDS tended to show worse prognosis than those with AE-ILDs as previously reported. In the present study, ARDS patients had significantly higher serum HO-1 and CRP levels at baseline compared with AE-ILD patients and serum HO-1 had positive correlation with serum LDH. These results indicate that ARDS patients had a stronger degree of systemic inflammatory response, pulmonary epithelial cell damage and endothelial cell damage with the consequent increase of vascular permeability reflecting oxidative/nitrosative stress response than AE-ILD patients [32‐34]. Though similar with HO-1, SODs, catalase, and GPx had been described as the endogenous antioxidants, HO-1 characterized as stress response protein with relatively small-molecular weight (32 kDa, much smaller than KL-6 (5000 kDa)), rapid response against stimuli, and various physiological activities including the antiapoptotic, anti-inflammatory, vasodilatory, anticoagulant, antioxidant, and antiproliferative reactions caused by HO-1 metabolites [7, 15‐17]. Therefore, we speculate that serum HO-1 reflects pulmonary cellular damage induced by ROS and RNS more closely and directly.

Ongoing and persistent oxidative stress leads to poor prognosis [35]. In patients with ILDs, persistently high ethane levels, a product of lipid peroxidation that has been proposed as a biomarker of oxidative stress, may correlate with poor prognosis [35]. Cancer cells with persistent Nrf2 activation often develop Nrf2 addiction and show malignant phenotypes, leading to poor prognoses [36]. In the present study, although serum HO-1 levels tended to decrease 2 weeks after the start of treatment in both ARDS and AE-ILD patients, serum HO-1 levels remained persistently high. Furthermore, in patients treated with intravenous corticosteroids, no significant decrease of serum HO-1 levels was observed. Systemic corticosteroids are able to block nuclear translocation of nuclear factor- kB, the main pathway of inflammatory cytokine synthesis, through their interaction with the glucocorticoid receptor, however, the use of corticosteroid in ARDS is not recommended routinely [37, 38]. Also, retrospective data derived from AE-IPF patients treated with corticosteroid alone did not show any reduction in mortality rate over the short term [39]. Several evidence in the animal models suggest that corticosteroid exposure can cause to increase oxidative stress [40, 41]. Although the exact mechanism by which corticosteroid increase oxidative stress is not well known, several hypothesis have been reported that glucocorticoids could bind to mitochondrial glucocorticoid receptors and activate mitochondrial function to generate ROS or ROS is generated by the activation of the protein kinase C (PKC) β / p66^shc signaling pathway by glucocorticoid in the cell [40, 41]. Therefore, in patients with ARDS and AE-ILDs, the use of corticosteroids may be harmful from the point of view of oxidative stress increase, and the potential antioxidant treatment such as N-acetylcysteine needs to be examined in the future [42].

Composite approaches have been developed using peripheral blood biomarkers and physiological and radiographic measurements to provide more accurate prognostic information [26, 43, 44]. For example, APACHE II score is frequently used to measure disease severity in intensive care unit patients with ARDS [26]. The composite scoring system, which is based on serum LDH, Krebs von den Lungen-6, P/F ratio, and extent of abnormal high resolution computed tomography findings, is useful for predicting 3-month mortality in AE-IPF patients [43]. We previously demonstrated that the Charlson Comorbidity Index score, sex, and serum LDH are important for predicting 3-month mortality in AE-ILD patients [44]. In the present study, we found that composite parameters including serum HO-1, P/F ratio, sex, and age which was characterised as objective biomarkers had acceptable AUC for prediction of 3-month mortality in ARDS and AE-ILD patients. Our data suggest that composite parameters including serum HO-1 and other clinical parameters could predict the prognosis of ARDS and AE-ILDs more accurately than serum HO-1 alone. To verify the utility and reproducibility of this composite parameters, large-scale, multi-institutional prospective collaborative research is essential.

There are several limitations to the present study. First, the study enrolled only a small number of patients from a few institutions. Our findings need to be confirmed in a multi-centre, prospective study. Second, the various endogenous oxidative stress markers such as not only HO-1 but also SODs, catalase, GPx, and myeloperoxidase have been reported. It is necessary to evaluate which markers are the most reliable for predicting the prognosis of ARDS and AE-ILDs. Third, clinical diagnoses among ARDS and AE-ILD patients were heterogeneous. Actually, in both patients with ARDS and AE-ILDs, our data showed better prognosis than the previously reported cohort data. In the future, it is necessary to evaluate the clinical significance of serum HO-1 in patients with DAD histologically.

Oxidative stress assessed by serum HO-1 is persistently high among enrolled patients for 2 weeks after diagnosis despite treatment with corticosteroids. Also, composite parameters including serum HO-1, P/F ratio, sex, and age had acceptable AUCs for prediction of 3-month mortality in ARDS and AE-ILD patients.