Kidney International, Vol. 63, Supplement 84 (2003), pp. S69–S71
Do circulating cytokines really matter in sepsis? CIRO TETTA, RINALDO BELLOMO, VINCENT D’INTINI, CONCETTA DE NITTI, PAOLA INGUAGGIATO, ALESSANDRA BRENDOLAN, and CLAUDIO RONCO Renal Research, Division of Medicine, Fresenius Medical Care, Bad Homburg, Germany; Department of Intensive Care, Austin & Repatriation Medical Centre, Melbourne, Australia; and Department of Nephrology, St. Bortolo Hospital, Vicenza, Italy
Do circulating cytokines really matter in sepsis? Sepsis remains the major cause of mortality worldwide, claiming millions of lives each year. The past decade has seen major advances in the understanding of the biological mechanisms involved in this complex process. Unfortunately, no definitive therapy yet exists that can successfully treat sepsis and its complications. In this review, we will address the significance of circulating cytokines in the pathophysiology of sepsis and its relevance to new approaches in extracorporeal therapies.
Sepsis is a systemic immune response that leads to multiple organ failure [1]. Initially described as mainly due to the overproduction of pro-inflammatory factors, its pathogenesis is now viewed as much more complex. Initially, the concept of blood purification by extracorporeal therapies stemmed from the assumption that nonspecific removal of several inflammatory mediators would improve outcome in septic shock [2]. However, useful convective removal of mediators from the human septic circulation has not been achieved to date, although many cytokines have a molecular weight below the theoretical cut-off point of commercial membranes currently in use [3]. The recent finding that the ultrafiltration dose is correlated to outcome in critically ill patients with acute renal failure strongly suggests the concept of a “sepsis dose” (removal of sepsis-associated mediators) in contrast to a “renal dose” in critically ill patients without systemic inflammation [4]. Based on the provocative title of this review, we will briefly address the complexity of the host response in sepsis, particularly in respect to extracorporeal therapies, with the prejudice that looking at circulating cytokines has only little value from a pathophysiologic viewpoint. Furthermore, we will discuss the concept that testing innovative extracorporeal techniques on the basis of their effect on circulating cytokines may overlook more
Key words: sepsis, septic shock, cytokines, endotoxin, LPS, hemofiltration, plasmapheresis, adsorption, multiorgan failure, plasma filtration.
2003 by the International Society of Nephrology
clinically significant biological effects at cell or organ level. WHAT IS THE RELEVANCE OF CIRCULATING CYTOKINES IN SEPSIS ? The sensitivity of monoclonal antibodies and assays to detect cytokines in plasma, the bound versus free cytokine ratio, and cytokine renal clearance are the most important factors influencing cytokine plasma levels. In fact, the presence or absence of detectable levels of cytokines within biological fluids reflects a complex balance between enhancing and inhibitory signals acting on producer cells, production and catabolism, cytokine binding to target cells, and the modulation of their receptors on the cell surface [5]. Furthermore, their presence does not necessarily parallel their activity, and a possible interplay between a given cytokine and its relative inhibitor (if known) should be considered [5]. Despite the fact that high plasma levels may reflect increased production, these levels do not necessarily represent enhanced bioactivity. There are several factors that may help to explain the incongruities seen in the cytokine-bioactivity story. One such factor is genotypic predisposition. Endotoxin-induced cytokine production differs among individuals. This genetically determined trait is referred to as endotoxin responsiveness [5]. However, subjects who are both highly endotoxin-responsive (i.e., genetically inclined to produce larger amounts of pro-inflammatory cytokines such as tumor necrosis factor-␣), or less endotoxin-responsive (i.e., lack of TNF␣-production but increased anti-inflammatory production, such as interleukin-10 [IL-10]) are predisposed to poorer prognoses [6]. In intensive care medicine, blocking one mediator has not led to measurable outcome improvements in patients with sepsis [7]. Furthermore, the time point in the septic process of therapeutic intervention seems to be crucial. As the network acts like a cascade, early intervention would seem most beneficial. On the other hand, sepsis does not fit a one-hit-model. Neither single-mediatordirected nor one-time interventions, therefore, seem ap-
S-69
S-70
Tetta et al: Plasma cytokines in sepsis
propriate. One of the major criticisms attributed to continuous blood purification therapy in sepsis—its lack of specificity—could turn out to be a major strength. Unspecific removal of soluble mediators, be they pro- or antiinflammatory, without completely eliminating their effect may be the most logical approach to a complex and long-running process like sepsis. The concept of cutting peaks of soluble mediators (e.g., through continuous hemofiltration) is a paradigm we call “the peak concentration hypothesis” [8]. BLOOD PURIFICATION BY EXTRACORPOREAL THERAPY: DOES REMOVAL OF CIRCULATING CYTOKINES REALLY MATTER? For several years, the issue of the capability of hemofiltration to remove inflammatory mediators has remained controversial. Numerous ex vivo, as well as animal and human studies, have shown that synthetic filters in common use in hemofiltration can extract nearly every substance involved in sepsis to a certain degree [9]. On the other hand, significant clinical benefits in terms of hemodynamic improvement have been achieved even without measurable decreases in cytokine plasma levels [10]. Obviously, the removal of substances other than those measured cytokines was responsible for the achieved effect. Alternatively, bioactive substances including some of the measured cytokines were removed, causing the observed beneficial effect. In this context, a step in further clarifying the impact of extracorporeal therapy on cell function has been taken by measuring a more downstream event integrating several cytokine influences: monocyte responsiveness [11, 12]. In spite of some encouraging results, the achievable clinical benefit with continuous renal replacement therapies (using conventional filters and flow rates) in sepsis has been disappointing [13]. Consequently, we sought to improve the efficiency of soluble mediator removal by increasing the amount of plasma water exchange (i.e., increasing ultrafiltration rates) [8]. However, apart from increasing ultrafiltration rates, higher removal rates of middle molecular weight molecules could be achieved by enlarging the pore size of membranes. Ex vivo models have proved useful in precisely calculating sieving coefficients and clearance values using large pore synthetic membranes. Two publications illustrate the advantages provided by examining membrane handling as well the limitations of this approach [14, 15]. In their paper, Cole et al [14], using a large pore membrane (nominal cut-off: 100 kD), found an average sieving coefficient of 0.30 for interleukin-(IL-) 8, 0.56 for TNF␣, 0.61 for IL-1, and 1.34 for IL-6; the respective membrane clearances were 24.2, 44.9, 48.5, and 106.8 mL/min. These results are 2 to 10
times greater than the best reported ex vivo sieving coefficients for commercially available membranes with nominal cut-off points of approximately 30 kD [16]. The results from these authors were less than those achieved by using a plasma filtration membrane (0.70 for IL-8, 1.22 for TNF␣, 1.48 for IL1, and 0.55 for IL-6) [15]. However, it is interesting to note that the ultimate membrane clearances of cytokines were greater in the study with a reduced nominal cut-off [14], since with the latter membrane, a greater filtrate flow could be achieved by a higher transmembrane pressure than those recorded with a plasma filtration membrane [15]. However, we, as the authors of the two above-mentioned publications, feel the need to strongly advise the reader about the limitations of ex vivo studies of this kind. Although in both a significant reduction of different cytokines was claimed, total body clearance in the in vivo situation is a completely different story. Using an experimental model of acute endotoxemia in the rabbit, we showed that nonselective adsorption of cytokines and other pro-inflammatory mediators could improve survival in the absence of any significant change in TNF␣-dependent plasma activity [17]. Much more effective than the effect on circulating cytokines is the impact on the functional responses of cells implicated in the pathogenesis of sepsis. In a very recent study [18], we showed that nonselective removal of mediators could restore leukocyte responsiveness in patients with septic shock. Immunomodulating substances (with molecular weight in the range of 5 to 50 kD) may be eliminated by diffusion, adsorption, or convection, depending on the rather variable cut-off of highly permeable membranes (range from 30 to 40 kD) [9]. In a randomized, prospective study on the effect of coupled plasma filtrationadsorption (CPFA) in human septic shock, Ronco et al [18] showed that the increase in mean arterial pressure was remarkably higher with CPFA than with conventional mixed convective-diffusive continuous therapy (CVVHDF). In these patients, the increase of mean arterial pressure was achieved using norepinephrine at much a lower dose in the CPFA than in conventionally treated patients. Of interest, these hemodynamic changes occurred in parallel with significant immunomodulatory changes. Although no changes in TNF and IL-10 plasma levels could be observed (despite complete adsorption of these cytokines by the adsorbent), TNF production induced ex vivo by endotoxin increased more in CPFA (5-fold) than in CVVHDF. IL-10 appeared to be an important but not an exclusive mediator of reduced leukocyte responsiveness. CONCLUSIONS An array of mostly acutely produced mediators plays a strategic role in the septic syndrome. Defining the
Tetta et al: Plasma cytokines in sepsis
severity of sepsis, the effect of therapeutic intervention and the course of the disease on the basis of circulating cytokines may only allow a gross perception of the sepsis process. In this context, the efficacy of new therapies should be targeted to changes in the biologic expression of different cytokine production, in the pro- versus antiinflammatory imbalance, and ultimately on how the biologic changes in the plasma may relate in target organs whenever accessible. In the future, integrated research projects among basic scientists, clinical researchers, and industries will be needed to concentrate all efforts toward innovative therapies for sepsis and septic shock. Reprint requests to Ciro Tetta, M.D., Renal Research, Division of Medicine, Fresenius Medical Care, Else Kroener Strasse 1, D–61352 Bad Homburg, Germany. E-mail: ciro.tetta@fmc–ag.com
REFERENCES 1. Wenzel RP: Treating sepsis. New Engl J Med 347:966–968, 2002 2. Schetz M: Non-renal indications for continuous renal replacement therapy. Kidney Int 56[Suppl 72]: S88–S94, 1999. 3. Schetz M: Evidence-based analysis of the use of hemofiltration in sepsis and MODS. Curr Opin Intens Care 3:434–441, 1997 4. Ronco C, Bellomo R, Homel P, et al: Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: A prospective randomised trial. Lancet 356:26–30, 2000 5. Cavaillon JM, Munoz C, Fitting C, et al: Circulating cytokines: The tip of the iceberg? Circ Shock 38:145–152, 1992 6. Westerndorp RGJ, et al: Genetic influence on cytokine production and fatal meningococcal disease. Lancet 349:170–173, 1999
S-71
7. Zeni F, Freeman B, Natanson C: Anti-inflammatory therapies to treat sepsis and septic shock: a reassessment. Crit Care Med 25(7): 1095–1100, 1997 8. Ronco C, Ricci Z, Bellomo R: Importance of increased ultrafiltration volume and impact on mortality: Sepsis and cytokine story and the role of continuous veno-venous hemofiltration. Cur Opn Nephrol Hypert 10:755–761, 2001 9. Silvester W: Mediator removal with CRRT: complement and cytokines. Am J Kidney Dis 30(Suppl 4):S38–43, 1997 10. Heering P, Morgera S, Schmitz FJ, et al: Cytokine removal and cardiovascular hemodynamics in septic patients with continuous venovenous hemofiltration. Intensive Care Med 23:288–296, 1997 11. Volk HD, Reinke P, Krausch D, et al: Monocyte deactivation: Rationale for a new therapeutic strategy in sepsis. Intensive Care Med (Suppl 4):S474–S481, 1996 12. Cavaillon JM, Adib-Conquy M, Cloez-Tayarani I, Fitting C: Immunodepression in sepsis and SIRS assessed by ex vivo cytokine production is not a generalized phenomenon: A review. J Endotoxin Res 7(2):85–93, 2001 13. De Vriese AS, Vanholder RC, Pascual M, et al: Can inflammatory cytokines be removed efficiently by continuous renal replacement therapies? Intensive Care Med 25:903–910, 1999 14. Cole L, Bellomo R, Davenport P, et al: The effect of coupled haemofiltration and adsorption on inflammatory cytokines in an ex vivo model. Nephrol Dial Transplant 17:1–8, 2002 15. Tetta C, Cavaillon JM, Schultze M, et al: Removal of cytokines and activated complement components in an experimental model of continuous plasma filtration coupled with sorbent adsorption. Nephrol Dial Transplant 13:1458–1464, 1998 16. Bouman CSC, van Olden RW, Stontenbeek CP: Cytokine filtration and adsorption during pre- and postdilution hemofiltration in four different membranes. Blood Purif 16:261–268, 1998 17. Tetta C, Gianotti L, Cavaillon JM, et al: Coupled plasma filtration-adsorption in a rabbit model of endotoxic shock. Crit Care Med 28:1526–1533, 18. Ronco C, Brendolan A, Lonnemann G, et al: A pilot study on coupled plasma filtration with adsorption in septic shock. Crit Care Med 30(6):1250–1255, 2002