Biomarkers in critically ill patients with systemic inflammatory response syndrome or sepsis supplemented with high-dose selenium

Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

Contents lists available at ScienceDirect

Journal of Trace Elements in Medicine and Biology journal homepage: www.elsevier.com/locate/jtemb

Clinical Studies

Biomarkers in critically ill patients with systemic inflammatory response syndrome or sepsis supplemented with high-dose selenium Helena Brodska a , Jiri Valenta b , Karin Malickova a , Pavel Kohout c , Antonin Kazda a , Tomas Drabek d,∗ a

Institute of Clinical Biochemistry, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic Department of Anaesthesiology and Intensive Care, First Faculty of Medicine, Charles University in Prague and General University Hospital in Prague, Czech Republic c Department of Internal Medicine, Nutrition and Dietetic Center, Thomayer University Hospital, Prague, Czech Republic d Department of Anesthesiology, University of Pittsburgh School of Medicine, Pittsburgh, PA, United States b

a r t i c l e

i n f o

Article history: Received 14 January 2015 Accepted 25 February 2015 Keywords: Selenium Systemic inflammatory response syndrome Sepsis Prealbumin Albumin Cholesterol Selenoenzymes Glutathione peroxidase Outcome Mortality

a b s t r a c t Objective: Low levels of selenium (Se) and glutathione peroxidase (GSHPx), a key selenoenzyme, were documented in systemic inflammatory response syndrome (SIRS) and sepsis, both associated with high mortality. Se supplementation had mixed effects on outcome. We hypothesized that Se supplementation could have a different impact on biomarkers and 28-day mortality in patients with SIRS vs. sepsis. Methods: Adult patients with SIRS or sepsis were randomized to either high-dose (Se+, n = 75) or standarddose (Se−, n = 75) Se supplementation. Plasma Se, whole blood GSHPx activity, C-reactive protein (CRP), procalcitonin (PCT), prealbumin, albumin and cholesterol levels were measured serially up to day 14. Results: There was no difference in mortality between Se− (24/75) vs. Se+ group (19/75; p = 0.367) or between SIRS and septic patients (8/26 vs. 35/124; p = 0.794). There was a trend to reduced mortality in SIRS patients in the Se+ vs. Se− group (p = 0.084). Plasma Se levels increased in the Se+ group only in patients with sepsis but not in patients with SIRS. Plasma Se levels correlated with GSHPx. In SIRS/Se+ group, Se correlated only with GSHPx. In SIRS/Se− group, Se correlated with cholesterol but not with other biomarkers. In sepsis patients, Se levels correlated with cholesterol, GSHPx and prealbumin. Cholesterol levels were higher in survivors in the Se− group. Conclusions: Se levels correlated with GSHPx activity and other nutritional biomarkers with significant differences between SIRS and sepsis groups. High-dose Se supplementation did not affect mortality but a strong trend to decreased mortality in SIRS patients warrants further studies in this population. © 2015 Published by Elsevier GmbH.

Introduction Sepsis remains the leading cause of death in critically ill patients, with mortality rates of ∼20–55%. Severe sepsis is a leading cause of death in the United States and the most common cause of death among critically ill patients in non-coronary intensive care units (ICU) [1]. Similar mortality rates from sepsis have been described in multiple developed countries [2–9]. Selenium (Se) is an important micronutrient affecting multiple physiologic functions. Se, incorporated in selenoenzymes, plays a key role in protection against oxidative stress, a principle

∗ Corresponding author at: Department of Anesthesiology, University of Pittsburgh School of Medicine, 200 Lothrop Street, Pittsburgh, PA 15213, United States. Tel.: +1 412 647 4725; fax: +1 412 624 0943. E-mail address: [email protected] (T. Drabek). http://dx.doi.org/10.1016/j.jtemb.2015.02.005 0946-672X/© 2015 Published by Elsevier GmbH.

phenomenon in inflammatory response and sepsis, along with other pathological conditions [10]. Se levels decrease early in inflammation. This decrease is probably caused by destruction of the enzymes as well as redistribution of Se, and its consumption [11]. Several research groups [12–15] previously reported low Se concentrations in critically ill patients, especially when associated with infection. Se supplementation might thus be expected to have some advantages in patients with severe sepsis. Indeed, Se supplementation has been shown to increase plasma levels of Se and selenoenzymes in a matter of days [16–18]. Several small-scale studies found beneficial effects with Se supplementation in the critically ill. Unfortunately, these salutary effects were not consistent. A meta-analysis by the Cochrane group published in 2004 [19] found no evidence to support use of Se supplementation for primary or secondary prevention of sepsis in critically ill patients. In contrast, three more recently published meta-analyses reported a

26

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

reduction in the risk of death in patients receiving high-dose Se supplementation, especially in the most severely ill [20–22]. Angstwurm et al. [16] found that adjuvant treatment of patients with high-dose sodium selenite might reduce mortality rates in patients with severe sepsis or septic shock. In our own study of critically ill patients with systemic inflammatory response syndrome (SIRS) and sepsis, we were unable to reproduce these results. The principal findings of our study were that high-dose Se substitution in patients with SIRS/sepsis increased plasma Se, but did not reduce 28-day mortality. Negative correlations were demonstrated between plasma Se, biomarkers of inflammation and severity of status at admission [18]. It is possible that the case mix that included both patients with SIRS and sepsis masked the potential benefits of Se supplementation in our patient population. Thus, in this a priori planned sub-study, we evaluated our data selectively for patients in SIRS vs. patients in sepsis, to report secondary outcome endpoints, specifically albumin, prealbumin and cholesterol, and glutathioneperoxidase (GSHPx). In critically ill patients, these “nutritional” biomarkers traditionally viewed as biomarkers of nutrition reflect the metabolic response to stress and not nutrition status, becoming “inflammatory” biomarkers [23]. Our hypotheses were that (1) Se supplementation will increase plasma Se levels; (2) Se supplementation will decrease mortality; (3) Se supplementation will correlate with nutritional biomarkers and markers of antioxidant defense; and (4) biomarkers of nutrition will be higher in survivors vs. non-survivors. These outcome parameters will be evaluated separately in patients with sepsis vs. patients with SIRS as significant differences between the outcomes of these two subgroups could be expected, as well as in the groups with different Se supplementation protocols.

Disease categories included SIRS, sepsis, severe sepsis, septic shock. For the purpose of this study, we dichotomized the patient population into “SIRS” and “sepsis”, the latter comprised of the sepsis, severe sepsis and septic shock categories. The details of the laboratory test methods have been published in greater detail previously [18]. The primary outcome of the study was 28-day mortality as recommended by the Food and Drug Administration (FDA) for pivotal trials of new therapies in sepsis. The study was powered to detect differences between Se− vs. Se+ groups (for details please see Electronic Supplemental Material to ref [18]). Statistical analysis The data were analyzed using SPSS 22.0 for Windows software (SPSS, Chicago, IL, USA). Discrete variables are expressed as counts (percentage) and continuous variables as means ± standard deviation (SD) or median and interquartile range [IQR] unless stated otherwise. Categorical data were compared using the 2 test with Yates’s correction, Fisher’s exact test or the Cochran–Armitage trend test, as appropriate. A Shapiro–Wilk’s test was used to verify the normality of distributions of continuous variables. Continuous variables conforming to a normal distribution were compared using Student’s t-test; otherwise, the Mann–Whitney U test was applied. A Pearson’s or Spearman’s rank-order correlation was run to assess the relationship between two variables according to the normality of data. The power of D0 or D1 biomarkers to predict the overall 28-day mortality rate was assessed with receiver operator characteristic curve analysis. A p-value of <0.05 was considered statistically significant. Results

Methods The study was approved by the Joint Institutional Ethics Committee of the First Faculty of Medicine, Charles University in Prague, and the General University Hospital in Prague, Czech Republic. A separate consent was obtained from the Institutional Review Board of the University of Pittsburgh to process the de-identified data. The study enrolled patients older than 18 years admitted for SIRS or sepsis with an initial sequential organ failure assessment (SOFA) score of >5. An informed consent was obtained from the patient or legal representative prior to the enrollment. This was a prospective, randomized, open-label single center study carried out at the University Hospital mixed ICU in Prague from March 2004 until April 2009. In brief, eligible patients with SIRS, sepsis or septic shock were randomized to either high-dose (Se+, n = 75) or standard-dose (Se−, n = 75) Se supplementation. Patients in the Se+ group received 1000 ␮g of Se on the first day, 500 ␮g/day on subsequent days in the form of Na-selenite pentahydrate (Na2 SeO3 ·5H2 O), in which 100 ␮g of Se corresponds to 333 ␮g of Na-selenite (Selenase T; Biosyn, Germany). Se was administered as a 30-min infusion for 5–14 days, according to the duration of the hospital stay. The patients in the Se− group did not receive any extra Se. Both groups received standard Se supplementation, as Na-selenite, added to the parenteral nutrition (<75 ␮g/day, or 0.38–0.95 mmol/day, respectively). Plasma Se, whole blood GSHPx activity, C-reactive protein (CRP), procalcitonin (PCT), prealbumin, albumin and cholesterol levels were measured at baseline (D0, on the day of admission) before the first dose of Na-selenite, and on the days 1, 3, 5, 7, 10 and 14. Severity of clinical status was assessed on admission (D0) using an Acute Physiology and Chronic Health Evaluation (APACHE) II, and on days 1, 3, 5, 7, 10 and 14 using SOFA scores, grading the underlying disabilities and the severity of current illness, respectively.

The baseline characteristics of the patient population are shown in Table 1. There were no differences between Se− and Se+ groups, respectively. Outcome – hypothesis 1 – plasma Se levels Plasma Se levels were similar at baseline in Se− vs. Se+ groups. Se levels increased only in the Se+ group but remained low in the Se− group (Fig. 1). Plasma levels increased in the Se+ group only in the patients with sepsis but not in patients with SIRS (Fig. 2). There were no differences between plasma Se levels in SIRS vs. sepsis groups in individual days. Outcome – hypothesis 2 – mortality The overall mortality of the patients in the study was 28.7% (43 out of 150 patients). There was no difference in mortality between Se− group (32.0%, 24/75) vs. Se+ group (25.3%, 19/75; p = 0.367). There was no difference in mortality between SIRS and septic patients irrespective of Se supplementation (8/26 vs. 35/124; p = 0.794). Also, there was no difference in mortality in SIRS or sepsis patients evaluated separately, although there was a trend to reduced mortality in SIRS patients in the Se+ vs. Se− group (p = 0.084; Table 2). Outcome – hypothesis 3 – correlation with nutritional biomarkers and GSHPx In patients with SIRS and sepsis together, Se levels correlated with levels of cholesterol, prealbumin and GSHPx, but not with albumin (Table 3, top panel). Plasma Se levels correlated with whole blood GSHPx activity in the Se+ group. The correlations were stronger with increasing severity of disease (Fig. 3). In a subgroup

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

27

Table 1 Demographic characteristics of the patient population. Group

p

Se− n = 75

Se+ n = 75

Age (mean ± SD) Gender (M/F) APACHE II (mean ± SD) (median) SOFA (median, IQR)

60 ± 15 51/24 28 ± 11 (26) 11 (8,14)

60 ± 16 46/29 30 ± 10 (28) 12 (10,14)

Sepsis severity – n (%) SIRS Sepsis Severe sepsis Septic shock

15 (20%) 24 (32%) 28 (37%) 8 (11%)

11 (15%) 20 (27%) 33 (44%) 11 (15%)

0.900 0.393 0.190 0.360

APACHE, Acute Physiology and Chronic Health Evaluation; SOFA, sequential organ failure assessment; SIRS, systemic inflammatory response syndrome.

Table 2 Mortality in individual subgroups according to SIRS vs. sepsis status and Se− vs. Se+ randomization. Non-survivors SIRS p = 0.084 Se− Se+ Total Sepsis p = 0.969 Se− Se+ Total

Survivors

Total

7 1 8

8 10 18

15 11 26

17 18 35

43 46 89

60 64 124

prealbumin in both Se− and Se+ groups, and with albumin only in the Se− group. These correlations were stronger in the Se+ group (Table 3, bottom panel). There was a statistically significant correlation between Se levels and cholesterol levels in patients supplemented with Se (p < 0.01).

Fig. 1. Plasma selenium levels in patients with SIRS/sepsis according to Se supplementation.

analysis, in SIRS patients in the Se+ group, Se correlated only with GSHPx. In SIRS patients in the Se− group, Se correlated with cholesterol but not with other biomarkers (Table 3, middle panel). In sepsis patients, Se levels correlated with cholesterol, GSHPx and

Outcome – hypothesis 4 – biomarkers of inflammation and nutrition will be higher in survivors vs. non-survivors SIRS and sepsis together On admission (D0), lower Se was associated with increased 28day mortality (area under curve [AUC] 0.374; p = 0.016; Fig. 4). Also, neither biomarkers of inflammation (CRP, PCT, GSHPx) nor

Fig. 2. Plasma selenium levels during in patients with SIRS (left panel) or sepsis (right panel) according to Se supplementation. Plasma Se levels increased only in the sepsis group.

28

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

Table 3 Correlation between Se and nutritional biomarkers in patients with SIRS/sepsis (top panel), in SIRS only (middle panel) or sepsis only (bottom panel). ALB

CHOL

GSHPx

PREALB

Se

280 −.077 .192 287

.324** .000 287 .286** .000 325

.197* .055 98 .664** .000 119

.246** .000 279 .337** .000 276

1.000 . 313 1.000 . 331

Correlations – Se− vs. Se+ group in SIRS Se− Correlation coefficient Sig. (2-tailed) N Correlation Coefficient Se+ Sig. (2-tailed) N

.108 .294 97 −.180 .131 72

.359** .000 97 .009 .942 72

.383 .079 22 .638** .006 17

.106 .300 97 .161 .176 72

Correlations – Se− vs. Se+ group in sepsis Se− Correlation coefficient Sig. (2-tailed) N Se+ Correlation coefficient Sig. (2-tailed) N

.151* .046 175 .013 .850 211

.282** .000 182 .379** .000 249

.271* .018 76 .639** .000 100

.323** .000 174 .407** .000 201

Correlations – Se− vs. Se+ group in SIRS/sepsis combined Se− Correlation coefficient Sig. (2-tailed) N Correlation coefficient Se+ Sig. (2-tailed) N

.155** .010

1.000 . 98 1.000 . 72 1.000 . 207 1.000 . 254

Statistically significant correlations based on Spearman’s rho test are highlighted. * Correlation is significant at the 0.05 level (2-tailed). ** Correlation is significant at the 0.01 level (2-tailed).

nutrition (albumin, prealbumin, cholesterol) indicated association with 28-day mortality when assessed at admission. On D1, both Se (AUC 0.387; p = 0.031) and GSHPx (AUC 0.300; p = 0.026) negatively correlated with mortality. The inflammatory biomarkers CRP (AUC 0.613; p = 0.031) and PCT (AUC 0.639; p = 0.008) showed positive associations. These correlations were even more pronounced when data from D0 were combined with data from D1: Se (AUC 0.380; p = 0.001), GSHPx (AUC 0.313; p = 0.003) and PCT (AUC 0.614; p = 0.002). In contrast, none of the nutritional biomarkers showed correlation with outcome.

Fig. 3. Correlation between plasma selenium levels and whole blood GSHPx activity according to the severity of status in the Se+ group.

SIRS and sepsis separately We found significant differences between selected outcome parameters in patients with SIRS vs. sepsis. Lower Se on admission (D0) showed a marginal statistically significant association with mortality in patients with SIRS (AUC 0.253; p = 0.049; Fig. 4). However, this association failed to reach statistical significance in patients with sepsis (AUC 0.396; p = 0.073; Fig. 4). Similarly, lower Se on D1 was associated with higher mortality in patients with SIRS (AUC 0.205; p = 0.018) but not in patients with sepsis (AUC 0.421; p = 0.172). No other significant associations with mortality

Fig. 4. Receiver operating characteristics of Se on admission for predicting 28-day mortality. Left panel, both SIRS and sepsis patients (AUC 0.374; p = 0.016); middle panel, patients with SIRS (AUC 0.253, p = 0.049), right panel, patients with sepsis (AUC 0.396, p = 0.073).

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

29

Fig. 5. Plasma cholesterol levels in survivors vs. non-survivors in the Se− (left panel) and Se+ groups, respectively. The boxplots represent 25–75% interquartile range. The line across the box represents the median. The whiskers represent min–max values. The round and asterisk markers represent outliers (“out” and “far out”).

were found in subgroups (SIRS or sepsis) for other variables i.e. albumin, cholesterol, prealbumin or GSHPx. In pooled data from all measurements, lower Se levels were associated with mortality in patients with SIRS (AUC 0.278; p < 0.001) but not in sepsis (AUC 0.455; p = 0.079). In contrast, prealbumin levels were associated with mortality not in patients with SIRS but only in patients with sepsis (AUC 0.417; p = 0.003). This was similar to cholesterol that showed negative association with mortality only in patients with sepsis (AUC 0.446; p = 0.041). Inflammatory biomarkers showed overall better associations, namely CRP in septic patients (AUC 0.610; p < 0.001) and PCT in both patients with SIRS (AUC 0.666; p = 0.002) and sepsis (AUC 0.655; p < 0.001). Plasma cholesterol levels (from all measurements) in the Se− group were higher in survivors vs. non-survivors (p < 0.001; Fig. 5). In contrast, there were no differences in cholesterol levels between survivors and non-survivors in the Se+ group (p = 0.546; Fig. 5).

Discussion While the overall mortality from sepsis had declined over the last decade worldwide [24], sepsis remains associated with high mortality. The initial criteria for sepsis and its stages, originally postulated in 1991 [25], were revisited in 2001 [26]. It was concluded that while SIRS remains a useful concept, the diagnostic criteria for SIRS are overly sensitive and non-specific. The inclusion of patients diagnosed with SIRS in reports focused on sepsis complicates interpretation of results. It is becoming clear that these represent distinct entities. There are ongoing efforts to better characterize differences in pathophysiological mechanisms operating in SIRS vs. sepsis that would unveil potential targets for future therapies. Multiple similarities but also distinct characteristic features were identified [27]. Different outcomes in critically ill patients with SIRS, sepsis or septic shock have been reported in a large descriptive study in Italian ICUs as early as in the 1990s [28]. Linder et al. [29] in a longitudinal study showed that long-term mortality of younger previously healthy patients with severe sepsis/septic shock is worse than that of patients with non-septic critical illness and of the general population, suggesting important differences in trajectories of these conditions.

Multiple biomarkers were tested in an effort to differentiate between SIRS and sepsis early in the disease course, including PCT [30,31], CRP [32], sRAGE [33], interleukins [31] and others, often assembled into a “diagnostic panel” [31,34]. Other types of biomarkers based on nutrition parameters are also used in critical care setting, e.g. cholesterol [35–38] and its fractions [39,40], prealbumin [41–43], or albumin [41]. It was proposed that these nutritional parameters are influenced by inflammatory mediators, acting as biomarkers of inflammation rather than as nutritional parameters [37]. Se is an essential nutrient in the diet with anti-inflammatory properties. Many of the benefits of Se are thought to be due to antioxidant and redox-regulating properties of certain selenoproteins [44]. Selected selenoproteins, namely GSHPx [45] or selenoprotein P [46] have also been used as biomarkers in inflammatory conditions. Sodium selenite shows both direct oxidant effects and indirect anti-oxidant properties in an experimental model of sepsis [47]. The lack effect of Se supplementation in our study is corroborated by the disappointing results from other studies. Andrews et al. [48] failed to show improvement in mortality in critically ill patients supplemented with Se. Heyland et al. [49] in a large multicenter study showed that addition of the antioxidants (including 500 ␮g Se) into the artificial nutrition for critically ill patients had no effect on 28-day mortality. In fact, another post hoc analysis of the data from the aforementioned study indicated that there is potential for harm from antioxidants in patients with multi-organ failure (MOF) that included renal dysfunction upon study enrollment [50]. These results are corroborated by the lack of effect of Se supplementation in septic patients with renal failure in whom GSHPx synthesis is impaired, while patients with lung injury may still show benefit [51]. Similarly, immune-modulating high-protein nutrients (including 285 ␮g Se) in medically critically ill were associated with an increased mortality at 6 months [52]. In contrast, three recent meta-analyses confirmed the hypothesis that high-dose supplementation Se may reduce mortality in critically ill septic patients [20,22,53]. These conflicting results suggest that further studies focused on identifying the most promising target patient population and optimizing the therapeutic regimen may be needed. In our study, we have found low Se level and low GSHPx levels on admission. Our results are consistent with findings of others

30

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

[14,45,54–57]. In patients with SIRS, Manzanares et al. reported that SIRS and MOF were associated with early decreases in Se and GSHPx. Low Se and GSHPx correlated with severity of disease, and Se levels were predictive for ICU mortality. High-dose supplementation quickly restored Se levels and also increased whole blood GSHPx activity [45]. Forceville et al. documented that standard Se supplementation did not restore Se levels. Only mild increase was seen in survivors, while prolonged decrease was associated with increase in mortality [14]. No increase in plasma Se levels was seen in controls in our study. Low plasma Se levels were also found in patient with SIRS [45,56]. The trend for improved survival in Se+ group in patients with SIRS in our study is surprising since Se supplementation did not result in increased levels of plasma Se. Also, GSHPx remained low in Se+ group in patients with SIRS. It could be hypothesized that other pathophysiological mechanisms played a role [58]. The anti-oxidant effect may not be the key effect in this subpopulation. Future studies could selectively target this patient population and should be appropriately powered to test the hypothesis that Se improves mortality in patients with SIRS. We have documented a significant correlation between low Se level on admission and increased 28-day mortality. This interaction has been also observed by others [55]. Duncan et al. [59] reported that plasma Se correlated with severity of systemic inflammatory reaction and markers of inflammation, namely CRP. The results from our study are in line with these reports and the correlations between Se levels and PCT or CRP have been reported by our research group previously [18]. There has been a growing evidence for the role of lipid profiles as markers of severity of sepsis. However, the results for total cholesterol predictive value remain mixed. While some studies support the association between early total cholesterol levels and mortality from sepsis [36,39], others do not support this association [40,60]. In septic patients, admission cholesterol levels did not correlate with severity of disease, but recovered in those whose clinical condition improved, while they stayed low in those who fared poorly [37]. Along with albumin, cholesterol could thus serve as a biomarker of illness severity in the delayed phases of the disease. Indeed, a recent study identified cholesterol as a more valuable predictor of outcome in septic patients than CRP or PCT [35], traditional biomarkers used in sepsis staging and prognostication [61]. However, there remains skepticism about prognostic value of nutritional parameters in critically ill [62]. Our findings did not show any correlation between admission cholesterol levels and 28-day mortality. However, pooled data from all measurements show that cholesterol levels were higher in survivors vs. non-survivors in the Se− group. There were no differences between survivors vs. non-survivors in the Se+ group. While serial assessment of cholesterol levels could provide an additional insight into the trajectory of the disease, we did not find that its initial value would facilitate the prognostication process. A more detailed analysis of cholesterol fractions, namely HDL-C, could yield better outcome, as suggested by others. Low prealbumin levels have been documented in both survivors and non-survivors of critical illness [41]. Prealbumin identified patients at highest risk for metabolic losses associated with stress hypermetabolism as serum prealbumin levels did not respond early to nutrition support because of the delayed return to anabolic status [42]. Also, CRP to prealbumin ratio correlated with severity of disease and mortality [43]. Several studies documented an association with higher prealbumin levels and better outcomes in critically ill patients [62]. In our study, Se levels correlated with prealbumin levels only in patients with sepsis but not in SIRS. Early prealbumin levels were not associated with 28-day mortality. In our study, albumin failed to show any associations with severity of disease or mortality. Our results are supported by others

that did not observe differences between septic patients vs. controls [41]. This may be explained by the long half-life of albumin. The evaluation of albumin levels could be biased by simultaneous administration of albumin as a volume replacement during sepsis as recently recommended [63]. In a recent meta-analysis, the use of albumin-containing solutions for the resuscitation of patients with sepsis was associated with lower mortality compared with other fluid resuscitation regimens [64]. However, it should be noted that addition of albumin to the therapy in a recent multi-center trial did not have an impact on 28-day mortality or 90-day mortality [7]. Lower blood Se levels were associated with lower serum albumin and enzyme activities in cancer patients [65]. Lower albumin levels were also associated with increased mortality in critically ill patients [55]. However, recent systematic review reported that only two out of ten high-quality studies found any association between albumin levels and improved outcome [62]. Our study has several limitations. In our original concept, we enrolled patients with both SIRS and sepsis. The proportion of patients with SIRS was small, comprising only 17% of the patient population. Thus, our study might have been underpowered for detecting differences in this particular subgroup. However, we felt it important to report our findings that could be viewed as preliminary, but certainly inspiring future research ideas in this area. We report the patient stratification into SIRS/sepsis categories based on the day of admission but we are aware that some patients transitioned into sepsis during their hospital stay. We decided to maintain the category as initially diagnosed. In conclusion, patients with both SIRS and sepsis had low Se levels and GSHPx activity on admission. High-dose Se supplementation increased plasma Se levels after D1. Surprisingly, the increased Se levels were documented only in the sepsis subgroup but not in the SIRS subgroup. Standard-dose supplementation in the Se− group did not increase the low Se levels documented in both SIRS and sepsis. The overall mortality was not different between Se− vs. Se+ group or its subgroups, but there was trend toward improved mortality in Se+ group with SIRS. We previously reported a trend to improved mortality in the most severely ill patients. Plasma Se levels correlated with some assessed biomarkers and GSHPx, with better correlation in septic patients. Lower Se values on admission were associated with 28-day mortality. This was driven by data from the SIRS rather than sepsis group. Inflammatory biomarkers, but not nutritional biomarkers assessed on D1, were associated with mortality. Cholesterol levels were higher in survivors vs. non-survivors in the Se− group. Future studies may target either patients with severe SIRS or patients in severe sepsis or septic shock, using high-dose Se supplementation with bolus dosing and sufficient duration of treatment.

Conflict of interest The authors report no conflict of interest.

Acknowledgement Supported by projects PRVOUK P25/LF1/2 (2012–2016) of Charles University in Prague and RVO VFN 64165 of the Ministry of Health of the Czech Republic.

References [1] Mayr FB, Yende S, Angus DC. Epidemiology of severe sepsis. Virulence 2014;5:4–11. [2] Zhou J, Qian C, Zhao M, Yu X, Kang Y, Ma X, et al. Epidemiology and outcome of severe sepsis and septic shock in intensive care units in mainland China. PLoS ONE 2014;9:e107181.

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32 [3] Engel C, Brunkhorst FM, Bone HG, Brunkhorst R, Gerlach H, Grond S, et al. Epidemiology of sepsis in Germany: results from a national prospective multicenter study. Intensive Care Med 2007;33:606–18. [4] Brun-Buisson C, Doyon F, Carlet J, Dellamonica P, Gouin F, Lepoutre A, et al. Incidence, risk factors, and outcome of severe sepsis and septic shock in adults. A multicenter prospective study in intensive care units. French ICU Group for Severe Sepsis. JAMA 1995;274:968–74. [5] Blanco J, Muriel-Bombin A, Sagredo V, Taboada F, Gandia F, Tamayo L, et al. Incidence, organ dysfunction and mortality in severe sepsis: a Spanish multicentre study. Crit Care 2008;12:R158. [6] Kaukonen KM, Bailey M, Suzuki S, Pilcher D, Bellomo R. Mortality related to severe sepsis and septic shock among critically ill patients in Australia and New Zealand, 2000–2012. JAMA 2014;311:1308–16. [7] Caironi P, Tognoni G, Masson S, Fumagalli R, Pesenti A, Romero M, et al. Albumin replacement in patients with severe sepsis or septic shock. N Engl J Med 2014;370:1412–21. [8] Investigators P, Yealy DM, Kellum JA, Huang DT, Barnato AE, Weissfeld LA, et al. A randomized trial of protocol-based care for early septic shock. N Engl J Med 2014;370:1683–93. [9] Jawad I, Luksic I, Rafnsson SB. Assessing available information on the burden of sepsis: global estimates of incidence, prevalence and mortality. J Glob Health 2012;2:010404. [10] Robberecht H, Van Cauwenbergh R, Hermans N. Blood selenium levels and factors influencing concentration values. Trace Elem Electroly 2012;29:172–88. [11] Valenta J, Brodska H, Drabek T, Stach Z, Zima T, Kazda A. Selenium: an important trace element and therapeutic adjunct in critical care. Trace Elem Electroly 2012;29:246–55. [12] Kazda A, Brodska H, Vinglerova H, Blaha J, Valenta J, Stritesky M, et al. Selenium and zinc plasmatic levels in intensive care patients. Crit Care 2004;8(Suppl 1):P265. [13] Sakr Y, Reinhart K, Bloos F, Marx G, Russwurm S, Bauer M, et al. Time course and relationship between plasma selenium concentrations, systemic inflammatory response, sepsis, and multiorgan failure. Br J Anaesth 2007;98:775–84. [14] Forceville X, Vitoux D, Gauzit R, Combes A, Lahilaire P, Chappuis P. Selenium, systemic immune response syndrome, sepsis, and outcome in critically ill patients. Crit Care Med 1998;26:1536–44. [15] Hawker FH, Stewart PM, Snitch PJ. Effects of acute illness on selenium homeostasis. Crit Care Med 1990;18:442–6. [16] Angstwurm MW, Engelmann L, Zimmermann T, Lehmann C, Spes CH, Abel P, et al. Selenium in Intensive Care (SIC): results of a prospective randomized, placebo-controlled, multiple-center study in patients with severe systemic inflammatory response syndrome, sepsis, and septic shock. Crit Care Med 2007;35:118–26. [17] Manzanares W, Biestro A, Torre MH, Galusso F, Facchin G, Hardy G. Highdose selenium reduces ventilator-associated pneumonia and illness severity in critically ill patients with systemic inflammation. Intensive Care Med 2011;37:1120–7. [18] Valenta J, Brodska H, Drabek T, Hendl J, Kazda A. High-dose selenium substitution in sepsis: a prospective randomized clinical trial. Intensive Care Med 2011;37:808–15. [19] Avenell A, Noble DW, Barr J, Engelhardt T. Selenium supplementation for critically ill adults. Cochrane Database Syst Rev 2004:CD003703. [20] Alhazzani W, Jacobi J, Sindi A, Hartog C, Reinhart K, Kokkoris S, et al. The effect of selenium therapy on mortality in patients with sepsis syndrome: a systematic review and meta-analysis of randomized controlled trials. Crit Care Med 2013;41:1555–64. [21] Manzanares W, Dhaliwal R, Jiang X, Murch L, Heyland DK. Antioxidant micronutrients in the critically ill: a systematic review and meta-analysis. Crit Care 2012;16:R66. [22] Landucci F, Mancinelli P, De Gaudio AR, Virgili G. Selenium supplementation in critically ill patients: a systematic review and meta-analysis. J Crit Care 2014;29:150–6. [23] McClave SA, Kozar R, Martindale RG, Heyland DK, Braga M, Carli F, et al. Summary points and consensus recommendations from the North American Surgical Nutrition Summit. JPEN J Parenter Enteral Nutr 2013;37: 99S–105S. [24] Levy MM, Dellinger RP, Townsend SR, Linde-Zwirble WT, Marshall JC, Bion J, et al. The Surviving Sepsis Campaign: results of an international guidelinebased performance improvement program targeting severe sepsis. Crit Care Med 2010;38:367–74. [25] 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;20:864–74. [26] Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, et al. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Intensive Care Med 2003;29:530–8. [27] Fry DE. Sepsis, systemic inflammatory response, and multiple organ dysfunction: the mystery continues. Am Surg 2012;78:1–8. [28] Salvo I, de Cian W, Musicco M, Langer M, Piadena R, Wolfler A, et al. The Italian SEPSIS study: preliminary results on the incidence and evolution of SIRS, sepsis, severe sepsis and septic shock. Intensive Care Med 1995;21(Suppl. 2): S244–9. [29] Linder A, Guh D, Boyd JH, Walley KR, Anis AH, Russell JA. Long-term (10-year) mortality of younger previously healthy patients with severe sepsis/septic shock is worse than that of patients with nonseptic critical illness and of the general population. Crit Care Med 2014;42:2211–8.

31

[30] Hohn A, Schroeder S, Gehrt A, Bernhardt K, Bein B, Wegscheider K, et al. Procalcitonin-guided algorithm to reduce length of antibiotic therapy in patients with severe sepsis and septic shock. BMC Infect Dis 2013;13:158. [31] 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. [32] Vincent JL, Donadello K, Schmit X. Biomarkers in the critically ill patient: Creactive protein. Crit Care Clin 2011;27:241–51. [33] Brodska H, Malickova K, Valenta J, Fabio A, Drabek T. Soluble receptor for advanced glycation end products predicts 28-day mortality in critically ill patients with sepsis. Scand J Clin Lab Invest 2013;73:650–60. [34] Punyadeera C, Schneider EM, Schaffer D, Hsu HY, Joos TO, Kriebel F, et al. A biomarker panel to discriminate between systemic inflammatory response syndrome and sepsis and sepsis severity. J Emerg Trauma Shock 2010;3:26–35. [35] Biller K, Fae P, Germann R, Drexel H, Walli AK, Fraunberger P. Cholesterol rather than procalcitonin or C-reactive protein predicts mortality in patients with infection. Shock 2014;42:129–32. [36] Memis D, Gursoy O, Tasdogan M, Sut N, Kurt I, Ture M, et al. High C-reactive protein and low cholesterol levels are prognostic markers of survival in severe sepsis. J Clin Anesth 2007;19:186–91. [37] Lopez Martinez J, Sanchez Castilla M, Ordonez Gonzalez FJ, Temprano Vazquez S, Garcia de Lorenzo A, del Nogal Saez F. The usefulness of cholesterol as a nutritional-metabolic marker in the septic patient. Nutr Hosp 1995;10: 24–31. [38] Dunham CM, Fealk MH, Sever 3rd WE. Following severe injury, hypocholesterolemia improves with convalescence but persists with organ failure or onset of infection. Crit Care 2003;7:R145–53. [39] Barlage S, Gnewuch C, Liebisch G, Wolf Z, Audebert FX, Gluck T, et al. Changes in HDL-associated apolipoproteins relate to mortality in human sepsis and correlate to monocyte and platelet activation. Intensive Care Med 2009;35:1877–85. [40] 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;33:1688–93. [41] Boles JM, Garre MA, Youinou PY, Mialon P, Menez JF, Jouquan J, et al. Nutritional status in intensive care patients: evaluation in 84 unselected patients. Crit Care Med 1983;11:87–90. [42] Devakonda A, George L, Raoof S, Esan A, Saleh A, Bernstein LH. Transthyretin as a marker to predict outcome in critically ill patients. Clin Biochem 2008;41:1126–30. [43] Pinilla JC, Hayes P, Laverty W, Arnold C, Laxdal V. The C-reactive protein to prealbumin ratio correlates with the severity of multiple organ dysfunction. Surgery 1998;124:799–805, discussion 05-6. [44] Mattmiller SA, Carlson BA, Sordillo LM. Regulation of inflammation by selenium and selenoproteins: impact on eicosanoid biosynthesis. J Nutr Sci 2013;2:e28. [45] Manzanares W, Biestro A, Galusso F, Torre MH, Manay N, Pittini G, et al. Serum selenium and glutathione peroxidase-3 activity: biomarkers of systemic inflammation in the critically ill. Intensive Care Med 2009;35:882–9. [46] Forceville X, Mostert V, Pierantoni A, Vitoux D, Le Toumelin P, Plouvier E, et al. Selenoprotein P, rather than glutathione peroxidase, as a potential marker of septic shock and related syndromes. Eur Surg Res 2009;43:338–47. [47] Forceville X, Touati S, Le Toumelin P, Ducros V, Laporte F, Chancerelle Y, et al. Elements of margin of safety, toxicity and action of sodium selenite in a lipopolysaccharide rat model. J Trace Elem Med Biol 2014;28:303–10. [48] Andrews PJ, Avenell A, Noble DW, Campbell MK, Croal BL, Simpson WG, et al. Randomised trial of glutamine, selenium, or both, to supplement parenteral nutrition for critically ill patients. BMJ 2011;342:d1542. [49] Heyland D, Muscedere J, Wischmeyer PE, Cook D, Jones G, Albert M, et al. A randomized trial of glutamine and antioxidants in critically ill patients. N Engl J Med 2013;368:1489–97. [50] Heyland DK, Elke G, Cook D, Berger MM, Wischmeyer PE, Albert M, et al. Glutamine and antioxidants in the critically ill patient: a post hoc analysis of a large-scale randomized trial. JPEN J Parenter Enteral Nutr 2014. [51] Kocan L, Vaskova J, Vasko L, Simonova J, Simon R, Firment J. Selenium adjuvant therapy in septic patients selected according to Carrico index. Clin Biochem 2014;47:44–50. [52] van Zanten AR, Sztark F, Kaisers UX, Zielmann S, Felbinger TW, Sablotzki AR, et al. High-protein enteral nutrition enriched with immune-modulating nutrients vs standard high-protein enteral nutrition and nosocomial infections in the ICU: a randomized clinical trial. JAMA 2014;312:514–24. [53] Huang TS, Shyu YC, Chen HY, Lin LM, Lo CY, Yuan SS, et al. Effect of parenteral selenium supplementation in critically ill patients: a systematic review and meta-analysis. PloS ONE 2013;8:e54431. [54] Iglesias SB, Leite HP, Paes AT, Oliveira SV, Sarni RO. Low plasma selenium concentrations in critically ill children: the interaction effect between inflammation and selenium deficiency. Crit Care 2014;18:R101. [55] Stefanowicz F, Gashut RA, Talwar D, Duncan A, Beulshausen JF, McMillan DC, et al. Assessment of plasma and red cell trace element concentrations, disease severity, and outcome in patients with critical illness. J Crit Care 2014;29:214–8. [56] Oakes EJ, Lyon TD, Duncan A, Gray A, Talwar D, O’Reilly DS. Acute inflammatory response does not affect erythrocyte concentrations of copper, zinc and selenium. Clin Nutr 2008;27:115–20. [57] Woth G, Nagy B, Merei A, Ernyey B, Vincze R, Kaurics Z, et al. The effect of Na-selenite treatment on the oxidative stress-antioxidants balance of multiple organ failure. J Crit Care 2014;29(883):e7–11.

32

H. Brodska et al. / Journal of Trace Elements in Medicine and Biology 31 (2015) 25–32

[58] Pierre JF, Heneghan AF, Lawson CM, Wischmeyer PE, Kozar RA, Kudsk KA. Pharmaconutrition review: physiological mechanisms. JPEN J Parenter Enteral Nutr 2013;37:51S–65S. [59] Duncan A, Talwar D, McMillan DC, Stefanowicz F, O’Reilly DS. Quantitative data on the magnitude of the systemic inflammatory response and its effect on micronutrient status based on plasma measurements. Am J Clin Nutr 2012;95:64–71. [60] Lekkou A, Mouzaki A, Siagris D, Ravani I, Gogos CA. Serum lipid profile, cytokine production, and clinical outcome in patients with severe sepsis. J Crit Care 2014;29:723–7. [61] Tschaikowsky K, Hedwig-Geissing M, Braun GG, Radespiel-Troeger M. Predictive value of procalcitonin, interleukin-6, and C-reactive protein for survival in postoperative patients with severe sepsis. J Crit Care 2011;26:54–64.

[62] Ferrie S, Allman-Farinelli M. Commonly used nutrition indicators do not predict outcome in the critically ill: a systematic review. Nutr Clin Pract 2013;28:463–84. [63] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med 2013;41:580–637. [64] Delaney AP, Dan A, McCaffrey J, Finfer S. The role of albumin as a resuscitation fluid for patients with sepsis: a systematic review and meta-analysis. Crit Care Med 2011;39:386–91. [65] Robinson MF, Godfrey PJ, Thomson CD, Rea HM, van Rij AM. Blood selenium and glutathione peroxidase activity in normal subjects and in surgical patients with and without cancer in New Zealand. Am J Clin Nutr 1979;32:1477–85.