Journal of Critical Care (2011) 26, 435.e1–435.e7

Increased myeloperoxidase enzyme activity in plasma is anindicator of inflammation and onset of sepsisNikhil Kothari MDa,⁎, Ravi S. Keshari MSc b, Jaishri Bogra MDa, Monica Kohli MDa,Haider Abbas MDa, Anita Malik MDa, Madhu Dikshit PhDb, Manoj K. Barthwal PhDb

aDepartment of Anaesthesia, Chatrapati Shahuji Maharaj Medical University, Lucknow, UP, IndiabPharmacology Division, Central Drug Research Institute, Lucknow, UP, India

0d

Keywords:Myeloperoxidase;Leukocyte count;SIRS;Sepsis

AbstractIntroduction: Circulating lipopolysaccharides released from bacteria may activate both neutrophils andmonocytes. The activated neutrophils release myeloperoxidase (MPO), a specific enzyme with strongoxidative activity. The aim of this study was to evaluate MPO enzyme activity in plasma of critically illpatients and to check the hypothesis that these concentrations in plasma would be higher in sepsis andsystemic inflammatory conditions, as neutrophils release their contents before proliferating in responseto stress.Material and Methods: Blood samples were collected from 105 critically ill patients admitted to theintensive care unit, consisting of those with systemic inflammatory response syndrome (n = 42), sepsis(n = 37), and septic shock (n = 26). Plasma MPO enzyme activity was determined by o-dianisidine-H2O2 method, modified for 96-well plates.Results: The plasma MPO enzyme activity in sepsis patients was significantly higher than that in thecontrol group (mean, 2.4 ± 1.8 in sepsis and 1.86 ± 1.2 nmol per milligram protein per 10 minutes insystemic inflammatory response syndrome vs 0.32 ± 0.11 nmol per milligram protein per 10 minutes inhealthy controls). Mean plasma lactate levels in sepsis (7.8 ± 1.2 mmol/L) and shock patients (9.5 ± 1.2mmol/L) and cytokines like tumor necrosis factor–α, interleukin-8, and interleukin-1β were simul-taneously evaluated to establish onset of inflammation and sepsis. These results show that neutrophilactivation occurring during inflammation and sepsis could be detected by plasma MPO concentration.Conclusion: The plasmaMPO concentrations may be a marker of the neutrophil proliferation and severityof inflammation.© 2011 Elsevier Inc. All rights reserved.

1. Introduction phagocytosis, and production of reactive oxygen interme-

Various functions of neutrophils have been described insepsis [1], including adherence, chemotaxis, degranulation,

⁎ Corresponding author.E-mail address: drnikhilkothari@rediffmail.com (N. Kothari).

883-9441/$ – see front matter © 2011 Elsevier Inc. All rights reserved.oi:10.1016/j.jcrc.2010.09.001

diates [2-5]. In sepsis, the expected and appropriateinflammatory response to an infectious process becomesamplified, leading to organ dysfunction or risk for secondaryinfection [6]. As a clinical syndrome, sepsis occurs when aninfection is associated with the systemic inflammatoryresponse syndrome (SIRS). A continuum exists from alow-grade systemic response associated with a self-limited

435.e2 N. Kothari et al.

infection to a marked systemic response with solitary ormultiorgan dysfunction, that is, severe sepsis. Myelopero-xidase (MPO) is a heme enzyme of neutrophils azurophilicgranules with a strong oxidative activity [7]. Myelopero-xidase present in the primary granules of polymorphonuclearleukocytes (PMNs) has been demonstrated to participate inthe microbicidal activity of these cells. Together with themembranous NADPH oxidase, MPO is involved in theformation of reactive oxygen species (ROS) and oxidationof biological material [8,9]. In stimulated PMNs, NADPHoxidase reduces molecular oxygen to superoxide anionradical. This species and its dismutation product hydrogenperoxide are substrates for MPO. Myeloperoxidase has alsobeen suggested to regulate the respiratory activity of PMNsduring phagocytosis. During phagocytosis, PMNs undergo aburst in respiration; and through a series of single electrontransfer, oxygen is reduced to superoxide anion (SO2

−) andhydrogen peroxide (H2O2) [10]. Although H2O2 alone ismicrobicidal, its bactericidal activity is greatly potentiated bythe heme enzyme. Myeloperoxidase constitutes 5% of thehuman PMN by weight and is believed to represent a majorpathway for O2-dependent microbicidal activity. Thestimulation of PMNs results in a sudden increase in oxygenconsumption, with the production of ROS and the release ofenzymes such as elastase and MPO. In addition to itsperoxidase activity, MPO catalyzes H2O2 and chloride anion(Cl−) reaction, forming hypochlorous acid (HOCl) [11].Hypochlorous acid is a powerful oxidant and reacts withamines to form chloramines, which contribute to bacterialdestruction inside the phagolysosome of the neutrophils [12].Increased plasma MPO levels are a marker of neutrophilproliferation and degranulation in humans. This work wasplanned to test the hypothesis that patients with SIRS andsepsis would have neutrophils activation resulting in higherMPO enzyme activity in plasma.

2. Material and methods

This study was carried out to assess MPO levels in plasmaof 105 critically ill patients admitted to the intensive care unit(ICU) and managed following Surviving Sepsis guidelines

Table 1 Patients demographic characteristics (mean ± SD)

Control (n = 80) SIRS

Age (y) 38.5 ± 8 38.3 ±Male/female ratio 55/25 24/18Temperature (°C) 37 ± 0.9 38.4 ±Heart rate (beats/min) 98 ± 2.3 121 ±Respiratory rate (breaths/min) 18 ± 3.2 29 ±MAP (mm Hg) 78 ± 10.2 89.4 ±APACHE II score – 25.23SOFA score – 10.89

MAP, mean arterial pressure.

[13]. Human ethical approval was granted by the institutionalreview board. Informed consents were obtained from controlsubjects and patients or their relatives. The control group wascomposed of the healthy relatives accompanying the patient.A total of 80 controls were taken; among them, 55 were menand 25 were women, with a mean (SD) age of 38.5 (8) years.Among the patient group, 67 were men and 38 were women,with a mean (SD) age of 37.4 (6) years (Table 1). Theinclusion criteria were the following: (1) clinical evidenceof infection, (2) hyperthermia (temperature N38°C) orhypothermia (temperature b35°C), (3) tachycardia (N100beats per minute), (4) tachypnea (N30 breaths per minute) orneed for mechanical ventilation, and (5) evidence ofinadequate organ function or perfusion within 12 hours ofenrollment. The exclusion criteria were the following: (1)patients older than 80 years, (2) cardiac failure (class III orIV), (3) liver insufficiency (Child C), and (4) immunosup-pression (positive HIV, HBs Ag virus serologic result,cancer). Patients were followed up throughout their stays inthe ICU. Age, sex, primary site of infection, infection-relatedorganisms, and severity indexes including Acute Physiologyand Chronic Health Evaluation Scores (APACHE II) andSequential Organ system Failure Assessment score (SOFA)were recorded for each patient's entry into the ICU. Theplasma of these patients was tested for MPO levels at thetime of study entry and then after every 24 hours until theirstay in ICU. Besides MPO, plasma levels of tumor necrosisfactor (TNF)–α, interleukin (IL)-1β, and IL-8 as markers ofinflammation were also assessed using standard enzyme-linked immunosorbent assay (ELISA) technique. Bloodlactate level was assessed for detection of metabolicacidosis due to inadequate tissue perfusion suggestinganaerobic metabolism in tissues. All the samples and infor-mation used in the study were coded, and patient con-fidentiality was preserved according to the guidelines forstudies of human subjects.

2.1. Blood sample collection

First blood sampling was performed before antimicrobial,adrenergic, or steroid therapy. Blood samples were collectedfrom central venous catheter (9 mL) into tubes containing

(n = 42) Sepsis (n = 37) Septic shock (n = 26)

4 36.2 ± 7 39.1 ± 426/11 17/9

1.3 38.3 ± 2.6 36.4 ± 1.118 118 ± 29 122 ± 268 22 ± 6 26.0 ± 516.3 88.3 ± 12.7 82.8 ± 23.5± 6.6 28.03 ± 6.2 30.0 ± 6.3± 3.4 13.52 ± 3.2 13.11 ± 2.8

CONTROL SIRS SEPSIS SHOCK0

1

2

3

**

**

**

MP

O (

nmol

/10

min

utes

/mg

prot

ein)

Fig. 1 Mean plasma MPO levels in nanomoles per 10 minutesper milligram protein. Differences in mean plasma levels amongeach group was statistically significant (⁎⁎P b .01).

435.e3MPO as an indicator of inflammation and onset of sepsis

1 mL trisodium citrate upon admission of patient to ICUand subsequently. Plasma was separated by centrifuge at13000 rpm for 15 minutes. The plasma was stored at −70°Cfor assessment of MPO enzyme activity, along with plasmaTNF-α, IL-1β, and IL-8 levels. Repeated freeze thaw ofsamples was avoided to prevent degradation of plasmacytokines levels. Plasma levels of TNF-α, IL-1β, and IL-8were assessed using standard ELISA technique.

2.2. MPO assay

Myeloperoxidase enzyme activity was determined byo-dianisidine-H2O2 method, modified for 96-well plates [14].Briefly, plasma samples (20 μL) were added to 0.53 mmol/Lo-dianisidine dihydrochloride Sigma-Aldrich, USA. and0.15 mmol/L H2O2 in 50 mmol/L potassium phosphatebuffer (pH 6.0). After incubation for 10 minutes at roomtemperature, the change in absorbance was measured at460 nm (ɛ = 10 062/M/cm). Results were expressed as unitsof MPO per milligram protein per 10 minutes, whereby1 unit of MPO was defined as the amount of enzymedegrading 1 nmol H2O2 per minute at 37°C [15].

2.3. Statistical analysis

All data were obtained in duplicate, and results ofcalculations are reported as means and standard deviationup to 2 decimal points. The data were analyzed by Bartletttest for nonparametric analysis of variance with Newman-Keuls multiple comparison posttest. The relation betweenplasma MPO value and total leukocyte count (TLC) wastested by determining the Pearson correlation coefficient (r).A P value b .05 was considered significant. All statisticalanalyses were performed with the Graph Pad InStat 5.0 demoprogram (Graph Pad Software, San Diego, CA).

3. Results

Among 105 patients admitted to the ICU during theperiod April 2009 to May 2010, 42 patients had SIRS, 37patients were in sepsis, 12 patients were in a state of septicshock, and 14 patients were initially in the SIRS group butlater developed sepsis. Sepsis was diagnosed on the basis ofmetabolic acidosis and specific culture reports from variouspossible sites of infection, including blood culture report.There were 67 male and 38 female patients, with a mean(SD) age of 37.4 (6) years (Table 1). We monitored theenzymatic activity of MPO in plasma as a marker ofinflammation and sepsis. Other standard markers likeTNF-α, IL-1β, and IL-8 were estimated for detecting onsetof inflammatory activity. The MPO specific activity,expressed as nanomoles of H2O2 degraded per milligramprotein per 10 minutes (Fig. 1), was increased 4- to 6-fold inplasma samples of SIRS and sepsis patients as compared

with that of healthy controls (mean, 2.4 ± 1.8 in sepsis and1.86 ± 1.2 in SIRS vs 0.32 ± 0.11 in healthy controls; range,1.2-6.2 in sepsis and 0.14-4.3 in SIRS). Pearson correlationanalysis identified a positive correlation between an increasein MPO specific activity and severity of sepsis assessed bySOFA and APACHE II scores recorded at the time ofadmission in ICU (Fig. 2; r2 = 0.82, P b .01). In the SIRSgroup, MPO enzyme activity increased with severity ofinflammation, showing a positive correlation (r2 = 0.62,P b .01), whereas in patients of septic shock, no suchcorrelation was observed (r2 = 0.021, P = .45), indicatingreduced MPO activity in severe sepsis. In state of septicshock, MPO activity was only 0.8 times increased, whichwas then correlated with reduced neutrophil count in casesof advanced sepsis (Fig. 3; mean, 0.6 ± 0.1; range, 0.1 to2.5 nM per milligram protein per 10 minutes).This was thenconfirmed by reduced TLCs in patients of advanced sepsisdue to bone marrow suppression.

The MPO enzyme activity was then compared with TLC,determined by automated cell counter. The overall increase inMPO activity in sepsis patients was associated with a cor-responding increase in TLC (Fig. 3; Pearson correlation coeffi-cient: r = 0.9,P b .001 for sepsis and r = 0.86,P b .01 for SIRS),whereas in patients of shock, cell count decreased with theprogression of disease; thus, low MPO activity was observed.

Detection of metabolic acidosis in arterial blood gasanalysis was evaluated further by blood lactate estimation(Fig. 5). Blood lactate levels were increased in the SIRSgroup (5.2 ± 1.1 mmol/L) with range from 4.9 to 5.5mmol/L.Similarly, blood lactate levels in sepsis (7.8 ± 1.2 mmol/L)and septic shock group (9.5 ± 1.2 mmol/L) were significantlyhigh (95% confidence interval [CI]: sepsis, 7.4-8.2; shock,8.9-10.1 mmol/L). In the control group, mean values were0.93 ± 0.3 mmol/L (95% CI, 0.8-1.0). Plasma cytokine levelsevaluated by ELISA were also elevated in SIRS, sepsis, andseptic shock patients (Fig. 4). In SIRS, mean (SE) value ofTNF-α, IL-8, and IL-1β were 77.99 (14.6), 68.25 (27.4), and

SIRS

0 2 4 6 80

10

20

30

40

(MPO nmol/10mins/mg protein)

APA

CH

ESIRS

0 2 4 6 80

10

20

30

(MPO nmol/10mins/mg protein)

SOFA

SEPSIS

0 2 4 60

10

20

30

40

(MPO nmol/10mins/mg protein)

APA

CH

E

SEPSIS

0 2 4 6 80

10

20

30

(MPO nmol/10mins/mg protein)

SOFA

SHOCK

0 1 2 30

10

20

30

(MPO nmol/10mins/mg protein)

SOFA

SHOCK

0 1 2 30

10

20

30

40

(MPO nmol/10mins/mg protein)

APA

CH

E

Fig. 2 Plasma MPO levels increased as the severity of disease increased; it correlated linearly with disease severity scores (APACHE IIand SOFA).

435.e4 N. Kothari et al.

99.52 (14.3) pg/mL, respectively; in sepsis, mean (SE)values were 187.1 (88.3), 225.9 (56), and 175.8 (10.7) pg/mL, respectively; in septic shock, mean (SE) values were107 (15.03), 252.9 (74.14), and 320.2 (33.06) pg/mL,respectively. It was observed that the difference in plasmavalues of cytokines between control, sepsis, and septicshock groups was statistically significant (P b .05).

4. Discussion

Polymorphonuclear leukocytes are the first cell type inhuman beings that is activated in host immune defense

against infection [16]. These cells driven by chemotacticgradients migrate to inflammatory loci, where they recog-nize and phagocytose bacteria and other extrinsic micro-organisms by release of hydrolytic enzymes and bactericidalproteins prestored in granules as well as newly generatedROS [17,18]. Vascular leakage and recruitment of circu-lating PMNs to the site of injury represent the earlyphase of the host defense mechanism and response to tissueinjury or sepsis. This response is common to all organs andtissues. Clinically, the increased number of PMNs in bloodis generally used to determine the development of inflam-mation/sepsis [19]. A more feasible and quantitativeapproach is the use of the biochemical assay of PMNs-associated MPO enzyme activity. This enzyme is highly

SIRS

0 2 4 6 80

10

20

30

(MPO nmol/10mins/mg protein)

TL

C

SEPSIS

0 2 4 6 80

10

20

30

(MPO nmol/10mins/mg protein)

TL

C

SHOCK

01230

2

4

6

(MPO nmol/10mins/mg protein)

TL

C

Fig. 3 Plasma MPO levels increased with TLC as the severity ofdisease increased; it correlated linearly with disease severity.Inverse correlation was seen in advanced stages of septic shock.

CONTROL SIRS SEPSIS SHOCK0

100

200

300

400 TNFIL8IL1b

*

*

**

****

*

#

#

#

(pg/

ml)

Fig. 4 Mean plasma levels of cytokines (TNF-α, IL-8, and IL-1β)in picograms per milliliter, assessed using standard ELISAtechnique. Differences in mean plasma levels among each groupwere statistically significant (⁎, ⁎⁎, #P b .05).

CONTROL SIRS SEPSIS SHOCK0

5

10

15

**

**

**

**

(mm

ol/L

)

Fig. 5 Blood lactate levels increased with severity of inflam-mation (SIRS) and with progression of sepsis and septic shock(⁎⁎P b .05). Values reported up to 95% CI.

435.e5MPO as an indicator of inflammation and onset of sepsis

enriched in the azurophilic granules of PMNs recruitedto injured tissue to mediate the acute phase of the inflam-matory response [20].

In this study, we report the association between severityof inflammation and plasma MPO enzyme activity. Neu-trophils are considered major contributors to the tissuedamage that occurs in inflammatory diseases. Myeloperox-idase, a major granule enzyme in neutrophils, accounts for5% of the total neutrophil protein and is responsible for theproduction of HOCl oxidant. Activated neutrophils produceROS (O2

− and H2O2) via NADPH oxidase as part of theirantipathogen response [21-23]. The release of ROS andHOCl by neutrophils may cause damage to importantbiological structures such as proteins, carbohydrates, lipids,and nucleic acids and may enhance inflammatory responses.

Development of sepsis and inflammatory response inpatients was further assessed by detecting markers ofinflammation (TNF-α, IL-1β, and IL-8) in patient bloodsamples [24]. These inflammatory cytokines significantlyincreased in SIRS, sepsis, and septic shock groups (Fig. 4),thus establishing the episodes of inflammation [25]. How-ever, increased lactate levels, as a marker of tissue hypoxiain critically ill patients, remain a matter of debate. Giventhe many processes that may affect the ultimate concen-tration of lactate, both individual lactate levels and thechange in level over time may well reflect the generalhomeostasis of the critically ill patient [26]. In our studygroups, blood lactate levels were found to be significantlyraised in sepsis and septic shock patients (Fig. 5). Ourdata in this study validate the observations that MPOactivity and cell count are increased in patients with SIRSand sepsis (Fig. 3). The finding of a positive correlationbetween the increase in MPO activity and TLC suggeststhat trauma- or bacteria-induced inflammatory or infec-tive responses are pathological and that patient withsepsis sustain MPO-dependent protein oxidative damage.

435.e6 N. Kothari et al.

Myeloperoxidase activity was expected to increase withthe progression of sepsis and development of septic shock,but it was constantly observed to be on the lower side inpatients with advanced stages of septic shock [27]. BecauseMPO enzyme activity reflects the neutrophil function,blood levels of neutrophils were then evaluated to look forTLC [28,29]. Neutrophil count was found to be low inshock patients; further evaluation revealed pancytopenia inthese patients due to bone marrow suppression in advancedstages of septic shock [30]. Evaluation of plasma MPOactivity in sepsis and septic shock patients could help usin assessing neutrophil status and its functioning; ourdata further point toward the use of plasma levels ofMPO enzyme activity as biomarkers of inflammatoryoxidative pathology.

5. Conclusion

The study reported here reveals that neutrophil activationoccurs in a high percentage during oxidative stress incritically ill patients, especially during SIRS and sepsis; thesecells proliferate during stress, and MPO is released intoblood. Owing to the high variability of MPO levels, we couldnot establish a cutoff point in MPO enzyme activity levels todistinguish survivors from nonsurvivors. In summary, wedemonstrate that plasma MPO enzyme activity is a goodbiomarker of inflammatory responses in patients with SIRSand sepsis.

Acknowledgments

We gratefully acknowledge the release of clinicalresearch fellowship by ICMR, New Delhi, for MD PhDproject. We are thankful to Director of CDRI forlaboratory work support. Furthermore, we thank ProfShally Awasthi, Faculty-In-charge, Research CellCSMMU, Lucknow, for her constant moral support andguidance to research fellows.

References

[1] Regel G, Nerlich ML, Dwenger A, Seidel J, Schmidt C, Sturm JA.Phagocytic function of polymorphonuclear leukocytes and the RES inendotoxemia. J Surg Res 1987;42:74-84.

[2] Veuczio RF, Westenfelder GO, Phaio JP. The adherence ofpolymorphonuclear leukocytes in patients with sepsis. J Infect Dis1982;145:351-6.

[3] Duignan JP, Collins PB, Johnson AH. The association of impairedneutrophil chemotaxis with postoperative surgical sepsis. Br J Surg1986;73:238-40.

[4] Wenisch C, Graninger W. Are soluble factors relevant for polymor-phonuclear leukocyte dysregulation in septicemia. Clin Diagn LabImmunol 1995;2:241-5.

[5] Solomkin JS, Cotha LA, Brodt JK. Regulation of neutrophilsuperoxide in sepsis. Arch Surg 1985;120:93-8.

[6] Zimmerman JJ, Millard JR, Farrin C. Septic plasma suppressessuperoxide anion synthesis by normal homologous Polymorphonu-clear leukocytes. Crit Care Med 1989;17:1241-6.

[7] Edwards SW. Biochemistry and physiology of the neutrophil. In:Cassatella MA, editor. The neutrophil: an emerging regulator ofinflammatory and immune response. London: Cambridge UniversityPress; 1994. p. 299.

[8] Sengeloev H, Kjeldsen L, Borregaard N. Properties, functionand secretion of human myeloperoxidase. J Immunol 1993;150:1535-43.

[9] Babior BM, Kipnes RS, Curnutte JT. The particulate superoxide-forming system from human neutrophils. Properties of the system andfurther evidence supporting its participation in the respiratory burst.J Clin Invest 1976;58(4):989-96.

[10] Butterfield DA, Koppal T, Howard B, Subramaniam R, Hall N,Hensley K, et al. Structural and functional changes in proteins inducedby free radical-mediated oxidative stress and protective action of theantioxidants N-tert-butylalpha-phenylnitrone and vitamin E. Ann NYAcad Sci 1998;854:448-62.

[11] Winterbourn CC, Kettle AJ. Biomarkers of myeloperoxidase-derivedhypochlorous acid. Free Radic Biol Med 2000;29:403-9.

[12] Woods AA, Linton SM, Davies MJ. Detection of HOCl mediatedprotein oxidation products in the extracellular matrix of humanatherosclerotic plaques. Biochem J 2003;370:729-35.

[13] Dellinger RP, Levy MM, Carlet JM. Surviving sepsis campaign:international guidelines for management of severe sepsis and septicshock. Crit Care Med 2008;36:296-327.

[14] Bradley PP, Priebat DA, Christensen RD, Rothstein G. Measurementof cutaneous inflammation: estimation of neutrophil content with anenzyme marker. J Investig Dermatol 1982;78:206-9.

[15] Van der Poll T, Opal SM. Host-pathogen interactions in sepsis. LancetInfect Dis 2008;8:32-43.

[16] Monisha D, Gladys PM, Sandoval RG, Moreno R, Garg NJ. Increasedmyeloperoxidase activity and protein nitration are indicators ofinflammation in patients with Chagas' disease. Clin Vaccine Immunol2009;16(5):660-6.

[17] Panasenko OM, Spalteholz H, Schiller J, Arnhold J. Myeloperoxidase-induced formation of chlorohydrins and lysophospholipids fromunsaturated phosphatidylcholines. Free Rad Biol Med 2003;34:553-62.

[18] Bian K, Kamisaki Y, Murad F. Proteomic modification by nitric oxide.J Pharmacol Sci 2006;101:271-9.

[19] Cinel I, Dellinger RP. Advances in pathogenesis and management ofsepsis. Curr Opin Infect Dis 2007;20:345-52.

[20] Stehr SN, Knels L, Weissflog C. Effects of IgM enrichedsolution on polymorphonuclear neutrophil function, bacterialclearance, and lung histology in endotoxemia. Shock 2007;29:167-72.

[21] Babior BM, Kipnes RS, Curnutte JT. Biological defense mechanisms:the production by leukocytes of superoxide, a potential bactericidalagent. J Clin Invest 1973;52:741.

[22] Segal AW, Abo A. The biochemical basis of the NADPH oxidase ofphagocytes. Trends Biochem Sci 1993;18:43.

[23] Robinson JM, Badwey JA. The NADPH oxidase complex ofphagocytic leukocytes: a biochemical and cytochemical view.Istochem Cell Biol 1993;103:163.

[24] Cannon JG, Tompkins RG, Gelfand JA, Michie HR, Stanford GG, vander Meer JW, et al. Circulating interleukin-1 and tumor necrosis factorin septic shock and experimental endotoxin fever. J Infect Dis1990;161(1):79-84.

[25] Van Zee KJ, DeForge LE, Fischer E, Marano MA, Kenney JS, RemickDG, et al. IL-8 in septic shock, endotoxemia and after IL-1administration. J Immunol 1991;146(10):3478-82.

[26] Bakker J, Jansen TC. Don't take vitals, take a lactate. Intensive CareMed 2007;33:1863-5.

435.e7MPO as an indicator of inflammation and onset of sepsis

[27] McCredie JA, Austin TW, Holliday RL, Sibbald WJ. Predictive valuefor survival of lymphocytes and polymorphonuclear leukocytes inpatients with sepsis. Can J Surg 1979;22(5):447-51.

[28] Wenisch C, Fladerer P, Patruta S, Krause R. Assessment of neutrophilfunction in patients with septic shock: comparison of methods. ClinDiagn Lab Immunol 2001:178-80.

[29] Monserrat J, de Pablo R, Reyes E, Díaz D, Barcenilla H, Zapata MR,et al. Clinical relevance of the severe abnormalities of the T cellcompartment in septic shock patients. Critical Care 2009;13:R26.

[30] Cadi P, Claessens YE, Cariou A, Safran D. Severe bone marrownecrosis associated with septic shock in the intensive care unit. Ann FrAnesth Reanim 2004;23(5):501-4.