Sepsis has been recently defined as a life-threatening organ dysfunction caused by a dysregulated host response to infection. It is a leading cause of mortality and its incidence is increasing worldwide (1). Mechanisms of disease and the role of many biomarkers are still not fully understood, and further studies are required to determine if these biomarkers have clinical implications for diagnosis, treatment, or prognostic purposes (2).
Advanced glycation end products (AGEs) are a group of heterogeneous compounds, some of them fluorescent, produced by nonenzymatic glycation of proteins and lipids. Modification of organic matrix proteins by AGEs deteriorates the structural integrity and physiological function of multiple organ systems (3, 4).
The size of the pool of AGEs in the body is associated with their increased ingestion and production, and their decreased degradation and renal clearance (5). While plasma AGEs reflect their current circulating levels, tissue AGEs reflect accumulated levels. The main causes of AGEs elevation comprise: aging; diet; smoking; chronic hyperglycaemia; and chronic kidney disease (CKD), but acute elevation of AGEs can also occur from oxidative stress and inflammation (3).
Furthermore, the interaction of AGEs with their cell surface receptor (RAGE) sets off a cascade of events including oxidative stress generation and inflammation. AGE-RAGE binding results in intracellular signaling, leading to the activation of the transcription factor NFκB, which plays a key role in expression of cytokine and inflammatory mediators (6).
A soluble form of RAGE (sRAGE), consisting of the extracellular domain without the transmembrane and cytoplasmic domains of RAGE, has been implicated in several diseases. Nevertheless, its role in sepsis is controversial since sRAGE can propagate inflammation, but it also can prevent ligands from interacting with cellular RAGE, acting as a decoy with anti-inflammatory properties (7).
Studies about the implication of AGE-RAGE axis in sepsis have focused on sRAGE (8–11) rather than on plasma AGEs (12, 13), while few works have additionally included analysis of cellular RAGE (7) or skin AGEs (14). Despite these efforts, data is still insufficient and sometimes conflicting. The aim of the present study was to investigate the behavior of plasma and skin AGEs and sRAGE during the septic process, focusing on the time courses of these specific biomarkers. The primary outcome was mortality at day 28.
PATIENTS AND METHODS
Study design and population
This is an observational prospective single-center study, where patients were consecutively recruited between February 2015 and February 2017 from a 15-bed medical/surgical Intensive Care Unit (ICU), at the Clinical University Hospital of Santiago de Compostela, Spain. The inclusion criteria were severe sepsis or septic shock according to the 2001 Consensus Conference (15), and patients were enrolled within the first 24 h of ICU admission. Exclusion criteria were age less than 18 years, known immunosuppression, or dark skin (Fitzpatrick skin color classes V–VI), since skin autofluorescence measurements with skin reflectance below 6% have been shown to lack accuracy (5). This study was conducted in accordance with the amended Declaration of Helsinki. Local Institutional Review Board approved the protocol (CEIG 2016/104) and written informed consent was obtained from patients or from their legal representatives if the clinical situation prevented it.
First 24 h of ICU admission were considered baseline or day 1. Skin autofluorescence (SAF) and plasma autofluorescence (PAF) were assessed during the first 5 days. sRAGE was determined on days 1, 3, and 5. In order to explore their kinetics, we calculated delta values (ΔPAF, ΔSAF and ΔsRAGE), as the difference between the SAF, PAF, or sRAGE value on a specific day and the value on day 1. Details on SAF, PAF, and sRAGE measurement are provided in Supplemental Digital Content 1, https://links.lww.com/SHK/A901(16–18). Clinical and routine laboratory data, including HbA1c and inflammatory mediators (TNF-α, IL-6, IL-8), were obtained, and daily SOFA score (19) was calculated from day 1 to 5. Mortality was assessed during a 28-day follow-up period.
First a descriptive analysis was performed, categorical variables were expressed with frequencies and percentages, and continuous variables with mean and standard deviation, or median and interquartile range, depending on their adjustment to a normal distribution (Kolmogorv–Smirnov test with Lilliefors correction). Differences between groups were determined with chi-squared, Fisher, t-Student, or Mann–Whitney tests. In addition, a correlation analysis was carried out to assess the relationship between two continuous variables. Finally, a bivariate logistic regression (LR) was performed and then, to assess the relationship of ΔPAF, ΔSAF, and sRAGE with mortality at day 28, we performed a multivariate LR with those variables considered relevant in the bivariate LR. Statistical significance was considered from P values < 0.05 [IBM SPSS statistics software v19.0].
Ninety consecutive patients meeting the criteria of severe sepsis or septic shock were enrolled. The main baseline characteristics stratified by mortality at day 28 are presented in Table 1.
Skin and plasma autofluorescence on day 1 (SAF1 and PAF1)
SAF1 values correlated with patient's age (r = 0.262, P = 0.013). These measured SAF1 values were higher (P < 0.001) than the calculated values adjusted to age (2.40 [2.20–2.60] AU) after applying the reference formula for healthy European control subjects [SAF = (0.024 × age) + 0.83] (20).
Diabetic patients with poor metabolic control (HbA1c > 7.5%) showed higher SAF1 values (Fig. 1A), while medical history of CKD was associated with higher values of SAF1 (Fig. 1B) and PAF1 (Fig. 1C).
Although there was a trend to have higher levels of SAF1 and PAF1 in non-survivors, in the bivariate analysis, neither showed association with 28-day mortality (Table 1 and Supplemental Digital Content 2, https://links.lww.com/SHK/A902).
Soluble RAGE on day 1 (sRAGE1)
sRAGE1 levels correlated with inflammatory mediators TNF-α (r = 0.267, P = 0.017), IL-6 (r = 0.385, P = 0.001), and IL-8 (r = 0.363, P = 0.002).
Correlation was found between sRAGE1 and APACHE-II (r = 0.323, P = 0.007) and between sRAGE1 and SOFA1 (r = 0.215, P = 0.004). With regard to SOFA score components, sRAGE1 was associated only with respiratory failure, and was significantly higher in those patients meeting criteria of moderate and severe acute respiratory distress syndrome (ARDS) according to the Berlin definition (PaO2/FiO2 ≤ 200 mm Hg) (21) (Fig. 2A). Pulmonary sepsis was the most common cause of ARDS, occurring in 16 out of 35 cases (45.7%).
Additionally, sRAGE1 showed association with 28-day mortality (Table 1 and Supplemental Digital Content 2, https://links.lww.com/SHK/A902). When analyzing patients with or without ARDS, sRAGE1 were significantly higher in non-survivors at day 28 only in the ARDS group (Fig. 2B).
Kinetics of PAF, SAF, and sRAGE
Among values for PAF and SAF on days 1 to 5, and for sRAGE on days 1, 3, and 5 stratified by mortality at day 28, we only found significant association between sRAGE1 and 28-day mortality (Table 1 and Supplemental Digital Content 2, https://links.lww.com/SHK/A902). No correlation was found between sRAGE and SAF or PAF levels at any time of the study. However, sRAGE1 showed a negative correlation with ΔPAF4-1 (r = −0.250, P = 0.019) and ΔPAF5-1 (r = −0.246, P = 0.024).
As can be observed in Figure 3, analysis of delta values found that ΔPAF3-1 [−4.0 (−15.5 – 1.9) vs. 2.8 (−2.5 – 6.5), P = 0.003], ΔPAF4-1 [−5.5 (−21.5 to −0.8) vs. 0.5 (−3.0 – 6.5), P = 0.002] and ΔPAF5-1 (−7.3 ± 13.8 vs. 2.3 ± 12.3, P = 0.016) showed a strong association with 28-day mortality. Furthermore, an association was found between ΔSAF5-1 and 28-day mortality (−0.5 ± 0.7 vs. 0.9 ± 0.6, P = 0.007). We did not find associations with ΔsRAGE 3-1 or ΔsRAGE 5-1 and mortality at day 28.
To verify that there was no influence of continuous renal replacement therapy (CRRT) in ΔPAF and ΔSAF values, Mann–Whitney test was applied, showing that delta values were independent of the CRRT (Supplemental Digital Content 3, https://links.lww.com/SHK/A903).
An LR was performed to calculate the odds ratios and their 95% confidence intervals [ORs (CI95%)] associated with mortality (Table 2). First, the bivariate LR showed a significant crude OR for ΔPAF3-1, ΔPAF4-1, ΔPAF5-1, ΔSAF5-1, and sRAGE1. Second, a multivariate LR was performed adjusting for the variables with significance P < 0.01 in the bivariate analysis (APACHE II, CRRT, and mechanical ventilation): ΔPAF2-1 became significant, ΔPAF3-1, ΔPAF4-1, ΔPAF5-1 and ΔSAF5-1 retained their significance, while sRAGE1 lost its significance with regard to the bivariate LR. The relationship between AGEs, sRAGE levels, and mortality is summarized in Figure 4.
To the best of our knowledge, this is the first study to demonstrate an association between the kinetics of plasma and skin AGEs, and the mortality of patients with severe sepsis and septic shock. Our main finding is that those patients with a greater decrease in plasma AGEs during the first 5 days of sepsis, and those with a greater decrease in skin AGEs on the fifth day of sepsis showed higher mortality.
These interesting findings deserve further discussion. Regarding chronic influences on AGEs, we have detected their impact on baseline levels of PAF and SAF, as would be expected. We found a good correlation between SAF and age. Koetsier et al. (20) described a linear increase of SAF with age of approximately 0.023 AU per year, providing the reference values for controls. SAF values of our patients, some of them with diabetes mellitus (DM) and CKD, were higher than the calculated values. In previous studies, SAF was found to be higher in critically ill patients compared with healthy controls, even after excluding DM and CKD; suggesting an acute accumulation of skin AGEs during critical illness (14, 22). We found that DM patients with poor metabolic control showed higher values of SAF, which is in agreement with the concept of SAF as a marker of metabolic memory. We also found that history of CKD was associated with higher values of SAF and PAF. Lavielle et al. (23) compared patients with acute kidney injury (AKI) and with CKD, finding a rapid rise of SAF in AKI patients, where SAF is lower in the AKI group compared with the CKD group, and its values were related to the duration of renal failure, reflecting the accumulation of AGEs in the skin.
It seems reasonable that we did not find association between plasma AGEs at day 1 and mortality, since our case-mix includes young patients without previous medical history and elderly patients with history of DM and CKD. Previously, Andrades et al. (13) studying early sepsis in 36 patients, 7 without organ failure, described higher levels of plasma non-fluorescent AGEs (CML and CEL) in survivors. Meertens et al. (14) reported the first study to sequentially measure SAF over 7 days, plus circulating CML, CEL, and sRAGE, in 45 patients with multiple organ dysfunction, 10 non-septic. This study found higher levels of CML in non-survivors; serial measurements of the variables were not associated with mortality. These conflicting results can be a consequence of analyzing different AGEs with different methods, at different moments of the septic process and with a different case-mix of patients.
Studies define two different actions for sRAGE, as a marker of tissue RAGE expression that sustains inflammation, and as a decoy receptor decreasing inflammation. This second action is based on the finding that exogenous sRAGE blocked the harmful effect of AGEs in animal models; however, it has been questioned whether endogenous sRAGE can exert the same biological effect (24). In our study, sRAGE levels on day 1 correlated with early inflammatory markers (TNF-α, IL-6, and IL-8), suggesting a close involvement of sRAGE in inflammatory responses. Additionally, we found association between sRAGE1 and 28-day mortality, and we recorded higher levels in non-survivors, which has been described before (9, 10). We have found an inverse correlation between sRAGE1 and delta PAF values, meaning that the higher the sRAGE1, the greater the decrease of plasma AGEs. Altogether we can suggest a link between bad prognosis and high levels of sRAGE1, which are responsible for the ligation to plasma AGEs, therefore decreasing their levels and subsequently the pool of skin AGEs.
We have found that sRAGE1 is associated with the presence of ARDS, disease severity, and mortality. While some studies (9, 10) found sRAGE elevated in septic patients and associated with outcome, others found sRAGE elevated in acute lung injury/ARDS regardless of the presence of severe sepsis (25). Ware et al. (26) compared 100 patients with severe sepsis and no ARDS, with 100 patients who had severe sepsis and ARDS, indicating sRAGE gives strong discrimination for the diagnosis of ARDS. It is significant that RAGE is constitutively expressed at high levels in the lung, even in a physiologic state; where persistent activation of RAGE can alter the protective innate immunity provoking acute lung injury (27). Recent work by Jabaudon et al. found sRAGE to be a good biomarker for ARDS, since it associates with lung injury severity, and shows good diagnostic and prognostic values (27–30).
We calculated delta values, for changes from day 1, providing information on the kinetics of the AGEs related to sepsis. Delta values are useful as they indicate the relative changes of plasma and skin AGEs during the septic process, and are less influenced by previous chronic conditions. In the non-survivors group, delta values of plasma AGEs were negative, meaning a decrease of plasma AGEs was detected in the early course of sepsis, from the second to the fifth day. It makes sense that a decrease in plasma AGEs subsequently translates into a delayed decrease in skin AGEs, which we detected at the fifth day with an interesting adjusted odds ratio (Table 2). These results have an added interest due to the non-invasive nature of delta values recorded for skin AGEs.
Importantly, no difference was found in delta values of PAF and SAF between those patients who received CRRT or not, showing that our results are independent of any “washing” effect related to continuous dialysis. We did not find published studies on the impact of CRRT on plasma AGEs, but in conventional hemodialysis, Ramsauer et al. (31) observed a SAF reduction 1 week after using a glucose-free dialysate, while protein-bound PAF was reduced independently of the dialysate. This decrease of SAF was explained as a decrease in reversible bound glucose degradation products in the skin, e.g., Amadori products.
Altogether, our results suggest that sRAGE levels are more elevated at the beginning of the septic process in those patients with severe sepsis and septic shock, whereas circulating AGEs are only related to age, diabetes mellitus, and CKD. However, we hypothesize that the elevated sRAGE levels could bind AGEs in plasma, helping in their reduction. This would explain a decrease of PAF during the first days of the septic process and a decrease in SAF at day 5.
This study has limitations. First, we provided meaningful statistic association, but we do not demonstrate a mechanistic causal relationship between plasma and skin AGEs, sRAGE, and mortality. Second, we measured total levels of circulating sRAGE, which includes sRAGE generated from the cleavage of cell-surface RAGE, and endogenous secretory RAGE, which have been suggested to be functionally different (29). In any case, our study provides valuable information on the behavior of plasma and skin AGEs and sRAGE in sepsis, which has not been reported before.
This study shows consistently that those septic patients with a significant decrease of plasma and skin AGEs during the first days of sepsis, suffered higher mortality. Also, sRAGE1 levels correlated with decreases in PAF levels, and are associated with 28-day mortality. Therefore, we can suggest that high levels of sRAGE at day 1 are at least partially responsible for the ligation to circulating AGEs, which then decreases their levels, and causes a subsequent decrease of skin AGEs. How this process leads to a worse outcome remains unclear. Further research should clarify the role of the AGE-RAGE axis in the maladaptive response of sepsis.
The authors thank Christopher Phillips for his English-language review.
1. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, Bellomo R, Bernard GR, Chiche JD, Coopersmith CM, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA
315 (8):801–810, 2016.
2. Pool R, Gomez H, Kellum JA. Mechanisms of organ dysfunction in sepsis. Crit Care Clin
34 (1):63–80, 2018.
3. Yamagishi S-I, Fukami K, Matsui T. Evaluation of tissue accumulation levels of advanced glycation end products by skin autofluorescence: a novel marker of vascular complications in high-risk patients for cardiovascular disease. Int J Cardiol
4. Ramsauer B, Engels G, Arsov S, Hadimeri H, Sikole A, Graaff R, Stegmayr B. Comparing changes in plasma and skin autofluorescence in low-flux versus high-flux hemodialysis. Int J Artif Organs
38 (9):488–493, 2015.
5. Da Moura Semedo C, Webb M, Waller H, Khunti K, Davies M. Skin autofluorescence, a non-invasive marker of advanced glycation end products: clinical relevance and limitations. Postgrad Med J
93 (1099):289–294, 2017.
6. Senatus LM, Schmidt AM. The AGE-RAGE axis: implications for age-associated arterial diseases. Front Genet
7. Hofer S, Uhle F, Fleming T, Hell C, Schmoch T, Bruckner T, Weigand MA, Brenner T. RAGE-mediated inflammation in patients with septic shock. J Surg Res
202 (2):315–327, 2016.
8. Creagh-Brown BC, Quinlan GJ, Evans TW, Burke-Gaffney A. The RAGE axis in systemic inflammation, acute lung injury and myocardial dysfunction: an important therapeutic target? Intensive Care Med
36 (10):1644–1656, 2010.
9. Bopp C, Hofer S, Weitz J, Bierhaus A, Nawroth PP, Martin E, Büchler MW, Weigand MA. sRAGE is elevated in septic patients and associated with patients outcome. J Surg Res
147 (1):79–83, 2008.
10. 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
73 (8):650–660, 2013.
11. Hamasaki MY, Barbeiro HV, de Souza HP, Machado MC, da Silva FP. sRAGE in septic shock: a potential biomarker of mortality. Rev Bras Ter Intensiva
26 (4):392–396, 2014.
12. Nakamura T, Sato E, Fujiwara N, Kawagoe Y, Suzuki T, Ueda Y, Yamada S, Shoji H, Takeuchi M, Ueda S, et al. Circulating levels of advanced glycation end products (AGE) and interleukin-6 (IL-6) are independent determinants of serum asymmetric dimethylarginine (ADMA) levels in patients with septic shock. Pharmacol Res
60 (6):515–518, 2009.
13. Andrades MÉ, Lorenzi R, Nagai R, Moreira JC, Ritter C, Dal-Pizzol F. Plasma glycation levels are associated with severity in sepsis. Eur J Clin Invest
42 (10):1055–1060, 2012.
14. Meertens JH, Nienhuis HL, Lefrandt JD, Schalkwijk CG, Nyyssönen K, Ligtenberg JJ, Smit AJ, Zijlstra JG, Mulder DJ. The course of skin and serum biomarkers of advanced glycation endproducts and its association with oxidative stress, inflammation, disease severity, and mortality during ICU admission in critically ill patients: results from a prospective pilot study. PLoS One
11 (8):e0160893, 2016.
15. Levy MM, Fink MP, Marshall JC, Abraham E, Angus D, Cook D, Cohen J, Opal SM, Vincent JL, Ramsay G. SCCM/ESICM/ACCP/ATS/SIS. 2001 SCCM/ESICM/ACCP/ATS/SIS International Sepsis Definitions Conference. Crit Care Med
31 (4):1250–1256, 2003.
16. Meerwaldt R, Graaff R, Oomen PHN, Links TP, Jager JJ, Alderson NL, Thorpe SR, Baynes JW, Gans ROB, Smit AJ. Simple non-invasive assessment of advanced glycation endproduct accumulation. Diabetologia
47 (7):1324–1330, 2004.
17. Münch G, Keis R, Wessels A, Riederer P, Bahner U, Heidland A, Niwa T, Lemke HD, Schinzel R. Determination of advanced glycation end products in serum by fluorescence spectroscopy and competitive ELISA. Eur J Clin Chem Clin Biochem
35 (9):669–677, 1997.
18. Raposeiras-Roubín S, Rodiño-Janeiro BK, Paradela-Dobarro B, Grigorian-Shamagian L, García-Acuña JM, Aguiar-Souto P, Jacquet-Hervet M, Reino-Maceiras MV, González-Juanatey JR, Álvarez E. Fluorescent advanced glycation end products and their soluble receptor: the birth of new plasmatic biomarkers for risk stratification of acute coronary syndrome. PLoS One
8 (9):e74302, 2013.
19. Vincent JL, Moreno R, Takala J, Willatts S, De Mendonça A, Bruining H, Reinhart CK, Suter PM, Thijs LG. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med
22 (7):707–710, 1996.
20. Koetsier M, Lutgers HL, de Jonge C, Links TP, Smit AJ, Graaff R. Reference values of skin autofluorescence. Diabetes Technol Ther
12 (5):399–403, 2010.
21. Definition Task Force ARDS, Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, Camporota L, Slutsky AS. Acute respiratory distress syndrome: the Berlin Definition. JAMA
307 (23):2526–2533, 2012.
22. Greven WL, Smit JM, Rommes JH, Spronk PE. Accumulation of advanced glycation end (AGEs) products in intensive care patients: an observational, prospective study. BMC Clin Pathol
23. Lavielle A, Rubin S, Boyer A, Moreau K, Rajaobelina K, Combe C, Rigalleau V. Skin autofluorescence in acute kidney injury. Crit Care Lond Engl
21 (1):24, 2017.
24. Nakamura K, Yamagishi SI, Matsui T, Adachi H, Takeuchi M, Imaizumi T. Serum levels of soluble form of receptor for advanced glycation end products (sRAGE) are correlated with AGEs in both diabetic and non-diabetic subjects. Clin Exp Med
7 (4):188–190, 2007.
25. Jabaudon M, Futier E, Roszyk L, Chalus E, Guerin R, Petit A, Mrozek S, Perbet S, Cayot-Constantin S, Chartier C, et al. Soluble form of the receptor for advanced glycation end products is a marker of acute lung injury but not of severe sepsis in critically ill patients. Crit Care Med
39 (3):480–488, 2011.
26. Ware LB, Koyama T, Zhao Z, Janz DR, Wickersham N, Bernard GR, May AK, Calfee CS, Matthay MA. Biomarkers of lung epithelial injury and inflammation distinguish severe sepsis patients with acute respiratory distress syndrome. Crit Care
17 (5):R253, 2013.
27. Jabaudon M, Berthelin P, Pranal T, Roszyk L, Godet T, Faure JS, Chabanne R, Eisenmann N, Lautrette A, Belville C, et al. Receptor for advanced glycation end-products and ARDS prediction: a multicentre observational study. Sci Rep
8 (1):2603, 2018.
28. Guo WA, Knight PR, Raghavendran K. The receptor for advanced glycation end products and acute lung injury/acute respiratory distress syndrome. Intensive Care Med
38 (10):1588–1598, 2012.
29. Jabaudon M, Blondonnet R, Roszyk L, Pereira B, Guérin R, Perbet S, Cayot S, Bouvier D, Blanchon L, Sapin V, et al. Soluble forms and ligands of the receptor for advanced glycation end-products in patients with acute respiratory distress syndrome: an observational prospective study. PLoS One
10 (8):e0135857, 2015.
30. Jabaudon M, Blondonnet R, Pereira B, Cartin-Ceba R, Lichtenstern C, Mauri T, Determann RM, Drabek T, Hubmayr RD, Gajic O, et al. Plasma sRAGE is independently associated with increased mortality in ARDS: a meta-analysis of individual patient data. Intensive Care Med
44 (9):1388–1399, 2018.
31. Ramsauer B, Engels GE, Graaff R, Sikole A, Arsov S, Stegmayr B. Skin- and Plasmaautofluorescence in hemodialysis with glucose-free or glucose-containing dialysate. BMC Nephrol
18 (1):5, 2017.