Editorial II

Role of the kidney in perioperative inflammatory responses

R. C. Baker1, M. A. Armstrong2, S. J. Allen3 and W. T. McBride3

1Department of Surgery The Queen’s University of Belfast Belfast UK 2Department of Immunobiology The Queen’s University of Belfast Belfast UK 3Department of Clinical Anaesthesia The Royal Group of Hospitals Trust Grosvenor Road Belfast BT12 6BA UK

Perioperative inflammatory responses are relatively well characterized. Increases in the plasma proinflammatory cytokines interleukin (IL)-1ß, tumour necrosis factor {alpha} (TNF-{alpha}) and IL-8 are later accompanied by increases in the anti-inflammatory cytokines IL-10, IL-1 receptor antagonist (IL-1ra) and TNF-{alpha} soluble receptors (TNF-sr)—the so-called phased anti-inflammatory response (PAIR).13 Proinflammatory cytokines act locally and in health are rarely detectable in plasma. In contrast, anti-inflammatory cytokines such as IL-1ra and TNF-sr are present in plasma in measurable concentrations. The molecular weights of IL-1ß, IL-8 and monomeric TNF-{alpha} are less than 20 kDa, rendering rapid glomerular filtration possible. As the biologically active form of TNF-{alpha} is a trimer (51 kDa) in equilibrium with the dimeric and monomeric forms, rapid filtration of the monomeric form may be expected to reduce the concentration of the biologically active trimer. This may be one reason why the proinflammatory cytokines are rarely detected in health and why, during an inflammatory response, their elevation in plasma is not sustained.3 In contrast, the anti-inflammatory cytokines are larger than 20 kDa and thus less readily filtered by the kidneys. Not surprisingly, IL-1ra and TNF-srs are present constitutively in plasma at measurable concentrations and provide an immediate control mechanism should elevated TNF-{alpha} or IL-1ß spill over into the plasma from a local tissue site of inflammation. If the kidney is filtering these proinflammatory cytokines, how does it safely do so, given their known tubulotoxic properties? Does the kidney sacrifice itself for the greater good of the body? Recently, we demonstrated a rapid perioperative increase in urinary anti-inflammatory cytokines (IL-1ra and TNF-srs), suggesting a possible intra-renal protective mechanism during proinflammatory cytokine disposal.3 Although involved in managing the inflammatory response, the kidney is especially vulnerable to these same proinflammatory processes, particularly if they jeopardize adequate tubular oxygenation or exaggerate pro-apoptotic or pro-necrotic stimuli.4 5

Let us consider two stages in the pathogenesis of acute renal failure and evaluate the important impact of inflammatory processes in each.

Stage 1 (transient hypotension). In hypotension, glomerular perfusion pressure falls. This is further accentuated by afferent arteriolar vasoconstriction.6 This leads to reduced tubular energy expenditure through reduced work of protein and ion reabsorption. Urine is of reduced volume but of ‘good quality’. At this stage, if there is a background inflammatory response, hypotension and the leucocyte adhesion potential may be accentuated, although adequate peritubular blood flow will minimize the deleterious consequences of this. Proinflammatory cytokine disposal is still possible.

Stage 2 (persistent hypotension). If hypotension persists, efferent arteriolar vasoconstriction restores glomerular perfusion pressure and normalizes urinary output. Sodium and protein transport energy requirements are restored in the absence of the restoration of tubular oxygen delivery. The reduced flow secondary to hypotension and efferent arteriolar vasoconstriction predisposes to margination of neutrophils along the capillary endothelial walls. Furthermore, the hypoxia resulting from reduced perfusion leads to tubular capillary endothelial adhesion molecule up-regulation and results in increased neutrophil adhesion potential.7 This now sets up a vicious circle of hypoxia and further neutrophil sludging.8

Adhering granulocytes liberate vasoconstrictive and tubulotoxic mediators. These include nitric oxide (NO), superoxide radicals and platelet activating factor. In the face of this inflammatory onslaught, tubular capillary endothelium reacts as if a local infective response has struck, needing urgent containment. Consequently, constitutive dilatory endothelial NO synthase (eNOS) is down-regulated (leading to vasoconstriction) and vasoconstrictive endothelin-1 is released, leading to further vasoconstriction. A key element in this process is neutrophil adhesion.8

It is apparent that any evolving systemic septic or proinflammatory response is going to seriously increase leucocyte adhesion potential within the tubular capillaries. For example, marked up-regulation of granulocyte adhesion molecules occurs during cardiopulmonary bypass,9 abdominal aortic aneurysm surgery10 and sepsis.11 Any superimposed hypotension in this context may exaggerate the already heightened adhesion potential within the renal capillaries. At this point, tubular hypoxia is unavoidable.

During all of the above, the kidney is still trying to contain the inflammatory response by filtering out proinflammatory mediators as well as many other substances, many of which may be tubulotoxic. For example, a host of filtered toxins, such as haemoglobin, myoglobin, radiocontrast medium12 and antibiotics provide pro-apoptotic and pro-necrotic stimuli. Filtered inflammatory mediators may be tubulotoxic. For TNF-{alpha} and IL-1ß a direct tubulotoxic mechanism has been suggested on the basis of in vitro demonstration of cytotoxic NO concentrations arising from activation of inducible NOS (iNOS) in cultured proximal tubular cells.13 14 In addition to these pronecrotic effects in vitro evidence suggests direct pro-apoptotic mechanisms are involved.15 Recently, correlations were demonstrated between plasma concentrations of both TNF-{alpha} and IL-8 and proximal tubular injury in cardiac surgery [as measured by the urinary N-acetyl-ß-D-glucosaminidase/creatinine ratio].3

What happens to the control mechanism of the inflammatory response when the kidney starts to fail?

Acute renal failure is associated with a high incidence of multiple organ failure (MOF) and death. For example, in cardiac surgery the incidence of acute renal failure requiring dialysis (ARF-D) is 1–5%1618 Mortality in such circumstances is high, ranging from 28%19 to 63.7%.20 Although this high mortality may be due directly to the deleterious effect on non-renal organs of the precipitating renal insult, there are theoretical possibilities that ARF-D itself has deleterious consequences for other organs by adversely modulating the inflammatory response. If the kidney is so important in controlling inflammatory responses, how can patients in chronic renal failure survive major surgery?

During cardiac surgery, the risk of perioperative mortality in patients with preoperative dialysis-dependent chronic renal failure (CRF-D) ranges from zero21 to 11.4%.22 Although preoperative CRF-D may lead to an expectation of greater perioperative morbidity and a longer hospital stay,17 this relatively limited increase in mortality of CRF-D patients may suggest a mechanism of proinflammatory control despite the absence of functioning kidneys. Patients in chronic renal failure, irrespective of dialysis dependence, cannot rely on immediate filtration of proinflammatory mediators as a primary anti-inflammatory mechanism. Their survival during an acute proinflammatory response may rely partly on secondary control mechanisms which, over time, have developed to compensate for the loss of renal immunomodulatory function.23 Consequently, chronic renal failure patients have a wide range of anti-inflammatory strategies in place.2426 These include increased plasma TNF-sr 1 and 2,27 28 increased IL-1ra,29 increased granulocyte inhibitory protein24 30 31 and uraemia. It is important to distinguish between acute and chronic renal dysfunction as risk factors for the development of MOF (Fig. 1).



View larger version (46K):
[in this window]
[in a new window]
 
Fig. 1 Conceptualized diagram representing two patients undergoing a simultaneous inflammatory insult of identical intensity, such as major cardiovascular surgery. Patient 1 has normal preoperative renal function and patient 2 has chronic renal impairment. (Time A) During the initial phase of the insult in both patients, local cytokine production spills into the systemic compartment. In patients 1 and 2 this is countered by two strategies: strategy 1 is renal excretion of proinflammatory cytokines; strategy 2 consists of anti-inflammatory mechanisms that are constitutively active in plasma. The notional quantity of proinflammatory cytokines safely controlled by these plasma anti-inflammatory mechanisms is represented by the box labelled ‘Safe systemic cytokine capacity’. To compensate for a defect in strategy 1, patient 2 has preoperatively developed strategy 2, as indicated by an increased safe systemic cytokine capacity. At this time point, organs in both patients are still ‘safe’. (Time B) In both patients, the kidneys undergo significant injury. This may be due either to the inflammatory response itself or to an additional hypotensive episode. Strategy 1 has now been knocked out and if the spillover of inflammatory mediators continues at a constant rate, strategy 2 is in danger of being overwhelmed, resulting in multiple organ failure (MOF). However, because patient 2 has compensated for a defective strategy 1 preoperatively by the use of the more effective strategy 2 (box labelled ‘Increased safe systemic cytokine capacity’), the impact of acute renal failure on inflammatory-mediated organ failure may be delayed. (Time C) The safe systemic cytokine capacity has been exceeded in both patients and proinflammatory mediators can have deleterious effects on other organs. Patients with normal renal function preoperatively who develop acute renal failure in the presence of an ongoing inflammatory response are at particular risk of multiple organ failure.

 
Application

No longer can the kidney be regarded as a passive bystander in the evolving inflammatory response: it is rather an organ actively involved in the control of the response, even to the extent of sustaining injury in the process. Consequently, clinical strategies of renal protection should also be considered as indirectly protecting other organs against inflammatory-mediated MOF. For example, volume replacement and avoidance of hypotension may reduce the danger of tubular capillary stasis and thus lessen the already heightened risk of tubular damage arising from leucocyte adhesion when there is a background of an ongoing inflammatory response. Furthermore, such measures, by preserving renal ability to remove proinflammatory cytokines, may help restrict the perioperative plasma changes in pro- and anti-inflammatory cytokines to physiologically tolerable limits. The effectiveness of perioperative inflammatory limitation strategies in minimizing renal injury awaits elucidation. When the kidneys fail, it takes time for the remaining plasma anti-inflammatory strategies to be overwhelmed and the resulting uncontrolled inflammatory responses to effect organ damage. It is conceivable that the patient who presents with normal renal function preoperatively has heightened vulnerability compared with the patient with chronic renal dysfunction. This is because the patient with chronic renal dysfunction has developed a greater plasma anti-inflammatory background to compensate for reduced ability to remove the proinflammatory load. Patients with normal renal function preoperatively who develop acute renal failure in the presence of an ongoing inflammatory response are at particular risk of MOF. Renal protective strategies should be considered as indirectly anti-inflammatory.

References

1 McBride WT, Armstrong MA, Crockard AD, McMurray TJ, Rea JM. Cytokine balance and immunosuppressive changes at cardiac surgery: contrasting response between patients and isolated CPB circuits. Br J Anaesth 1995; 75: 724–33[Abstract/Free Full Text]

2 McBride WT, Armstrong MA, Gilliland H, McMurray TJ. The balance of pro and anti-inflammatory cytokines in plasma and bronchoalveolar lavage (BAL) at paediatric cardiac surgery. Cytokine 1996; 8: 724–9[ISI][Medline]

3 Gormley SM, McBride WT, Armstrong MA et al. Plasma and urinary cytokine homeostasis and renal dysfunction during cardiac surgery. Anesthesiology 2000; 93: 1210–6; discussion 5A[ISI][Medline]

4 Lieberthal W, Koh JS, Levine JS. Necrosis and apoptosis in acute renal failure. Semin Nephrol 1998; 18: 505–18[ISI][Medline]

5 Lieberthal W, Levine JS. Mechanisms of apoptosis and its potential role in renal tubular epithelial cell injury. Am J Physiol 1996; 271: F477–88[Abstract/Free Full Text]

6 Conger J. Hemodynamic factors in acute renal failure. Adv Renal Replace Ther 1997; 4: 25–37[Medline]

7 Rabb H, O’Meara YM, Maderna P, Coleman P, Brady HR. Leukocytes, cell adhesion molecules and ischemic acute renal failure. Kidney Int 1997; 51: 1463–8[ISI][Medline]

8 Lieberthal W. Biology of ischemic and toxic renal tubular cell injury: role of nitric oxide and the inflammatory response. Curr Opin Nephrol Hypertens 1998; 7: 289–95[ISI][Medline]

9 Grunenfelder J, Zund G, Schoeberlein A et al. Expression of adhesion molecules and cytokines after coronary artery bypass grafting during normothermic and hypothermic cardiac arrest. Eur J Cardiothorac Surg 2000; 17: 723–8[Abstract/Free Full Text]

10 Galle C, De Maertelaer V, Motte S et al. Early inflammatory response after elective abdominal aortic aneurysm repair: a comparison between endovascular procedure and conventional surgery. J Vasc Surg 2000; 32: 234–46[ISI][Medline]

11 Lush CW, Kvietys PR. Microvascular dysfunction in sepsis. Microcirculation 2000; 7: 83–101[ISI][Medline]

12 McCullough PA, Wolyn R, Rocher LL, Levin RN, O’Neill WW. Acute renal failure after coronary intervention: incidence, risk factors, and relationship to mortality. Am J Med 1997; 103: 368–75[ISI][Medline]

13 McLay JS, Chatterjee P, Nicolson AG et al. Nitric oxide production by human proximal tubular cells: a novel immunomodulatory mechanism? Kidney Int 1994; 46: 1043–9[ISI][Medline]

14 Chatterjee PK, Hawksworth GM, McLay JS. Cytokine-stimulated nitric oxide production in the human renal proximal tubule and its modulation by natriuretic peptides: A novel immunomodulatory mechanism? Exp Nephrol 1999; 7: 438–48[ISI][Medline]

15 Meldrum DR, Donnahoo KK. Role of TNF in mediating renal insufficiency following cardiac surgery: evidence of a postbypass cardiorenal syndrome. J Surg Res 1999; 85: 185–99[ISI][Medline]

16 Chertow GM, Lazarus JM, Christiansen CL et al. Preoperative renal risk stratification. Circulation 1997; 95: 878–84[Abstract/Free Full Text]

17 Bhattacharyya N, Cheung AH, Dang CR et al. Open heart surgery in patients with end-stage renal disease. Am J Nephrol 1997; 17: 435–9[ISI][Medline]

18 Fortescue EB, Bates DW, Chertow GM. Predicting acute renal failure after coronary bypass surgery: cross-validation of two risk-stratification algorithms. Kidney Int 2000; 57: 2594–602[ISI][Medline]

19 Conlon PJ, Stafford-Smith M, White WD et al. Acute renal failure following cardiac surgery. Nephrol Dial Transplant 1999; 14: 1158–62[Abstract]

20 Chertow GM, Levy EM, Hammermeister KE, Grover F, Daley J. Independent association between acute renal failure and mortality following cardiac surgery. Am J Med 1998; 104: 343–8[ISI][Medline]

21 Hosoda Y, Yamamoto T, Takazawa K et al. Coronary artery bypass grafting in patients on chronic hemodialysis: surgical outcome in diabetic nephropathy versus nondiabetic nephropathy patients. Ann Thorac Surg 2001; 71: 543–8[Abstract/Free Full Text]

22 Franga DL, Kratz JM, Crumbley AJ et al. Early and long-term results of coronary artery bypass grafting in dialysis patients. Ann Thorac Surg 2000; 70: 813–8[Abstract/Free Full Text]

23 Tran DD, Visser JJ. Possible protective effect of chronic renal failure against sepsis and endotoxin-induced organ failure and mortality in acute renal failure. Nephron 1995; 69: 172–3[ISI][Medline]

24 Hörl WH. Genesis of the uraemic syndrome: role of uraemic toxins. Wien Klin Wochenschr 1998; 110: 511–20[ISI][Medline]

25 Pereira BJ, Shapiro L, King AJ et al. Plasma levels of IL-1 beta, TNF alpha and their specific inhibitors in undialyzed chronic renal failure, CAPD and hemodialysis patients. Kidney Int 1994; 45: 890–6[ISI][Medline]

26 Pereira BJ. Cytokine production in patients on dialysis. Blood Purif 1995; 13: 135–46[ISI][Medline]

27 Ward R, McLeish KR. Soluble TNF alpha receptors are increased in chronic renal insufficiency and hemodialysis and inhibit neutrophil priming by TNF alpha. Artif Organs 1996; 20: 390–5[ISI][Medline]

28 van de Kar NC, Sauerwein RW, Demacker PN et al. Plasma cytokine levels in hemolytic uremic syndrome. Nephron 1995; 71: 309–13[ISI][Medline]

29 Descamps-Latscha B, Herbelin A, Nguyen AT et al. Balance between IL-1 beta, TNF-alpha, and their specific inhibitors in chronic renal failure and maintenance dialysis. Relationships with activation markers of T cells, B cells, and monocytes. J Immunol 1995; 154: 882–92[Abstract/Free Full Text]

30 Ziesche R, Roth M, Papakonstantinou E et al. A granulocyte inhibitory protein overexpressed in chronic renal disease regulates expression of interleukin 6 and interleukin 8. Proc Natl Acad Sci USA 1994; 91: 301–5[Abstract]

31 Haag-Weber M, Hörl WH. Dysfunction of polymorphonuclear leukocytes in uremia. Semin Nephrol 1996; 16: 192–201[ISI][Medline]