Prohormones for prediction of adverse medical outcome in community-acquired pneumonia and lower respiratory tract infections

The assessment of disease severity and prediction of outcome in lower respiratory tract infections (LRTI) and, in particular, community-acquired pneumonia (CAP), is essential for the appropriate allocation of health care resources and for optimized treatment decisions. These include hospital or intensive care unit admission, the extent of diagnostic work-up, the choice and route of antimicrobial agents and the evaluation for early discharge. In an attempt to optimize and lower unnecessary hospital admission rates, professional organizations have developed prediction rules and propagated guidelines to stratify patients with CAP based on predicted risks for mortality [1‐3]. The pneumonia severity index (PSI) is a well validated scoring system in North America based on 19 prognostic parameters [4]. The CURB65 score, a more simplified assessment tool developed by the British Thoracic Society, focuses on only five predictors [5, 6]. This score is easier to calculate, but has a lower prognostic accuracy. Both risk scores were validated for the prediction of mortality only. Their ability to predict other important adverse disease outcomes including the need for ICU admission and complications due to the infection has not been established. Patients with PSI risk classes 1, 2 and 3 should be considered as candidates for outpatient treatment, but still a high percentage of subjects in these risk classes may experience unexpected complications indicating the need for improvement of these scores [7].

To improve the accuracy of clinical severity scores, prohormones have been proposed as biomarkers that provide more detailed and complementary information [8‐25]. Several biomarkers have been related to disease severity and outcome in LRTI and sepsis, including levels of the cardiac hormone atrial-natriuretic peptide (ANP) [13‐17], the stress- and volume-dependent antidiuretic hormone (ADH, vasopressin) [21‐25], the endothelium derived hormones endothelin-1 (ET-1) [11, 18‐20] and adrenomedullin (ADM) [8‐12], and procalcitonin (PCT) a specific marker of bacterial infections [26‐35].

The simultaneous measurement of a panel of prohormones each reflecting a specific pathophysiological pathway could enhance risk stratification in patients with CAP and other LRTI. We therefore validated the usefulness of five previously reported prohormones for predicting serious complications in patients with CAP and other LRTI enrolled in the multicenter ProHOSP study [31, 34].

In this large community-based sample of patients with CAP and other LRTI from a multicenter study [34], five prohormones from distinct biologic pathways were specific predictors for short term serious complications with moderate improvement of clinical risk scores. Thereby, this study validates a series of previous smaller trials demonstrating a clinical utility of prohormones for an optimized risk prediction in LRTI [8‐25].

Meaningful statistical assessment of the potential clinical utility of a biomarker is challenging. In addition to classical performance measures like two group comparisons and ROC curves, more clinically meaningful statistical approaches have been put forward [44, 48]. We performed several different statistical analyses to investigate the added value of biomarkers to clinical scores; more specifically, we assessed the addition of prohormones to PSI and CURB65 scores per se and to a multivariate regression model based on CURB65 covariates. We measured the prognostic performance of these models by several different quantities (AUC, Brier score and reclassification methods). Thereby, some prohormones, namely proADM, improved both clinical risk scores and were superior per se for serious complications prediction. The incorporation of a combination of biomarkers reflecting systemic inflammation, endothelial dysfunction, stress and cardiac function to the clinical risk scores improved their prognostic accuracy for prediction of short term complication rate and to a lesser extent mortality. When comparing the biomarkers to models based on raw clinical predictors included in the CURB65 score, the improvement was less extensive as shown by a relatively small increase in the AUC, but reclassification methods still established highly significant improvements of the model due to addition of the prohormones. Thus, as demonstrated previously for biomarkers in cardiovascular disease [44], prohormones significantly improve classification of patients into pre-defined risk groups.

The combination of clinical predictors and prognostic biomarkers has been suggested as a promising approach to optimize the prognostic certainty and thus the management of LRTI patients [49]. The information on the disease driven host-response mirrored in the circulating level of a biomarker may provide insights into the pathophysiology and prognosis of a disease process. As a quantifiable tool it facilitates risk stratification and monitoring of therapy as a surrogate outcome measure. In the future, a panel of biomarkers might help in delineating distinct populations of patients with discrete pathologies - a prerequisite to enable the targeted application of specific biologically rational therapies. In this trial, we validate the prognostic performance of five promising, rapidly measurable prohormones [8‐25]. ADM is one of the most potent vasodilating agents with immune modulating, metabolic and bactericidal properties [40, 50]. Atrial-natriuretic peptide, a member of the family of natriuretic peptides regulates a variety of physiological parameters [51]. In septic states, ANP levels may mirror both, the inflammatory cytokine response correlated with the severity of infection, as well as the presence of disease-relevant comorbidities, namely heart failure and renal dysfunction [41, 52]. Copeptin, stoichiometrically cleaved from the vasopressin precursor, has hemodynamic and osmoregulatory effects, and mirrors the individual stress response [53]. Endothelin-1 is an important vasoconstrictor and correlates with disease severity and short term outcome [11, 18‐20]. Unfortunately, these mature hormones are difficult to measure with high reliability because they are not stable at room temperature and have a rapid clearance from the circulation limiting their use in clinical routine. For this reason new sandwich immunoassays have been recently introduced that measure the more stable precursor fragments (proANP, Copeptin (proADH), proET-1 and proADM) [8, 25, 39‐41]. Unlike the mature peptides, these precursors can be detected for hours in the circulation. Because of the stoichiometric generation, these prohormones correlate with the release of the active peptides, a condition similar to that of insulin and C-peptide. Thus, these precursor peptides can be used to indirectly measure the release of the mature hormone under physiological and pathological conditions.

We focused our analysis on initial risk assessment and initial prohormone levels, but also explored the utility of repeated biomarker measurements. We used Cox proportional hazards regression models with time-dependent covariates (in addition to the baseline biomarker) and found that this model significantly improves upon the model with baseline covariates only. Moreover, we found that the baseline value of the biomarker is no longer significant after adjustment for the current biomarker value suggesting that the absolute value of the current biomarker value contains most information regarding future prognosis and the baseline value (as well as the change in the biomarker from baseline to follow-up) are less relevant. Further research is needed to derive clinical decision rules based on time-updated biomarker values.

The development of sepsis from a localized infection is a dynamic continuum and in the majority a sequelae of CAP [54]. The severity of a disease determines the consumption of costly and limited health-care resources. An early and adequate diagnosis and risk assessment is, thus, pivotal for optimized risk-adapted care of patients with severe infections. Scoring systems, such as the PSI, are well validated prognostic tools to determine mortality risks and rely mostly on age as the main driver of mortality [4]. However, calculation of the PSI in daily practice is time consuming which limits its dissemination and implementation in routine care [55]. In addition, the PSI is not a validated predictor for the clinically relevant rate of serious complications. Other clinical prediction rules have focused to predict eligibility for ICU admission. Multiple ICU prediction rules have been proposed including the Infectious Disease Society of America/American Thoracic Society (IDSA/ATS) criteria, the SMART-COP and scores based on the PIRO (Predisposition, insult/infection, response, and organ dysfunction) concept [56‐60].

We focused our analysis on a combined endpoint of serious complications, which included mortality, ICU admission and disease-specific complications. The strength of this approach is the clinical relevance for initial site-of-care decisions as patients experiencing one of these serious complications should arguably not be managed in the outpatient setting. However, heterogeneity of this combined endpoint makes prognostication more challenging as shown by the lower AUCs in ROC curves in this study when compared to mortality prediction alone. While age and comorbidities are major drivers of mortality, extent and severity of infection and organ failure may be the most important predictors for ICU admission. In this regard, combination of clinical parameters and biomarkers seems a promising approach.

As a limitation of this study, our findings may not unconditionally be applied to a general LRTI population because of selection bias in regard to exclusion criteria of the underlying randomized controlled trial. Since the PCT-guided group in the ProHOSP trial was non-inferior to the guidelines group with respect to the risk of adverse outcomes, treatment assignment was not considered any further in this analysis. Switzerland has previously been shown to have very low rates of ICU-acquired nosocomial infections and related mortality; thus country-specific differences may limit generalizability and external validation is warranted [61]. Further, covariates for risk score determination (but not biomarker values) were relatively frequently missing in our dataset. We used multiple imputations to deal with missing variables, but this methodology may not be correct on an individual patient basis and may explain some of the PSI and CURB65 miscalibration observed within this study. Although we provided web-based guidelines based on ATS criteria for ICU admission of patients, the final decision for ICU admission was left to the treating physician team. Other clinical risk scores have been suggested for prediction of ICU admission [56‐60]. However, as not all covariates were prospectively collected we did not compare biomarkers with these scores.

Previous studies have demonstrated the clinical and scientific impact of the biomarker PCT on the antibiotic management of LRTI [26‐32] but up to now, no study has investigated the clinical utility of a prognostic biomarker on the management of patients with LRTI. Because of its high prevalence and associated large need of health care resources, accurate prognostication and improved site-of-care decisions have high relevance for public health, both for primary and hospital care. The ultimate clinical utility of a biomarker is defined by the degree it improves clinical decision making and adds timely information beyond that of readily available information from clinical examination. Observational studies alone cannot provide such information, but may help to provide a rationale for future intervention studies. These are now warranted to show whether biomarker measurement improves risk prognostication and thus the clinical management of patients with LRTI.

Our data suggest that the addition of a panel of prohormones of distinct biological pathways, particularly proADM, to established CAP risk scores improves the risk stratification for serious complications and mortality in CAP patients. If these results are validated, the incorporation of one or several biomarkers in clinical practice for the prediction of adverse prognosis could help to optimize admission decisions in patients with LRTI and CAP upon hospital admission and in the emergency department.