Journal of Critical Care (2013) 28, 291295

Impact of serum C-reactive protein measurements in the first 2 days on the 30-day mortality in hospitalized patients with severe community-acquired pneumonia: A cohort study
William Nseir a,b,, Raymond Farah b,c , Julnar Mograbi a , Nicola Makhoul b,d
a

Internal Medicine Department and Infectious Diseases Unit, Holy Family Hospital, Nazareth, Israel Faculty of Medicine in the Galilee, Bar-Ilan University, Safed, Israel c Internal Medicine Department B, Ziv Medical Center, Safed, Israel d Respiratory Intensive Care Unit and Eliachar Research Laboratory, Western Galilee, Hospital- Nahariya, Israel
b

Keywords:
Community-acquired pneumonia; C-reactive protein (CRP); Fractional decrease in CRP; 30-day mortality

Abstract Introduction: The purpose of the study is to evaluate the impact of daily consecutive measurements of C-reactive protein (CRP) in the initial 2 days of hospitalization on the 30-day all-cause mortality in patients with severe community-acquired pneumonia (CAP). Methods: We used 4 different thresholds of fractional decrease (FD) in CRP at the second day of admission (CRP2) of 25%, 30%, 40%, and 60%. In addition, we studied the association of each of these thresholds with the 30-day all-cause mortality. Results: The mean age was 64 20; males, 59%. The 30-day mortality rate was 18% (20/111). The mean serum CRP levels at the first day of all study group and CRP2 were 203 98 vs 146 92 mg/L, respectively, P = .05. The mean FD in CRP2 levels among the survivors was 33 %, whereas among the nonsurvivors, was 7%, P b .001. Multiple regression analysis revealed that FD less than 25% in CRP2 was associated with 30-day all-cause mortality, odds ratio of 3.07 (95% confidence interval, 2.84-5.03), P = .002, compared with those with FD more than 25% in CRP2. Conclusions: Fractional decrease less than 25% in CRP levels at the second day was significantly associated with 30-day all-cause mortality in hospitalized patients with severe CAP. 2013 Elsevier Inc. All rights reserved.

1. Introduction
Corresponding author. Department of Internal Medicine, Infectious Diseases Unit, Holy Family Hospital, Nazareth, Faculty of Medicine in the Galilee, Bar-Ilan University, Israel, POB, 8. Tel.: +97246508942; fax: +97246508942. E-mail address: w.nseir@yahoo.com (W. Nseir). 0883-9441/$ see front matter 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2012.09.012

Lower respiratory infections, which result in pneumonia, are the third leading cause of death worldwide [1]. The estimated incidence of community-acquired pneumonia (CAP) is 2 to 12 cases per 1000 population per year [2]. Most cases of CAP are managed outside hospital, but

292 approximately 20% require hospital admission; out of these patients, approximately 10% develop severe CAP [3]. Accurate diagnosis and appropriate management of CAP are crucial for reducing medical costs, length of hospital stay, mortality rates, and antibiotic resistance; the latter has become a public health problem on a global scale [4]. In the absence of generally accepted criteria for determining the severity and the ensuing management of CAP, the search has intensified for reliable biomarkers for diagnosis, risk prediction, and disease management [5]. C-reactive protein (CRP) is an acute phase protein synthesized by the liver, primarily in response to interleukin-6. Its measurement is inexpensive and widely accessible. C-reactive protein levels have been shown to discriminate between pneumonia and healthy status [6] as well as between pneumonia and exacerbations of chronic obstructive pulmonary disease, heart failure, and asthma [7-9]. Consecutive measurements of CRP levels at days 3 and 7 also have been shown to distinguish between causative pathogens [6,10], including bacterial and nonbacterial causes [11]. Recently, Ramirez et al [12] reported that inflammatory biomarkers identified patients needing intensive care unit (ICU) admission, including those with delayed ICU admission. Regarding to process of medical decision in cases of severe CAP, the timing of the CRP measurements after the fourth day of admission seems to be delayed for making any change in medical management. Therefore, the rationale of this study was to examine the impact of early measurements of CRP despite the half lifetime of CRP of 19 hours. Against this background, the current study was to evaluate the impact of consecutive measurements of serum CRP levels in the first 2 days of hospitalization on the 30-day all-cause mortality.

W. Nseir et al. 109 neutrophils/L), HIV infection, intravenous drug use, presence of a severe infection other than pneumonia, on antibiotic use, Pneumonia Severity Index (PSI) less than class IV, CURB-65 (Confusion, Urea, Respiratory rate, Blood pressure, and Age >65) score less than 3 points, a life expectancy less than 1 month because of an underlying disease, rheumatoid arthritis, an active autoimmune disease, or patients admitted directly to an ICU. On admission, demographic data, clinical signs, symptoms, and laboratory data including calculation of the PSI class and CURB-65 score were recorded [13,14]. Community-acquired pneumonia was defined as present in cases with at least 2 symptoms of acute lower respiratory tract infection with onset before hospital admission and a new or progressive pulmonary infiltrate on chest radiograph. A standard follow-up period of 30 days was established for all the participants. Serum concentration of CRP was measured by particleenhanced immunoturbimetric assay (Tina-quant CRP latex; Roche Diagnostics Corporation, Indianapolis, Ind). The reference range for this assay is less than 5 mg/L. For each patient, CRP was measured at days 1, 2, 3, 4, and 5. C-reactive protein 1 was taken within 6 hours from the admission to emergency department. A unique protocol was used to treat CAP (a -lactam and macrolide). To emphasize, we did not use steroids as adjunctive therapy in our protocol; however, we used steroid according to the medical needs of each patient. All patients had received the antibiotic therapy within 6 hours of their presentation to the emergency department.

2.3. Outcome
The main outcome was to evaluate the impact of daily consecutive measurements of CRP in the initial 2 days of hospitalization on the 30-day all-cause mortality in patients with severe CAP.

2. Patients and methods

2.1. Design and setting
This was a prospective observational cohort study that was conducted between November 2009 and December 2010 in internal medicine department and respiratory ICU, Western Galilee Hospital, Nahariya, Faculty of Medicine in the Galilee, Bar-Ilan University, Israel. All patients gave written informed consent before the enrollment to the study. The study was approved by the local medical ethics committee.

2.4. Data presentation and statistical analysis

The data were statistically analyzed using WinSTAT (Kalamia, Cambridge, Mass), the statistics add-in for Microsoft Excel. Continuous variables are expressed as the mean SD. Initially, the study participants were divided into 2 groups: those who had survive the 30-day survivors and those who died nonsurvivors and differences between the 2 groups were assessed by the 2 test for categorical variables and the Student t test for continuous variables. We used 4 different thresholds of fractional decrease (FD) in CRP at the second day of admission (CRP2) of 25%, 30%, 40, and 60%. Fractional decrease was calculated according to the following formula: (CRP1 CRP2)/CRP1 100%. Spearman rank correlation and univariate regression analysis were used to determine the strength of the relationship between the different variables for 30-day all-cause mortality, namely, age, sex, PSI class of 4 or higher, CURB-65 score of 3 points

2.2. Selection of participants, data collection, and definitions

Patients of either sex with severe CAP that required admission in internal medicine department and older than 18 years were consecutively included. We excluded patients with interstitial pneumonia, cystic fibrosis, a history of colonization with gram-negative bacteria due to structural damage to the respiratory tract, severe neutropenia (b 0.5x

Impact of serum C-reactive protein measurements

268 patients were admitted due to CAP Reasons for exclusion: 66 patients with PSI class< IV 54 patients with CURB-65 score <3points 157 patients were excluded 12 patients had terminal illnesses 6 patients were immunosuppressed 8 patients with pulmonary structural disorders 11 patients refused to participate 111 patients were included in the study

293

Fig. 1

Study flowchart.

or higher, the 4 thresholds of FD in CRP2, and FD in CRP5, smoking, and chronic obstructive pulmonary disease (COPD). A variable associated with P b .05 in univariate analysis was used after that for feature analysis. A multiple regression analysis was done to determine the association between the variables and the 30-day all-cause mortality. Statistical significance was set at 5%.

between the survivors and nonsurvivors are shown in Table 2. In 72 (79%) of 91 patients from the survivors had reduced greater than 25% of their CRP in day 2 from the initial CRP. In contrast, among the nonsurvivors, 1 patient had FD in CRP2 greater than 25%; P b .001. To the fifth day, from 111 patients, 6 patients died, 32 patients discharged, and 73 patients remained hospitalized.

3.2. Outcome analysis

3. Results
3.1. Characteristics of the study population
A total of 268 patients with CAP were admitted to the Internal Medicine Department during the study period; 111 patients met the study criteria (Fig. 1). We excluded 157 patients: 66 patients had PSI class less than 4, 54 patients had CURB-65 score less than 3 points, 12 patients had terminal illnesses, 6 patients were immunosuppressed, 8 patients had pulmonary structural disorders, and 11 patients had refused to participate. Demographic, clinical characteristics, and comparisons between survivors (n = 91) and nonsurvivors (n = 20) are presented in Table 1. The nonsurvivors were older. There were no significant differences between the 2 groups regarding the sex, COPD as comorbidity, and smoking. The mean PSI class, CURB-65 score, number of positive blood cultures, the need for mechanical ventilation, transfer to ICU, and hospital length of stay in days were significantly differs statistically between the 2 groups. The decline in CRP levels during the hospitalization among the survivors vs nonsurvivors was statistically significant. The differences in the thresholds of FD in CRP of day 2 from the initial CRP greater than 25%, 30%, 40%, and 60% The main outcome was to evaluate the impact of daily consecutive measurements of CRP in the initial 2 days of hospitalization on the 30-day all-cause mortality in patients with severe CAP. To examine the associations of the different variables including FD in CRP at the second day of hospitalization with the 30-day all-cause mortality, we performed multiple regression analysis (Table 3), which revealed that age, odds ratio (OR) of 0.44 (95% confidence interval [CI], 0.28-1.25), P = .63; male sex, OR of 2.12 (95% CI, 1.88-2.95), P = .03; FD in CRP2 less than 25%, OR of 3.07 (95% CI, 2.84 -5.03), P = .002; PSI class of 4 or higher, OR of 1.45 (CI, 0.82-2.31), P = .05; CURB-65 of 3 points or higher, OR of 5.79 (95% CI, 5.21-6.39), P b .001, were associated with 30-day allcause mortality in patients with severe CAP.

4. Discussion
In this study of hospitalized patients with severe CAP, we found that the FD less than 25% in CRP levels at the second day of hospitalization was significantly associated with 30day all-cause mortality. According to our knowledge, this is

294

W. Nseir et al.

Table 1 Demographic and clinical variables of the 111 study participants and comparisons between survivors and nonsurvivors of severe CAP All patients (N = 111) Age, y Male sex, n (%) PSI (class) CURB-65 score ( points) CRP mg/dL in day 1(means) CRP in day 2 (means) CRP mg/dL in day 3 (means) CRP mg/dL in day 4 (means) CRP mg/dL in day 5 (means) CRP mg/dL at discharge (means) The FD of CRP2 from CRP1 (%) COPD, n (%) Smoking, n (%) Positive blood cultures, n (%) Needed MV during hospitalization, n (%) Transfer to ICU, n (%) Hospital LOS, in days (means) 64 20 66 (59%) 4.5 0.55 2.62 0.86 203 98 146 92 125 80 93 71 86 77 43 41 28% 14 (12%) 47 (42%) 19 (17%) 22 (20%) 27 (24%) 7.5 6 Survivors (n = 91, 82%) 62 20 54 (59%) 4.37 0.55 2.38 0.6 193 95 128 82 108 68 75 56 67 54 33 26 33% 11 (12%) 35 (38%) 11 (12%) 10 (11%) 13 (14%) 6.7 4.5 Nonsurvivors (n = 20, 18%) 71 14 12 (60%) 4.8 0.41 3.7 0.9 246 99 228 91 205 84 174 74 165 106 82 65 7% 3 (15%) 9 (45%) 8 (40%) 12 (60%) 14 (70%) 11 9.3 Pa .03 .20 .001 b .001 .02 b .001 b .001 b .001 b .001 b .001 b .001 .07 .08 b .001 b .001 b .001 .001

MV, mechanical ventilation; LOS, length of stay. a For comparisons among survivors and nonsurvivors.

the first study investigating the association of CRP levels measurements of the first 2 days and the 30-day all-cause mortality in patients with severe CAP. Several studies have shown that CRP is a good marker of CAP diagnosis as well as an useful tool for assessing the severity and effectiveness of treatment [6,15,16]. Moreover, measurement of CRP has demonstrated effectiveness in distinguishing between CAP, which requires or not requires antibiotics [5,7,8]. However, in a study of elderly hospitalized patients with CAP, CRP at admission was not associated with the need for transfer to an ICU or with mortality [17]. Although singular measurements at hospital admissions have demonstrated the usefulness of CRP in the diagnosis of pneumonia, consecutive measurements support the role for this biomarker in the therapeutic management of pneumonia. In a prospective study of 294 hospitalized patients with CAP, Menndez et al [18] found that low levels of CRP at 72 hours after treatment initiation, together with clinical criteria, might

Table 2 The FDs greater than 25%, 30%, 40%, and 60% in CRP levels at the second day (CRP2) from the CRP levels of first day of hospitalization among of the survivors and nonsurvivors and the comparisons between them in each threshold of CRP2 Survivors, Nonsurvivors, P a n = 91 n = 20 FD in FD in FD in FD in
a

improve the prediction of absence of severe complications of CAP. Furthermore, Bruns et al [10] found that consecutive CRP measurements during the first week of hospitalization have demonstrated usefulness for follow-up of antibiotic treatment for CAP and a delayed normalization of CRP levels is associated with a higher risk of receiving inappropriate antibiotic treatment for severe CAP. Ruiz Gonzalez et al [19] found that CRP levels at day 4 were useful in distinguishing between failure to respond to antibiotic treatment for CAP and slow response to treatment. Moreover, Chalmers et al [20] reported that a decrease in CRP levels of less than 50% within the first 4 days of hospitalization was associated with an increased risk for 30day mortality and a need for mechanical ventilation or inotropic support. Previous studies based on the levels of CRP at day 3 or later in making medical decision. Therefore, we believe that our finding of this threshold of FD less than 25% in CRP at day 2 may provide a useful tool for identifying high-risk patients. Numerous clinical reports have been published regarding steroid use in severe pneumonia with conflicting results
Table 3 The results of multiple regression analysis on the association between the different variables and the 30-day allcause mortality of 111 hospitalized patients with severe CAP Variables Age Male sex FD b 25% in CRP2 PSI class 4 CURB-65 3 points OR (95% CI) 0.44 (0.28-1.25) 2.12 (1.88-2.95) 3.07 (2.84-5.03) 1.45 (0.82-2.31) 5.79 (5.21-6.39) P .63 .03 .002 .05 b .001

CRP2 CRP2 CRP2 CRP2

N 25%, N 30%, N 40%, N 60%,

n (%) n (%) n (%) n (%)

72 62 41 17

(79%) (68%) (45%) (19%)

1 (5%) 0 (0%) 0 (0%) 0 (0%)

b .001 b .001 .002 .03

For comparisons among survivors and nonsurvivors.

Impact of serum C-reactive protein measurements [21-24]. We did not use steroid as adjuvant therapy in our protocol therapy for CAP. However, we used steroids as needed, especially among COPD patients (COPD patients were 12% of all patients), although the effect of steroids on CRP levels is inconclusive. Salluh et al [25] examined the impact of systemic corticosteroids on the clinical course and outcomes in 111 patients with severe CAP, and they found that adjuvant corticosteroids did not influence hospital mortality. The same results have been shown by Perren et al [26] who in addition found that corticosteroids do not modify the time-dependent decay of CRP in case that CAP is adequately treated. The present study has several limitations. It was performed in 1 center, with relatively small number of patients, and we did not consider the causative pathogens in our interpretation of CRP levels [27].

295
[9] Bafadhel M, Clark TW, Reid C, et al. Procalcitonin and C-reactive protein in hospitalized adult patients with community-acquired pneumonia or exacerbation of asthma or COPD. Chest 2011;139: 1410-8. [10] Bruns AH, Oosterheert JJ, Hak E, et al. Usefulness of consecutive C-reactive protein measurements in follow-up of severe communityacquired pneumonia. Eur Respir J 2008;32:726-32. [11] Muller B, Harbarth S, Stolz D, et al. Diagnostic and prognostic accuracy of clinical and laboratory parameters in community-acquired pneumonia. BMC Infect Dis 2007;7:10. [12] Ramirez P, Ferrer M, Marti V, et al. Inflammatory biomarkers and prediction for intensive care unit admission in severe communityacquired pneumonia. Crit Care Med 2011;39:2211-7. [13] Lim WS, van der Eerden MM, Laing R, et al. Defining community acquired pneumonia severity on presentation to hospital: an international derivation and validation study. Thorax 2003;58:377-82. [14] Fine MJ, Auble TE, Yealy DM, et al. A prediction rule to identify lowrisk patients with community-acquired pneumonia. N Engl J Med 1997;336:243-50. [15] Gracia Vazquez E, Martinez JA, Mensa J, et al. C-reactive protein levels in community-acquired pneumonia. Eur Respir J 2003;21: 702-5. [16] Requejo HI, Cocoza AM. C-reactive protein in the diagnosis of community-acquired pneumonia. Braz J Infect Dis 2003;7:C241-4. [17] Thiem U, Niklaus D, Sehlhoff B, et al. C-reactive protein, severity of pneumonia and mortality in elderly, hospitalized patients with community-acquired pneumonia. Age Ageing 2009;38:693-7. [18] Menndez R, Martinez R, Reyes S, et al. Stability in communityacquired pneumonia: one step forward with markers? Thorax 2009;64: 987-92. [19] Ruiz-Gonzlez A, Falguera M, Porcel JM, et al. C-reactive protein discriminating treatment failure from slow responding pneumonia. Eur J Intern Med 2010;21:548-52. [20] Chalmers JD, Singanayagam A, Hill AT. C-reactive protein is an independent predictor of severity in community-acquired pneumonia. Am J Med 2008;121:219-25. [21] Confalonieri M, Urbino R, Potena A, et al. Hydrocortisone infusion for severe community-acquired pneumonia: a preliminary randomized study. Am J Respir Crit Care Med 2005;171:242-8. [22] Mikami K, Suzuki M, Kitagawa H, et al. Efficacy of corticosteroids in the treatment of community-acquired pneumonia requiring hospitalization. Lung 2007;185:249-55. [23] Confalonieri M, Meduri GU. Glucocorticoid treatment in communityacquired pneumonia. Lancet 2011;377:1982-4. [24] Lippi G, Meschi T, Cervellin G. Inflammatory biomarkers for the diagnosis, monitoring and follow-up of community-acquired pneumonia: clinical evidence and perspectives: review article. Eur J Intern Med 2011;22:460-5. [25] Salluh JI, Soares M, Coelho LM, et al. Impact of systemic corticosteroids on the clinical course and outcomes of patients with severe community-acquired pneumonia: a cohort study. J Crit Care 2011;26:193-200. [26] Perren A, Cerutti B, Lepori M, et al. Influence of steroids on procalcitonin and C-reactive protein in patients with COPD and community-acquired pneumonia. Infection 2008;36:163-6. [27] Chiang WC, Teoh OH, Chong CY, et al. Epidemiology, clinical characteristics and antibacterial resistance patterns of communityacquired pneumonia in 1702 hospitalized children in Singapore. Respirology 2007;12:254-61.

5. Conclusions
Fractional decrease less than 25% in CRP levels in the second day of hospitalization was significantly associated with 30-day all-cause mortality in patients with severe CAP. Further prospective studies with larger number of patients are needed to assess the clinical impact of this fall of CRP in the second day among patients with severe CAP.

References
[1] World Health Organization (WHO). The top 10 causes of death (2008). Fact Sheet No. 310. Updated June 2011. http://www.who.int/ mediacentre/factsheets/fs310/en/index.html. [2] Baudouin SV. The pulmonary physician in critical care.3: critical care management of community acquired pneumonia. Thorax 2002;57: 267-71. [3] British Thoracic Society Standards of Care Committee: BTS Guidelines for the management of community acquired pneumonia in adults. Thorax 2001;56:IV1-IV64. [4] Hawkey PM, Jones AM. The changing epidemiology of resistance. J Antimicrob Chemother 2009;64(Suppl 1):i3-i10. [5] Christ-Crain M, Opal SM. Clinical review: the role of biomarkers in the diagnosis and management of community acquired pneumonia. Crit Care 2010;14:203. [6] Almirall J, Bolbar I, Toran P, et al. Contribution of C-reactive protein to the diagnosis and assessment of severity of community-acquired pneumonia. Chest 2004;125:1335-42. [7] Justo D, Lachmi S, Saar N, et al. C-reactive protein velocity following antibiotics in patients with chronic obstructive pulmonary disease exacerbation and community acquired pneumonia. Eur J Intern Med 2009;20:518-21. [8] Joffe E, Justo D, Mashav N, et al. C-reactive protein to distinguish pneumonia from acute decompensated heart failure. Clin Biochem 2009;42:1628-34.