ORIGINAL ARTICLE
Role of Combined Procalcitonin and Lipopolysaccharide- binding Protein as Prognostic Markers of Mortality
in Patients with Ventilator-associated Pneumonia
Cleophas M. Rumende, Dinajani Mahdi
Department of Internal Medicine, Faculty of Medicine, University of Indonesia - Cipto Mangunkusumo Hospital, Jakarta, Indonesia.
Correspondence mail:
Division of Pulmonology, Department of Internal Medicine, Faculty of Medicine, University of Indonesia - Cipto Mangunkusumo Hospital, Jl. Diponegoro no. 71, Jakarta 10430, Indonesia. email: [email protected].
ABSTRAK
Tujuan: untuk mengetahui manfaat pemeriksaan prokalsitonin (PCT) sebagai prediktor dan petanda prognostik serta lipopolysaccharide-binding protein (LBP) sebagai petanda prognostik ventilator-associated pneumonia (VAP). Metode: penelitian dengan desain kohort prospektif kelompok tunggal mengambil subyek secara konsekutif dari pasien yang dirawat di ICU/HCU RSCM tahun 2006–2007. Didapatkan sebanyak 35 pasien dengan VAP. Analisis data menggunakan chi-square atau menggunakan uji Fisher exact. Berdasarkan penelitian sebelumnya, kami menggunakan nilai cut-off 5 ng/ml dan 0,5 ng/ml untuk PCT, 30 mg/ml dan 25 mg/ml untuk LBP setelah 3 hari dan 7 hari pemberian antibiotik. Receiving operating curve digunakan untuk mengetahui sensitifitas dan spesifisitas dari PCT dan LBP sebagai petanda infeksi.
Hasil: pada penelitian ini tidak didapatkan hubungan yang bermakna antara kadar PCT visit-1 dengan risiko terjadinya VAP (p=0,240). Setelah 3 hari pengobatan pasien dengan kadar PCT >5 ng/ml dan LBP
>30 ug/ml mempunyai prognosis yang lebih buruk (p<0,05) dengan sensitifitas 88,5%, spesifisitas 53,2%
dan AUC 0,69. Prognosis yang buruk juga didapatkan bila setelah 7 hari pengobatan didapatkan kadar PCT
>0,5 ng/ml dan LBP >25 ug/ml (p<0,05) dengan sensitifitas 96,3%, spesifisitas 66,7% dan AUC 0,81%.
Kesimpulan: pemeriksaan prokalsitonin dan LBP secara simultan bermanfaat sebagai petanda prognostik pada pasien VAP dengan nilai sensitivitas dan spesifisitas yang cukup baik.
Kata kunci: ventilator-associated pneumonia, prokalsitonin, lipopolysaccharide-binding protein.
ABSTRACT
Aim: to investigate the role of combined Lipopolysaccharide-Binding Protein (LBP) and Procalcitonin (PCT) ) as prognostic marker of mortality in patients with Ventilator-Associated Pneumonia (VAP).
Methods: this prospective cohort study was held in ICU/HCU of Cipto Mangunkusumo hospital between 2006 to 2007 by taking the subjects consecutively. Thirty five patients with VAP were studied. For analysing the data, chi-square or its alternative Fisher exact test were used. Based on a previous study for evaluation, we used cut off pants of 5 ng/ml and 0.5 ng/ml for PCT and 30 µg/ml and 25 µg/ml for LBP after three-day and seven-day treatment respectively. Receiver operating curve was made to determine the sensitivity and specificity of PCT and LBP as infection markers. Results: 35 patients participated in this study. After three days of therapy, if the level of PCT >5 ng/mL and LBP >30 µg/mL the prognosis would be bad (p<0.05) with a sensitivity of 88.5%, specificity of 53.2% and AUC value 0.69. Poor prognosis was also found if after seven day therapy PCT level was >0.5 ng/mL and LBP level >25 µg/mL (p<0.05) with sensitivity of 96.3%, specificity of 66.7% and AUC value 0.81.
INTRODUCTION
Ventilator-associated pneumonia (VAP) is a nosocomial infection commonly found in the intensive care unit (ICU) patients with the incidence of 11.7 per 1000 treatments on ventilator. The frequency of VAP in patients using ventilator is 8-28% with 24 to 50%
mortality rate.1,2 Ventilator-associated pneumonia is usually encountered among patients using ventilators for more than 48 hours due to bacterial infection.1,3-6 Lipopolysaccharide-binding protein (LBP) is a new marker useful as an infection marker to measure the severity of the infection and to evaluate the result of treatments, while Procalcitonin has been established as an infection marker due to its level increases in bacterial infections.7,8 In VAP, macrophage alveoli that have phagocytized microorganism will release pro-inflammation cytokine such as TNF-α and IL-1β that will stimulate macrophage cells and lung neuroendocrine to produce PCT. The increase of PCT in blood can also be the result of lipopolysaccharide (LPS) and peptidogly can stimulate in macrophages.6 Lipopolysaccharide- binding protein (LBP) is one of acute protein phase which is increased in the presence of IL-1 and IL-6 as a result of macrophage exposed to microorganism.9-12 To distinguish between SIRS with other septic conditions Charles-Eduard Luyt used PCT with limit value of 1.5 ng/mL and 0.5 ng/mL after three days and seven days of treatment, respectively.13 Prucha et al. reported the different levels of initial LBP between SIRS and severe sepsis/septic shock patients using a limit value of 29.8 µg/mL. After seven days of treatment there were differences of LBP level between patients survivals and non-survivals of 11.9 µg/mL and 26.4 µg/mL.14-16
The objective of this study is to investigate the role of PCT in conjunction with LBP as prognostic markers of mortality in VAP patients.
METHODS
This prospective cohort study was conducted from August 2006 to August 2007 at Cipto Mangunkusumo Hospital. Inclusion criteria:
patients with VAP who were admitted in ICU/
HCU. Diagnosis of VAP was established based on American Thoracic Society Guidelines. It was defined as a new or progressive infiltrate on chest radiography associated with at least two of the following: purulent tracheal secretions, fever (body temperatures 38oC), leukocytosis/- penia (leukocyte count >11,000/µL or,<3000/µL) and positive sputum culture. VAP patients were eligible for the study if they were intubated for mechanical ventilation for at least 48 hour. During the period of the study there were 35 patients with VAP. As prognostic markers, the evaluation of PCT and LBP were performed when VAP occured, then after 3 days and 7 days of treatment.
Quantitative PCT examination was carried out using BRAHM PCT sensitive KRYPTOR, while LBP used a device called Immulite DPC.
Antibiotic selection was given based on ATS/
IDSA 2005. All the data were analysed by using SPSS software with significance level accepted at p <0.05. For analysing the data, Chi-square or its alternative Fisher exact test were used. Based on a previous study we used a cut-off of 5 ng/ml and 0.5 ng/ml for PCT and 30 µg/ml and 25 µg/
ml for LBP to evaluate after three-day and seven- day treatment, respectively. Receiver operating curve was made to determin the sensitivity and specificity of either PCT or LBP as prognostic marker of VAP patients mortality.
RESULTS
From August 2006 to August 2007, as many as 129 patients were identified, 46 patients with VAP but those meeting the inclusion criteria were 35 patients. (Table 1)
Conclusion: examination of combined PCT and LBP can be taken as a good prognostic markers to predict mortality in patients with VAP.
Key words: ventilator-associated pneumonia, procalcitonin, lipopolysaccharide-binding protein.
After three days treatment for VAP, by using a cut-off value of 5 ng/mL it is found that patients with PCT level >5 ng/mL shows worse prognosis (p=0.003). This limit value shows the 80% sensitivity and 30% specificity with area under the curve (AUC) of 0.75.
After seven days treatment with using a cut-off value of 0.5 ng/mL it was shown that patients with PCT level >0.5 ng/mL after seven days of treatment would have worse prognosis (p<0.0001). This limit value shows 81.35%
sensitivity and 16.7% specificity with AUC of 0.884.
After three days treament by using limit value of 30 µg/mL it is shown that patient with LBP level
>30 µg/mL shows worse prognosis (p=0.036).
This cut-off value shows 60% sensitivity and 25%
specificity with AUC of 0.675.
After seven days of treatment, using cut-off value of 25 µg/mL it is shown that patients with LBP level >25 µg/mL have a worse prognosis (p=0.005). As a prognostic marker, visit-4 LBP cut-off value shows 73.3% sensitivity and 25%
specificity with AUC of 0.74.
Combined analysis of the two levels of PCT visit-2 (cut-off value 5 ng/mL) and LBP visit-2 (cut-off value 30 µg/mL) to evaluate the prognosis shows that patients with PCT and LBP levels above aforementioned values have worse prognosis (p=0.021). (Table 2) The limit value of PCT and LBP show 88.5% sensitivity and 53.2% specificity with AUC of 0.693. (Figure 1)
Table 1. The clinical characteristics of patients
Characteristics Value
Sex n (%)
- Male 24 (69)
- Female 11 (31)
Age (years)
- Mean (SD) 38,3 (SD 14)
- Confidence Interval CI95% 14-77
Reintubation; n (%) 23 (65)
Risk factor n (%)
- COPD 2 (5,7)
- Reintubation 23 (65)
- NGT 28 (80)
CPIS*
- Mean (SD) 6,54 (SD 2)
- Confidence Interval CI95% 2-10 SAPS II*
- Mean (SD) 45,20 (SD 14)
- Confidence Interval CI95% 0-71 Time of Diagnostic VAP
- Mean (SD) 4,03 (SD 3)
- Confidence Interval CI95% 2-14
*CPIS : Clinical Pulmonary Infection Score.
*SAPS : Symplified Acute Pulmonary Infection Score.
Table 2. Association between the combined levels of PCT visit-2 (limit value 5 ng/mL) and LBP visit-4 (limit value 30 µg/mL) with patients prognosis
Observation Prediction
RR 95%CI P
value Recovery Dead
Recovery 17 3
6.476 1.318;31.831 0.021
Dead 7 8
Table 3. Association between the combined levels of PCT visit-4 (cut-off value 0.5 ng/mL) and LBP visit-4 (cut-off value 25 µg/mL) with patients prognosis
Observation Prediction
RR 95%CI P
value Recovery Dead
Recovery 17 3
15.583 2.910;
83.455 0.001
Dead 4 11
sensitivity
1-specificity
AUC 0.693
Sensitivity 88.5%
Specificity 53.2%
P value 0.056
PCT visit-2 LBP visit-2
Figure 1. ROC curve for PCT visit-2 and LBP visit-2
For visit-4 by combining PCT cut-off value of 0.5 ng/mL and LBP limit value 25 µg/mL it was shown that patients with PCT and LBP results above aforementioned levels shows worse prognosis (p=0.001). (Table 3) Visit-4 PCT and LBP cut-off value shows 96.3% sensitivity and 66.7% specivicity with AUC of 0.815. (Figure 2)
DISCUSSION
Using the cut-off value of 5 ng/mL, it appeared that patients with PCT level >5 ng/
mL after three days treatment showed worse prognosis with sensitivity of 80%, specificity of 30% and AUC of 0.75 (Table 3 and Figure 1). Other studies use a lower limit value. Luyt EC, in his study in 2003 used a limit value of PCT 1.5 ng/mL show 74% sensivity and 84%
specificity.16 To distinguish between SIRS with other septic conditions Stephan Harbarth used a lower limit value of PCT 1.1 ng/mL and show 97% sensivity and 78% specificity with AUC 0.92.17 The difference in limit value occurred because of differences in patient characteristics in terms of type of surgery whether elective or acute, and the severity of the disease if there were comorbidities with other illnesses.
Carroll et al. (1997) found that sepsis patients with initial LBP levels >46 µg/mL had a higher mortality rate.13 In 2003, Prucha et al. report his study about LBP levels in 68 patients with critical illness. There are differences in levels between SIRS patients with severe sepsis/septic shock using a limit value of 29.8 µg/mL show 74.2%
sensitivity and 50% specificity. After seven days of treatment, it was found that the differences in levels of LBP among patients who recovered and who died of 11.9 µg/mL and 26.4 µg/mL.14 In this study, the initial limit value of LBP used after three days treatment are almost the same as Prucha et al. namely 30 ug/mL. Patients with LBP levels ≥30 µg/mL have a worse prognosis (p=0.036), with 60% sensitivity, 25% specificity
(p=0.080) and AUC of 0.675. After seven days of treatment using the limit value LBP 25 µg/mL, a worse prognosis was shown in patients with LBP ≥25 µg/mL (p=0.005) with a sensitivity of 73.3%, specificity 25% (p=0.016) and AUC of 0.742. Patients with LBP level >30 µg/mL and LBP levels 7 days post-treatment >25 µg/
mL have a worse prognosis. Procalcitonin as a prognostic marker showed better sensitivity compared to LBP after three days and seven days treatment respectively 80% and 81.3% vs. 60%
and 73% with both specificity were equally low (30% and 16.7% for PCT and 25% for LBP).
Increased PCT is an indication of the degree of host immune response to infection produced by macrophages. While an increase in LBP is an indication of the severity of organ involvement occurred in VAP especially because LBP is produced by lung epithelial cells and alveolar type 2.9,10 To increase the specificity, analysis was performed by combining the two markers and the sensitivity was increased to 88.5% and 96.3%, the specificity increased to 53.2% and 66.7% after three days and seven days treatment respectively.
The specialty of this study is the use of LBP as a new infection marker combined with PCT as preexisting one to predict the mortality of VAP patient, and the limitation is the relatively small sample conducted in only one centre.
CONCLUSION
Combined examination of Procalcitonin and LBP can be used as prognostic marker of mortality in patients with VAP. The best prognostic performance was obtained using combination of these markers after seven days of treatment.
ACKNOWLEDGMENTS
The author would like to acknowledge Prof DR Dr Dinajani Mahdi SpPD KAI SH as a promotor of this study and also to Prof DR Dr Sarwono Waspadji SpPD KEMD, Prof Dr Nelwan SpPD KPTI, Prof Dr Roesli Tahir Sp An (Alm), Prof DR Dr Zul Dahlan SpPD KP (Alm), Dr Adang Bachtiar BSc for supporting the this study.
1.0
0.8
0.6
0.4
0.2 0.6
0.0
0.0 0.2 0.4 0.6 0.8 1.0
sensitivity
1-specificity
AUC 0.815
Sensitivity 96.3%
Specificity 66.7%
P value 0.015
PCT visit-4 LBP visit-4
Figure 2. ROC curve for PCT visit-4 and LBP visit-4
REFERENCES
1. Chastre J, Fagon J. Ventilator- associated pneumonia.
Am J Respir Crit Care Med. 2002;165:867-94.
2. Markowicz P, Wolff M, Djedaini K. Multicenter prospective study of ventilator-associated pneumonia during acute respiratory distress syndrome. Incidence, prognosis, and risk factor. ARDS Study Group. Am J Respir Crit Care Med. 2000;161:1942-8.
3. Koeman M, Van der Ven AJAM, Ramsay G, Hoepelman IM, Bonten MJM. Ventilator-associated pneumonia: recent issue on pathogenesis, prevention and diagnosis. J Hosp Infect. 2001;49:155-60.
4. Niederman MS, Craven DE. Guideline for the management of adult with hospital-acquired, ventilator- associated, and healthcare- associated pneumonia. Am J Respir Crit Care Med. 2005;171:388-409.
5. SalyersAA, Bacteria pathogenesis. A molecular approach. In: Whitt, editor. Wasington DC: ASM Press;
2002. p. 247-62.
6. Brooks GF, Carroll KC. Pathogenesis of bacterial infection. In: Butel JS, Morse SA, eds. Medical microbiology. Toronto: Mc Graw Hill; 2004. p .147-58.
7. Gattas DJ, Cook DJ. Procalcitonin as a diagnostic test for sepsis: Health technology assessment in the ICU.
J Crit Care. 2003;18:52-8.
8. Christ-crain M, Jaccard-stolz D, Bingisser R. Effect of procalcitonin guided therapy on antibiotic use and outcome in lower respiratory tract infection:
cluster-randomized, single-blinded trial. Lancet.
2004;363:600-7.
9. Gutsmann T Muller M, Carroll S, Mackanzie RC, Wiese A, Seydel U. Dual role of lipopolysaccharide- binding protein in neutralization of LPS-induced of mononuclear cells. Infect Immun. 2001;69:6942-50.
10. Van Amersfoort ES, Van Berkel TJC, Kuiper J.
Receptors, mediators, and mechanism involved in bacterial sepsis and septic shock. Clin Microbiol Rev.
2003;16:374-414.
11. Monick MM, Hunninghake GW. Activation of second messenger pathway in alveolar macrophage by endotoxin. Eur Respir J. 2002;20:210-22.
12. Appelmelk BJ, Lynn WA. The cause of sepsis: bacterial cell components that trigger the cytokine cascade. In:
Dhainaut JF, Thijs LG, Park G, eds. Septic shock.
Toronto: WB Sunders company limited; 2000. p. 21- 33.
13. Luyt CE, Guerin V, Combes A, Truillet JL. Procalcitonin kinetics as a prognostic marker of ventilator-associated pneumonia. Am J Respir Crit Care Med. 2005;171:48- 53.
14. Prucha M, Herold I, Zazula R, Dubska L, Dostal M.
Significance of lipopolysaccharide-binding protein (an acute phase protein) in monitoring critically ill patients.
Crit Care. 2003;7:R154-9.
15. Delevaux I, Andre M, Colombier M. Can procalcitonin measurement help in differentiating between bacterial infection and other kind of inflammatory processes?
Ann Rheum Dis. 2003;62:337-40.
16. Brunkhorst FM, Wegscheider K, Forycki ZF, Brunkhorst R. Procalcitonin for early diagnosis and differentiation of SIRS, sepsis, severe sepsis, and septic shock. Intensive care Med. 2000;26:S148-52.
