Introduction

Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive, fibrotic interstitial lung disease with a poor prognosis which is frequently diagnosed in older ages (1). Patients with IPF may experience sudden onset respiratory worsening, most of which is idiopathic and defined as acute exacerbation of IPF (IPF-AE). It is usually challenging to differentiate IPF-AE from heart failure, bilateral pneumonia (PNA), or pulmonary embolism (2). Cough, increased expectoration, dyspnea, hypoxemia, and newly formed parenchymal infiltrations on chest imaging are non-specific for IPF-AE; moreover, these observations can also be seen in bacterial PNA (3). Procalcitonin (PCT), the precursor of calcitonin, is produced by neuroendocrine cells when confronted with microbial toxins and inflammatory mediators. It is a highly specific biomarker for severe bacterial infections in suspected cases of sepsis (4,5). The objective of the study was to investigate the diagnostic value of PCT in the differential diagnosis of acute exacerbation and pneumonia in IPF patients.

Material and Methods

The data of IPF inpatients were retrospectively analyzed at a tertiary referral hospital between January 2015 and June 2018. Age, gender, presenting symptoms, duration of hospitalization, vital signs, biochemical markers [including C-reactive protein (CRP) and PCT levels], and radiological findings were recorded. Patients with incomplete data such as absence of thorax computerized tomography/high-resolution computerized tomography (CT/HRCT) and those who were hospitalized for other conditions such as pulmonary embolism, decompensated heart failure, pneumothorax, and acute renal failure were excluded from the study. Since the study was retrospective, patient consent was not obtained.

The diagnosis of IPF-AE was made according to the criteria suggested by the ATS/ERS/JRS/ALAT committee and the International Working Group Report by Collard et al. (6,7). The presence of fever (>37.5 °C), pneumonic consolidation in thoracic imaging, and positive sputum and/or blood culture were suggestive of PNA. On the other hand, if the patients had progressive dyspnea in the previous month, recently developed bilateral ground-glass opacities, and/or consolidations with a usual interstitial PNA pattern without heart failure or fluid retention, the diagnosis was IPF-AE (7).

CRP and PCT levels were measured using an immunoturbidimetric assay via Beckman Coulter AU Analyzer (US).

Study approval was taken from the Karabük University Ethics Committee on 03.01.2018 with verdict number 1/2.

Power Analysis

In the a priori power (G*Power Version 3.0) analysis in independent groups based on the data (effect size d=0.926, a=0.05, power=0.80) of a study by Ding et al. (8) on procalcitonin-guided antibiotic use in IPF-AE, it was calculated that at least 16 independent control and 16 experimental subjects were required for a study investigating procalcitonin testing. Each group of this independent study consisted of 22 subjects.

Statistical Analysis

The minimum (min), maximum (max), and mean ± standard deviation (SD) was calculated for all variables. The SPSS for Mac 20.0 package program (SPSS Inc, Chicago, IL, USA) was used to analyze the data. Frequency and percentage for discrete data and mean ± SD for continuous data were used in descriptive statistics. Normally distributed continuous data were analyzed by the Kolmogorov-Smirnov test while all nominal data were analyzed by the chi-square test. Non-parametric distributions were compared with the Mann-Whitney U test, and mean values ​​with parametric distribution between groups were analyzed with Student’s t-test. Receiver operating characteristic (ROC) analysis was used to determine the optimal cut-off values of the PCT and CRP to predict PNA. A p-value <0.05 was considered as statistically significant.

Results

Fifty-seven IPF patients were evaluated; 13 of them were excluded due to pulmonary embolism (n=1), decompensated heart failure (n=3), absence of HRCT scans (n=3), and failure to obtain laboratory work-up (n=6). A total of 44 eligible patients were randomized into two groups at a 1:1 ratio. The mean age was 62.9±7.6 years in the pneumonia in patients with IPF (IPF-PNA) group and 67.3±6.6 years in the IPF-AE, with a statistically significant difference (p=0.049). The number of male patients was significantly higher in the IPF-PNA group (p=0.007) than in the IPF-AE.

The PCT levels were significantly higher in the IPF-PNA group than in the IPF-AE [0.253 ng/mL (0.10-12) vs. 0.05 ng/mL (0-0.15), p<0.001]. Additionally, the CRP levels were significantly higher in the IPF-PNA group than in the IPF-AE [129 mg/L (33.6-413) vs. 37.9 mg/L (4-152), p<0.001] (Figure 1). Also, the hospital stay was significantly higher in the IPF-PNA group (15.18±8.3 days vs. 8.54±2.5 days, p=0.001). Demographical data and laboratory results are summarized in Table 1.

ROC analysis was performed for identification of the PCT and CRP cut-off values to predict PNA in worsening IPF (Figure 2). When 0.1065 ng/mL was taken as the cut-off for PCT, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were 0.955, 0.773, 0.808, and 0.944, respectively. When 35 mg/L was taken as the cut-off for CRP, the results were 0.955, 0.500, 0.656, and 0.917, respectively (ROC graphic, Table 2).

Regarding sputum analyses, the causative pathogens identified in the IPF-PNA group were Klebsiella pneumoniae (n=3), Pseudomonas aeruginosa (n=1), and Escherichia coli (n=1).

Discussion

In our study, the PCT and CRP levels were significantly higher in the IPF-PNA group when compared with the IPF-AE group. Also, the IPF-PNA group had a longer hospital stay. When the cut-off values of the CRP and PCT levels were specified, PCT was found to be a more sensitive biomarker for differentiating between IPF-PNA and IPF-AE.

The fact that both IPF-PNA and IPF-AE groups have similar clinical, laboratory, and radiologic findings makes it difficult to decide on antibiotic use. The differentiation between patients who require antibiotic treatment and those who do not remains a challenge, as reliable clinical and/or microbiological parameters that are easily accessible during sampling are still inadequate. At this stage, the delayed availability of culture results, low sensitivities of some tests, and the contamination risk, especially in sputum cultures, pose challenges in making decisions regarding antibiotic usage. Inflammatory markers like CRP or white blood cells (WBC) are faster and more easily accessible indicators for indicating infection in daily practice, yet lack specificity for bacterial infections. This scenario has prompted the exploration of new biomarkers that are more discriminative. One of these biomarkers mentioned is PCT.

Procalcitonin is produced by the medullary C glands of the thyroid and neuroendocrine cells of the lung in the healthy population (9,10). It is a member of the CAPA protein family and is a 116 amino acid polypeptide (11). It rapidly converts into the 32 amino acid hormone calcitonin, which plays an important role in calcium hemostasis, and therefore presents in very small amounts in circulation (<0.01 ng/mL) (12). Depending on the type and severity of infection, varying amounts of PCT are released into the circulation from other parenchymal tissues and differentiated cell types (13,14). PCT serum levels are increased in bacterial, fungal, and parasitic infections, but do not rise or slightly increase during viral infections and non-infectious inflammatory reactions (14). It is also critical to differentiate between infectious and non-infectious exacerbations in IPF patients because bacterial PNA stands as a well-known leading cause of mortality in patients with IPF (7,15). Additionally, the treatments for these two conditions are very different from each other; while antibiotherapy is required for infectious exacerbations, systemic glucocorticoids are suggested for non-infectious exacerbations (16). Ding et al. (8) showed that PCT-guided antibiotic use in IPF-AE reduced antibiotic exposure and duration of antibiotic treatment, indicating that PCT is a useful biomarker in the management of these patients. In the light of this information, PCT, which gives quick result and provides practical approach, was preferred in our study design. The outcomes of our study also demonstrate that PCT can be employed as a biomarker to differentiate between IPF-AE and IPF-PNA.

Procalcitonin emerges as a more specific biomarker for the identification of bacterial infections. PCT is produced in response to endotoxins or bacterial infections, triggered by cytokines like interleukin (IL)-1b, tumor necrosis factor (TNF)-a, and IL-6 (17). The elevation of PCT levels is attenuated by interferon (INF)-ɣ released in viral infections, rendering it lower in viral infections and more specific for bacterial infections, allowing for the differentiation between bacterial and viral illnesses. A meta-analysis performed by Simon et al. (18) showed that PCT is more sensitive and specific than CRP for the differential diagnosis of viral and bacterial infection. It also helps in planning to start or cease antibiotic treatments (19). Because unnecessary and prolonged use of antibiotics increases antibiotic resistance and the risk of fungal infections (20,21). The widespread use of antibiotics is also a public health issue due to increased healthcare costs (22). However, avoiding antibiotics when PNA is suspected, is as dangerous as the overuse of antibiotics (23,24). The role of PCT has been researched in many different infectious diseases. ProHOSP study has shown that using PCT may reduce antibiotic use in emergency departments in infections of lower respiratory tract such as community-acquired pneumonia (CAP), acute bronchitis, and acute exacerbations of chronic obstructive pulmonary disease (25). Jain et al. (26) have illustrated that PCT may be helpful in antibiotic decision-making for CAP. Moisa et al. (27) evaluated hematological biomarkers in distinguishing viral and bacterial sepsis and found that PCT was the best discriminative parameter among all measured parameters. Another retrospective study showed that PCT and CRP levels in non-neutropenic lung cancer patients with tumor fever to differentiate localized bacterial infection from bacteremia, and compared with CRP, PCT was to be more accurate in differentiating tumor fever from bacterial infections (28). Wang et al. (29) found that PCT was a more useful biomarker than CRP to differentiate acute radiation pneumonitis from bacterial PNA in patients with lung cancer. In a study examining WBC, CRP, and PCT levels of patients with cryptogenic organizing pneumonia (COP), another interstitial lung disease, were compared with CAP patients; PCT was shown to be a more useful biomarker in differentiating COP from CAP (30). In our study, which aimed to differentiate pneumonia from exacerbation according to inflammatory biomarkers measured on the first day of hospitalization, PCT was found to be more sensitive than CRP in predicting exacerbation in patients with IPF, which is consistent with the literature.

A systematic review assessing PCT’s role as a predictor of sepsis has indicated that surpassing threshold levels in PCT could be a superior indicator compared to CRP. Nonetheless, it’s noted that the cut-off values for patients can vary due to diverse surgical procedures, medication use, and immune system conditions (31). Ding et al. (8) have evaluated the feasibility of PCT in navigating antibiotic use in IPF-AE. One group of patients with IPF-AE was given antibiotics according to daily routine practice, whereas the other group was treated with the guidance of PCT levels. This study concluded that the use of 0.25 ng/mL as a threshold level for PCT decreased the length of antibiotic treatment and the number of patients receiving antibiotics (8). Sim
et al. (3) have retrospectively evaluated 21 cases with interstitial lung disease who had recently increased dyspnea and worsened radiological findings. In nine of 21 patients, the cause of deterioration was bacterial PNA. They concluded that a cut-off value of 0.1 ng/mL for PCT may clearly indicate IPF-AE and facilitate clinical decision-making. At this threshold, the sensitivity, specificity, and NPV of serum PCT level are 88.9%, 100.0%, and 92.3%, respectively (3). In our study, when 0.1065 ng/mL was taken as the cut-off for PCT, the results were 95.5%, 77.3%, and 94.4%, respectively. Our findings are consistent with the previous study. Further large-scale studies are still needed to identify the exact cut-off values in more homogenous patient populations.

IPF is more common in older adult men and Kärkkäinen et al. (32) have reported that this is a risk factor for mortality in IPF-PNA patients. As far as we know, there is no data regarding the higher frequency of IPF-PNA cases in males compared to IPF-AE cases. A statistically significant predominance of male-sex in the IPF-PNA group was seen in this study.

Study Limitations

The limitations of the study can be described as the retrospective structure and the small sample size. Given the retrospective nature of our study, decisions regarding antibiotic initiation and cessation were not solely based on PCT. In order to comprehensively understand the role of this biomarker in guiding the clinical decision-making process, prospective studies are imperative. Additionally, it’s noteworthy that most patients received PNA diagnoses through multidisciplinary evaluations and clinical and radiological assessments. Although this approach proves satisfactory in diagnosis, the absence of comprehensive microbiological confirmation constitutes another limitation of our study.

Conclusion

PCT and CRP levels were higher in the IPF-PNA than in the IPF-AE. Compared with CRP, PCT was found to be a more sensitive biomarker in differentiating IPF-PNA from IPF-AE. Future studies with larger study populations are essential on this issue.

Information: The article was presented as an oral presentation at TÜSAD 40. Ulusal Kongresi 13-16 October 2018, Kemer, Antalya, SS-129.

Ethics

Ethics Committee Approval: Study approval was taken from the Karabük University Ethics Committee on 03.01.2018 with verdict number 1/2.

Informed Consent: Since the study was retrospective, patient consent was not obtained.

Peer-review: Externally peer-reviewed.

Authorship Contributions

Surgical and Medical Practices: B.D., E.T., B.Z.Y., D.D., E.G.U.C., D.T., H.Ç., E.Ç., Concept: B.D., E.T., B.Z.Y., D.D., D.T., F.E.G., H.Ç., E.Ç., Design: B.D., E.T., B.Z.Y., D.D., E.G.U.C., F.E.G., H.Ç., E.Ç., Data Collection or Processing: B.D., E.G.U.C., D.T., F.E.G., Analysis or Interpretation: B.D., E.T., B.Z.Y., D.D., E.G.U.C., D.T., F.E.G., E.Ç., Literature Search: B.D., E.T., E.G.U.C., D.T., F.E.G., H.Ç., Writing: B.D., E.T., B.Z.Y., D.D., H.Ç., E.Ç.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: The authors declared that this study received no financial support.