•  
  •  
 
The University of Louisville Journal of Respiratory Infections

Abstract

The goals of this study were to investigate the relationship of systemic and local cytokine responses with time to clinical stability (TCS) in patients with community-acquired pneumonia (CAP) and to develop a model to integrate multiple cytokine data into “cytokine response profiles” based on local vs. systemic and pro- vs. anti-inflammatory cytokine patterns in order to better understand their relationships with measures of CAP severity and outcomes. Forty hospitalized patients enrolled through the Community Acquired Pneumonia Inflammatory Study Group (CAPISG) were analyzed. Based on the ranked distribution of the levels of eight different pro-inflammatory cytokines and chemokines (IL-1b, IL-6, IL-8, IL-12p40, IL-17A, IFNg, TNFa and CXCL10) in plasma and sputum on hospital admission, a “pro-inflammatory cytokine score (PICS)” was defined. PICS in plasma and sputum were plotted against each other and quadrants used to define profiles based on the four possible high/low combinations. A similar approach was used to contrast sputum PICS vs. anti-inflammatory cytokines (IL-1ra and IL-10). Some of the “profiles” thus defined were found to group patients with common etiologic characteristics and/or associate with similar measures of disease severity and/or clinical outcomes, suggesting the predictive value of the use of cytokine data in CAP patients.

Body Text

INTRODUCTION

Despite the availability and use of antimicrobial therapy, community-acquired pneumonia (CAP) remains one of the major causes of infectious disease-related death in the world [1,2]. In CAP, a local immune response to the invading microorganisms induces the recruitment and activation of leukocytes to the lungs [3,4]. Even in the presence of antimicrobial therapy, this response plays a key role in determining whether or not the offending microorganisms escape control and further spread beyond the lungs, eliciting a vigorous systemic inflammatory response (5,6).

While the initial inflammatory response has a protective role, excessive or uncontrolled inflammation may result in deleterious consequences for the host [6-8]. In fact, an association between an exaggerated systemic inflammatory response (i.e., “cytokine storm”) with poor clinical outcomes has been established by several investigators [6, 9, 10] and it is generally agreed that failure to control excessive inflammation may lead to organ damage and a higher mortality risk [6, 9, 10-12]. The potentially deleterious role of disproportionate inflammation has prompted several studies to explore the therapeutic use of immunomodulatory corticosteroids in CAP patients. Unfortunately, results have been inconsistent and their real efficacy so far remains controversial [13-17]. While the reasons for the discrepancies are not clear, it has been suggested that although only a subset of CAP patients may actually benefit from corticosteroid use (those with the most severe inflammation), most studies have enrolled patients without consideration of their inflammatory status [18].

Based on previous studies [19] we have hypothesized that patients who are able to mount strong local inflammatory responses might be able to have better clinical outcomes. However, while most studies in CAP patients have focused on the association between disease severity or clinical outcomes and the systemic inflammatory response (measured in serum or plasma), data on associations with the local inflammatory response, such as measured in bronchoalveolar lavage (BAL) or sputum, are limited. Thus, the aims of this study were to investigate the association of both, local and systemic cytokine response patterns, with the time to clinical stability (TCS) and to explore potential means to integrate the overall cytokine response (plasma and sputum) in the evaluation of CAP patients.

MATERIALS AND METHODS

This was a secondary analysis of data previously collected from the Community Acquired Pneumonia Inflammatory Study Group (CAPISG). The design of our study has been previously described [19]. Briefly, this study included 40 patients with CAP hospitalized at the University of Louisville Hospital or the Louisville’s Veteran Administration Hospital from 01/04/2011 to 01/08/2012. The study was approved by both the University of Louisville Human Subjects Program Protection Office and the Robley Rex Veterans Affairs Medical Center Institutional Review Boards (Approvals #: 07.0182 and 0009, respectively). The demographic and clinical data of the patient population in this study has already been reported [19].

Criteria for CAP.

CAP was defined as evidence of a new pulmonary infiltrate at chest radiograph associated with at least one of the following: 1) new or increased cough; 2) fever or hypothermia; and 3) leukocytosis, left shift, or leukopenia. Pneumonia was defined as community-acquired if a patient had no history of hospitalization during the 2 weeks prior to admission. The inclusion and exclusion criteria and full case report forms for this study can be found at the Community Acquired Pneumonia Organization study site at www.caposite.com.

Microbiologic Analysis.

Testing of sputum samples and blood cultures; serology for respiratory viruses and atypical organisms as well as urinary antigens for Legionella spp. and Streptococcus pneumoniae were performed according to standard clinical practice. The identification of microorganisms and susceptibility testing were performed according to standard methods [20]. Cases in which no specific etiologic agent was isolated or identified were classified as CAP of unknown etiology.

Severity of Disease and Time to Clinical Stability.

The Pneumonia Severity Index (PSI) was used to define CAP severity [21]. Hospitalized patients were considered to have severe CAP if their PSI was Risk Class IV or V (91 points or higher). Patients with a PSI Risk Class of I-III (points) were defined as having non-severe CAP. Accordingly, 21 patients were classified as having severe and 19 non-severe CAP. The time to clinical stability (TCS) was defined based on the American Thoracic Society (ATS)/Infectious Diseases Society of America (IDSA) 2007 guidelines as improved clinical signs (cough and shortness of breath), lack of fever for at least 8 hours, improving leukocytosis (decreased at least 10% from the previous day), and tolerating oral intake [22]. For this study, patients were divided into two groups based on a TCS equal or less than 3 days (TCS≤3d) and a TCS greater than three days (TCS>3d). Accordingly, 24 patients were classified as having a TCS≤3d and 16 patients a TCS>3d.

Samples.

Plasma samples: Blood samples were obtained on the day of admission at the hospital. Venous blood was collected using sodium citrate Vacutainer tubes. Following centrifugation at 300 x g for 10 min, the plasma was separated by aspiration, aliquoted and stored frozen at -80°C until assayed.

Sputum samples: Sputum samples were collected from a total of 15 CAP patients (7 with non-severe, 8 with severe CAP) on the day of admission to the hospital. Sputum samples were processed following the method described by Pizzichini et al [23]. The cell-free supernatants were aliquoted and stored frozen at -80°C until used for the measurement of cytokine levels.

Cytokine measurements.

The concentrations of ten different cytokines and chemokines in plasma and sputum samples were determined using Milliplex MAP High Sensitivity Human Cytokine Magnetic bead kits (HSCYTMAG-60SK, EMD Millipore, Billerica, MA) according to the manufacturer’s instructions. The panel included eight important pro-inflammatory cytokines and chemokines: interleukin (IL)-1β, IL-6, CXCL8 (IL-8), IL-12p40, IL-17A, interferon (IFN)γ, tumor necrosis factor (TNF)α and CXCL10 (IP-10); and two main anti-inflammatory cytokines: IL-1 receptor antagonist (IL-1ra) and IL-10.

Statistics.

Statistical analysis and graphs were performed using GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA). Data distribution was analyzed using the D’Agostino and Pearson omnibus normality test. Due to a non-normal distribution of cytokine data, statistical comparisons between the groups were performed using the Wilcoxon-Mann-Whitney test. Unpaired t-tests were used to compare normally-distributed data. GraphPad was used to construct ROC curves. The Youden’s Index (J) was calculated as J = (specificity+sensitivity)-1. p-values of ≤0.05 were considered statistically significant.

RESULTS

Cytokine Patterns in Plasma and Sputum and Association with the TCS.

In order to associate cytokine patterns with time to clinical stability (TCS), we compared the plasma and sputum cytokine levels in patients divided in two groups based on their TCS (≤ and > 3 days) as indicated in the Methods section. Figure 1 shows the distribution of plasma concentration values for the ten different cytokines measured. Despite the overlap in values between the two groups, patients in the TCS>3d group had statistically significantly higher plasma levels (pFigure 2), there was a tendency for higher cytokine levels in the TCS≤3d group in the case of the pro-inflammatory cytokines, particularly IL-1b (p=0.11), TNFa, (p=0.06) and IFNg (p=0.08), but no statistically significant differences were recorded.

Fig. 1 Plasma cytokine distributions in CAP patient groups relative to the TCS. Blood samples from patients with TCS≤3d (n=26) or TCS>3d (n=16) were obtained shortly after admission to the hospital and assayed for the indicated cytokines and chemokines using a bead-based multiplex assay as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Due to non-normal distribution of cytokine scores, groups were compared using the Wilcoxon-Mann-Whitney test. *p < 0.05.

Fig. 2 Sputum cytokine distributions profiles in CAP patient groups relative to the TCS. Sputum samples from patients with TCS≤3d (n=8) or TCS>3d (n=7) were obtained shortly after admission to the hospital and processed and assayed for the indicated cytokines and chemokines using a bead-based multiplex assay as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Due to non-normal distribution of cytokine scores, groups were compared using the Wilcoxon-Mann-Whitney test.

Pro-Inflammatory Cytokine Ratio (PICR).

In order to integrate systemic and local inflammatory cytokine response profiles in CAP patients, a ratio based on the relative levels of the eight pro-inflammatory cytokines and chemokines measured in plasma vs. sputum was calculated. As a first step, the distributions of the concentration levels for all cytokines measured in plasma (n=40) and sputum samples (n=15) were analyzed and ranked in quartiles. The cytokine values (in pg/mL) for the 25% percentile, median and 75% percentile for each of the measured cytokines in plasma and sputum are shown in Table 1 and Supplementary Figs. 1-2).

Table 1 Distribution of ranked cytokine values in plasma and sputum

Supplementary Fig. 1 Distribution of cytokine values in plasma as shown in Table 1. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Whiskers represent the 10 and 90 percentiles. Cytokine data from all 40 patients was included.

Supplementary Fig. 2 Distribution of cytokine values in sputum as shown in Table 1. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Whiskers represent the 10 and 90 percentiles. Cytokine data from all 15 patients with sputum samples was included.

Next, for the 15 patients having both plasma and sputum data available, each of the eight pro-inflammatory cytokines was given a score of 1 through 4, according to the quartile it belonged (e.g., 1 for 75%). Then, the eight cytokine scores in plasma and sputum were added and the total defined as the “Pro-inflammatory cytokine score (PICS)” for the corresponding fluid. Finally, the PICS in plasma was divided by that in the sputum to obtain the Pro-inflammatory cytokine ratio (PICR). The values thus calculated for each of the 15 patients are shown in Table 2, along with their relevant clinical information. The TCS≤3d group (n=8) included five patients in PSI risk classes I-III and three in PSI risk classes IV-V; whereas the TCS>3d group (n=7) included two patients in PSI risk classes I-III and five in PSI risk classes IV-V.

Table 2 Characteristics of CAP patients with both Plasma and Sputum Cytokine Data

Integration of Cytokine Response Profiles

In order to compare the combined cytokine response patterns in plasma and sputum among the patients and to investigate their association with measures of CAP outcomes (TCS) and severity (PSI), we developed an initial model based on a score, the PICR, calculated based on the ranked cytokine scores in plasma divided by sputum. The calculated PICR values for each patient are shown in Table 2. In addition, Figure 3 shows the distribution of PCIR for the patients grouped by TCS (≤ and >3 days) or by TSI group (classes I-III and IV-V). Comparison of the groups indicated that there were statistically significant differences in PICR values in both cases. Patients in the TCS>3d group had significantly higher PICR compared to the TCS≤3d group (p=0.009) and patients in PSI class IV-V had higher PICR values compared to those in PSI classes I-III (p=0.042).

Fig. 3Distribution of PICR scores in patients grouped by TCS (≤3d or >3d) or by PSI risk class (I-III and IV-V). PICR scores were calculated by dividing pro-inflammatory cytokine scores for plasma by those in sputum as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Groups were compared using the unpaired t-test. *p

Predictive value of PICR scores

Based on the statistically significant differences found, ROC curves were constructed in order to analyze the ability of PICR scores to predict CAP patient outcomes or severity. As shown in Figure 4, ROC curves analyzing the ability of PICR to discriminate between those patients with a TCS≤3d vs. >3d showed area under the curve (AUC) values of 0.893 (p=0.011), consistent with a very good level of accuracy. The cut-point optimizing specificity and sensitivity (Youden’s Index) based on this graph is a PICR of 0.96, which would give a sensitivity of 85.7% and a specificity of 87.5%. PICR values were less accurate, however, at discriminating patients with non-severe vs. severe CAP, as defined by PSI class (AUC: 0.786; p=0.064).

Fig. 4 ROC curves comparing the accuracy of PICR values to separate patients by TCS (≤3d or >3d) or PSI (I-III and IV-V) group. AUC values are indicated in the graphs.

DISCUSSION

Results from this study indicate that, similarly to patients severe CAP, patients with a TCS>3 days have higher median plasma levels of several pro-inflammatory (IL-6, IL-8) and anti-inflammatory (IL-10) cytokines compared to those with shorter TCS. Moreover, our exploratory results suggest that it may be possible to analyze and integrate cytokine response “patterns” in CAP patients based on sputum and plasma cytokine levels assayed at hospital admission. These “patterns” or “scores” (i.e., PICR) may potentially be used as predictive tools for measures of outcomes and disease severity, identifying patients at risk to have a delayed TCS or a more severe disease. With one exception (patient 4), all of the patients in the TCS≤3d group had PICR values less than 1, suggesting a relatively stronger local compared to systemic response. The only exception was a patient with Haemophilus influenzae who had strong systemic and local cytokine responses, yet had a TCS of 2d. On the other hand, five out of the seven patients with a TCS>3d had PICR values > 1.00, with the two exceptions being patients with infections by Pseudomonas aeruginosa and Acinetobacter baumannii.

Thanks to the current multiplex technology, it is now possible to measure multiple cytokines in the same sample, resulting in the generation of a substantial amount of data. However, it is not yet clear what the best approach is to integrate all of the cytokine information. Even in our small study, it was evident from comparing the different individual cytokine patterns that considerable variability existed in the cytokine responses of the different patients. Our results suggest, indeed, that analysis of the relationship between the local and systemic cytokine responses may have some value in predicting CAP disease severity and outcomes and that it might help to make predictions and guide potential therapeutic anti-inflammatory and/or immuno-modulatory interventions even upon admission.

Much of the evidence linking the deleterious effects of an over-abundant inflammatory response with the severity of CAP has been based on circulating, and thus systemic, cytokine levels [5-8]. However, local inflammatory responses have been less often studied and assumed to be reflected in the systemic cytokine levels. In a previous study [19], as well as the present study, the distribution of the levels of the different pro-inflammatory cytokines in plasma supports a more pronounced systemic inflammatory response in patients with severe CAP or with a TCS>3d. However, cytokine responses in the sputum did not follow the same trend as in plasma, suggesting that these two compartments (lung vs. systemic) do not necessarily mirror each other.

Our model to integrate the local and systemic inflammatory phenotypes in CAP patients was based on the ranked levels of eight pro-inflammatory cytokines/chemokines. When analyzing the data, it was clear that not all cytokines in the same patient followed the same trends nor that all patients had the same cytokine patterns. Given the different etiologies and underlying factors in the different CAP patients, this was not entirely surprising. Indeed, there were differences of inflammatory phenotypes when comparing patients with CAP due to different etiological agents, suggesting that the inflammatory pattern might be influenced by the etiologic agent, in agreement with previous reports [24]. The effect of the etiologic agent in our model remains to be further investigated. Moreover, while the pro-inflammatory cytokine response was the initial focus of our model for this study, the involvement of anti-inflammatory cytokine responses needs also to be considered.

Clearly, the main limitation of this study is the sample size. Due to the limited size in this study, our exploratory model needs to be tested and validated against a more substantial number of patients. In this study, we only evaluated eight pro-inflammatory cytokines. Since other cytokines may play an important role in defining inflammatory phenotypes, including anti-inflammatory cytokines, future studies should explore additional pro- and anti-inflammatory cytokines. Additional limitations that may have had affected the measurements include the fact that our model is based on a mathematical analysis of cytokine concentrations. However, given the potential interrelationships among the different cytokines, their effect in vivo is much more difficult, if not impossible, to measure with the current knowledge. Regarding the patients, pre-hospital admission of steroids and/or antibiotics may have influenced their responses and will have to be documented for future studies. Moreover, pre-existing comorbidities, some of which may induce enhanced systemic inflammation, may have also the potential to influence cytokine response patterns. Thus it will be important to characterize how these variables may affect systemic and local cytokine responses and thus disease severity and patient outcomes.

A main strength of this study is the fact that both local and systemic cytokine responses were measured at the time of the patient’s admission, allowing the comparison of the two and their correlation with measures of disease severity and clinical outcomes. While still preliminary, our results suggest the feasibility to use cytokine data at hospital admission to help identify those patients at higher risk. After further validation, such an approach may facilitate a better understanding of the relationship between local and systemic inflammatory responses and a more individualized therapeutic approach to modulate the inflammatory response in hospitalized patients with CAP.

REFERENCES

  1. G. L. Armstrong, L. A. Conn, and R. W. Pinner, “Trends in infectious disease mortality in the united states during the 20th century,” JAMA, vol. 281, no. 1, pp. 61–66, 1999.
  2. J. Almirall, I. Bolibar, J. Vidal, G. Sauca, P. Coll, B. Niklasson, M. Bartolome, and X. Balanzo, “Epidemiology of community-acquired pneumonia in adults: a population-based study,” European Respiratory Journal, vol. 15, no. 4, pp. 757–763, 2000.
  3. A. Craig, J. Mai, S. Cai, and S. Jeyaseelan, “Neutrophil recruitment to the lungs during bacterial pneumonia,” Infection and immunity, vol. 77, no. 2, pp. 568–575, 2009.
  4. U. Kolling, F. Hansen, J. Braun, L. Rink, H. Katus, and K. Dalhoff, “Leukocyte response and anti-inflammatory cytokines in community acquired pneumonia,” Thorax, vol. 56, no. 2, pp.121–125, 2001.
  5. G. Antunes, S. Evans, J. Lordan, and A. Frew, “Systemic cytokine levels in community-acquired pneumonia and their association with disease severity,” European Respiratory Journal, vol. 20, no. 4, pp. 990–995, 2002.
  6. J. Bordon, S. Aliberti, R. Fernandez-Botran, S. M. Uriarte, M. J. Rane, P. Duvvuri, P. Peyrani, L. C. Morlacchi, F. Blasi, and J. A. Ramirez, “Understanding the roles of cytokines and neutrophil activity and neutrophil apoptosis in the protective versus deleterious inflammatory response in pneumonia,” International Journal of Infectious Diseases, vol. 17, no. 2, pp. e76–e83, 2013.
  7. R. Menéndez, R. Martinez, S. Reyes, J. Mensa, X. Filella, M. A. Marcos, A. Martinez, C. Esquinas, P. Ramirez, and A. Torres, “Biomarkers improve mortality prediction by prognostic scales in community-acquired pneumonia,” Thorax, vol. 64, no. 7, pp. 587–591, 2009.
  8. P. Ramírez, M. Ferrer, V. Martí, S. Reyes, R. Martínez, R. Menéndez, S. Ewig, and A. Torres, “Inflammatory biomarkers and prediction for intensive care unit admission in severe community-acquired pneumonia,” Critical care medicine, vol. 39, no. 10, pp. 2211–2217, 2011.
  9. D. Rittirsch, M. A. Flierl, and P. A. Ward, “Harmful molecular mechanisms in sepsis,” Nature Reviews Immunology, vol. 8, no. 10, pp. 776–787, 2008.
  10. J. R. Tisoncik, M. J. Korth, C. P. Simmons, J. Farrar, T. R. Martin, and M. G. Katze, “Into the eye of the cytokine storm,” Microbiology and Molecular Biology Reviews, vol. 76, no. 1, pp. 16–32, 2012.
  11. S. Chollet-Martin, P. Montravers, C. Gibert, C. Elbim, J. M. Desmonts, J. Y. Fagon, and M. A. Gougerot-Pocidalo, “Subpopulation of hyperresponsive polymorphonuclear neutrophils in patients with adult respiratory distress syndrome. role of cytokine production.” Am Rev Respir Dis, vol. 146, pp. 990–996, 1992.
  12. J. Grammes and O. Soehnlein, “Contribution of neutrophils to acute lung injury,” Molecular medicine, vol. 17, no. 3, p. 293, 2011.
  13. M. Confalonieri, R. Urbino, A. Potena, M. Piattella, P. Parigi, G. Puccio, R. Della Porta, C. Giorgio, F. Blasi, R. Umberger et al., “Hydrocortisone infusion for severe community-acquired pneumonia: a preliminary randomized study,” Am J Respir Crit Care Med, vol. 171, no. 3, pp. 242–248, 2005.
  14. C. Garcia-Vidal, E. Calbo, V. Pascual, C. Ferrer, S. Quintana, and J. Garau, “Effects of systemic steroids in patients with severe community-acquired pneumonia,” European Respiratory Journal, vol. 30, no. 5, pp. 951–956, 2007.
  15. S. Fernández-Serrano, J. Dorca, C. Garcia-Vidal, N. Fernández-Sabé, J. Carratalà, A. Fernández-Agüera, M. Corominas, S. Padrones, F. Gudiol, and F. Manresa, “Effect of corticosteroids on the clinical course of community-acquired pneumonia: a randomized controlled trial,” Critical Care, vol. 15, no. 2, p. R96, 2011.
  16. S. C. Meijvis, H. Hardeman, H. H. Remmelts, R. Heijligenberg, G. T. Rijkers, H. van Velzen-Blad, G. P. Voorn, E. M. van de Garde, H. Endeman, J. C. Grutters et al., “Dexamethasone and length of hospital stay in patients with community-acquired pneumonia: a randomised, double-blind, placebo-controlled trial,” The Lancet, vol. 377, no. 9782, pp. 2023–2030, 2011.
  17. R. Siemieniuk, M. O. Meade, P. Alonso-Coello, M. Briel, N. Evaniew, M. Prasad, P. E. Alexander, Y. Fei, P. O. Vandvik, M. Loeb et al., “Corticosteroid therapy for patients hospitalized with community-acquired pneumonia. a systematic review meta-analysis.” Ann Intern Med, vol. 163, pp. 519–528, 2015.
  18. A. Torres, Martí, O. Sibila Vidal, M. Ferrer, E. Polverino, R. Menéndez, J. Mensa Pueyo, A. Gabarrús, J. Sellarés, M. I. Restrepo, A. Anzueto et al., “Effect of corticosteroids on treatment failure among hospitalized patients with severe community-acquired pneumonia and high inflammatory response: a randomized clinical trial.” JAMA, vol. 313, no. 7, pp. 677–686, 2015.
  19. R. Fernandez-Botran, S. M. Uriarte, F. W. Arnold, L. Rodriguez-Hernandez, M. J. Rane, P. Peyrani, T. Wiemken, R. Kelley, S. Uppatla, R. Cavallazzi et al., “Contrasting inflammatory responses in severe and non-severe community-acquired pneumonia,” Inflammation, vol. 37, no. 4, pp. 1158–1166, 2014.
  20. A. Balows, W. J. Haussler, K. L. Herrmann, H. D. Isenburg, and H. J. Shadomy, Manual of Clinical Microbiology, 5th ed. Washington, DC: American Society of Microbiology, 1991, pp. 147–150.
  21. C. Marti, N. Garin, O. Grosgurin, A. Poncet, C. Combescure, S. Carballo, and A. Perrier, “Prediction of severe community-acquired pneumonia: a systematic review and meta-analysis,” Critical Care, vol. 16, no. 4, p. R141, 2012.
  22. S. Aliberti, A. M. Zanaboni, T.Wiemken, A. Nahas, S. Uppatla, L. C. Morlacchi, P. Peyrani, F. Blasi, and J. Ramirez, “Criteria for clinical stability in hospitalised patients with community-acquired pneumonia,” European Respiratory Journal, vol. 42, no. 3, pp. 742–749, 2013.
  23. E. Pizzichini, M. Pizzichini, A. Efthimiadis, S. Evans, M. M. Morris, D. Squillace, G. J. Gleich, J. Dolovich, and F. E. Hargreave, “Indices of airway inflammation in induced sputum: reproducibility and validity of cell and fluid-phase measurements.” American journal of respiratory and critical care medicine, vol. 154, no. 2, pp. 308–317, 1996.
  24. R. Menéndez, J. M. Sahuquillo-Arce, S. Reyes, R. Martínez, E. Polverino, C. Cillóniz, J. G. Córdoba, B. Montull, and A. Torres, “Cytokine activation patterns and biomarkers are influenced by microorganisms in community-acquired pneumonia,” CHEST, vol. 141, no. 6, pp. 1537–1545, 2012.

DOI

10.18297/jri/vol1/iss3/5/

jri008-Fig_1.pdf (38 kB)
Fig. 1 Plasma cytokine distributions in CAP patient groups relative to the TCS. Blood samples from patients with TCS≤3d (n=26) or TCS>3d (n=16) were obtained shortly after admission to the hospital and assayed for the indicated cytokines and chemokines using a bead-based multiplex assay as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Due to non-normal distribution of cytokine scores, groups were compared using the Wilcoxon-Mann-Whitney test. *p < 0.05.

jri008-Fig_2.pdf (40 kB)
Fig. 2 Sputum cytokine distributions profiles in CAP patient groups relative to the TCS. Sputum samples from patients with TCS≤3d (n=8) or TCS>3d (n=7) were obtained shortly after admission to the hospital and processed and assayed for the indicated cytokines and chemokines using a bead-based multiplex assay as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Due to non-normal distribution of cytokine scores, groups were compared using the Wilcoxon-Mann-Whitney test.

jri008-Table_1.png (43 kB)
Table 1 Distribution of ranked cytokine values in plasma and sputum

jri008-Sup_Fig_1.pdf (42 kB)
Supplementary Fig. 1 Distribution of cytokine values in plasma as shown in Table 1. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Whiskers represent the 10 and 90 percentiles. Cytokine data from all 40 patients was included.

jri008-Sup_Fig_2.pdf (50 kB)
Supplementary Fig. 2 Distribution of cytokine values in sputum as shown in Table 1. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Whiskers represent the 10 and 90 percentiles. Cytokine data from all 15 patients with sputum samples was included.

jri008-Table_2.png (68 kB)
Table 2 Characteristics of CAP patients with both Plasma and Sputum Cytokine Data

jri008-Fig_3.pdf (124 kB)
Fig. 3 Distribution of PICR scores in patients grouped by TCS (≤3d or >3d) or by PSI risk class (I-III and IV-V). PICR scores were calculated by dividing pro-inflammatory cytokine scores for plasma by those in sputum as indicated in Materials and Methods. Box plots depict the 25-75% interquartile range and the horizontal bar depicts the median. Groups were compared using the unpaired t-test. *p < 0.05; ** p<0.01

jri008-Fig_4.pdf (19 kB)
Fig. 4 ROC curves comparing the accuracy of PICR values to separate patients by TCS (≤3d or >3d) or PSI (I-III and IV-V) group. AUC values are indicated in the graphs.

jri008-Fig_1-thumbnail.png (41 kB)
jri008-Fig_2-thumbnail.png (36 kB)
jri008-Table_1-thumbnail.png (63 kB)
jri008-Sup-Fig_1-thumbnail.png (35 kB)
jri008-Sup-Fig_2-thumbnail.png (34 kB)
jri008-Table_2-thumbnail.png (48 kB)
jri008-Fig_3-thumbnail.png (22 kB)
jri008-Fig_4-thumbnail.png (23 kB)
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.