Prognosis of community acquired pneumonia (CAP): Value of triggering receptor expressed on myeloid cells-1 (TREM-1) and other mediators of the inflammatory response

www.elsevier.com/locate/issn/10434666 Cytokine 38 (2007) 117–123

Prognosis of community acquired pneumonia (CAP): Value of triggering receptor expressed on myeloid cells-1 (TREM-1) and other mediators of the inflammatory response Alicia Tejera a, Francisco Santolaria a,*, Marı´a-Luisa Diez b, Marı´a-Remedios Alema´n-Valls a, Emilio Gonza´lez-Reimers a, Antonio Martı´nez-Riera a, Antonio Milena-Abril b a

Servicio de Medicina Interna, Hospital Universitario de Canarias, Universidad de La Laguna 38320, Tenerife, Spain b Laboratorio de Bioquı´mica Hospital Universitario de Canarias, Spain Received 11 June 2006; received in revised form 25 April 2007; accepted 3 May 2007

Abstract TREM-1 is an activating receptor expressed on the surface of neutrophils and mature monocytes when stimulated by bacteria or fungi, leading to amplification of the inflammatory response. Our objective is to analyze the prognostic value of serum sTREM-1 levels and other mediators of the inflammatory response, in patients hospitalized for CAP, and to compare its prognostic value with those of advanced age, pneumonia severity scores, Charlson index, nutritional status and severity of sepsis. Methods. We included 226 patients with CAP, 145 males and 81 females, median age of 74 years. The following tests were performed: arterial blood gases and chest radiography, nutritional assessment, assessment of the severity of the sepsis, Pneumonia Severity Index (PSI) and CURB-65, and mediators of inflammation: TNF alfa, IL-6, IL-10, IL-1ra, LBP, sCD14, CRP, and sTREM-1. Mortality during admittance was defined as the sole end point. Results. Twenty-eight of the two-hundred and twenty-six patients died (12.4%). On univariate analysis advanced age, dehydration, increased Na, low BMI, handgrip strength, serum albumin, prealbumin, IGF-1, lymphocyte count, conscious drowsiness, tachypnea, decreased PaO2, hypotension, creatinine, ASAT, LDH, severity of sepsis, a high PSI or CURB65, TNFa, IL-6, IL-10, IL-1ra, and sTREM-1 were related to mortality. Variables with an independent value were IGF-1, CURB-65, TREM-1, advanced age and IL-6. Conclusions. This study confirms the usefulness of TREM-1 in the diagnosis and prognosis of patients with CAP, which is independent of advanced age, other inflammation markers such as IL-6, severity index for CAP such as CURB-65 or PSI, severity of sepsis and nutritional status including IGF-1. Ó 2007 Elsevier Ltd. All rights reserved. Keywords: TREM-1; Cytokine; Pneumonia Severity Index (PSI); Community acquired pneumonia (CAP); Prognosis

1. Introduction Prognosis of community acquired pneumonia (CAP) has been well-established in recent years by different scales such as the Pneumonia Severity Index (PSI) reported by Fine or CURB-65 (confusion, urea nitrogen, respiratory rate, low blood pressure and age above 65 years) [1–5]. These severity scores help to decide which patients should *

Corresponding author. Fax: +34 922 653808. E-mail address: [email protected] (F. Santolaria).

1043-4666/$ - see front matter Ó 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.cyto.2007.05.002

be hospitalized or may be treated at home, and which patients should be admitted to critical care units. Other not so well-studied prognostic factors in CAP are those related to nutritional status [6–10] and more recently, mediators of inflammatory response, such as pro-inflammatory and anti-inflammatory cytokines. IL-6, in addition to being the cytokine most frequently and consistently associated with severity and poor prognosis in sepsis [11–13], is also the cytokine most studied in relation to severity of CAP [14–19]. In recent years, a member of the immunoglobulin superfamily, the triggering

118

A. Tejera et al. / Cytokine 38 (2007) 117–123

receptor expressed on myeloid cells-1 (TREM-1) has been shown to be a useful diagnostic tool in ventilator-related pneumonia [20,21], and in the diagnosis and prognosis of sepsis in critical care [22–24]. Reduction of serum values of TREM-1 are also useful to assess the antibiotic response in patients with CAP [25]. TREM-1 is a trigger receptor expressed on the surface of neutrophils and mature monocytes when activated by bacteria and fungi. Therefore, inflammatory responses are amplified by a pathway mediated by TREM-1. Moreover, TREM-1 is shed from the membrane of activated phagocytes and its soluble form, named sTREM-1, can be detected in the bronchoalveolar lavage of patients with ventilator-related pneumonia or in the blood of patients with sepsis. The objective of this study is to analyse the prognostic value (relationship with severity and mortality during admittance) of serum sTREM-1 levels and other mediators of the inflammatory response, lipopolysaccharide binding protein (LBP), soluble receptor CD14 (sCD14), proinflammatory (TNFa and IL-6), and anti-inflammatory (IL-10 and IL-1 receptor antagonist (IL-1ra)) cytokine and C reactive protein (CRP) in patients hospitalized for CAP. Also, to compare its prognostic value with those of advanced age, severity pneumonia scores, such as PSI or CURB-65, co-morbidity Charlson index, nutritional status and severity of sepsis. 2. Patients and methods Two-hundred and twenty-six patients admitted for pneumonia, 145 males and 81 females, with a median age of 74 years (25th–75th percentile: 61–82 years) were included. All the patients were hospitalized at the Hospital Universitario de Canarias, 183 were attended for conventional care, at the internal medicine unit, while the other 43 received critical care. Patients were included after evaluation in the emergency room. CAP was defined as an acute illness associated with 1 or more of the following: new cough with or without sputum production, pleuritic chest pain, dyspnea, fever or hypothermia, altered breath sounds on auscultation, leukocytosis, plus the presence of a new infiltrate on a chest radiograph, in patients who had not been hospitalized within the previous month and in whom no alternative diagnosis had emerged during follow-up. Mortality was considered as a sole end point during admittance. 2.1. Clinical evaluation At the emergency room, patients underwent a complete clinical history and physical examination. Basic chemistry and haematology tests, arterial blood gases determinations, and chest radiography were performed. Hospital admission and type of care (conventional or critical) was decided by the team on duty; microbiological studies were indicated by the patient’s physician.

2.2. Nutritional assessment We inquired about anorexia and weight loss. Weight and height were recorded on admission, with further calculation of body mass index as weight/height2. We also determined the mid-arm circumference (MAC) and the triceps skinfold (TSF) by a Holtain lipocaliper and then calculated the mid-arm muscle area (MAMA) [26]. Anthropometric parameters were compared with those of the population of the Western Canary Islands for adults of the same sex and age [27]. We determined serum prealbumin, albumin, transferrin and IGF-1, in order to assess visceral proteins. Total lymphocyte count and CD4 and CD8 populations were determined in order to assess the T cell response. Subjective nutritional evaluation included examination of the muscle masses of the upper and lower limbs and of the temporal muscle, defining two degrees of atrophy (severe, moderate), and absence of atrophy. We assigned 2, 1 and 0 points to each category, respectively. Bichat’s fat and subcutaneous fat atrophy, recorded by physical examination of the fat loss on the cheek and abdomen, respectively, were classified in the same way. Thus, we defined a subjective nutritional score (SNS) based on the sum of the assigned points. As previously reported, a score of 0–2 was considered normal, one of 3–4 points as mild malnutrition and higher than four points as severe malnutrition [28,29]. To assess muscle function, hand grip strength was determined by dynamometry. 2.3. Severity of infection The following categories were considered: sepsis was defined by the clinical signs of SIRS together with evidence of infection; severe sepsis was defined when this was associated with hypoperfusion or organ dysfunction, such as metabolic acidosis, oliguria, or an acute alteration in mental status; and septic shock when hypotension (systolic blood pressure of <90 mm Hg or a reduction of P40 mm Hg from baseline) persisted despite adequate fluid therapy or when vasoactive amines were needed [30]. 2.4. Severity of CAP The PSI was assessed according to Fine, and classified as low risk (grades I, II and III), moderate risk (grade IV), and high risk (grade V) [1,2]. Another prognostic index, the CURB-65, was determined as a five point score, assigning one point for each of the following: Confusion, Urea >7 mmol/l, Respiratory rate P30/min, low systolic (<90 mm Hg) or diastolic (660 mm Hg) Blood pressure, age P65 years. Patients with four or more points were considered at high risk [3–5]. Also, Co-morbidity Charlson index and Glasgow coma scale were performed [31,32].

A. Tejera et al. / Cytokine 38 (2007) 117–123

2.5. Mediators of the inflammatory response Blood samples were taken at 08.00 am and frozen at 40 °C for further determination, by chemiluminiescent enzyme immunometric assay (IMMULITE analyzer), of: IGF-1 sensitivity: 20 ng/ml, Diagnostic Products Corporation (DPC), Los Angeles, USA; TNF alfa, sensitivity of 1.7 pg/ml, DPC, Los Angeles, USA; IL-6, sensitivity: 2 pg/ml, DPC, Los Angeles, USA; IL-10, sensitivity of 1 pg/ml, DPC, Los Angeles, USA; high sensitivity CRP, sensitivity of 0.02 mg/dl, DPC, Los Angeles, USA; and LBP, sensitivity of 0.2 lg/ml, DPC, Los Angeles, USA; by enzyme immunoassay was determined IL-1ra, sensitivity of 4 pg/ml, BioSource Europe SA, Nivelles. Belgium; and sCD14, sensitivity of 0.24 ng/ml, Immuno-Biological Laboratories (IBL), Hamburg, Germany. sTREM-1 was determined by a hand-made enzyme immunoassay. Capture antibody of sTREM-1 (R&D Inc., Minneapolis, USA) was diluted to 4 lg/ml and distributed in a 96-well plates at a volume of 100 ll per well. After overnight incubation at room temperature, wells were aspirated and thoroughly (three times) washed with wash buffer (400 ll). Plates were blocked by adding 300 ll of Reagent Diluent to each well and incubated for one hour. Then 100 ll of serum or standard concentrations of sTREM-1 (15.1–4000 pg/ml) were added to the wells. After incubation for 2 h, wells were washed three times, and 100 ll of a 400 ng/ml dilution of sTREM-1 detection antibody was added per well and incubated for 2 h. Then, attached antibodies were signalled by streptavidin-HRP. Concentrations of sTREM-1 in each well were estimated by the optical density detected at 450 nm after addition of one 1:1 solution of H2O2:tetramethylbenzidine as a substrate. All determinations were performed in duplicate. The sTREM-1, cytokines, sCD14, LBP and IGF-1 were determined for 32 healthy controls (16 male and 16 female), median age of 63 years (25th–75th percentile: 54–71 years). 2.6. Statistical analysis (SPSS12.0) As cytokines are not normally distributed, we performed the non-parametric tests Kruskal–Wallis and Mann–Whitney’s U test, Spearman correlation, v2, and Fisher’s exact test when necessary. The predictive value regarding mortality of continuous variables was determined by receiver operating characteristics (ROC) curves with calculation of the area under the receiver operating characteristics curve (AURC). Multivariate analysis (stepwise logistic regression) was performed in order to discern which parameters yield independent predictive value on survival. Data are presented as median, 25th and 75th percentiles. 3. Results We studied 226 patients admitted for CAP, 28 of whom (12.4%) died during admittance while the remaining 198 were discharged. Eighty-one were women and 145 men,

119

with a mortality of 9.9% and 13.8%, respectively (non-significant differences). Table 1 shows which clinical characteristics were related to mortality. Age was related to mortality especially in these 39 patients aged 85 or more, who presented a mortality of 33.3% (p < 0.001). Fortyeight patients were dehydrated with a mortality of 29.2% (p < 0.001). Forty-one patients had been previously diagnosed of dementia, with a mortality of 22% (p < 0.073). Forty-four patients had chronic renal failure with a mortality of 25% (p < 0.01). 3.1. Nutrition Malnutrition was related to short-term prognosis. Patients with a BMI under 20 kg/m2 have a mortality of 20.7%, between 20 and 25 kg/m2, 15.2%, and above 25 kg/m2 6.2% (p = 0.032). A MAMA under 60% of the median of the control population was associated with a mortality of 21.5% (p = 0.005). When serum albumin was above 3.5 g/dl no patient died, between 3.5 g/dl and 2.8 g/dl, mortality was 9.3% and with serum albumin equal to or below of 2.7 g/dl, mortality was 21.1% (p < 0.001). Mortality was also associated with a decrease in CD4 or CD8 lymphocytes. IGF-1 was related to all nutritional data: it significantly correlated with BMI (p = 0.004), SNS (p < 0.001), MAC (p < 0.001), MAMA (p = 0.001), hand grip strength (p < 0.001), serum albumin (p < 0.001), and prealbumin (p < 0.001) levels and CD4 lymphocyte count (p = 0.046). On the other hand, IGF-1 was negatively related to IL-6 (p = 0.036) and TNF (p = 0.002). IGF-1 was closely related to mortality (p < 0.0001). None patient died with IGF-1 above 104 ng/ml. 3.2. Sepsis evolution Only one patient did not fulfill the SIRS criteria, he was analysed together with the 98 patients who were diagnosed of sepsis, with a mortality of 4%, 114 were diagnosed of severe sepsis, mortality 18.4%, nine of septic shock, mortality 22%, and four of multi-organ failure with a mortality of 25% (p = 0.009). Twenty-eight patients presented hypotension on admission with a mortality of 28.6% (p = 0.01). Patients with mental alterations, assessed by Glasgow coma score (14 or less) showed a raised mortality (19%), whereas this was only 4.8% in patients with normal consciousness. 3.3. Severity of pneumonia (PSI and CURB-65) Twenty-two patients were admitted with a PSI grade I, and three patients with a PSI grade II, none of them died. Only one out of 31 patients with grade III died. Eighty-two patients with a grade IV (moderate risk) showed a mortality of 11% and 88 at grade V (high risk) showed a mortality of 20.5%. Regarding CURB-65 index, 17 patients showed 0 points with 0% mortality; 33 patients had one point, 3%

120

A. Tejera et al. / Cytokine 38 (2007) 117–123

Table 1 Data with prognostic value at admittance of patients with CAP Alive at discharge

Age (years) SNS BMI (kg/m2) MAMA (%) Hand-grip (lbs) Prealbumin (mg/dl) Albumin (g/dl) IGF-1 (ng/ml) Lymphocyte/mm3 CD4 count/mm3 CD8 count/mm3 Glasgow score Breath rate Pa O2 (mm Hg) Sat O2 (%) Systolic BP (mm Hg) Creatinine (mg/dl) BUN (mg/dl) ASAT (UI/l) LDH (UI/l) PSI CURB-65

Dead

UMW

AURC (95% CI)

(n) median (25th–75th)

(n) median (25th–75th)

z

p

(198) (198) (196) (198) (198) (190) (195) (197) (198) (176) (176) (198) (198) (190) (190) (198) (198) (198) (196) (195) (198) (198)

(28) (28) (25) (26) (28) (21) (22) (28) (28) (20) (20) (28) (28) (27) (27) (28) (27) (28) (24) (22) (28) (28)

2.93 2.58 2.84 2.76 4.04 3.50 3.57 3.88 3.12 2.55 2.70 3.82 3.63 2.70 2.92 2.79 2.39 2.07 3.03 3.46 4.32 5.01

0.003 0.010 0.004 0.006 0.000 0.001 0.001 0.000 0.002 0.011 0.007 0.000 0.000 0.007 0.004 0.005 0.017 0.039 0.002 0.001 0.000 0.000

73 ( 80.25–57) 2.0 (0–4) 25.6 (22.2–29.1) 70.5 (57.3–90.9) 20.3 (0–86.7) 12 (9–18) 3.2 (2.8–3.6) 74 (49.9–100.2) 1020 (650–1534) 486 (316–742) 276 (164–426) 15 (14–15) 24 (20–30) 63.1 (54.1–74) 93.4 (89.3–96.3) 120 (100–140) 0.9 (0.7–1.3) 22 (15–33.3) 26 (18–49) 401 (328–491) 115 ( 87–147) 2 (1.75–3)

81 (69–87.5) 2.5 (1.25–6.75) 22.4 (19.6–25.5) 57.2 (43.5–72.5) 0 (0–0) 7 (5.5–11.5) 2.7 (2.5–3.1) 41.75 (27.2–71.9) 547 (336–1061) 243 (155–621) 124 (69–452) 14 (12.5–14) 32 (24–36) 56 (48.4–65.1) 89.9 (83.2–93.5) 100 (85.5–120) 1.2 (0.8–1.9) 29.5 (19.5–47) 43.5 (25.5–205) 576 (422–820) 155 (122–179) 4 (3–4)

0.671 0.644 0.675 0.667 0.724 0.733 0.732 0.726 0.682 0.674 0.685 0.704 0.711 0.661 0.674 0.662 0.641 0.621 0.690 0.725 0.752 0.784

(0.560–0.782) (0.536–0.753) (0.570–0.779) (0.553–0.781) (0.641–0.807) (0.616–0.849) (0.642–0.822) (0.628–0.824) (0.564–0.801) (0.536–0.813) (0.531–0.838) (0.600–0.808) (0.605–0.816) (0.553–0.769) (0.566–0.781) (0.557–0.767) (0.517–0.765) (0.502–0.740) (0.580–0.800) (0.599–0.851) (0.669–0.836) (0.699–0.869)

UMW, U of Mann–Whitney; AURC, area under ROC Curve; SNS, Subjective Nutritional Score; BMI, body mass index; MAMA, mid-arm muscle area; PSI, Pneumonia Severity Index; CURB-65, confusion, urea >7 mmol/l, respiratory rate P30/min, low systolic (<90 mm Hg) or diastolic (660 mm Hg), blood pressure, age P65 years.

mortality; 72 patients had two points, 4.2% mortality; 58 patients had three points, 13.8% mortality; 39 patients had four points, 35.9% mortality; finally seven patients reached 5 points, 28.6% mortality. Both indexes, PSI and CURB-65, are useful to predict mortality with an area under ROC curve of 0.752 and 0.784, respectively. 3.4. Mediators of the inflammatory response Table 2 shows that all the mediators studied: LBP, sCD14, pro-inflammatory cytokines such as TNFa, IL-6, anti-inflammatory ones, such as IL-10, IL-1ra, acute phase reactant such as CRP and the TREM-1 were raised in CAP patients compared with controls. These differences were highly significant (p < 0.0001 for all the mediators) and also when we excluded those patients with CAP who died.

In Table 3 we can see that TNFa, IL-6, IL-10, IL-1ra and sTREM-1 had prognostic value regarding mortality, whereas LBP, sCD14 and CRP did not. Of all of them, sTREM-1 had the best prognostic yield, with the highest AURC (0.740) only surpassed by the multiple data composed scales PSI and CURB-65. Regarding the mediators of inflammatory response, sTREM-1 was related to IL-6 (p = 0.006), CRP (p = 0.001) and IL-1ra (p = 0.026). Moreover, sTREM-1 was correlated with age (p = 0.001), neutrophil count (p < 0.0001), platelet count (p = 0.049), SNS (p = 0.040), serum BUN and creatinine levels (p = 0.001), and negatively with serum albumin levels, (p = 0.046) and prothrombin activity (p = 0.031). Blood culture was performed on 149 patients (66%), and was positive in only 15 of them. However, patients with positive blood samples showed significantly higher levels

Table 2 All the inflammatory markers evaluated are increased at admittance of patients with CAP

LBP (lg/ml) s CD14 (ng/ml) sTREM-1 (pg/ml) TNFa (pg/ml) IL-6 (pg/ml) IL-1ra (pg/ml) IL-10 (pg/ml) CRP (mg/ml)

CAP

Control

UMW

Median (25th–75th) (n = 226)

Median (25th–75th) (n = 32)

z

p

114.5 (73.2–200) 6152 (4828–7664) 34 (19.9–50.6) 11.4 (8.2–19.3) 23.5 (9.45–81.4) 567 (287–1199) 5.5 (5.2–7.1) 154 (68–291)

25.1 (16.3–34.6) 3753 (3327–4420) 12.25 (6.25 –23.6) 5.4 (4.25–7.28) 2.4 (2.2–3.75) 179 (132–227) 5 (5–5) 2.25 (0.96–4.1)

7.63 6.78 4.34 6.57 8.08 6.69 7.59 8.97

<0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

A. Tejera et al. / Cytokine 38 (2007) 117–123

121

Table 3 Inflammatory markers with independent prognostic value at admittance of patients with CAP Alive at discharge (n = 198)

LBP (lg/ml) s CD14 (ng/ml) sTREM-1 (pg/ml) TNFa (pg/ml) IL-6 (pg/ml) IL-1ra (pg/ml) IL-10 (pg/ml) CRP (lg/ml)

Dead (n = 28)

UMW

AURC

Median (25th–75th)

Median (25th 75th)

z

p

111.5 (73–200) 6068 (4786–7618) 31.3 (18.9–47.5) 11.2 (7.8–18.8) 22.3 (9.1–67.6) 545 (280–986) 5.5(5.1–6.8) 144 (66–280)

169 (78.7–200) 6696 (4920–7889) 55 (38.3 –90.4) 14.8 (10.4–25.1) 81.9 (16.7–166) 962 (559–1905) 5.8 (5.3–15.9) 216 (90–300)

1.59 0.60 4.11 2.70 2.44 2.71 2.52 1.45

NS NS <0.0001 0.007 0.015 0.007 0.024 NS

of TNF (p < 0.001), IL-6 (p < 0.001), IL-10 (p = 0.014), IL-1ra (p < 0.001), LBP (p = 0.018), and sCD14 (p = 0.007), but not of sTREM-1. sTREM-1 increased in patients with severe sepsis and septic shock, especially in those who died, with a median of 45.1 pg/ml in sepsis, 56.5 pg/ml in severe sepsis and 68 pg/ml in septic shock or MOF. Median values for those who survived were 27.5 pg/ml in sepsis, 34.5 pg/ml in severe sepsis and 28.7 pg/ml in septic shock or MOF. The relationship between short-term mortality and TREM-1 was independent on sepsis severity (p < 0.0001). In the same way, the relationship between IL-6 (p = 0.049), TNF (p = 0.001), IL-10 (p = 0.005), and mortality was independent of severity of sepsis. Among the 183 patients treated with conventional care, mortality was 11.5%, whereas among the 43 who received critical care, mortality was 16.3% (non significant differences). When we analysed separately critical and conventional care, the relationships between mediators of the inflammatory response and death persist, with significant differences between dead and alive patients subjected to critical care in sTREM-1 (p = 0.007) and IL-10 (p = 0.038), and in conventional care patients, in sTREM-1 (p = 0.001), TNFa (p = 0.007), IL-6 (p = 0.011), and IL1ra (p = 0.012). The relationship between mortality and sTREM and IL-6 was not lost in different age groups, so we found significant differences both in patients over and under 85-years-old. 3.5. Multivariate analysis As many data were related to mortality we performed a logistic regression analysis to determine which of them had an independent prognostic value. We included the following variables: age (85 years or more), dehydration, Subjective Nutritional Score, hand grip (dynamometry), Glasgow coma score, severity of sepsis, PSI, CURB-65, TNFa, IL-6, sTREM-1 and IGF-1. Variables with predictive independent value were IGF-1 less than 37.5 ng/ml, (Relative Risk 10.2; Confidence interval (CI) 95%: 3.2–32.5), CURB-65 above 3 (Relative Risk 3.3; CI 95%: 1.2–9), TREM-1, greater than 50 pg/ml, (Relative Risk 7.0; CI 95%: 2.3– 21), age, 85 years or older, (Relative Risk 6.2; CI 95%: 2.1–18.3) and IL-6 over 80 pg/ml (Relative Risk 2.9; CI

0.740 0.658 0.645 0.658 0.631

(0.644–0.837) (0.559–0.756) (0.530–0.760) (0.543–0.773) (0.522–0.741)

95%: 1.01–8.2). With these five data, the area under the ROC curve increases to 0.917 (0.857–0.977). 4. Discussion CAP remains an important cause of mortality despite of effective antibiotics; 12.4% in our 226 patients hospitalized for CAP. The inclusion of patients in the study was not limited by advanced age, immunosuppressant or the type of care received (43 patients admitted for critical care). For this reason, although our results have a slightly higher mortality than other studies, they extrapolate better to dayto-day clinical activity. However, when we stratify by PSI, our mortality rates, 11% in grade IV and 20,5% in grade V, are similar to those currently reported [1,33–35]. Both Fine’s PSI and CURB-65 are closely related to mortality. It is noteworthy that CURB-65, with only five items: confusion, increase of BUN, tachypnea, low blood pressure and age, all them included in PSI score, is slightly better than PSI with a slightly higher area under the ROC curve. However, PSI is considered better to diagnose low grade CAP, and also to discern which patients can be treated as outpatients [3]. Malnutrition is another prognostic sign that is not included in the PSI. In our study, patients who died showed lower BMI, MAC, MAMA, and a more impaired subjective nutritional assessment. Handgrip strength, a parameter related with muscle function and with malnutrition, has been shown to be related to mortality in CAP hospitalized patients [8]. In our study, dynamometry was related to nutrition, especially muscle area, and to mortality, in the same way as serum albumin and IGF-1. The close relationship between IGF-1 and mortality is noteworthy [36]. Moreover, the highest value of IGF-1 among patients who died was 104 ng/ml; a value which is around the 75th percentile of the surviving patients. A low CD4 and CD8 lymphocyte count was also related to malnutrition and to a poor prognosis. Severe sepsis or septic-shock are frequent among patients hospitalized for CAP, 56% in our study [37]. Mediators and markers of the inflammatory response such as cytokine and CRP have been extensively studied in recent years in sepsis and CAP. Among the cytokines, IL-6 has shown a better correlation with severity of CAP and

122

A. Tejera et al. / Cytokine 38 (2007) 117–123

prognosis. However, most studies were performed with a small number of CAP patients, in whom severity scored as APACHEII or SIRS, rather than mortality, was assessed [15–19]. In our study, with a large number of patients, we found that both pro-inflammatory, -TNF, IL-6- and anti-inflammatory cytokines –IL-10 and IL-1ra – are related to mortality during admittance. All these cytokines have been related to mortality in septic patients [11–13]. However, CRP, CD14 and LBP although increased in patients with CAP, were not related to shortterm mortality. TREM-1 is a receptor on cell membranes of neutrophils and monocytes that, when triggered, contributes to intensification of the inflammatory process. In this sense, we found that sTREM-1 was related to both inflammatory and antinflammatory cytokines, LBP and CRP. As soluble TREM-1 is detached from neutrophils, sTREM-1 correlates with the neutrophil count as mentioned by Routsie C (2005) [24]. TREM-1 has been mainly studied in critical care septic patients. It is useful to diagnose the septic process, especially in the case of ventilator-associated pneumonia, both in the bronchio-alveolar aspirate and in peripheral blood. Moreover, it has been related to mortality in critical care septic patients. In two studies in critical care septic patients, Gibot S (2004 and 2005) reports that sTREM-1 is decreased at admission in patients who died, but shows a progressive increase until death, whereas it increased, on admission, in patients who survived, but progressively decreases later [22,23]. A correct interpretation of any result should take into account the moment in which the analysis was performed. In this sense, Routsie C (2005) in ventilator-associated pneumonia patients, reports that an increased TREM-1 at the time of diagnosis is a poor prognostic sign [24]. Our results agree with the latter, as we found raised serum levels of sTREM-1 in patients who died during admittance. Most patients, 183 (81%), in our study, received conventional care whereas 43 (19%) were admitted to critical care units. In both cases we found that initially increased sTREM-1 levels were related to short-term mortality. Relationship with mortality of sTREM-1 is not lost with age, so we found significant differences both in patients over and under 85 years old. However, although sTREM-1 is increased in patients with severe sepsis and related to pro-inflammatory and anti-inflammatory cytokines, it was not increased in patients with bacteraemia. The high area under the curve of sTREM-1 is noteworthy, 0.740, which is only exceeded by those of the composite indices of PSI and CURB-65, of 0.752 and 0.784, respectively. To assess the relative prognostic value of sTREM-1, we performed a multivariate analysis and found that TREM-1 prognostic value is independent of those of advanced age, CURB-65, IGF-1 and IL-6. When these five data are analysed together, the area under the curve raises to 0.917. TREM-1 is a membrane molecule that is expressed on the surface of neutrophils when triggered by bacteria. Its soluble form, called sTREM-1, is present in the sera of

patients with sepsis. The precise role of sTREM-1 in the pathogenesis of sepsis is yet undefined. The present study, in CAP, confirms its usefulness in the diagnosis and prognosis of septic patients. Moreover, its prognostic value is independent of advanced age, other markers of inflammation such as IL-6, severity indexes for CAP such as CURB65 or PSI, severity of sepsis and nutritional status including IGF-1. In conclusion, we believe that its relation with mortality, unrelated to other well-known prognostic factors, may confer a special prognostic relevance to the assessment of sTREM-1 in septic pneumonia patients. References [1] 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. [2] Fine MJ, Stone RA, Singer DE, Coley CM, Marrie TJ, Lave JR, et al. Processes and outcomes of care for patients with communityacquired pneumonia: results from the Pneumonia Patient Outcomes Research Team (PORT) cohort study. Arch Intern Med 1999;159:970–80. [3] Aujesky D, Auble TE, Yealy DM, Stone RA, Obrosky DS, Meehan TP, et al. Prospective comparison of three validated prediction rules for prognosis in community-acquired pneumonia. Am J Med 2005;118:384–92. [4] Buising KL, Thursky KA, Black JF, Macgregor L, Street A, Kennedy M, et al. Reconsidering what is meant by severe pneumonia: A prospective comparison of severity scores for community acquired pneumonia. Thorax 2006;61:419–24. [5] Laterre PF, Garber G, Levy H, Wunderink R, Kinasewitz GT, Sollet JP, et al. Severe community-acquired pneumonia as a cause of severe sepsis: data from the PROWESS study. Crit Care Med 2005;33:952–61. [6] Riquelme R, Torres A, el-Ebiary M, Mensa J, Estruch R, Ruiz M, et al. Community-acquired pneumonia in the elderly. Clinical and nutritional aspects. Am J Respir Crit Care Med 1997;156:1908–14. [7] Hedlund J. Community-acquired pneumonia requiring hospitalisation. Factors of importance for the short- and long-term prognosis. Scand J Infect Dis 1995;97(suppl):1–60. [8] Vecchiarino P, Bohannon RW, Ferullo J, Maljanian R. Short-term outcomes and their predictors for patients hospitalized with community-acquired pneumonia. Heart Lung 2004;33:301–7. [9] Loeb M, High K. The effect of malnutrition on risk and outcome of community-acquired pneumonia. Respir Care Clin N Am 2005;11:99–108. [10] Masotti L, Ceccarelli E, Cappelli R, Barabesi L, Guerrini M, Forconi S. Length of hospitalization in elderly patients with communityacquired pneumonia. Aging (Milano) 2000;12:35–41. [11] Rodrı´guez-Gaspar M, Santolaria F, Jarque A, Gonzalez-Reimers E, Milena A, de la Vega MJ, et al. Prognostic value of cytokines in sirs general medical patients. Cytokine 2001;15:232–6. [12] Selberg O, Hecker H, Martin M, Klos A, Bautsch W, Kohl J. Discrimination of sepsis and systemic inflammatory response syndrome by determination of circulating plasma concentrations of procalcitonin, protein complement 3a, and interleukin-6. Crit Care Med 2000;28:2793–8. [13] Spittler A, Razenberger M, Kupper H, Kaul M, Hackl W, BoltzNitulescu GH, et al. Relationship between Interleukin-6 plasma concentration in patients with sepsis, monocyte phenotype, monocyte phagocytic properties, and cytokine production. Clin Infect Dis 2000;3:1338–42. [14] Glynn P, Coakley R, Kilgallen I, Murphy N, O’Neill S. Circulating interleukin 6 and interleukin 10 in community acquired pneumonia. Thorax 1999;54:51–5.

A. Tejera et al. / Cytokine 38 (2007) 117–123 [15] Marik PE. The clinical features of severe community-acquired pneumonia presenting as septic shock. Norasept II Study Investigators. J Crit Care 2000;15:85–90. [16] Antunes G, Evans SA, Lordan JL, Frew AJ. Systemic cytokine levels in community-acquired pneumonia and their association with disease severity. Eur Respir J 2002;20:990–5. [17] Kolsuz M, Erginel S, Alatas O, Alatas F, Metintas M, Ucgun I, et al. Acute phase reactants and cytokine levels in unilateral communityacquired pneumonia. Respiration 2003;70:615–22. [18] Fernandez-Serrano S, Dorca J, Coromines M, Carratala J, Gudiol F, Manresa F. Molecular inflammatory responses measured in blood of patients with severe community-acquired pneumonia. Clin Diagn Lab Immunol 2003;10:813–20. [19] Igonin AA, Armstrong VW, Shipkova M, Lazareva NB, Kukes VG, Oellerich M. Circulating cytokines as markers of systemic inflammatory response in severe community-acquired pneumonia. Clin Biochem 2004;37:204–9. [20] Gibot S, Cravoisy A, Levy B, Bene MC, Faure G, Bollaert PE. Soluble triggering receptor expressed on myeloid cells and the diagnosis of pneumonia. N Engl J Med 2004;350:451–8. [21] Determann RM, Millo JL, Gibot S, Korevaar JC, Vroom MB, van der Poll T, et al. Serial changes in soluble triggering receptor expressed on myeloid cells in the lung during development of ventilator-associated pneumonia. Int Care Med 2005;31:1495–500. [22] Gibot S, Kolopp-Sarda MN, Bene MC, Cravoisy A, Levy B, Faure GC, et al. Related articles, links plasma level of a triggering receptor expressed on myeloid cells-1: its diagnostic accuracy in patients with suspected sepsis. Ann Intern Med 2004;141:9–15. [23] Gibot S, Cravoisy A, Kolopp-Sarda MN, Bene MC, Faure G, Bollaert PE, et al. Time-course of sTREM (soluble triggering receptor expressed on myeloid cells)-1, procalcitonin, and C-reactive protein plasma concentrations during sepsis. Crit Care Med 2005;33:792–6. [24] Routsi C, Giamarellos-Bourboulis EJ, Antonopoulou A, Kollias S, Siasiakou S, Koronaios A, et al. Does soluble triggering receptor expressed on myeloid cells-1 play any role in the pathogenesis of septic shock?. Clin Exp Immunol 2005;142:62–7. [25] Chao WC, Wang CH, Chan MC, Chow KC, Hsu JY, Wu CL. Predictive value of serial measurements of sTREM-1 in the treatment

[26]

[27]

[28]

[29]

[30] [31] [32] [33]

[34]

[35]

[36]

[37]

123

response of patients with community-acquired pneumonia. J Formos Med Assoc 2007;106:187–95. Blackburn GL, Bistrian BR, Maini BS, Schlamm HT, Smith MF. Nutritional and metabolic assessment of the hospitalized patient. JPEN 1977;1:11–22. Gonza´lez-Hermoso F, Porta R, Herrero A. Resultados del estudio antropome´trico de la poblacio´n canaria adulta. Nutr Hosp 1990;5:238–47. Herna´ndez-Plasencia D, Santolaria F, Herna´ndez-Garcı´a M, Gonza´lez Reimers E, Batista N, Jorge J, et al. Subjective Nutritional Assessment and short-term prognosis. J Nutr Med 1991;2:151–62. Tormo A, Santolaria F, Gonza´lez-Reimers E, Oramas J, Rodrı´guez Rodrı´guez E, Rodrı´guez Moreno F, et al. Short-term prognostic value of subjective nutritional assessment in general medical patients. J Nutr Med 1994;4:287–95. Bone RC. Toward an epidemiology and natural history of SIRS. JAMA 1992;268:3452–5. Charlson M, Szatrowski TP, Peterson J, Gold J. Validation of a combined comorbidity index. J Clin Epidemiol 1994;47:1245–51. Jennett B, Teasdale G. Aspects of coma after severe head injury. Lancet 1977;8017:878–81. Marras TK, Gutierrez C, Chan CK. Applying a prediction rule to identify low-risk patients with community-acquired pneumonia. Chest 2000;118:1339–43. Dedier J, Singer DE, Chang Y, Moore M, Atlas SJ. Processes of care, illness severity, and outcomes in the management of communityacquired pneumonia at academic hospitals. Arch Intern Med 2001;161:2099–104. Roson B, Carratala J, Dorca J, Casanova A, Manresa F, Gudiol F. Etiology, reasons for hospitalization, risk classes, and outcomes of community-acquired pneumonia in patients hospitalized on the basis of conventional admission criteria. Clin Infect Dis 2001;33:158–65. Newman AB, Kupelian V, Visser M, Simonsick EM, Goodpaster BH, Kritchevsky SB, et al. Strength, but not muscle mass, is associated with mortality in the health, aging and body composition study cohort. J Gerontol A Biol Sci Med Sci 2006;61:72–7. Dremsizov T, Clermont G, Kellum JA, Kalassian KG, Fine MJ, Angus DC. Severe sepsis in community-acquired pneumonia: when does it happen, and do systemic inflammatory response syndrome criteria help predict course?. Chest 2006;129:968–78.