Introduction

Acute renal failure is commonly associated with severe sepsis : the combination demonstrates a mortality as high as a 68 percent ₁. As yet, no single intervention has been conclusively shown to decrease the incidence of acute renal failure in this clinical setting. The endogenous homeostatic peptide, recombinant b-natriuretic peptide (rBNP) is a physiologically and therapeutically attractive renoprotective agent due to its properties of renal vasodilation, natriuresis, and inhibition of renin secretion. We present a case in which rBNP was used as adjunctive therapy to prevent incipient sepsis related renal failure.

Case report

A 52 year old man was admitted to our intensive care unit (ICU) with grade 4 graft versus host disease and 100% total body surface area skin sloughing on post-procedure day 25 after a peripheral blood stem cell transplant. On arrival, he was sedated and intubated in order that adequate pain control be achieved during planned aggressive skin debridement and dressing. On ICU day 10, despite meticulous skin care and reverse isolation, the patient was noted to have green discoloration of some of his exposed dermis and he developed fever, hypotension, oliguria, and increased leukopenia. The diagnosis of sepsis syndrome was made and blood cultures later grew pseudomonas aeruginosa. Broad spectrum antiobiotics were administered and an infusion of norepinephrine .01 mcg/kg/min was begun to maintain hemodynamic stability. In consideration of the patient's already very high risk of acute renal failure (baseline creatinine of 1.4 mg/dL and blood urea nitrogen 80 gm/dL) and the potential renovascular constrictive properties of norepinephrine, an infusion of rBNP was begun at .005 mcg/kg/min without a loading dose. The introduction of rBNP was not associated with hemodynamic instability or increased pressor requirements and perfusion pressure remained stable at a mean of 70 mmHg. Serum creatinine and urine output were found to improve despite the presence of septic shock and vasopressor support, and the infusion was discontinued at 14 hours (figure 1). Upon cessation of the rBNP infusion, renal function was preserved and no further deterioration was witnessed over the following week.

Discussion

Hypotension and hypoperfusion are commonly understood to be the common denominator in precipitating acute renal failure in the setting of sepsis , however local renovascular pathology may also be the direct result of endotoxin, other inflammatory mediators, and disordered neurohormonal mechanisms . For example, platelet activating factor, levels of which correlate with severity of sepsis-induced ARF₃, decreases glomerular filtration rate (GFR) by increasing both afferent and efferent arteriolar resistances₄, ₅. The effect is attenuated by PAF antagonism₆ and potentially exacerbated by endotoxin ,leukotrienes ₂,₇,₈ and the hyperreninemic milieu of sepsis₉,₁₀.

Given the complexity of the pathophysiology of sepsis-related ARF, success with putative renoprotective agents has been very disappointing, none more so than so called “renal dose”dopamine. Although there are no trials which specifically address renal outcomes in septic patients on dopamine, two recent meta-analyses and one large randomized, placebo controlled trial₁₁,₁₂,₁₃ have debunked the notion that dopamine itself alters renal outcomes in any positive fashion and it has largely fallen from clinical favor as preserving renal function. On the contrary, it may in fact be harmful. Alternatively, a number of diuretic agents have been employed in attempts to prevent ARF. Of these, furosemide and mannitol are the most studied. Mannitol with volume loading and bicarbonate have been advocated in rhabdomyolysis, but this strategy did not reduce the risk of acute renal failure, need for hemodialysis, or mortality in a retrospective analysis of 350 trauma patients with creatine kinase levels between 5000 and 30000 of whom 130 received a protocolized infusion of bicarbonate and mannitol₁₄. Furosemide has been widely used to convert oliguric to nonoliguric renal failure and to protect against ARF despite almost three decades of evidence that it can, in fact, cause ARF₁₅. Indeed, multiple studies of furosemide as a reno-protectant have shown no significant alteration of renal recovery, need for dialysis, or mortality₁₆, ₁₇ and its use to alter the course of renal insult has been abandoned. Nor has mediator removal has not met with better success. In the absence of a clear pharmacologic agent, mediator removal by continuous hemofiltration has generated considerable interest in the prevention of renal failure in sepsis. However, recent evidence has not demonstrated improved prognosis or organ function despite the institution of hemofiltration ₁₉,₂₀.

Unlike the previously studied agents human b-type recombinant natriuretic peptide provides an endogenous molecule which is part of the kidney's homeostatic response to hemodynamic aberration, as occurs in acute renal failure₂₁. By selectively increasing intracellular cGMP in renovascular endothelium renovascular resistance is reduced and perfusion increased : an effect which is resistant not only to antagonism by norepinephrine but enhanced by r-BNP inhibition of renin and aldosterone secretion₂₁, ₂₂. Ultimately, by selectively vasodilating the afferent arteriole and constricting the efferent arteriole, GFR is preserved, if not increased. This has been supported clinically by the administration of administration of closely related atrial natriuretic peptides, including urodilatin , in the setting of acute renal failure, toxic nephropathies etc, where GFR was found to improve and the need for dialysis was reduced ₂₁,₂₃. While no other putative renoprotective agent demonstrates the reno-specific mechanisms of increased GFR secondary to direct renovascular activity while simultaneously antagonizing renovascular constrictors , the dose response relationship of natriuretic peptides has remained unclear as attested to by up to a 3000% variation in dosing in some studies leading to the potential of systemically untoward effects₂₄. However by applying an ultra low dose regimen , hemodynamic instability was avoided.

In conclusion, infusion of ultra low dose rBNP appears to have precipitated sustained increased urine output, subsequent decline of serum creatinine, overall improvement in renal function and the avoidance of hemodynamic instability despite the presence of shock and the requirement for vasopressor support. However compelling the result, on the basis of a single case it is impossible to make conclusions about the potential therapeutic role of rBNP. Further study of the potential reno-specific action of this peptide with particular re-evaluation of its dose response relationship therefore seems warranted.

Figure 1

Figure 1: Creatinine and urine output. The rBNP infusion was initiated at hour 5 and stopped at hour 15.

Correspondence to

Michael P. Hutchens, M.D., M.A. Department of Anesthesiology and Perioperative Medicine UHS-2 Oregon Health and Science University 3181 Sam Jackson Park Way Portland, OR, 97239 (o)503-494-5142 hutchenm@ohsu.edu