Associations Between Procalcitonin and Markers of Bacterial Sepsis

Transcription

1 Clinical Investigations 383 :383-7 Associations Between Procalcitonin and Markers of Bacterial Sepsis Veeresh K. Patil 1, Jaymin B. Morjaria 2, Francois De Villers 1, Suresh K. Babu 2 1 Inverclyde Royal Hospital, Greenock, Scotland, 2 University of Southampton, Southampton, United Kingdom Key words: procalcitonin; blood culture; bacterial sepsis; C-reactive protein; blood neutrophil; blood white cell count. Summary. Background. Bacterial sepsis with no bacterial isolates can be a difficult clinical conundrum, where other markers like C-reactive protein (CRP), white cell count (WCC), and neutrophilia are helpful to arrive at a diagnosis. Procalcitonin (PCT) has been shown to be a useful biomarker in bacterial sepsis. The aim of the study was to look at the association of PCT with bacterial cultures and compare this to currently used markers of bacterial sepsis. Material and Methods. WCC, neutrophil count, and CRP with PCT were compared in patients with a positive bacterial culture from blood/body fluid. The specificity and sensitivity of PCT were compared with those of CRP. Results. Of the 99 paired samples obtained, 25 cultures were positive for bacteria. There was a significant difference in CRP (P=.4) and PCT (P<.1) levels between culture-positive and culture-negative samples. PCT had a better sensitivity and specificity than CRP (84% and 64.9% vs. 69.6% and 52.9%, respectively), with a combined specificity (CRP and PCT) of 83.5%. Conclusions. PCT has a better association with bacterial sepsis and is superior to currently available biomarkers in the clinical setting. The rapid pharmacodynamics of PCT can serve as an early predictor of the diagnosis of bacterial sepsis while awaiting the bacterial culture results avoiding undue delay in the institution of antibiotics, hence, potentially improving the prognosis of patients with bacterial sepsis. Introduction Severe sepsis and septic shock affect millions of people every year and are associated with an everincreasing mortality (1). Bacterial sepsis is a treatable cause, but needs prompt identification, diagnosis, and treatment. Akin to other acute medical emergencies, the Surviving Sepsis Campaign (SSC) has suggested evidence-based recommendations in severe sepsis where the speed and appropriateness of the treatment given in the early hours significantly influence the consequential result. Patients with bacterial sepsis require antibiotics. In current routine practice, a variety of parameters of infection, such as temperature, heart rate, respiratory rate, white cell count (WCC), and serum C-reactive protein (CRP) levels, are being used. However, these parameters offer limited information to differentiate between the diagnosis of bacterial infections, nonbacterial infections, and noninfective inflammatory conditions. Bacterial cultures are considered the gold standard method for the detection of bacterial infection. Conventional microbiology cultures, despite their specificity and accuracy, are time consuming, and a negative result in many cases of bacterial Correspondence to S. K. Babu, Level C, Respiratory Medicine, Queen Alexandra Hospital, Cosham, Portsmouth, PO6 3LY, United Kingdom. ksb@soton.ac.uk sepsis does not exclude an infective etiology. This leaves the physician in a dilemma as to whether the patient needs antibiotics or not. Newer molecular technologies are emerging as promising tests for use into routine clinical laboratories. Hence, a test that would provide a prompt and dependable result to rule out bacterial sepsis would be valuable not only in deciding the appropriateness of prescribing antibiotics, but also would be cost effective in terms of hospital stay and medical costs (2). Procalcitonin (PCT) is a 116-amino acid prohormone of calcitonin that is physiologically produced and secreted in the thyroid gland. Normally, calcitonin, which is proteolytically processed from PCT, has a half-life of a few minutes in the circulation. PCT levels have been noted to be elevated in the circulation in patients with bacterial sepsis, but not in viral infections, allergic reactions, autoimmune conditions, and graft-versus-host disease (3 6). PCT in bacterial sepsis is considered to be synthesized from an extrathyroid source (7), and the elevated systemic levels are due to the failure of suitable proteolysis (6). PCT has a half-life of 24 to 3 hours in the circulation (6) in contrast to other markers of sepsis, namely tumor necrosis factor alpha (TNF-α) or interleukin 6 (IL-6), which return to baseline within 6 to 8 hours. CRP takes 12 to

2 384 Veeresh K. Patil, Jaymin B. Morjaria, Francois De Villers, Suresh K. Babu 24 hours to rise and 2 to 72 hours to reach a plateau and remains elevated for 3 to 7 days. This is 2 to 3 days longer than it takes PCT concentrations to normalize, offering PCT a natural advantage in monitoring the disease (8). PCT has been considered a marker of sepsis, but the evidence is not as clear-cut. A meta-analysis by Tang et al. suggested that PCT, in the critical care setting, was not a reliable marker to differentiate between sepsis and noninfectious causes of systemic inflammatory response syndrome (SIRS) (9), while other study by Uzzan et al. was contrary to this (1). We have earlier shown that a semiquantitative assay for PCT was useful in the diagnosis of systemic bacterial sepsis with a predominance of gram-positive organisms (11). Schuetz et al. have suggested that PCT can potentially help in discriminating gram-negative septicemia from blood contamination (12). The aim of this retrospective observational study was to evaluate the specificity of PCT in comparison with the current markers of inflammation currently used as adjuncts to diagnose bacterial infection/sepsis, namely WCC, neutrophil count, and CRP, and to examine if PCT can reliably recognize and identify bacterial sepsis compared with the gold standard of body fluid cultures. Material and Methods Consecutive acute admissions of adult patients with a presumed diagnosis of bacterial sepsis between July 27 and June 28 at the Inverclyde Royal Hospital, a district general hospital in Glasgow, were assessed in this observational study. The subjects had clinical evidence of infection associated with a temperature of >38 C or <36 C, tachycardia, tachypnea, or WBC of >12 cells/mm 3 or <4 cells/mm 3. The data were collected retrospectively from these patients. Serum PCT, WCC, neutrophil count, and CRP were examined and compared with culture reports available from these patients. PCT levels were assayed with a VIDAS- B-R-A-H-M-S enzyme-linked fluorescent assay (BioMérieux Inc, USA) as per the manufacturer s instructions. The blood samples for WCC, neutrophil count, and CRP and PCT levels were analyzed. The results were then compared with the results of cultures obtained from blood and peritoneal and pleural fluid. The cultures were done on liquid media (Robertsons meat broth) and incubated for 5 days before it was considered negative. A bactec 924 system was used for blood cultures and VITEK 2 (BioMérieux, Inc, USA) for identification of blood culture isolates and susceptibilities. If VITEK 2 identification/susceptibility failed or was considered inappropriate for a specific organism, then analytical profile index identification and disk diffusion/e-testing were used as per Clinical and Laboratory Standards Institute standards. Cultures from urine, sputum, tracheal aspirates, and wound swabs were excluded, as these are likely to represent local infections rather than systemic sepsis. Of note, all the analyses were conducted by the Inverclyde Royal Hospital laboratories, and the laboratory results were strictly quality controlled. The results were grouped based on a positive or negative bacterial culture. Serum PCT and CRP levels, WCC, and neutrophil count were compared between the groups. The data were analyzed using PASW Statistics 18 statistical package (SPSS, Chicago, USA) and SigmaPlot 11. (Systat Software Inc, San Jose, USA). The data were expressed as median (interquartile range [IQR]) due to their nonparametric nature. Similarly, nonparametric tests were used for comparing the data with a P value of <.5, which was considered statistically significant. Receiver operating characteristic (ROC) curves were constructed to determine the sensitivity, specificity, and area under the curve (AUC) for PCT and CRP in sepsis. Results A total of 348 PCT levels were analyzed in a period of 12 months. Paired (PCT and other markers of infection) and culture results (any) were available for 135 patients, out of which 99 had blood, pleural, or peritoneal fluid cultures; 36 patients had culture results from urine, sputum, tracheal aspirate, wound swab, and stools, which were excluded from the analysis (Fig. 1). Coagulase-negative Staphylococcus isolates were also excluded from the analysis to ensure that results likely due to contamination were excluded. The 99 paired results of blood, pleural, and peritoneal cultures were used for final analysis. The demographics are depicted in Table. The median total WCC, neutrophil count, and CRP and PCT levels were /L (IQR, ), /L (IQR, ), 158 mg/l (IQR, 9 247), and 2 ng/l (IQR, ), respectively. Of the 99 paired samples, 25 were from the patients who had a positive culture. It was noted that there were no significant differences in the total WCC or neutrophil count between the culturepositive and culture-negative groups (Table, Figs. 2 4). The median CRP level in the culture-positive group was 212 mg/l (IQR, ), while in the culture-negative group, it was 134 mg/l (IQR, ) (P=.4). The median PCT level in the culture-positive group was 8.66 ng/l (IQR, ), and in the culture-negative group, it was 1.3 ng/l (IQR, ) (P<.1). Of those with culture-positive results, gram-positive organisms were isolated in 13 of the 25 cultures (52%), gram-negative isolates in 4 of the 25 cultures (16%), and the remaining 8 cultures (32%) had mixed flora. The median PCT was 32.9 ng/l (IQR, ) in the gram-negative group, 2.8 ng/l (IQR, ) in the gram-positive group, and 4.2 ng/l (IQR, ) in the mixed or-

3 Associations Between Procalcitonin and Markers of Bacterial Sepsis 385 Table. Markers of Infection in all the Samples and in the Culture- and Culture- Groups Parameter All Samples n=99 Culture- n=25 Culture- n=74 Age, years 62. ( ) 66.5 ( ) 59. (45 71) Total WCC, 1 9 /L 9.9 ( ) 1.1 ( ) 11.6 ( ) Neutrophils, 1 9 /L 7.5 ( ) 8.7 ( ) 8.51 ( ) CRP, mg/l 158 (9 247) 212 ( ) 134 (68 238) PCT, ng/l 2 ( ) 8.66 ( ) 1.3 ( ) Values are median (interquartile range). WCC, white cell count; CRP, C-reactive protein; PCT, procalcitonin. P > < Total PCT levels in 12 months 6 P= Culture results available 213 No paired cultures available CRP, mg/l Included Blood cultures, positive pleural and peritoneal 36 Excluded Urine, sputum, tracheal aspirate, wound, stools (no blood culture) 74 (75%) culture negative 25 culture positive Fig. 3. Box-and-whisker plots for C-reactive protein (CRP) levels in the culture-positive and culture-negative groups 25 2 P<.1 Fig. 1. Consort diagram of procalcitonin and cultures considered in the study PCT, ng/l P=.43 P= /L 3 2 Fig. 4. Box-and-whisker plots for procalcitonin levels (PCT) in the culture-positive and culture-negative groups 1 WWC WWC Neutrophil Neutrophil Fig. 2. Box-and-whisker plots for white cell count (WCC) and neutrophil counts in the culture-positive and culture-negative groups ganism group. There were no significat differences comparing the groups. ROC curve analyses for PCT and CRP were performed to assess the sensitivity and specificity of these tests (Fig. 3). A PCT value of 2.42 ng/l demonstrated a sensitivity and specificity of 84% and 64.9%, respectively, and CRP levels of 15 mg/l had a sensitivity of 69.6% and a specificity of 52.9%. The combined specificity of PCT and CRP was 83.5%. The area under the ROC curve (AUC) for PCT was

4 386 Veeresh K. Patil, Jaymin B. Morjaria, Francois De Villers, Suresh K. Babu Sensitivity PCT CRP Specificity Fig. 5. Receiver operating characteristic (ROC) curves for procalcitonin (PCT) and C-reactive protein (CRP).78 (95% CI,.67.88), which was higher than for CRP, i.e.,.64 (95% CI,.52.76; P=.9). Discussion Blood culture is the most reliable way of identifying bacterial sepsis; however, the low positive detection rates and the time delay in obtaining results are major stumbling blocks, as an early diagnosis is key to bacterial sepsis management. PCT is believed to be a primitive form of antibacterial defense, which has emerged as a useful biomarker in the diagnosis of bacterial sepsis and bacteremia (13). In the context of initiating empirical or presumptive antibiotic treatment, PCT assays can be a useful diagnostic adjunct to physicians in an acute setting (14). The present study aimed to correlate currently used markers of inflammation, namely WCC, neutrophil count, and CRP and PCT levels with positive cultures, the gold standard for the diagnosis of bacterial sepsis. The study was conducted in consecutive adult admissions with an initial clinical diagnosis of sepsis in a district general hospital. The subjects had blood/body fluid cultures, PCT, and inflammatory indices monitored and were not confounded by age, site, or source of sepsis. Thus, this study acted as a real-life retrospective study in the day-to-day management of patients in a field situation. The study has shown that the PCT levels are strongly related to a positive bacterial culture from body fluids and blood. Furthermore, it has also been shown that other inflammatory markers like WCC and neutrophil count do not clearly show a relationship with a positive culture. While CRP was found to be elevated in patients with a positive culture, the sensitivity and specificity were 69.6% and 52.9% with a cutoff value of 15 mg/l, the ROC analysis for the identification of bacterial sepsis was better for PCT compared with CRP. Although still far from ideal, for all the biomarkers available currently in clinical use for bacterial sepsis, PCT is thought to have the best ROC curve. Interestingly, the combined specificity was 83.5%. While the optimal cutoff value for PCT from 25 studies determined from ROC curves ranged from.6 to 5 ng/l, our value of 2.42 ng/l gave us a reasonable sensitivity and specificity (1). There was a predominance of gram-positive organisms both in the current study and in our earlier study (11); however, other studies have reported variable results (15). The current study did not look at the levels of PCT in the various groups as the sample size was small and it would be difficult to interpret; however, the study by Charles and colleagues suggested that in critically ill patients, higher PCT values were associated with gram-negative sepsis (16). In the current study, the group with gram-negative bacteremia had the higher levels of PCT, but these differences were not statistically significant, which reflects the small numbers of gram-negative isolates we had. Furthermore, the data could have been skewed by the significant numbers of mixed cultures, thus, making it difficult to interpret. Two widely quoted reviews and meta-analyses have come to differing conclusions (9, 1). The study by Tang et al. suggested that PCT could not reliably differentiate sepsis from other noninfectious causes of SIRS in critically ill adult patients (9). The ProHOSP study concluded that PCT was helpful during follow-up and for predicting adverse events, hence, improving mortality (17). Uzzan and colleagues concluded that PCT represented a good biological diagnostic marker for sepsis, severe sepsis, or septic shock, which are difficult diagnoses in critically ill patients. It was superior to CRP and should be included in diagnostic guidelines for sepsis and in clinical practice in intensive care units (1). Recently, the Procalcitonin and Survival Study (PASS) has reported that the use of PCT does not have any effect on mortality, reduced antibiotic use, and length of stay in the intensive care unit (18). With no clear inferences drawn from these randomized studies regarding the role of PCT in sepsis, the jury is still open. The current study has limitations. Firstly, it is essential to acknowledge that patients with bacterial infections can have negative cultures. In the current study, we only looked at the positive cultures to establish a place for PCT as a biomarker of bacterial sepsis in everyday clinical use. Secondly, this work is a retrospective study of consecutive patients over a year, who had blood cultures, PCT, WCC, and CRP checked, and, hence, is subject to a recruitment bias. Despite having a selection bias, the reallife nature of this study makes it unique. Thirdly,

5 Associations Between Procalcitonin and Markers of Bacterial Sepsis 387 only those patients who had positive blood/body fluid cultures were considered in the analysis. This is because there is always a possibility that some of the patients excluded like patients with other positive cultures (tracheal aspirate, sputum, etc.) could have had sepsis from these sites or contamination, and, hence, provide spurious results. The aim of the current study was to establish the role of PCT in confirmed positive cultures, which remains the gold standard for bacteremia and bacterial sepsis. Moreover, 8 of the 25 positive cultures were mixed grampositive and gram-negative, which may have been considered as contamination and excluded, but we felt that based on the individual patient s clinical condition, the infections were valid and, hence, included them in the analysis. Conclusions The present study has shown that PCT correlates with blood culture results better than currently used tests, i.e., WCC, neutrophil count, and CRP. PCT is a good predictive biomarker of bacterial sepsis with a sensitivity of 85%, but in combination with CRP, the specificity improves to over 83%. While there was a predominance of gram-positive isolates, this may be due to the small sample size taken into consideration; however, this could possibly provide a useful guide for empirical antibiotic use. Nevertheless, our view is that the definite diagnosis of bacterial sepsis/ bacteremia may not rely on a single PCT result, but on a complete clinical and laboratory evaluation of the patient with PCT playing an important role. The rapid pharmacodynamics of PCT may correlate well with the onset of bacterial sepsis and can be a useful tool to assess whether an infection is evolving and responding favorably to antibiotics and detect superadded infection in patients hospitalized for another cause, hence, making it a desirable prognostic biomarker of bacterial sepsis. Of paramount importance is the fact that when using any biomarker of infection, be it PCT or another, the conclusions drawn from the results must always be taken in the context of the clinical picture. Contributions V.P., F.V., and K.S.B. contributed to the data collection. J.B.M., F.V., and K.S.B. contributed to analysis and interpretation of the data. J.B.M. and K.S.B. drafted the manuscript and revised it critically for intellectual content. V.P. and J.B.M. contributed equally to the study. Statement of Conflict of Interest The authors state no conflict of interest. References 1. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 28. Intensive Care Med 28;34(1): Christ-Crain M, Jaccard-Stolz D, Bingisser R, Gencay MM, Huber PR, Tamm M, et al. Effect of procalcitonin-guided treatment on antibiotic use and outcome in lower respiratory tract infections: cluster-randomised, single-blinded intervention trial. Lancet 24;363(949): Gendrel D, Raymond J, Coste J, Moulin F, Lorrot M, Guerin S, et al. Comparison of procalcitonin with C-reactive protein, interleukin 6 and interferon-alpha for differentiation of bacterial vs. viral infections. Pediatr Infect Dis J 1999;18(1): Delevaux I, Andre M, Colombier M, Albuisson E, Meylheuc F, Begue RJ, et al. Can procalcitonin measurement help in differentiating between bacterial infection and other kinds of inflammatory processes? Ann Rheum Dis 23; 62(4): Ugarte H, Silva E, Mercan D, De Mendonca A, Vincent JL. Procalcitonin used as a marker of infection in the intensive care unit. Crit Care Med 1999;27(3): Meisner M, Tschaikowsky K, Schmidt J, Schuttler J. Procalcitonin (PCT) Indications for a new diagnostic parameter of severe bacterial infection and sepsis in transplantation, immunosuppression and cardiac assist devices. Cardiovasc Eng 1996;1: Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993;341(8844): Schneider HG, Lam QT. Procalcitonin for the clinical laboratory: a review. Pathology 27;39(4): Tang BM, Eslick GD, Craig JC, McLean AS. Accuracy of procalcitonin for sepsis diagnosis in critically ill patients: systematic review and meta-analysis. Lancet Infect Dis 27;7(3):21-7. Received 18 May 212, accepted 2 July Uzzan B, Cohen R, Nicolas P, Cucherat M, Perret GY. Procalcitonin as a diagnostic test for sepsis in critically ill adults and after surgery or trauma: a systematic review and metaanalysis. Crit Care Med 26;34(7): Patil VK, Kabir S, Morjaria JB, DeVilliers F, Babu KS. The value of a semi-quantitative serum procalcitonin assay in patients with sepsis. Am J Respir Crit Care Med 29; 179:A Schuetz P, Mueller B, Trampuz A. Serum procalcitonin for discrimination of blood contamination from bloodstream infection due to coagulase-negative staphylococci. Infection 27;35(5): Schuetz P, Christ-Crain M, Muller B. Procalcitonin and other biomarkers to improve assessment and antibiotic stewardship in infections hope for hype? Swiss Med Wkly 29;139(23-24): Guven H, Altintop L, Baydin A, Esen S, Aygun D, Hokelek M, et al. Diagnostic value of procalcitonin levels as an early indicator of sepsis. Am J Emerg Med 22;2(3): Koivula I, Hämäläinen S, Jantunen E, Pulkki K, Kuittinen T, Nousiainen T, et al. Elevated procalcitonin predicts gram-negative sepsis in haematological patients with febrile neutropenia. Scand J Infect Dis 211;43(6-7): Charles PE, Ladoire S, Aho S, Quenot JP, Doise JM, Prin S, et al. Serum procalcitonin elevation in critically ill patients at the onset of bacteremia caused by either gram negative or gram positive bacteria. BMC Infect Dis 28;8: Schuetz P, Suter-Widmer I, Chaudri A, Christ-Crain M, Zimmerli W, Mueller B. Prognostic value of procalcitonin in community-acquired pneumonia. Eur Respir J 211; 37(2): Jensen JU, Hein L, Lundgren B, Bestle MH, Mohr TT, Andersen MH, et al. Procalcitonin-guided interventions against infections to increase early appropriate antibiotics and improve survival in the intensive care unit: a randomized trial. Crit Care Med 211;39(9):