Use of Laboratory Tests and Their Prognostic Value in Patients with Stable Chronic Obstructive Pulmonary Disease

Abstract

Chronic obstructive pulmonary disease (COPD) is not merely a lung disease as the name indicates. Patients with COPD experience associated complications with systemic inflammation, heart strain, muscle wasting, poor functional performance, and psychological issues. An assessment of these patients based only on lung function tests or functional capacity would be inadequate. Many studies have reported the significance and prognostic value of several laboratory tests. Troponin, C–reactive protein (CRP), hemoglobin, and carbon dioxide are older tests but compared with the newly developed tests, they are relatively inexpensive to measure and widely available. This article will review laboratory tests used for COPD and discuss their prognostic value. The laboratory tests that can identify high–risk patients will be discussed. We will explore the role of these tests in clinical practice.

Introduction

Chronic obstructive pulmonary disease (COPD) is one of the leading causes of morbidity and mortality in the United States.¹ With over 10 million cases of COPD diagnosed in 2018, COPD places a heavy burden on the health system.² Patients with COPD experience episodes of worsening symptoms and a reduction in lung function identified as exacerbations. Severe exacerbations will require hospitalization. The disease has a variable progression over time, and the major subtypes (emphysema, chronic bronchitis, and chronic obstructive asthma) have interrelationships between the closely related disorders that cause airflow limitation, this provides a foundation for understanding the spectrum of patient presentations.

The pulmonary function tests (PFTs), which usually include spirometry, measurement of lung volumes and diffusing capacity of carbon monoxide in addition to arterial blood gas analysis at rest have historically been used to measure the physiologic impairment in the lungs to support the clinical diagnosis of COPD. However, spirometry is only a single factor in determining the severity of the disease as nearly 10% of patients with clinically diagnosed COPD with emphysema findings on chest computed tomography (CT) scan had normal PFTs.³ The Global Initiative for Chronic Obstructive Lung Disease (GOLD) criteria, body mass index (BMI), the severity of airflow obstruction, dyspnea, and the exercise capacity index (BODE index) have also been used to identify high–risk patients and to guide treatment protocols.⁴ While these scores predict to some extent high–risk patients with COPD, important factors such as systemic inflammation, associated morbidities, and heart disease are not included in the existing scoring criteria.

Laboratory testing has been developed as simple reliable tests to identify patients with COPD at higher risk for decline, recurrent exacerbations, and death.⁵ Studying these tests may allow for phenotype recognition, better COPD stratification, and earlier interventions and tailored management.⁶ Multiple studies have been performed evaluating the association of troponin, C–reactive protein (CRP), carbon dioxide (CO₂), and hemoglobin levels in patients with stable COPD with all–cause mortality. This review explores the prognostic value and clinical utility of these tests.

High–Sensitivity Troponin and COPD

Troponin is a cardiac muscle marker of injury and is one of the cornerstones for acute coronary syndrome diagnosis. New troponin assays including high–sensitivity troponin (hs–TN) and troponin I are more specific for cardiac muscle injury. Elevated high–sensitivity troponin has been observed in non–cardiac primary conditions including renal failure, pulmonary embolism, and sepsis. The rise in troponin may indicate myocardial damage rather than infarction. ⁷ The suggested pathophysiology may be related to increased myocardial stress. Circulatory overload, hyperdynamic circulation and increased pressures may lead to cardiac muscle overstretching and troponin leak. Pulmonary vascular resistance elevation in COPD may lead to increased right ventricular afterload. Even a slight elevation in pulmonary vascular resistance may result in right ventricular impairment and increased wall thickness in patients with COPD.⁸ Increased systemic inflammation and a prothrombotic state may also contribute to increased pulmonary vascular resistance. COPD is a strong risk factor for cardiovascular events, independent of smoking, this is specifically significant in COPD chronic bronchitis patients. Having symptoms of chronic bronchitis alone increases the risk of coronary deaths by 50%.

In a previous study, elevated high–sensitivity troponin during COPD exacerbations predicted short–term mortality.⁹ Establishing the same relationship in stable patients would be more challenging. A review of the literature identified three studies discussing the troponin level in patients with stable COPD. The first study was published in 2016 by Neukamm and coworkers.¹⁰ Two–hundred seventy–five patients were selected and followed for a mean duration of 2.8 years. More than half of the patients (57%) were GOLD class III/IV. Patients with elevated troponin after adjusting for factors including age, kidney function, the presence of Q waves on electrocardiography (ECG), left ventricular hypertrophy, GOLD class, six–minute walk distance, and modified medical research counsel for dyspnea score (mMRC) had higher mortality rates. The hazard ratios (HR) for death were 1.7 (95% confidence interval [CI] 0.8–3.9) and 2.9 (95% CI 1.2–7.2) among patients with hs–TN concentrations of 5.0–13.9 and ≥ 14 ng/L, indicating higher mortality with higher troponin levels. The study excluded coronary artery disease (CAD) by the absence of Q waves on ECG, and this may have led to false negative results.¹¹

Adamson and colleagues^12,13 studied 1,599 subjects, a subset of patients with COPD from the Study to Understand Mortality and MorbidITy (SUMMIT) trial with measured serum troponin I levels. These patients were either diagnosed with CAD per definition or at risk and were followed up for median of 18 months. The higher troponin I groups (≥ 4.8 ng/L and ≥ 7.7 ng/L) were at greater risk of cardiovascular events and cardiovascular death with no change in the COPD exacerbation rate after adjustment for risk factors. Inhaled corticosteroids, long–acting beta–2 agonists, or combination therapy did not affect troponin levels.

The latest published input from the COPD and Systemic Consequences–Comorbidities Network (COSYCONET) cohort study by Waschki and colleagues¹⁴ analyzed 2,085 patients with stable COPD (median follow up was 36 months). The study included patients with GOLD stages I–IV. Ninety–six percent of the patients had detectable high sensitivity troponin I levels (hs–TNI). Only 1.8% of the patients had levels greater the reference range of 27 ng/L (cut–off to diagnose myocardial infarction). Patients with hs–TNI levels ≥ 6 ng/L had increased mortality compared with the lower troponin groups irrespective of the cardiovascular risk factors, but they did not have more COPD exacerbations. Furthermore, patients with a high BODE index and hs–TNI > 6 ng/L had a 2.6–fold higher risk of all–cause mortality compared with patients with low troponin.

These studies have shown that patients with COPD with higher baseline troponin levels are at higher risk for mortality, including cardiovascular death. The cut–off value for troponin elevation associated with increased risk was 4.8 ng/L. A proportional increases in all–cause and cardiovascular mortality was also observed. One of the challenges in these studies has been stratifying patients with a history of CAD, with risk factors (including age, sex, smoking, hypertension, and dyslipidemia), and with electrocardiogram or echocardiography without relying on accurate diagnostic testing including myocardial perfusion imaging. This approach may have underestimated CAD prevalence in study groups.

Adamson and colleagues¹³ demonstrated that inhaled bronchodilator therapy did not affect troponin levels. Only 60% of the study patients were on both antiplatelet and statin therapy, and up to 25% of the patients in the study by Waschki et al.¹⁴ were on statins. In one cohort study, successful treatment with statins showed a reduction in lung–related and all–cause mortality in patients with COPD.¹⁵ Daily aspirin use was also associated with a reduced rate of mild and moderate COPD exacerbations, less dyspnea, and better quality of life in a recent observational cohort study.¹⁶ Troponin levels decreased with statin therapy in selected patients with elevated low–density lipoprotein (LDL) but without a history of cardiovascular disease.¹⁷ More studies are needed to evaluate this correlation and the value of following elevated troponin levels in patients with COPD with inhaled bronchodilator, statin, or aspirin therapy.

In conclusion, testing for troponin can identify patients with COPD at high risk for exacerbations and mortality. Starting aspirin or statin therapy in this patient subgroup may be justified to reduce exacerbation risk and mortality.

Anemia and COPD

Anemia is generally defined as a hemoglobin level < 13 g/dL in men and < 12 g/dL in women. However, in patients with COPD, it is difficult to set a clinically meaningful cut–off point because COPD is generally considered a cause of polycythemia in the setting of chronic hypoxemia.¹⁸ The most common type of anemia seen in patients with COPD is anemia of chronic disease (ACD). Different studies have found the prevalence of anemia in patients with COPD to be around 21%–23%, making the prevalence of anemia in COPD comparable to the prevalence of anemia in other chronic diseases well known to be associated with anemia such as malignancy and chronic kidney disease.¹⁹ There are different hypothetical mechanisms to explain the development of ACD in patients with COPD. Specifically, the increased levels of cytokines and leukotrienes suppress the erythropoietin response and shorten erythrocyte life, leading to anemia.²⁰ Moreover, these cytokines and interferons increase ferritin levels, which trap iron and decrease its availability for hematopoiesis.²¹ Anemia is known as a significant predictor of mortality, the hospitalization rate, and the duration of hospital stay in patients with COPD on long –term oxygen therapy (LTOT).²² However, it is less clear how much anemia could affect survival and morbidity in patients with stable COPD. Therefore, we conducted a literature review and found three studies in which the authors discussed the significance of anemia in patients with stable COPD.

The first study was published in 2006 by Halpern and coworkers.²³ They retrospectively analyzed U.S. Medicare claims data for patients with COPD in the period between 1997 to 2001. They enrolled 132,424 patients with COPD in the study. They found that the prevalence of anemia among patients with COPD was 21%. The primary endpoint of the study was the costs to the Medicare system for all medical care. Patients with COPD and anemia cost 2.2 times more than patients with COPD without anemia for all Medicare payments ($1,466 vs. $649, p < 0.001). They also found the mortality rate among patients with COPD and anemia (262 per 1,000 person–years of follow–up) was approximately two–fold the mortality rate in patients with COPD without anemia (133 deaths per 1,000 person–years of follow–up; p < 0.001). In this study there were no specific measures of disease severity available. The next study was done by Boutou and colleagues in 2013. They included 294 outpatients with stable COPD; anemia was identified in 46 patients (15.6%). The authors reported a significant difference in the median survival and survival rates between outpatients with COPD regardless of whether they also had anemia. In patients with anemia, the median survival was 68.7 months compared with 79.8 months in patients without anemia (p = 0.035). The survival rates at three and five years were 0.65 and 0.52 for the patients with anemia and 0.8 and 0.68 for the patients without anemia, respectively. Anemia but not the hemoglobin concentration was found to be an independent predictor of mortality in the total population (HR 1.87, 95% CI 1.06–3.29).

Oh et al.²⁴ evaluated the prognostic role of anemia on the clinical course of outpatients with stable COPD in Korea in 2017. They included 407 patients with COPD enrolled in the Korean obstructive lung disease (KOLD) cohort at 16 hospitals in Korea recruited over nine years. They found anemia to be a significant factor that negatively impacts the survival of patients with COPD: The patients with COPD without anemia (mean survival period = 8.88 ± 0.15 years) survived longer than patients with COPD and anemia (mean survival period = 6.77 ± 0.84 years; p < 0.001). They also found that elderly patients, female patients, and patients with a lower BMI and a lower albumin level were more likely to have a significantly lower serum hemoglobin level (all p < 0.05). One major limitation of this study was the male–dominant sample (male = 97.3%) as smoking is far more common in Korean males than Korean females.

The most recently published cohort study was in 2018 by Park and coworkers.²⁵ They identified 7,114 patients with COPD in the period from 2003 to 2013 by using the National Health Insurance Service–Health Screening (NHIS) cohort in Korea. Of these patients, 469 (6.6%) were anemic. During the study period, the mortality rate in patients with COPD and anemia was 46.5% compared with 32.1% in patients with COPD without anemia (p < 0.001). In addition, anemia was identified as one of the determinants significantly associated with overall mortality in multivariate analysis. The HR of anemia for mortality was 1.31 (95% CI 1.11–1.54). The decrease in the hemoglobin level was closely related to the decrease in survival rate among patients with COPD and anemia. However, the authors could not exclude patients with comorbidities associated with anemia, which could be a potential confounder.

In summary, all these studies have clearly shown that anemia in patients with stable COPD is a major clinical concern because this has been associated with significantly higher mortality rates among patients with COPD. However, there has been no clear agreement about the role of the hemoglobin concentration in predicting mortality and survival in patients with COPD with or without anemia.

There appears to be several benefits to red blood cell transfusion in patients with COPD and chronic anemia. Schönhofer et al.²⁶ found that red blood cell transfusion in patients with COPD and anemia leads to a significant reduction in both the minute ventilation and the work of breathing compared with patients with anemia but not COPD. In addition, in patients with COPD and anemia who have difficulty in weaning from the ventilator, blood transfusion may lead to successful weaning.²⁷ LTOT showed a progressive but non–significant improvement in hemoglobin concentration in patients with COPD and chronic anemia.²⁸ Additional studies are needed to determine the impact of red blood cell transfusion or LTOT on the survival of patients with stable COPD and chronic anemia.

C–reactive Protein and COPD

C–reactive protein (CRP) is a non–specific acute phase protein and inflammatory biomarker that is elevated in conditions associated with tissue damage, inflammation, infection, and malignancy.²⁹ Given that COPD is a chronic inflammatory disease of the lungs that involves different inflammatory cells and mediators, several inflammatory markers including CRP have been found to be increased in the systemic circulation of patients with the disease.³⁰ The inflammatory process in COPD is not limited to the lungs; it also involves systemic inflammation manifested by several systemic complications common in patients with COPD, such as osteoporosis and cachexia.³¹ Elevated CRP levels in the presence of a major exacerbation symptom is useful in confirming the COPD exacerbation diagnosis.³² However, CRP levels were not helpful in predicting survival and mortality in the acute COPD exacerbation event.³³ Determining the role of CRP levels in predicting the prognosis in patients with stable COPD is less clear. Using a high–quality literature review, we identified several studies exploring the prognostic role of CRP in patients with stable COPD.

The first study was published in 2006 by Man and colleagues;³⁴ they measured CRP levels in 4,803 cigarette–smoking patients aged 35–60 years with mild–to–moderate airflow obstruction on spirometry using a highly sensitive enzyme linked immunosorbent assay. All–cause mortality increased significantly with increased CRP levels. In addition, the risk of CAD–and cancer–related mortality was significantly related to CRP levels. However, these results were limited to patients with mild–to–moderate disease but not to patients with more advanced disease. In 2007, Dahl et al.³⁵ selected 1,302 patients with COPD from the ongoing Copenhagen City Heart Study; they followed them for eight years and recorded COPD admissions and deaths. They found that the cumulative incidence of COPD hospitalizations and death was significantly higher in patients with COPD with high CRP values (> 3 mg/L) independent of lung function and smoking.

De Torres et al.³⁶ evaluated the association between CRP levels and survival in patients with moderate to very severe COPD. They followed 218 patients with stable COPD for a median of 36 months after measuring baseline CRP levels. Their published results revealed that a baseline CRP level was not significantly associated with survival status in patients with COPD with moderate to very severe disease. Although this study was a clinical rather than epidemiological one, it was limited by the relatively small population size. Liu and colleagues³⁷ investigated the role of CRP levels and BODE scores in predicting mortality in patients with COPD. Their prospective study included 125 patients with stable COPD. They found that both serum CRP concentrations and BODE scores were independent prognostic variables of mortality in the studied population. However, the combination of both precisely predicted the survival of patients with stable COPD. In 2016, Kim et al.³⁸ conducted a retrospective study to evaluate the factors associated with mortality in patients with COPD at high risk for acute COPD exacerbation. They identified 61 patients at high risk for acute COPD exacerbation out of 606 enrolled in the study. High–sensitivity CRP was one of several factors (including neutrophil percent, lymphocyte percent, and the number of comorbidities) that significantly predicted mortality in patients with COPD at high risk for acute COPD exacerbation (defined as hospitalization at least once in a year).

In 2017, Xiong and colleagues³⁹ enrolled 409 patients with stable COPD in a prospective, case–control study. They followed the patients every three months during the 24–month follow–up period with routine blood tests, pulmonary function tests, CRP, Erythrocyte Sedimentation Rate (ESR), procalcitonin, and the BODE index to evaluate the prognostic role of these variables. CRP was positively correlated with mortality. However, the relevance of CRP to mortality was inferior compared with other variables measured in routine blood tests. The patients with COPD included in this study were those with moderate to very severe disease, so the results may not apply to patients with mild disease. In 2018, Mendy et al.⁴⁰ analyzed data regarding 431 patients with COPD from the National Health and Nutrition Examination Survey (NHANES) conducted from 2007 to 2010 and the corresponding Mortality–Linked File; they aimed to determine the blood biomarkers that could predict mortality in patients with COPD. They found the CRP level to be a significant determinant of mortality in the studied population. Moreover, the role of the CRP level in predicting mortality could significantly improve with the addition of eosinophil and neutrophil counts.

Celli and colleagues⁴¹ carried out a sub–study of the randomized SUMMIT trial in 2019. They studied the significance of several serum biomarkers in predicting hospitalizations, exacerbations, FEV1 decline, and mortality in patients with moderate COPD. The study showed CRP as a significant factor associated with increased risk of death in patients with COPD. However, CRP was not associated with predicting hospitalizations, exacerbations, or FEV1 decline.

Based on the above–mentioned studies, there are conflicting results regarding the significance of the CRP level in predicting mortality in patients with stable COPD, especially regarding changes in COPD severity. However, the use of the CRP level in combination with other clinical variables and blood biomarkers could help to predict mortality in patients with stable COPD. Therefore, we recommend against the individual use of the CRP level in predicting mortality in patients with stable COPD, but we recommend combining it with other variables to improve the predictive value.

Hypercapnia and COPD

Elevated arterial CO₂ levels (hypercapnia), defined as an arterial partial pressure of CO₂ (PaCO₂) ≥ 45 mmHg in patients with COPD, is observed in stable patients and during exacerbations. According to one study, 28.4% of stable GOLD three and four patients had hypercapnia.⁴² Emphysema, airflow obstruction, and respiratory muscle weakness may lead to increased alveolar dead space and hypercapnia.⁴³ Emphysema patients usually are not hypercapnic until near death and are not frequently admitted for exacerbations of respiratory failure. On the other hand, chronic bronchitis patients usually retain CO2 and become hypoxemic, either acutely or chronically, and this leads to the complications of pulmonary artery hypertension and cor pulmonale. A distinct subset of patients with COPD had temporary, reversible hypercapnia following exacerbations.⁴⁴ Many questions followed. Do patients with COPD and hypercapnia have a worse prognosis compared with normocapnic patients? Furthermore, does hypercapnia represent the natural progression of COPD or a different phenotype? We accordingly conducted a literature review.

Costello and colleagues⁴⁴ studied 85 patients with COPD over five years. The analysis included arterial blood gas at admission and discharge, the number of exacerbations, and mortality. The patients were divided into three groups according to admission and discharge PaCO₂. Interestingly, the reversible hypercapnia group had similar five–year mortality compared with the normocapnia group. Only 24% of patients with reversible hypercapnia developed chronic hypercapnia during follow–up. Patients with irreversible hypercapnia had lower follow–up partial pressure of oxygen (PaO₂) and higher PaCO₂ values compared with the other groups with higher long–term mortality.

Dave and colleagues⁴⁵ studied 1,224 patients with COPD retrospectively, of which 587 had the alpha–1 antitrypsin deficiency (AATD) diagnosis. During a 10–year follow–up, patients with non–AATD COPD and stable hypercapnia did not show an increased mortality risk compared with normocapnic patients. On the other hand, patients with worsening stable–state hypercapnia had significantly increased mortality (HR 1.65, 95% CI 1.06–2.56, p = 0.03). Yang et al.⁴⁶ implemented a prospective cohort study to compare the survival of patients with COPD and chronic hypercapnia with those with COPD and normal PaCO₂. Between 1993 and 2006, they enrolled 275 patients with stable COPD, including 98 patients with normocapnia and 177 patients with chronic hypercapnia. They found that patients with normal PaCO₂ survived longer than those with hypercapnia (6.5 vs. 5.0 years, p = 0.016).

In 1998, Aida and colleagues⁴⁷ studied the prognostic value of hypercapnia in patients with chronic hypoxemic respiratory failure requiring LTOT, as hypercapnia was generally considered an ominous sign in chronic lung diseases at that time. Their study comprised patients in Japan from 1985–1993: 3028 patients with the sequela of pulmonary tuberculosis on LTOT and 4,552 patients with COPD on LTOT. There was an increase of ≥ 5 mmHg in PaCO₂ over the six to 18–month follow–up period that was associated significantly with a poorer prognosis compared with those with stable PaCO₂. However, there were no differences in mortality between patients with hypercapnia and normocapnia with the sequela of pulmonary tuberculosis on LTOT.

Ahmadi et al.⁴⁸ performed a national prospective study in Sweden between 2005 and 2009 to evaluate the prognostic role of PaCO₂ in patients with oxygen–dependent COPD. They found that PaCO₂ was an independent predictor of mortality (p < 0.001). It exhibited a U–shaped pattern in association with mortality: Mortality decreased with hypocapnia (PaCO₂ < 5.0 kPa) and hypercapnia (PaCO₂ > 7 kPa). In 2018, Brat et al.⁴⁹ implemented a non–interventional multicenter study on patients with COPD using the Czech Multicenter Research Database to investigate the respiratory parameters associated with increased mortality in GOLD category B patients with COPD, which represents the largest portion of patients with COPD in real–life studies. The results showed that hypercapnia (PaCO₂ > 7.0 kPa) was associated significantly with mortality in these patients (p < 0.001).

In conclusion, all of the above–mentioned studies agree that the elevation of arterial carbon dioxide is a significant predictor of mortality in patients with stable COPD. However, there is variable cut–off for PaCO₂ values that are associated with higher mortality. Therefore, we recommend using hypercapnia as a valid predictor of mortality in patients with stable COPD, but additional studies are needed to determine the PaCO₂ value that significantly impacts mortality rates in patients with stable COPD.

Conclusion

There have been a myriad of studies in which researchers have explored the significance of several blood laboratory tests in predicting mortality in patients with stable COPD. A high troponin level is a reliable predictor of mortality in patients with stable COPD, and we recommend considering starting these patients on aspirin or statin therapy to reduce mortality. Although anemia is significantly associated with higher mortality among patients with stable COPD, we recommend against the sole use of the hemoglobin concentration in predicting mortality in these patients. In addition, studies have shown that the use of the CRP level alone to assess the risk of mortality in patients with stable COPD is not significant. Hypercapnia is clearly associated with higher mortality rates among patients with stable COPD, but no clear PaCO₂ cut–off has been found to predict mortality in this population. Our literature review supports using CRP, troponin, anemia, and PaCO₂ to predict mortality in patients with stable COPD. However, we highly recommend pursuing the well–established process in the diagnosis and follow up utilizing the complete and focused history and physical examination, complete pulmonary function test (PFT) including routine spirometry, determination of lung volumes and measurement of diffusing capacity of carbon monoxide in addition to the use of noninvasive oximetry and blood gas analysis for follow ups. Currently, there is no indication in the guidelines to employ measurements of laboratory tests in a steady state COPD patient.