Arginine-related Therapies in Sepsis
As a result of the different hypotheses regarding the role of NO in the pathogenesis of sepsis, different arginine-related therapies have been used ( Table 2 ). These therapies are directed on one side at inhibition of excessive NO production by nonselective and selective NOS-2 inhibitors and, on the other side, at arginine supplementation or administration of NO. But what are the consequences of these therapies, and which is the most beneficial in sepsis?
NO can be administered by NO inhalation or by administration of NO donors, but few data are available from clinical studies. NO inhalation in an animal model of sepsis inhibited the decrease in cardiac output and selectively reduced pulmonary hypertension and improved arterial oxygenation and pH, with a marked attenuation of sympathetic activation.[110,111] The NO donor nitroglycerin markedly increased sublingual microvascular flow in septic patients, even though arterial and central venous pressure dropped temporarily, whereas pressure-guided resuscitation resulted in corrected blood pressure but depressed microcirculatory flow. In experimental animal models of sepsis, NO donors increase portal, mesenteric, and liver blood flow,[113,114,115,116,117] prevent lactic acidosis, and increase cardiac index and left ventricular stroke work index.[114,116] Arterial and pulmonary pressures were, however, not affected by the NO donor 3-morpholinosydnonimine (SIN-1), whereas renal blood flow was decreased.
Thus, beneficial effects of NO administration have been described on microcirculatory flow, but whether this results in better clinical outcome is still unknown. Therefore, we wondered whether the same beneficial effects of NO could also be applied to supplementation of the NO precursor arginine in sepsis?
Data on enteral l-arginine supplementation in septic patients are limited to immunonutrition (IMPACT [Novartis Nutrition, Minneapolis, MN], Immun-Aid [B Braun, Irvine, CA], and Perative [Abbott Laboratories, Abbott Park, IL]) containing l-arginine as a component.[118,119,120,121] Several reviews on immunonutrition in critical illness have been published.[122,123,124,125,126,127,128] However, conclusions regarding the benefits and potential use in sepsis are not uniform, and recently published Canadian guidelines for nutritional support in critically ill patients recommended against the use of arginine-supplemented diets in critically ill patients due to lack of treatment effect with respect to mortality and infections.[126,129] However, the observation that specifically low-arginine nutrition coincides with higher mortality whereas this effect was not observed for high-arginine nutrition[126,129] questions the validity of the recommendation against use of arginine-supplemented diets. Second, the study of Bower et al. is probably confounded by the fact that not all patients tolerated enteral nutrition, and that among these patients, the Acute Physiology and Chronic Health Evaluation score was higher in the immunonutrition group. Third, the major difficulty of drawing conclusions about arginine from arginine-containing immunonutrition is that, in addition to arginine, the formulae also contain nucleotides and ω-3 polyunsaturated fatty acids. Moreover, immunonutrition is, in general, supplied enterally and therefore depends on adequate intestinal function. Well-controlled studies with l-arginine as a monotherapy or as supplied in comparison with an isonitrogenous placebo are therefore needed. The only study using l-arginine infusion in septic patients as a monotherapy involved a bolus l-arginine injection, which induced transient systemic and pulmonary vasodilatory actions only at 1 min after administration, without further adverse effects. An increase in oxygen delivery and consumption was also observed. Further data on l-arginine treatment involve only animal studies. In these studies, l-arginine induced both systemic and pulmonary vasodilatation,[110,131] and, at least when given before the onset of sepsis, reduced production of inflammatory variables, enhanced cellular immunity,[132,133,134] and improved survival by modulating macrophage bacterial clearance mechanisms. Moreover, l-arginine increased albumin and liver protein synthesis, restored endothelial histologic injury, and slightly limited the increase in pulmonary arterial pressure and concomitant edema formation. In a hyperdynamic pig model, l-arginine increased muscle protein turnover, and protein turnover was reduced in the liver. We therefore suggested that l-arginine reduces the severity of the hepatic response to tissue injury and inflammation. When l-arginine infusion was begun even before initiation of endotoxin infusion in the same pig model,[139a] liver blood flow and oxygen delivery and consumption increased. l-arginine also normalized the intestinal motility pattern in endotoxemic pigs (Bruins et al., unpublished). However, effects on survival are not uniformly positive.[134,140] In Table 3 , the potential metabolic pathways that may benefit from arginine supplementation in sepsis are summarized. Beneficial effects of arginine supplementation that were observed in other clinical diseased states[141,142,143,144] may also apply to sepsis.
In conclusion, the previous discussion may point to a beneficial effect of arginine and NO, although the effect on survival is still questionable. Does the information about the effects of NO inhibition therefore help us in further understanding of the beneficial effects of arginine in sepsis?
In human sepsis, only nonselective NOS inhibitors have been used. Prolonged inhibition of NO synthesis in sepsis with NG-nitro-l-arginine methyl ester or NG-monomethyl-l-arginine increased systemic and pulmonary blood pressure and vascular resistance, decreased cardiac output,[145,146,147,148,149,150,151] did not change plasma proinflammatory cytokines, and reduced the requirement for norepinephrine.[150,152] A decrease in plasma nitrate and increase in plasma arginine was found during NG-monomethyl-l-arginine treatment by Watson et al., but the decrease in plasma nitrate was not confirmed by Avontuur et al. Moreover, oxygen delivery decreased, arterial oxygenation and oxygen extraction improved,[146,150,152] and pulmonary gas exchange improved. However, no effects of these NOS inhibitors were seen on renal, pulmonary, and liver function.[151,152,153,154,155,156] It was suggested that tissue oxygenation was not compromised because oxygen consumption and splanchnic oxygenation (as measured by tonometry) were unaffected and because arterial lactate and pH did not change.[146,153] Although short-term resolution of shock was higher in septic patients treated with NG-monomethyl-l-arginine, this did not improve long-term survival.[146,153,154,155,156,157] A subsequent international, multiple-center, phase III study investigating nonselective NOS inhibition (NG-monomethyl-l-arginine) in septic patients was terminated because of increased mortality in the treatment group. The use of nonselective NOS inhibitors in septic patients is therefore not recommended. Animal studies indicated improved systemic vascular response[109,131,138,158] but impaired organ perfusion[114,159,160,161,162] or reduced oxygen supply with tissue damage. Increased mortality was reported in several studies (see Symeonides and Balk and Kirkeboen and Strand for reviews). Using nonselective NOS-inhibitors, NO was suggested to have beneficial effects on microcirculation by anti-adhesive and improved erythrocyte deformability.[102,164] Due to the absence of an overall benefit of nonselective NOS inhibitors, selective NOS-2 inhibitors came into use, but they have only been applied in animals. These inhibitors in general prevent hypotension, but the increase in pulmonary pressure and vascular resistances were not influenced by NOS-2 inhibition. Moreover, cardiac contractility and tissue oxygen delivery and uptake improved, and the increase in monocyte reactive oxygen species production and metabolic derangements was prevented.[165,166,167,168] In addition, tyrosine nitration is prevented, the endotoxin-induced impairment in liver microvascular and in portal and hepatic arterial blood flow is reduced, and liver injury is prohibited. However, tissue hypoperfusion, ischemia and decreased oxygen delivery, and acidosis have also been reported. In addition, absence of changes in liver morphology and in hepatocellular injury, intestinal mucosal injury, renal dysfunction, and pancreatic injury were observed by others.[167,171,172,173] Short-term survival (6 to 7 hrs)[165,171,174] and long-term survival improved, but increased late mortality (2 days) has also been observed.
Differences between animal (mainly small vs. large mammals) and sepsis models (time and dose of administration) may contribute to the divergent results in experimental studies and contribute to differences between animal and human studies. Moreover, effectiveness of NO blockade may depend on the precise stage at which treatment is given. Thus, a uniform strategy for the use of NOS inhibitors is not present at the moment.
In conclusion, nonselective NOS inhibition seems not to be beneficial, and selective NOS-2 inhibition also did not induce overall beneficial effects. In summary, l-arginine administration as a monotherapy has beneficial effects in animal models, and more data in septic patients with regard to its metabolic effects and outcome are required. In general, the majority of experimental and clinical data regarding NO-donors do not support the suggested drawback of increasing NO production with l-arginine, and also, NOS inhibition is not beneficial overall.
Crit Care Med. 2004;32(10) © 2004 Lippincott Williams & Wilkins
Cite this: Sepsis: An Arginine Deficiency State? - Medscape - Oct 01, 2004.