a b s t r a c t

Introduction: community-acquired pneumonia is a frequent cause of hospitalization and a leading cause of mortality worldwide. Early diagnosis and the initiation of appropriate antibiotic therapy are essential to reduce pneumonia-related morbidity and mortality. CRP is a well-established biomarker in many clinical settings, but has been traditionally considered not specific enough to be a useful guide in the diagnostic process of pneumonia. There is still a need for more specific and practical markers in CAP for diagnosis. The aim of this study was to in- vestigate the diagnostic value of ischemia-modified albumin levels in the diagnosis of CAP in the Emergen- cy Department.

Methods: The study included 81 patients admitted with CAP and 81 control patients. Initial hour levels of IMA and CRP were measured. The IMA mean levels were compared between the study and control group. Correlation analyses were performed to investigate the association of serum IMA levels with CRP. Results: Mean levels of IMA were 0.532 +- 0.117 IU/ml in the study group and 0.345 +- 0.082 IU/ml in the control group. IMA levels were significantly higher in the study group compared to the control group. The IMA level of

0.442 IU/ml had sensitivity of 75.3% and specificity of 91.3% and was positively correlated with CRP levels (r = 0.506; p b 0.05).

Conclusion: Blood IMA levels significantly increase in adult patients presenting with CAP. IMA may be considered as a novel biomarker in the diagnosis of CAP.

(C) 2017

Introduction

community-acquired pneumonia is defined as an infection of the lung parenchyma that is not acquired in a hospital, long-term care facility, or other recent contact with the health care system. CAP is a sig- nificant cause of morbidity and mortality in adults [1]. Some studies have also reported epidemiological and clinical data demonstrating that the incidence, prevalence, and mortality rate of CAP increase with ageing [2]. In clinical care, the mortality rate remains at 5-15% [3]. In the USA, the annual estimated costs for treating CAP exceed US$12 billion [4].

* Corresponding author.

E-mail addresses: [email protected] (M. Bolatkale), [email protected] (M. Duger), [email protected] (G. Ulfer), [email protected] (C. Can), [email protected] (A.C. Acara), [email protected] (T. Yigitbasi), [email protected] (E.C. Seyhan), [email protected] (M. Bulut).

Several risk scores are available for the evaluation of the prognosis of patients with CAP [5].The pneumonia severity index described by Fine et al. [6] in 1997 is widely used in the United States [7]. In Europe, CURB-65, a score covering the variables of acute confusion, serum urea, respiratory rate, blood pressure and age, is used to predict prognosis [7, 8]. biochemical markers of inflammation have also been discussed as potentially important prognostic variables. These include, the readily available C-reactive protein level and white blood cell count . However, the value of these markers remains unclear.

CRP is a well-established biomarker in many clinical settings, but has been traditionally considered not to be specific enough to be a useful guide in the diagnostic process of pneumonia. In fact, virtually all infec- tive, autoimmune, ischemic and neoplastic diseases can contribute to increased serum CRP values. Nevertheless, some studies have confirmed that it may have a good performance in defining pneumonia diagnosis and severity [9]. New, simple, sensitive and specific tests to diagnose CAP are therefore warranted to reduce the number of advanced Imaging techniques needed. Despite the use of various biochemical markers and

http://dx.doi.org/10.1016/j.ajem.2017.03.018

0735-6757/(C) 2017

probability calculation algorithms based on clinical findings for that purpose, there is still a need for more specific and practical markers in CAP diagnosis [10].

Ischemia modified albumin is an FDA-approved test and one of the newly investigated cardiac markers [11]. IMA results from the modification of N-terminus cobalt binding sites of albumin, which is caused by the release of free radicals from ischemic tissue. The forma- tion of this new albumin molecule, which has lost its ability to bind co- balt, is one of the earliest predictors of ischemia [12]. However, new studies have shown that IMA, which is evaluated as a cardiac ischemia marker, may also increase in different pathologies and affect other or- gans [13]. Previous studies have revealed that IMA levels are significant- ly increased in adults with severe sepsis [14]. However, there are no studies in literature regarding IMA levels in adult patients with CAP. In this study, the diagnostic value of IMA was investigated in patients who presented at the Emergency Department (ED) with CAP. The aim of the present study was to compare CRP and IMA in the diagnosis of community-acquired pneumonia and therefore the relationship of IMA and CRP with PSI was examined.

Methods

Approval for this prospective case-control study was granted by Medipol University Clinical Research Ethics Committee. The study was conducted in Medipol University Hospital Adult ED within 6 months of approval. (28.10.2015/E3200-509)

Study and control groups

The study group was formed of patients with a diagnosis of CAP who presented in ED and adult patients who experienced CAP in ED. The con- trol group comprised Healthy adults with no chronic disease. A written informed consent form was provided by patients (from patient himself/ herself if conscious and from his/her relatives if unconscious) who agreed to participate in the study. All patients in both groups were

>= 18 years old.

Gender, age, hemograms, CRP and IMA levels were recorded for both groups and the following values were examined such as vital signs, GCS of the patient, blood gases, glucose, BUN, creatinine, Na, K and PSI Scores of the study group. Morbidity and mortality were also noted on the study form.

Inclusion and exclusion criteria

Inclusion criteria

Study group: The radiological verification of a newly manifest infil- trate was required. Patients were enrolled who presented with clinically typical pneumonia, if at least two clinical symptoms were present sug- gestive of respiratory tract infection (i.e. dyspnea, cough, new or puru- lent sputum, fever N 38.0 ?C, pleuritic chest pain). Atypical cases presenting for example with mobility impairment, falls, confusion, in- continence or other signs of clinical deterioration were included when a newly manifest infiltrate was verified and no other cause could be de- tected which sufficiently explained the clinical condition of the patient. Control group: Patients without any exclusion criteria who present-

ed at the ED due to complaints not including any disease.

Exclusion criteria

Cases were excluded if they had stenosis-induced or Aspiration pneumonia, malignancy and were receiving chemotherapy or radio- therapy. Patients were also excluded if they had been hospitalized with- in the previous 4 weeks, lacked radiological confirmation of an infiltrate within the first 24 h after admission or were receiving palliative treat- ment modalities. Patients with neutropenia (b 1.0 x 10⁹ cells/l), human immunodeficiency virus infection, tuberculosis, fungal in- fection and those treated with steroids in a prednisone-equivalent

dosage of N 20 mg/day for >= 2 weeks were excluded [15,16]. Patients with acute or chronic conditions that may affect IMA levels such as trau- ma, acute ischemic heart disease/myocardial infarction, peripheral vas- cular disease, Mesenteric ischemia, acute ischemic cerebrovascular disease, pulmonary embolism, muscle diseases, and liver disease and patients who did not agree to participate in the study were excluded. Nursing home residents were not excluded.

Biochemical analysis

Blood samples from brachial veins of the study and control groups were collected in empty vacuum tubes to measure IMA and CRP levels. serum samples were obtained after suitable centrifugation and the sam- ples were stored frozen at -80 ?C until the day of serum IMA analysis. Blood gases were taken from the Radial artery in the study group (ABL 800 FLEX, Radiometer, Bronshoj, Denmark). Glucose, BUN, creatinine, Na and K were measured immediately in the study group (Cobas 6000 auto analyzer, Roche, Tokyo, Japan). Blood samples were collected in 2 ml EDTA tubes and analyzed on an automated hematology analyzer (XT-2000I; Symex, Osaka, Japan) in the study and control groups on the first day of presentation.

CRP levels were measured with the immunoturbidimetric method

immediately. Blood samples from the brachial veins of the study and control groups were collected in empty vacuum tubes which were ini- tially covered with gel to measure CRP (Cobas 6000 auto analyzer, Roche, Tokyo, Japan).

Measurement of serum IMA levels

Blood samples from the brachial veins of the study and control groups were collected in empty vacuum tubes which were initially cov- ered with gel to measure IMA levels. After the blood coagulated in the tubes, they were centrifuged at 1200g and 3000 rpm for 10 min, and then the upper remaining serum parts were collected into Eppendorf micro centrifuge tubes (Eppendorf AG, Hamburg, Germany) and stored at -80 ?C. All blood samples were analyzed in Medipol University Re- search Laboratory after the end of patient enrollment.

IMA levels were measured with the method described by Bar-Or et al. The albumin Cobalt binding test was analyzed according to the meth- od defined by Bar-Or et al. [12]. In this method 200 ml serum was added to the water solution of 50 ml 0.1% (w/v) cobalt chloridine (Sigma; CoCl₂.6H₂O). It was mixed gently and left for 10 min for sufficient co- balt-albumin binding. Then 50 ml dithiothreitol (DTT) (Sigma; 1.5 mg/ml H₂O) was added as a colorizing agent. After waiting for 2 min 1.0 ml 0.9% NaCl was added to stop the cobalt binding process of the albumin. Absorbance was then measured with a spectrophotometer at 470 nm (Shimadzu, model UV160U). Sample blanks without DTT were used as blinds. The results were reported as absorbance units (ABSU).

Statistical analysis

Statistical analyses of the data obtained in the study were made using SPSS 23.0 software (SPSS Inc., Chicago, IL, USA). Variables were stated as means with standard deviation or median with interquartile range. The Student’s t-test was used to compare mean values and the Mann-Whitney U test was used to compare median values. Frequencies were compared with the Chi-square and Fisher’s exact tests. Spearman’s and Pearson’s correlation tests were applied for correlation analyses. Simple correlation analyses were performed to investigate the associa- tion of serum IMA levels with CRP. The median IMA value was calculat- ed, and patients with pneumonia were classified into 2 groups, i.e., those above and equal to or below the median of IMA. To determine a cut-off value of IMA level for pneumonia, receiver operating character- istic (ROC) analysis was performed in sensitivity and specificity calcula- tions. A value of p b 0.05 was considered statistically significant.

Table 1

IMA, CRP, WBC, Hb, neutrophil, lymphocyte, thrombocyte values in the study and control groups.

	Study group (n = 81) (Mean +- SD)	Control group (n = 81) (Mean +- SD)	P value
IMA	0.532 +- 0.117	0.345 +- 0.082	b 0.001
CRP	55.33 +- 68.94	2.44 +- 2.29	b 0.001
WBC	10.49 +- 4.34	7.32 +- 1.80	b 0.001
Hb	13.43 +- 2.02	14.30 +- 1.18	b 0.001
Neutrophil	9.78 +- 3.69	4.27 +- 1.18	b 0.001
Lymphocyte	1.49 +- 0.69	2.12 +- 0.51	0.157
Thrombocyte	142.9 +- 148.6	229.2 +- 45	b 0.001

All values, except lymphocyte, were statistically significant in the study and control groups.

Results

The study included 81 patients admitted with CAP and 81 Control subjects. The mean age was 51.8 +- 18.2 years in the study group and 51 +- 14.9 years in the control group. The study group comprised 38 (47%) females and 43 (53%) males and the control group, 40 (49%) fe- males and 41 (51%) males (p = 0.875). Of the 81 patients in the study group, 24 were admitted to the ED of our hospital and 6 were admitted to the ICU. One of the patients in the ICU died and all others were discharged. The mean levels of IMA of the study group were higher than those of the control group p b 0.001 (Table 1; Fig. 1). The mean levels of CRP of the study group were higher than those of the control group p b 0.001 (Table 1).

The area under the ROC curve for IMA level was 0.905 (95% confi- dence interval [CI], 0.848-0.945) (Fig. 2). The IMA level had sensitivity of 75.3% and specificity of 91.3% at 0.442 IU/ml. The CRP level had sensi- tivity of 76.5% and specificity of 83.9% at 5 IU/ml (Table 2). Pairwise comparison of ROC curves for IMA and CRP difference between areas showed 0.00754 (95% confidence interval [CI], -0.0465-0.0616) and z statistic: 0,273 (p = 0.7845) (Fig. 2). IMA levels were positively corre- lated with CRP levels (r = 0.506; p b 0.05) in this study. When the PSI scores of the 81 patients in the study group were examined, there were 57 patients with PSI scores of 1-3 and 24 patients with PSI scores of 4-5. In the PSI scores 1-3 (n = 57), IMA levels were high (N 0.442 IU/ml) in 37 patients and CRP levels were high (N 0.5 IU/ml)

in 38 patients. In the PSI scores 4-5 (n = 24), the IMA and CRP levels were high in the 24 patients admitted to the ED, and in the 6 in the ICU.

Discussion

The results of this prospective case-control study indicated that serum IMA levels are significantly increased in patients with CAP, com- pared to healthy control subjects. To the best of our knowledge, this is the first study to investigate serum IMA levels in adult patients present- ing at ED with CAP. The present study has shown for the first time that IMA is a sensitive and specific, novel biomarker for the diagnosis of CAP in patients presenting at ED.

CAP is a frequent cause of hospitalization and a leading cause of mor- tality worldwide. The mortality rate for patients admitted to hospital with pneumonia is estimated to be 5-15% [3]. Acute inflammatory con- ditions, such as CAP are characterized by inflammation of the pulmo- nary parenchyma in response to an infectious insult, involving systemic and local release of cytokines and recruitment of neutrophils. Excessive cytokine production provokes a systemic inflammatory re- sponse that can lead to end-organ dysfunction and death. The severity of infection is related to the extent of the inflammatory response of the immune system [17].

First recorded in 1930, CRP is an acute-phase protein produced by the liver in response to most forms of infection, inflammation, and tis- sue injury, which is widely used as a sensitive, but non-specific, marker of systemic inflammation [18]. In response to infection or tissue inflam- mation, CRP production is rapidly stimulated by cytokines, particularly interleukin (IL)-6, IL-1 and tumor necrosis factor [19]. Although its exact function in vivo is not known, it probably has a role in opsonization of infectious agents and damaged cells. Thus, this protein has been considered a non-specific biomarker for early diagnosis and prognostic measurements [20]. However, increases of CRP are also seen in many pulmonary disorders, including infective pathologies such as pneumonia and non-infective pathologies such as malignancies, trauma and pulmonary thromboembolism [21,22]. Specifically, CRP ex- hibits superior diagnostic value for bacterial infections with high plasma concentrations. However, CRP levels remain normal or increase only slightly during most viral infections [23].

A recently published study [24] reported a considerable association between CRP and mortality, describing CRP as an independent risk

Fig. 1. Distribution of IMA index for each group.

Fig. 2. Receiver operating characteristic curves for IMA and CRP (IMA and CPR cut-off point is 0.442 IU/ml and N 0.5 IU/ml).

factor for complications and 30-day mortality. In a receiver operator characteristic curve (ROC) analysis, CRP produced a moderate perfor- mance with regard to mortality, while it outperformed CURB-65 and PSI with regard to complicated pneumonia [25]. Another recent paper

[24] confirmed a significant association between prognosis and CRP and WBC. In the early studies, however, the power of CRP and WBC to predict mortality was significantly lower than CURB-65 and procalcitonin. Moreover, CRP and WBC lost their independence as pre- dictors of prognosis in a multivariate analysis that was adjusted for pneumonia severity [24]. With regard to markers of inflammation, re- cent reviews have found that CRP had limited diagnostic value for pneu- monia in primary care when the probability of pneumonia was b 10% [26,27].

Albumin is the most plentiful serum protein and is a powerful extra- cellular antioxidant. IMA results from modification of the N-terminus cobalt binding sites of albumin caused by the release of free radicals from ischemic tissue. The formation of this new albumin molecule, which has lost its ability to bind cobalt, is one of the earliest predictors of ischemia [12]. The biochemical mechanism modifying the N-terminal region of albumin during ischemia is unclear, but reperfusion after an is- chemic event may damage serum albumin as much as, if not more than, ischemia itself [28]. These modifications to albumin may involve hypox- ia, acidosis, or free radical damage, most of which occur within minutes [12].

The hypothesis of this study was that IMA is increased in conditions that produce inflammation, hypoxia, and free radicals because of CAP. In a study by Sinha et al., it was shown that IMA rises early in ischemia and therefore can be measured earlier than other markers. Elevated levels have been seen in 6-10 min, making it useful in the diagnosis of patients admitted early [29].

Early diagnosis and the initiation of appropriate antibiotic therapy within 4 h are essential to reduce pneumonia-related morbidity and

Table 2

Sensitivity and specificity of IMA and CRP. Positive and negative predictive value of IMA and CRP (IMA and CPR cut-off point is 0.442 IU/ml and N 0.5 IU/ml).

Estimated value 95% Confidence interval

mortality [30]. This Early prediction by a biochemical marker of inflam- mation is important as it may improve our ability to risk stratify CAP pa- tients and guide therapeutic decisions [31]. Furthermore, this study has demonstrated that a raised IMA level on admission is an independent predictor of CAP. The results of the study have shown that this marker has a high value in diagnosis of CAP as it increases immediately after in- fection and inflammation and is known to have high negative and pos- itive predictive values.

The findings of this study showed that there is a significant increase in serum IMA levels in patients with CAP compared to healthy control subjects. Serum CRP, which has been most widely used as a CAP bio- marker, was positively correlated to serum IMA levels in patients with CAP. Thus, the hypothesis was confirmed that IMA may be useful as a di- agnostic biomarker for CAP because it can indicate the severity of illness in patients with CAP. If these results are confirmed by further studies, the use of IMA may improve current Diagnostic strategies for CAP.

Conclusion

Blood IMA levels significantly increase in adult patients with CAP. The results of this study support the diagnosis of CAP using IMA in the ED.

Study limitations

This study has some limitations, IMA is a new biomarker and levels are influenced significantly by a wide array of physiological variables, including exercise and hydration. However, it was not possible to con- trol all the variables that could influence IMA levels. There was no com- parison of other biochemical markers except CRP and IMA in the diagnosis of CAP. The contact time to hospital was accepted as the initial hour in this study. The period which started after the collection of the first blood samples was not standardized in every patient and the num- ber of subjects included in the study was limited.

References

1. Watkins Richard R, Lemonovich Tracy L. Diagnosis and management of community- acquired pneumonia in adults. Am Sam Physician 2011;83:1299-306.
2. Pinner RW, Teutsch SM, Simonsen L, Klug LA, Graber JM, Clarke MJ, et al. Trends in infectious diseases mortality in the United States. JAMA 1996;275:189-93.
3. Chalmers JD, Mandal P, Singanayagam A, Akram AR, Choudhury G, Short PM, et al. Severity assessment tools to guide ICU admission in community-acquired pneumo- nia: systematic review and meta-analysis. Intensive Care Med 2011;37:1409-20.
4. Colice GL, Morley MA, Asche C, Birnbaum HG. treatment costs of community-ac- quired pneumonia in an employed population. Chest 2004;125:2140-5.
5. Sirvent JM, De la Torre MC, Lorencio C, Tache A, Ferri C, Garcia-Gil J, et al. Predictive factors of mortality in severe community-acquired pneumonia: a model with data on the first 24h of ICU admission. Med Intensiva 2013;37:308-15.6.
6. Fine MJ, Auble TE, Yealy DM, Hanusa BH, Weissfeld LA, Singer DE, et al. A prediction rule to identify low-risk patients with community-acquired pneumonia. N Engl J Med 1997;336:243-50.
7. Ewig S, Torres A, Woodhead M. Assessment of pneumonia severity: a European per- spective. Eur Respir J 2006;27:6-8.
8. Lim WS, van der Eerden MM, Laing R, Boersma WG, Karalus N, Town GI, et al. Defin- ing community acquired pneumonia severity on presentation to hospital: an inter- national derivation and validation study. Thorax 2003;58:377-82.
9. Garcia Vazquez E, Martinez JA, Mensa J, Sanchez F, Marcos MA, de Roux A, et al. C- reactive protein levels in community-acquired pneumonia. Eur Respir J 2003;21: 702-5.
10. Woodhead M, Blasi F, Ewig S, Huchon G, Ieven M, Ortqvist A, et al. Guidelines for the management of adult lower respiratory tract infections. Eur Respir J 2005;26: 1138-80.

    		Lower limit	Upper limit	[11] Wudkowska A, Goch J, Goch A. Ischemia-modified albumin in differential diagnosis of acute coronary syndrome without ST elevation and unstable angina pectoris.
    Sensitivity of IMA	0.753086	0.642546	0.839248	Kardiol Pol 2010;68:431-7.
    Specificity of IMA	0.91358	0.824583	0.9616	[12] Bar-Or D, Lau E, Winckler JV. A novel assay for cobalt-albumin binding and its poten-
    True positive of IMA	0.897059	0.793493	0.954133	tial as a marker of myocardial ischemia-a preliminary report. J Emerg Med 2000;19:
    True negative of IMA	0.787234	0.688278	0.862228	311-5.
    Sensitivity of CRP	0.765432	0.655793	0.849477	[13] Lippi G, Montagnana M. Ischemia modified albumin in ischemic disorders. Ann

    Specificity of CRP 0.839506 0.737517 0.908475

    True positive of CRP 0.826667 0.718184 0.900943

    True negative of CRP 0.781609 0.677653 0.860221

    Thorac Cardiovasc Surg 2009;15:137.

    Erdem Said Sami, Yerlikaya Fatma Humeyra, Cicekler Humeyra, Gul Mehmet. Asso- ciation between ischemia-modified albumin, homocysteine, vitamin B12 and folic acid in patients with severe sepsis. Clin Chem Lab Med 2012;50:1417-21.

11. Thiem Ulrich, Niklaus David, Sehlhoff Bettina, Stuckle Christoph, Heppner Hans Jurgen, Endres Heinz Gerd, et al. C-reactive protein, severity of pneumonia and mor- tality in elderly, hospitalized patients with community-acquired pneumonia. Age Ageing 2009;38:693-7.
12. Bruns AHW, Oosterheert JJ, HaK E, Hoepelman AIM. Usefulness of consecutive C-re- active protein measurements in follow-up of severe community-acquired pneumo- nia. Eur Respir J 2008;32:726-32.
13. Delclaux C, Azoulay E. Inflammatory response to infectious pulmonary injury. Eur Respir J 2003;42:10s-4s.
14. Pepys MB, Hirschfield GM. C-reactive protein: a critical update. J Clin Invest 2003; 111:1805-12.
15. Epstein F. Acute phase proteins and other systemic responses to inflammation. N Engl J Med 1999;340:448-54.
16. Lobo SM, Lobo FR, Bota DP, Lopes-Ferreira F, Soliman HM, Melot C, et al. C-reactive protein levels correlate with mortality and organ failure in critically ill patients. Chest 2003;123:2043-9.
17. Liu A, Bui T, Van NH, Ong B, Shen Q, Kamalasena D. Serum C-reactive protein as a biomarker for early detection of bacterial infection in the older patient. Age Ageing 2010;39:559-65.
18. Smith RP, Lipworth BJ. C-reactive protein in simple community-acquired pneumo- nia. Chest 1995;107:1028-31.
19. Haran JP, Beaudoin FL, Suner S, Lu S. C-reactive protein as predictor of bacterial in- fection among patients with an influenza-like illness. Am J Emerg Med 2013;31: 137-44.
20. Kruger S, Ewig S, Marre R, Papassotiriou J, Richter K, von Baum H, et al. Procalcitonin predicts patients at low risk of death from community-acquired pneumonia across all CRB-65 classes. Eur Respir J 2008;31:349-55.
21. Chalmers JD, Singanayagam A, Hill AT. C-reactive protein is an independent predic- tor of severity in community-acquired pneumonia. Am J Med 2008;121:219-25.
22. Falk G, Fahey T. C-reactive protein and community-acquired pneumonia in ambula- tory care: systematic review of diagnostic accuracy studies. Fam Pract 2009;26: 10-21.
23. Engel MF, Paling FP, Hoepelman AI, van der Meer V, Oosterheert JJ. Evaluating the evidence for the implementation of C-reactive protein measurement in adult pa- tients with suspected lower respiratory tract infection in primary care: a systematic review. Fam Pract 2012;29:383-93.
24. Edwards SW, Hallett MB, Campbell AK. Oxygen-radical production during inflam- mation may be limited by oxygen concentration. Biochem J 1984;217:851-4.
25. Sinha MK, Roy D, Gaze DC, Collins PO, Kaski JC. Role of “ischemia modified albumin,” a new biochemical marker of myocardial ischemia, in the early diagnosis of acute coronary syndromes. Emerg Med J 2004;21:29-34.
26. Marrie TJ. Community-acquired pneumonia. Clin Infect Dis 1994;18:501-13.
27. Rodriguez AH, Aviles-Jurado FX, Diaz E, Schuetz P, Trefler SI, Sole-Violan J, et al. procalcitonin levels for ruling-out bacterial coinfection in ICU patients with in- fluenza: a CHAID decision-tree analysis. J Infect 2016;72:143-51.