SEARCH

SEARCH BY CITATION

Keywords:

  • Cholesterol ester transfer protein;
  • HDL;
  • lipoproteins;
  • mortality;
  • risk factor;
  • sepsis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

Eur J Clin Invest 2010; 40 (4): 330–338

Abstract

Background  The magnitude of lipoprotein level reduction during the acute-phase response may be associated with the severity and mortality of sepsis. However, it remains to be determined whether low lipoprotein levels can be considered a risk factor for developing sepsis. We aimed to investigate lipoprotein levels as risk factors for sepsis in hospitalized patients, and also describe sequential changes in lipoprotein and cholesterol ester transfer protein (CETP) levels during sepsis.

Design  This is a prospective cohort study and case–control analysis from selected hospitalized patients. Blood samples were collected at admission, and participants were monitored for severe sepsis. Total cholesterol, high density lipoprotein (HDL), low density lipoprotein, and triglyceride levels were compared between sepsis cases and controls. Cholesterol, apolipoprotein, phospholipid and CETP concentrations were monitored in the case group.

Results  Of 1719 enrolled patients, 51 developed severe sepsis and were paired with 71 controls by age, gender, presence of infection at admission and chronic disease. HDL cholesterol level at admission was a risk factor for severe sepsis (OR = 0·969; 95% CI: 0·944–0·995). Mean CETP levels diminished between hospital admission and day 3 of sepsis. The magnitude of this variation (ΔCETP) was more pronounced in non-survivors (0·78 ± 1·08 μg mL−1) than that in survivors (0·02 ± 0·58 μg mL−1, P = 0·01).

Conclusions  HDL cholesterol may have a protective effect against sepsis. Each 1 mg dL−1 increase in HDL decreased the odds of severe sepsis by 3% during hospitalization. The reduction of plasma CETP was associated with mortality.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

Sepsis persists as a great concern in modern medicine, as it is the main cause of death in hospitalized patients [1–3]. Concern over the impact of sepsis on public health prompted the launch of an international campaign aimed to reduce sepsis-related mortality [4]. The difficulties in sepsis treatment stem from the complex aetiopathology of this disease [5]. Recently, there has been increasing interest concerning the ability of lipoproteins, especially high density lipoprotein (HDL), to modulate the acute inflammatory response [6,7]. The mechanisms of modulation involve: (i) binding and neutralization of bacterial toxins, (ii) inhibition of adhesion molecule expression, (iii) stimulation of endothelial nitric oxide synthase (eNOS) production and (iv) protection of low density lipoproteins (LDL) against peroxidative damage [8].

Lipoprotein levels drop dramatically during the acute-phase response, and the magnitude of this reduction may be associated with the severity and mortality of sepsis [9,10]. However, it remains to be determined whether the fall in blood lipoprotein levels merely reflects the severity of acute inflammatory response, or rather that low lipoprotein levels can be considered a risk factor for developing sepsis [9,11,12]. This is an important distinction because in the first case HDL level is a prognostic marker, but in the latter HDL level can be monitored and can predict sepsis.

The involvement of plasma lipid transporters in systemic inflammation is also unclear. Lipopolysaccharide binding protein (LBP) is known as the main neutralizer of bacterial toxins and is part of the innate immune response [13]. Other proteins capable of binding toxins, such as phospholipid transfer protein and cholesterol ester transfer protein (CETP), come from the lipid transport protein family and are structurally and physiologically similar to LBP [14–16]. Experimental studies have reported a reduction in CETP activity during systemic inflammation [17,18]. In addition, an experimental model of sepsis in CETP transgenic mice suggested a protective effect of CETP through reduced sepsis mortality [19]. However, few data exist that detail CETP changes during sepsis in humans [20].

The main purpose of this study was to evaluate if the lipoproteins levels at hospital admission are a predictive risk factor for the development of severe sepsis. In addition, we aimed to describe in detail the sequential changes of lipoproteins and CETP levels during the acute-phase response to severe sepsis and the association between CETP changes and mortality.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

Participants and study design

A prospective cohort study and case–control analysis of selected patients was conducted at Londrina University Hospital, between May 31st and December 1st, 2005. The study group included all adult patients admitted to the hospital during the study period who gave informed written consent directly or through legal representatives. Exclusion criteria included the following: (i) hospitalization for less than 24 h, (ii) presence of severe sepsis on admission, (iii) pregnancy, (iv) use of parenteral nutrition or propofol, (v) hyperlipidaemia or use of lipid-lowering drugs, (vi) thyroid disorders, (vii) use of corticosteroids, (viii) immunosuppressive condition (i.e. AIDS, neutropenia, chemotherapy, cancer, transplantation and malnutrition) and (ix) liver cirrhosis. Patients with infections not meeting severe sepsis criteria on admission were not excluded. Fasting blood samples were obtained after an overnight fasting period of at least 12 h from all subjects within the first 24 h (D0) of hospital admission. If infection was present at admission or detected during hospitalization by the Infection Control Committee, according to CDC criteria [21], the patient was monitored daily to detect severe sepsis. Following the diagnosis of severe sepsis, disease severity was assessed by the Acute Physiology and Chronic Health Evaluation II (APACHE II) score [22]. Organ system function was monitored daily using the Sequential Organ Failure Assessment (SOFA) [23], and patients were followed for 28 days or until death. Blood samples were also obtained from septic patients of the case group on the mornings of days one (D1), three (D3), five (D5), seven (D7), fourteen (D14) and twenty-eight (D28) after severe sepsis diagnosis or until death. All serum samples were stored at −70 °C until analysis. These time-intervals between blood samplings were chosen based on the previous studies, which demonstrated early rapid decline of lipoprotein levels on the third day of sepsis and a slow recovery over the next 28 days [9,10]. Chronic disease was defined as a diagnosis of diabetes mellitus, chronic pulmonary obstructive disease, chronic renal failure or chronic heart failure. Severe sepsis was defined according to the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference [24], which defines suspected infection based on the presence of clinical data associated with two or more signs of systemic inflammation plus sepsis induced dysfunction of at least one organ or system. The control group was chosen among patients hospitalized because of trauma, elective surgery or diagnostic procedure, and like septic patients of the case group, their blood samples were collected at D0. All 51 patients who developed severe sepsis during follow-up study period were included in the case group, and were paired with one or more controls matched for age, gender, presence of infection at hospital admission and the presence of chronic disease. The control group was comprised of 71 patients to obtain homogeneity in the paired criteria between the groups. All demographic and diagnostic data were collected at hospital admission, and diagnoses at hospital admission were classified by ICD – 10 criteria (2007). The study protocol was approved by the Londrina University Hospital Ethics Committee.

Laboratory analyses

Total cholesterol, HDL, triglyceride and albumin levels were measured with commercially available kits (Dade Behring, Newark, NJ, USA) in a Dade Dimension XL® auto-analyzer (Newark, NJ, USA) in case and control patients. LDL cholesterol was estimated by the Friedewald formula [25] for those patients with triglyceride values bellow 400 mg dL−1. Two patients presented with serum triglyceride levels above 400 mg dL−1 and were not included in the analysis. In additional, apolipoprotein A1 (Apo A1) and apolipoprotein B (Apo B) levels were measured in case patients by turbidimetry (Randox®; Randox Laboratories, Antrim, UK) and determined with a chemistry analyzer Cobas Mira Plus® (Roche, Switzerland). Phospholipid and CETP levels were determined in case patients using commercially available kits (Wako Chemicals, Richmond, VA, USA).

Statistical analyses

To detect a 20% difference in HDL cholesterol concentration between sepsis cases and controls with 90% power and 95% confidence, we estimated that we would need 35 patients in each group. This estimate was extrapolated from another study [26], wherein a 20% change in HDL cholesterol levels was identified as a risk factor for nosocomial infection. Statistical analyses were performed with SAS (SAS Institute, Cary, NC, USA), Epi Info version 3·3·2 (CDC, Atlanta, GA, USA) and STATISTICA (Statsoft, Tulsa, OK, USA). The data are described as mean ± SD, or median (interquartiles) and proportions. Continuous variables were compared using either Student’s t-test or Mann–Whitney U-test, according to data distribution. Chi-square with Yates’ correction or Fisher’s exact test was used to evaluate the statistical significance of categorical variables. Multivariate logistic regression was performed to evaluate the effect of lipid profile on the risk of developing severe sepsis. Explanatory variables included total cholesterol, HDL cholesterol, LDL cholesterol and triglyceride levels. Forward-stepwise selection by logistic regression analysis was performed using an inclusion criterion of P = 0·05. Regression results were also presented as odds ratios (ORs) and their respective 95% CIs after determination of remaining variables. Calibration of the model was assessed by the Hosmer–Lemeshow goodness-of-fit test (P > 0·05). Changes in continuous variables over time were tested for significance using analysis of variance (anova) for repeated measures. Spearman’s rank-correlation coefficient was used to express the correlation between continuous variables. All statistical tests were two-tailed, and P < 0·05 was considered significant.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

During the study period, 5601 patients were admitted to the hospital, and 3775 were excluded from the subject group by the criteria described above. An additional 107 subjects were lost because of inadequate sample collection or patient’s request to leave the study. Thus, 1719 subjects were ultimately entered into the cohort. Infection was present at admission in 276 patients, and an additional 134 patients developed infection during hospitalization, resulting in 410 patients with infection. These 410 patients were monitored daily during the investigation period to detect severe sepsis criteria, which occurred in 51 patients (case group) (Fig. 1). There were no differences in gender, age, presence of infection at admission or prevalence of chronic diseases between case and control groups (Table 1). At admission, the most frequent diagnoses in the case group were peripheral thrombosis (11/51), heart failure (6/51), pulmonary infection (6/51) and stroke (5/51). In the control group, the most frequent diagnoses were trauma (12/71), pulmonary infection (8/71), peripheral thrombosis (7/71), inguinal or diaphragmatic hernia (7/71), heart failure (5/71) and gastrointestinal haemorrhage (4/71).

image

Figure 1.  Fluxogram of patient selection in this study.

Download figure to PowerPoint

Table 1.   Baseline characteristics of the study population
General characteristicsCases (N = 51)Controls (N = 71)P-value
  1. COPD, chronic obstructive pulmonary disease.

  2. *Student’s t-test.

  3. Chi-square with Yates’ correction.

  4. Fisher’s Exact test.

Age (years)
 Mean ± standard deviation65·81 ± 15·7165·12 ± 15·490·809*
 Median (interquartile range)69·4 (56·7–77·6)68·2 (56·7–76·6)
Gender
 Male26/51 (51%)44/71 (62%)0·152
Infection at hospital admission20/51 (39%)19/71 (26·7%)0·214
Chronic disease
 Diabetes mellitus16 (31·4%)23 (32·3%)0·926
 Heart failure4 (7·8%)9 (12·6%)0·581
 COPD4 (7·8%)6 (8·4%)0·829
 Chronic renal failure2 (3·9%)1 (1·4%)0·773

In the control group, infection foci present at hospital admission were lung (6/71; 8·4%), skin (6/71; 8·4%), urinary (4/71; 5·6%), abdominal (1/71; 1·4%), surgical site infection (1/71; 1·4%) and bacteraemia (1/71; 1·4%). In the case group, the main foci of infection were lung (35/51; 68·6%), abdominal (9/51; 17·6%), urinary (5/51; 9·8%), endocardial (1/51; 2%) and osteal (1/51; 2%). The mean APACHE II score in the case group was 21·84 ± 7·84, and the initial SOFA was 6·2 ± 3·5. Thirty (58·8%) patients in the case group died during follow-up. On the first day of sepsis, non-survivors had higher SOFA scores (7·0 ± 3·8) and thus more organ dysfunction than survivors (5·0 ± 2·7, P = 0·04).

At hospital admission, HDL cholesterol levels were significantly lower in the case group (33·07 ± 14·59 mg dL−1) compared with controls (40·33 ± 16·54 mg dL−1, P = 0·013). Among the lipoprotein blood measures analysed in the final multivariate logistic regression model, only HDL cholesterol (odds ratio 0·969; 95% CI: 0·944–0·995) showed a protective effect (Table 2). Specifically, for each 1 mg dL−1 increase in HDL cholesterol, there was a 3% decrease in the odds of severe sepsis during hospitalization. Figure 2 shows the linear regression curve of HDL and the occurrence of sepsis in the studied population. The sequential measurements of lipids and lipoproteins in the case group are shown in Fig. 3. There was a significant correlation between HDL cholesterol and Apo A1 (r = 0·87, P = 0·001) as well as LDL cholesterol and Apo B (r = 0·83, P = 0·001).

Table 2.   Analysis of cholesterol levels in case and control patients at admission as a risk factor for developing severe sepsis
VariableUnivariateMultivariate*
OR(95% CI)P-valueOR(95% CI)P-value
  1. OR, odds ratio; CI, confidence interval; LDLc, LDL cholesterol; HDLc, HDL cholesterol.

  2. *Multiple logistic regression (Hosmer and Lemeshow Goodness-of-Fit test P-value = 0·296).

  3. Wald chi-square test.

Total cholesterol0·9930·984–1·0020·144   
HDLc0·9690·945–0·9940·0110·9690·945–0·9940·011
LDLc0·9940·982–1·0060·338   
Triglycerides1·0000·996–1·0050·704   
image

Figure 2.  Correlation between HDL cholesterol levels and the occurrence of severe sepsis.

Download figure to PowerPoint

image

Figure 3.  Sequential measurements of cholesterol, triglyceride, apolipoprotein and phospholipid concentrations in 51 patients with severe sepsis.

Download figure to PowerPoint

When the case group was subdivided into survivors and non-survivors, both presented low levels of HDL cholesterol and Apo A1 on the first day of sepsis. These levels, however, had an increasing trend in survivors that did not reach statistical significance. There was no difference in lipoprotein, phospholipid or triglyceride concentrations between survivors and non-survivors (Fig. 4).

image

Figure 4.  Sequential measurements of apolipoprotein A1, HDL cholesterol, apolipoprotein B, and LDL cholesterol levels in survivors and non-survivors among 51 patients with severe sepsis.

Download figure to PowerPoint

In the sepsis case group, albumin levels declined significantly between hospital admission and the third day of sepsis (from 2·69 ± 0·70 to 1·74 ± 0·53 g dL−1, P < 0·001). CETP levels showed a similar decrease (from 1·52 ± 0·82 to 1·14 ± 0·67 μg mL−1, P = 0·01). However, there was no difference in blood measures between survivor and non-survivor subgroups. When the differences in CETP levels between D0 and D3 of sepsis (ΔCETP = D0 − D3) were calculated, the decrease observed was more pronounced in non-survivors (0·78 ± 1·08 μg mL−1) than that in survivors (0·02 ± 0·58 μg mL−1, P = 0·01) (Table 3). There was also a positive correlation between ΔCETP measures and the severity of sepsis measured by APACHE II score (r = 0·39, P = 0·013) as well as organ dysfunction measured by SOFA (r = 0·36, P = 0·032).

Table 3.   Distribution of plasma albumin and CETP levels among survivors and non-survivors in the sepsis case group
VariableNon-survivorsSurvivorsP-value
  1. Mean values at hospital admission (D0) and third day (D3) of severe sepsis.

  2. Albumin g dL−1; CETP μg mL−1.

  3. CETP, cholesterol ester transfer protein.

  4. *Delta(Δ) = D0 − D3.

  5. Student’s t-test.

  6. Mann–Whitney U-test.

Albumin day 02·79 ± 0·732·60 ± 0·680·373
Albumin day 31·67 ± 0·501·81 ± 0·560·424
Δ* albumin1·25 ± 1·000·79 ± 0·560·046
CETP day 01·63 ± 0·891·25 ± 0·490·162
CETP day 31·05 ± 0·651·23 ± 0·700·410
Δ* CETP0·78 ± 1·080·02 ± 0·580·010

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

This study shows that low HDL cholesterol concentration may be a risk factor for the development of severe sepsis after hospital admission. We also confirmed previous findings showing significant fluctuations in cholesterol, apolipoprotein, phospholipid and CETP levels during severe sepsis [9,10,13,17–19] and revealed a reduction in CETP levels associated with mortality.

We initially investigated whether total cholesterol, triglyceride or lipoprotein concentrations at hospital admission (D0) were different between case patients and controls and found that lower levels of HDL cholesterol were associated with increased risk of developing severe sepsis during hospitalization. HDL cholesterol participates in the innate immune response and apparently modulates favourable binding and neutralizing bacterial toxins. We speculate that the presence of enough HDL in plasma before the infectious insult can be protective, stemming from its function as a scavenger of bacterial toxins. Furthermore, hypocholesterolaemia has also been associated with increased risk of hospital infection in surgical patients [26], leading to a prolonged hospital stay [27]. Low serum LDL cholesterol has been shown to be associated with an increased risk of fever and sepsis [28].

The HDL is the major plasma lipoprotein present in human plasma, and Apo A1 is its major structural apoliprotein. During the acute phase response, Apo A1 is replaced by SAA in the HDL particle, and this change reduces HDL affinity for hepatocytes and increases its affinity for macrophages. This suggests that during inflammation, HDL is redirected from hepatocytes towards a macrophage scavenger pathway. Another consequence of the altered biophysical properties of HDL may be a more rapid turnover. The decline in HDL during sepsis may render the patient more susceptible to inflammatory stimuli, and this may lead to a vicious cycle that leads to progressive multiple organ dysfunction and death [7]. A recent study revealed several immunomodulatory mechanisms of HDL, confirming that this particle is more than a simple LPS scavenger. These findings include attenuation of adhesion molecule expression, eNOS activation, anti-oxidant activities and the protective effects of some of the enzymes associated with HDL [8].

Our results suggest that low basal serum HDL cholesterol may be considered a risk factor for the development of severe sepsis in hospitalized patients. The design and temporal sequence of data collection of this study supports the speculation that this risk has a causal association. However, sepsis has multiple risk factors that interact with each other, and causality is difficult to elucidate. Although the most probable bias in a case–control study is selection, we have minimized this impact in the current study – all patients who developed sepsis in the study period were already enrolled. Controls were paired within the defined study group to reduce differences caused by other risk factors. Despite the attempt to control for confounding biases by pairing cases and controls, the apparent risk association may be spurious because of unmeasured characteristics. Furthermore, although caution must be exercised in claiming biological plausibility, the current knowledge of physiological mechanisms of organ dysfunction in sepsis supports the possibility of such a causal association [8].

Our findings of changes in lipoprotein concentrations during severe sepsis are consistent with the previous reports that showed low cholesterol, apolipoprotein and phospholipid levels and increased triglyceride levels in the acute-phase response [9,10,20,29–34]. In our patients, total cholesterol levels remained persistently low during the entire 28-day follow-up period.

An association between low serum HDL cholesterol levels and increased mortality in severe sepsis has been reported [10]; however, we could not demonstrate this association in our patients. The trends in HDL cholesterol levels were quite different in our study because all patients presented with low cholesterol levels on day one of sepsis, and only the survivors showed a tendency of increased HDL concentrations. On the other hand, Chien et al. [10] found that non-survivors had lower levels of cholesterol at day one that increased over time, while survivors had unchanged HDL cholesterol concentrations. Our study cohort was not large enough to detect a difference in HDL concentration between survivors and non-survivors. Given the results we obtained, if we were to calculate the sample needed to detect a 20% difference in HDL cholesterol concentrations between survivors and non-survivors with an 80% power and a 95% confidence, we would need 136 patients for each time point interval. In our cohort, we failed to observe numerical differences between the mean level of HDL cholesterol between survivors and non-survivors before day 5, as opposed to what has been observed by other authors [10].

The role of CETP in the acute-phase response has yet to be clearly established. Experimental studies have shown that the addition of LPS to plasma resulted in a marked decrease in CETP concentration and activity, leading to changes in HDL levels considered as an adaptive response to preserve or increase HDL [17,18]. There is a direct affinity interaction between CETP and LPS, and CETP has been shown to inhibit cytokine release by macrophages in a dose-dependent manner [19]. A published case series reported a similar trend of plasma CETP changes in humans, with a transient decrease in CETP activity during clinical sepsis [20]. Thus, to our knowledge, this study is the first to document a reduction in CETP plasma concentration during clinical sepsis alongside an association with an increased rate of mortality. We speculate that this reduction in CETP concentration during sepsis, similar to that found after exposure to LPS, may help preserve HDL after exposure to LPS and the protective effects of this lipoprotein.

In conclusion, ours is the first prospective study to report that low HDL cholesterol concentrations may be considered as a risk factor for the development of severe sepsis in hospitalized patients. Plasma HDL had a protective effect, as each 1 mg dL−1 increase in HDL decreased the odds of severe sepsis during hospitalization by 3%. The ongoing variation in lipoprotein concentrations during hospital stay indicated profound fluctuations in total cholesterol, HDL and LDL cholesterol, triglycerides, phospholipids and CETP during severe sepsis. The level of plasma CETP reduction on the third day of sepsis was associated with mortality. Further research is needed to elucidate whether these associations are causal or whether they represent uncontrolled data confounded by unmeasured risk factors.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

We thank the physicians and staff of the Londrina University Hospital who facilitated this study. We thank Dr Claudia M. D. M. Carrilho and Dr Joseani P. Garcia (Infection Control Committee, University Hospital, Londrina State University) for research support and data collection. None of those acknowledged received any compensation for their contributions.

Source of funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

Post-graduate programme in medicine at the Health Science Center of Londrina State University, and University Hospital Technology and Development Support Foundation (HUTEC).

Address

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References

Department of Internal Medicine, Londrina University Hospital, Londrina State University (C. M. C. Grion, L. T. Q. Cardoso, A. J. F. Carrilho); Medical Student, Londrina University Hospital, Londrina State University (T. F. Perazolo, A. S. Garcia); Department of Pathology, Clinical Analysis and Toxicology, Londrina University Hospital, Londrina State University (D. S. Barbosa, H. K. Morimoto); Department of Statistics, Londrina University Hospital, University (T. Matsuo), Londrina, Paraná, Brazil.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Source of funding
  9. Address
  10. References
  • 1
    Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR. Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med 2001;29:130310.
  • 2
    Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 2003;348:154654.
  • 3
    Silva E, Pedro MA, Sogayar ACB, Mohovic T, Silva CLO, Janiszewski M et al. Brazilian Sepsis Epidemiological Study (BASES study). Crit Care 2004;8:R25660.
  • 4
    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: 2008. Crit Care Med 2008;36:296327.
  • 5
    Parrillo JE. Mechanisms of disease: pathogenetic mechanisms of septic shock. N Engl J Med 1993;328:14718.
  • 6
    Wu A, Hinds CJ, Thiemermann C. High-density lipoproteins in sepsis and septic shock: metabolism, actions, and therapeutic applications. Shock 2004;21:21021.
  • 7
    Van Leeuwen HJ, Van Beek AP, Dallinga-Thie GM, Van Strijp JAG, Verhoef J, Van Kessel KPM. The role of high density lipoprotein in sepsis. Neth J Med 2001;59:10210.
  • 8
    Murch O, Collin M, Hinds CJ, Thiemermann C. Lipoproteins in inflammation and sepsis. I. Basic Science. Intensive Care Med 2007;33:1324.
  • 9
    Van Leeuwen HJ, Heezius ECJM, Dallinga GM, Van Strijp JAG, Verhoef J, Van Kessel KPM. Lipoprotein metabolism in patients with severe sepsis. Crit Care Med 2003;31:135966.
  • 10
    Chien JY, Jerng JS, Yu CJ, Yang PC. Low serum level of high-density lipoprotein cholesterol is a poor prognostic factor for severe sepsis. Crit Care Med 2005;33:168893.
  • 11
    Van Amersfoort ES, Van Berkel TJC, Kuiper J. Receptors, mediators, and mechanisms involved in bacterial sepsis and septic shock. Clin Microbiol Rev 2003;16:379414.
  • 12
    Vincent JL, Abraham E. The last 100 years of sepsis. Am J Respir Crit Care Med 2006;173:25663.
  • 13
    Kitchens RL, Thompson PA, Munford RS, O’Keefe GE. Acute inflammation and infection maintain circulating phospholipid levels and enhance lipopolysaccharide binding to plasma lipoproteins. J Lipid Res 2003;44:2239348.
  • 14
    Gray PW, Flaggs G, Leong SR, Gumina RJ, Weiss J, Ooi CE et al. Cloning of the cDNA of a human neutrophil bactericidal protein: structural and functional correlations. J Biol Chem 1989;264:95059.
  • 15
    Kirschning CJ, Au-Young J, Lamping N, Reuter D, Pfeil D, Seilhamer JJ et al. Similar organization of the lipopolysaccharide-binding protein (LBP) and phospholipid transfer protein (PLTP) genes suggests a common gene family of lipid-binding proteins. Genomics 1997;46:41625.
  • 16
    Connolly DT, Krul ES, Heuvelman D, Glenn KC. Inhibition of cholesteryl ester transfer protein by apolipoproteins, lipopolysaccharides, and cholesteryl sulfate. Biochim Biophys Acta 1996;1304:14560.
  • 17
    Masucci-Magoulas L, Moulis P, Jiang XC, Richardson H, Walsh A, Breslow JL et al. Decreased cholesteryl ester transfer protein (CETP) mRNA and protein and increased high density lipoprotein following lipopolysaccharide administration in human CETP transgenic mice. J Clin Invest 1995;95:158794.
  • 18
    Hardardóttir I, Moser AH, Fuller J, Fielding C, Feingold K, Grünfeld C. Endotoxin and cytokines decrease serum levels and extra hepatic protein and mRNA levels of cholesteryl ester transfer protein in Syrian hamsters. J Clin Invest 1996;97:258592.
  • 19
    Casita PM, Barbeiro DF, Moretti AI, Quintão EC, Soriano FG. Human cholesteryl ester transfer protein expression enhances the mouse survival rate in an experimental systemic inflammation model: a novel role for CETP. Shock 2008;30:5905.
  • 20
    Levels JHM, Pajkrt D, Schultz M, Hoek FJ, Van Tol A, Meijers JCM et al. Alterations in lipoprotein homeostasis during human experimental endotoxemia and clinical sepsis. Biochim Biophys Acta 2007;1771:142938.
  • 21
    Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections. Am J Infect Control 1988;16:12840.
  • 22
    Knaus WA, Draper EA, Wagner DP, Zimmerman JE. APACHE II: a severity of disease classification system. Crit Care Med 1985;13:81829.
  • 23
    Vincent JL, Mendonça A, Cantraine F, Moreno R, Takala J, Suter PM et al. Use of the SOFA score to assess the incidence of organ dysfunction/failure in intensive care units: results of a multicenter, prospective study. Crit Care Med 1998;26:1793800.
  • 24
    Members of the American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference Committee: American College of Chest Physicians/Society of Critical Care Medicine Consensus Conference. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Crit Care Med 1992;20:86474.
  • 25
    Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem 1972;18:499502.
  • 26
    Delgado-Rodriguez M, Medina-Cuadros M, Martinez-Gallego G, Sillero-Arenas M. Total cholesterol, HDL-cholesterol, and risk of nosocomial infection: a prospective study in surgical patients. Infec Control Hosp Epidemiol 1997;18:918.
  • 27
    Gordon BR, Parker TS, Levine DM, Saal SD, Wang JCL, Sloan BJ et al. Relationship of hypolipidemia to cytokine concentrations and outcomes in critically ill surgical patients. Crit Care Med 2001;29:15638.
  • 28
    Shor R, Wainstein J, Oz D, Boaz M, Matas Z, Fux A et al. Low serum LDL cholesterol levels and risk of fever, sepsis, and malignancy. Ann Clin Lab Sci 2007;37:3438.
  • 29
    Alvarez C, Ramos A. Lipids, lipoproteins, and apoproteins in serum during infection. Clin Chem 1986;32:1425.
  • 30
    Memis D, Gursoy O, Tasdogan M, Süt N, Kurt I, Türe M et al. High C-reactive protein and low cholesterol levels are prognostic markers of survival in severe sepsis. J Clin Anesth 2007;19:18691.
  • 31
    Levels JHM, Lemaire LCJM, Van Den Ende AE, Van Deventer SJH, Van Lanschot JB. Lipid composition and lipopolysaccharide binding capacity of lipoproteins in plasma and lymph of patients with systemic inflammatory response syndrome and multiple organ failure. Crit Care Med 2003;31:164753.
  • 32
    Fraunberger P, Pilz G, Cremer P, Werdan K, Walli AK. Association of serum tumor necrosis factor levels with decrease of cholesterol during septic shock. Shock 1998;10:35963.
  • 33
    Sammalkorpi K, Valtonen V, Kerttula Y, Nikkilä E, Taskinen MR. Changes in serum lipoprotein pattern induced by acute infections. Metabolism 1988;37:85965.
  • 34
    Gallin JI, Kaye D, O’Leary WM. Serum lipids in infection. N Engl J Med 1969;281:10816.