All patients undergoing surgery—from simple surgery to an extremely complex operative procedure—suffer some perturbation in oxygen delivery to the kidneys. Postoperative acute kidney injury (AKI) portends an increase in overall morbidity, mortality, and hospital resource use. Thus the identification of patients with an increased risk for postoperative AKI is critical. The prognosis of postoperative AKI for a particular patient scheduled for a specific operative intervention may (1) assist the patient and the family in the decision to undergo a particular type of surgical procedure and (2) allow optimization of preoperative, intraoperative, and postoperative renal homeostasis. Several investigations suggest that the anticipation of postoperative AKI and its timely diagnosis are critical to the effective treatment of postoperative AKI, and this has potential to minimize the risk for temporary or chronic renal replacement therapies. Patient characteristics (e.g., advanced age, diabetes mellitus [DM], vascular disease), the type of operative procedure (e.g., endoscopic cholecystectomy versus aortic arch replacement), and intraoperative characteristics (e.g., the duration of hypotension or suprarenal cross-clamp placement, circulatory arrest), as well as medications and/or interventions that are known risk factors for AKI (aminoglycoside or radiocontrast dye), all affect the risk of developing postoperative AKI ( Box 8.1 )
Patient characteristics
- •
Advanced age
- •
Diabetes mellitus
- •
Left ventricular dysfunction
- •
Peripheral vascular disease
- •
Renovascular disease
- •
Sepsis
- •
Hepatic failure
Operative procedure
- •
Aortic surgery
- •
Cardiopulmonary bypass
- •
Trauma surgery
- •
Liver transplant
- •
Renal transplant
- •
Lung transplant
Perioperative renal insults
- •
Prolonged dehydration
- •
Prolonged bladder obstruction
- •
Hypoxia
- •
Hypotension
- •
Aminoglycoside exposure
- •
Myoglobin or hemoglobinuria
- •
Intravenous contrast dyes
Several chapters in this book address the association between the surgery and postoperative AKI; however, the specific goal of this chapter is to provide clinicians with methods to identify patients at increased risk for postoperative AKI. First, we outline the intrinsic effects of surgery and anesthesia on renal physiology that illustrate why all patients are at potential risk for postoperative AKI. Second, we discuss patient characteristics that are associated with an increase in risk for postoperative AKI. Third, we detail three types of surgical intervention—extracorporeal circulation, profound hypothermic circulatory arrest (HCA), and suprarenal aortic occlusion—that mechanically and profoundly reduce normal renal perfusion during an operative procedure and thus pose a significant risk for postoperative AKI. (Although this is not an exhaustive list of the mechanical perturbations that can affect renal blood flow [RBF] during an operative procedure, these interventions are chosen because they are used in a number of surgical procedures.) Fourth, because preexisting renal impairment is associated with an increased risk for postoperative AKI, we present strategies to identify patients with preexisting kidney disease or to quantify the severity of preexisting chronic kidney disease. Fifth, we conclude the chapter with a review of existing scoring systems for quantifying the risk for postoperative AKI and their respective utilities.
Influence of Surgery and Anesthesia on Renal Perfusion
The kidney is an elegant system of integrated processes that maintain fluid homeostasis and eliminate waste products. Although the kidney requires only 10% of the total corporeal oxygen consumption, the renal cortex receives 90% of the total RBF and extracts only 18% of the oxygen delivered. In contrast, the renal medulla receives only 10% of the RBF to the kidney, and it is the site of the costly energy- and oxygen-consuming processes that are responsible for reabsorbing tubular sodium and water. In the medulla, 79% of the oxygen delivered is extracted, resulting in a high arteriovenous oxygen gradient in the medulla of the kidney. Thus the medulla is exquisitely sensitive to reductions in RBF. A 40% reduction in RBF may lead to acute tubular necrosis, especially in the presence of other renal insults. Interventions that improve RBF (increased cardiac output [CO], fluid replacement) or decrease medullary oxygen consumption can potentially improve the tolerance to intermittent periods of ischemia. Other conditions, including exposure to radiocontrast dyes and an increased bilirubin concentration or myoglobinuria, can exacerbate the adverse response to renal hypoxia by increasing the osmotic load to the nephron and further increasing oxygen requirements.
Several types of drugs have been shown to be nephrotoxic through a variety of mechanisms. The perioperative use of aminoglycosides or non-steroidal anti-inflammatory drugs may precipitate AKI in the hypoxic kidney. The chronic use of angiotensin-converting enzyme (ACE) inhibitors may lead to an attenuation of the normal compensatory response to a decrease in renal perfusion. Aprotinin, which is an anti-inflammatory serine protease inhibitor that was used intraoperatively to reduce blood loss, is concentrated in the kidney, and as previous studies suggested, its use was associated with postoperative AKI and renal failure requiring renal replacement therapy. Further, a combination of ACE inhibitors and aprotinin can have a synergistic adverse effect on renal function in patients undergoing cardiac surgery with cardiopulmonary bypass (CPB).
AKI itself is defined by an abrupt loss of kidney function, resulting in the accumulation of metabolic waste and byproducts and the loss of fluid and electrolyte regulation. There has been a shift from terms such as “renal dysfunction” and “renal failure” to the more accepted term of “kidney injury” to better reflect the knowledge that a small degree of renal dysfunction may not necessarily result in complete organ failure but still have profound physiologic consequences and can result in significant morbidity and increased mortality. There are many—over 30—definitions of AKI using various clinical criteria. Indeed, the lack of consensus of a single clear and universally accepted definition of AKI has been a source of controversy and confusion in the field, as well as a barrier to research of the disease and advancement in knowledge of AKI.
Three of the most well-known definitions are the Acute Dialysis Quality Initiative RIFLE criteria, the Acute Kidney Injury Network (AKIN) criteria, and the Kidney Disease: Improving Global Outcomes (KDIGO) criteria, which have been developed over years as knowledge of AKI has improved. RIFLE criteria used the idea of stages of kidney injury, known as “risk,” “injury,” “failure,” “loss of kidney function,” and “end-stage renal disease,” from which the acronym was derived, and used increases in creatinine (Cr) and decreases in urine output (UO) as metrics. The AKIN criteria were developed later and revised the RIFLE criteria, removing the “loss of kidney function” and “end-stage renal disease” stages and simplifying the names of the first three stages to stages 1, 2, and 3. They also modified the time scale over which these perturbations occur, reducing the time to 48 hours from the RIFLE criteria’s original 7-day window. The KDIGO criteria are the most recent and allow correction of volume status and obstructive causes of AKI before classification. The KDIGO definition of AKI involves an increase in serum creatinine (sCr) by 0.3 mg/dL or more within 48 hours, or an increase in sCr to 1.5 times baseline or more within the prior 7 days, or UO less than 0.5 mL/kg/h for 6 hours. Further criteria are then used to stage AKI.
Anesthesia (both general and regional) and surgery, independently and synergistically, could impair renal homeostasis during operative procedures. Both types of anesthesia are associated with peripheral vasodilation, thus leading to a reduction of circulating blood volume and renal perfusion. Vasodilation and the resultant reduction in RBF are particularly a concern in a patient who has been nil per os (NPO) for 8 or more hours, was prescribed a chronic diuretic or an ACE inhibitor, has suffered recent vomiting or diarrhea, has undergone a recent bowel preparation, or is actively bleeding. No studies have demonstrated that general anesthesia is superior to regional anesthesia in limiting the likelihood for postoperative AKI.
Regarding general anesthesia in particular, there have been investigations into the possible adverse renal effects of inhalational anesthetics and their fluoride ion metabolic byproducts. Of the clinically used inhalational agents, sevoflurane and enflurane release the greatest number of fluoride moieties. However, the pharmacokinetics of these volatile agents greatly limits the likelihood of renal dysfunction secondary to fluoride exposure. Choice of an inhalational agent has been found to have little relevance for inducing postoperative AKI, as it was reflected by a lack of elevation of AKI biomarkers in the patients who underwent elective cardiac surgery. An interaction between sevoflurane and soda lime or barium hydroxyl carbon dioxide could also pose a potential risk for producing a potentially nephrotoxic haloalkene known as “compound A.” However, the metabolic pathways for this compound in humans limit the likelihood of compound A–induced nephrotoxicity. Indeed, there are no published reports of AKI induced by exposure to sevoflurane in surgical patients.
Patient Characteristics and the Risk of Postoperative Acute Kidney Injury
Patient-specific factors (identifiable in the preoperative period) that are associated with postoperative AKI fall generally into one of three categories: (1) patient characteristics suggesting impaired renal functional reserve, (2) physiologic findings associated with impaired renal perfusion, or (3) pathophysiologic processes that cause renal injury. Potential risk factors include advanced age; abnormal sCr values; DM; and any sign, symptom, or factor indicative of a reduced CO and thus compromised renal perfusion ( Fig. 8.1 ). However, the low incidence of postoperative AKI necessitating renal replacement therapy (< 2.0% even in high-risk populations) and the lack of sensitive, routine laboratory studies that identify AKI even when it does not result in an increase in Cr or blood urea nitrogen (BUN) have challenged investigators to identify factors that are associated with postoperative AKI.
Patient Characteristics Associated with Reduced Renal Reserve
Renal mass and overall renal function decrease with advancing age. By age 70 years, the number of functioning nephrons is reduced by half. Novis and colleagues reviewed 28 studies evaluating risk factors for postoperative AKI. Four of the five largest studies identified advanced age as a risk for postoperative AKI. In a study of 2400 patients undergoing elective coronary artery bypass grafting (CABG) surgery, Mangano and colleagues found that patients between 70 and 80 years old and those over 80 years old suffered a twofold and fourfold increase in the risk for postoperative AKI, respectively. Similarly, Chertow and colleagues, in their study of 42,723 US Department of Veterans Affairs (VA) cardiac surgery patients, found that 1.5% and 1.8% of CABG surgery patients between 70 and 80 years old and older than 80 years, respectively, required postoperative dialysis. In comparison, only 0.5% and 0.9% of patients 50 to 59 and 60 to 69 years of age, respectively, required renal replacement therapy. However, after adjusting for peripheral vascular disease, prior cardiac surgery, and other related variables, advanced age no longer remained a significant predictor of renal replacement therapy.
Patients with DM, especially those suffering from long-standing disease managed with insulin, tend to have reduced renal reserve. Type 1 DM and preoperative glucose values greater than 300 mg/dL were significantly and independently associated with postoperative AKI in the study of Mangano and colleagues. Those patients with type 1 DM and those with hyperglycemia were 1.8 and 3.7 times more likely, respectively, than patients without type 1 DM to develop postoperative AKI. Although some investigators have failed to demonstrate an independent association between DM and postoperative AKI, many other investigators have been able to develop risk indices that included DM as a significant predictor of postoperative AKI.
Recent research has focused on the role of obesity and its association with increased risk of postoperative AKI. Several studies have shown a two- to threefold increase in the rate of postoperative AKI in patients with obesity, and a potentially larger increase in risk when other components of metabolic syndrome were also present. This relationship has been seen in both cardiac and noncardiac surgery patients. Interestingly, some studies have found a paradoxical effect between obesity and improved outcomes, which has been ascribed to several possible metabolic and pathophysiologic mechanisms, all of which unfortunately remain poorly studied and therefore poorly understood.
Reduced renal functional reserve is best identified by measuring glomerular filtration rate (GFR). However, most studies rely on sCr values as a potential measure of renal reserve. Because Cr values are substantially affected by nonrenal factors (e.g., age, sex, muscle mass), they cannot be considered adequate measures of GFR. Some studies have used the Cockcroft-Gault equation to estimate GFR. Studies using sCr values, derived values for GFR, or true creatinine clearance (CrCl) assessments have almost uniformly identified reduced preoperative renal reserve, manifested as an increase in Cr or as a reduction in GFR, as the most common and one of the most important patient characteristics associated with postoperative AKI ( Table 8.1 ).
| Renal reserve | Remaining nephron (%) | Glomerular filtration rate (mL/min) | Signs/symptoms | Laboratory abnormalities | Risk of dysfunction or failure |
|---|---|---|---|---|---|
| Normal | > 50 | 125 | None | None | Minimal |
| Decreased renal reserve | 40 | 50–80 | None | None | Mild |
| Renal insufficiency | 20–40 | 20–50 | Nocturia | Moderate increase in BUN/creatinine, unless stressed | Moderate |
| Uremia | 5–10 | < 20 | Uremic syndrome | Multiple | Severe |
Several studies have emphasized the additive risk associated with even mild preoperative kidney disease on postoperative AKI. Weerasinghe and colleagues studied 1427 patients with no known kidney disease who were scheduled for elective primary CABG surgery. They reported that patients with a minimal elevation in Cr values (Cr > 130 μmol/L) were at a substantial risk for postoperative renal failure requiring renal replacement therapy.
Patient Characteristics Suggesting Reduced Renal Blood Flow
Many patients presenting for surgery—particularly cardiac or major vascular surgery—have signs and symptoms that suggest a reduced CO and thus reduced RBF. Such signs and symptoms include rales, rhonchi, edema, jugular venous distention, abnormal heart sounds, a need for pharmacologic or mechanical ventricular support, and echocardiographic evidence of a significantly reduced ejection fraction (< 35%). Even in the absence of apparent preexisting renal dysfunction (i.e., with normal Cr values), these patients may have a greatly reduced ability to maintain normal GFR and fluid homeostasis, especially with the additional insults of surgery and other interventions.
The study by Chertow and colleagues of 43,642 cardiac surgery patients and the report by Thakar and colleagues of 33,217 cardiac surgery patients both emphasize the prognostic value of signs, symptoms, interventions, and laboratory assessments indicating reduced left ventricular function and thus compromised renal perfusion. The preoperative presence of any of these factors usually indicates that there is an increased risk for postoperative AKI.
Recent studies have further explored the association between preoperative hemodynamic compromise and postoperative AKI. A meta-analysis by Brienza and colleagues that included 20 studies with a total of 4220 patients suggested that perioperative hemodynamic optimization may reduce the risk of postoperative AKI. Adequate fluid resuscitation combined with inotropic support in the preoperative period could potentially be an effective way to minimize the risk for postoperative AKI, particularly in higher-risk patients.
Several studies assessed the association of postoperative AKI with factors that might be expected to be associated with reduced RBF, such as renal artery stenosis. In one of the studies, Conlon and colleagues assessed the association of preexisting renal artery stenosis (> 50% stenosis) with postoperative AKI, but they found no significant association between preexisting renal artery stenosis and postoperative AKI.
Other Preoperative Patient Variables Associated with Kidney Injury
A patient may suffer various pathophysiologic processes in the preoperative period that expose the kidney to toxic insults, such as increased levels of metabolic byproducts or naturally occurring biomolecules, thus putting additional stress on renal function. These phenomena cause an increase in renal oxygen demand. For example, hepatic failure or obstruction leads to an increase in bilirubin and other incompletely metabolized moieties. As these metabolites accumulate, the renal parenchyma compensates for hepatic insufficiency and must detoxify, excrete, concentrate, or secrete these toxic substances. This often requires an increase in oxygen demand, and thus, it may limit the kidney’s ability to withstand any further renal insults that are inflicted during the perioperative period. Similarly, patients suffering massive trauma, hemorrhage, or extensive burn injuries have increased renal oxygen requirements, as the kidneys must filter increased plasma levels of myoglobin and hemoglobin. Both sepsis and gut ischemia are associated with endotoxin release and may lead to a reduction in renal perfusion, and perhaps more important, an accelerated inflammatory response reaction. During the acute inflammatory phase, the kidney is recruited to help manage many of the inflammatory kinins, and it is exposed to activated leukocytes. Filtered inflammatory mediators may be tubulotoxic. Hypotension reduces GFR, and efferent arterioles constrict to compensate. Ultimately, the hypotension, the medullary ischemia, and the sludging of activated neutrophils lead to an increased neutrophil adhesion potential. Adherent neutrophils release vasoconstrictive and tubulotoxic mediators, further increasing renal oxygen demand and reducing oxygen delivery.
Other Preoperative Patient Factors Associated with Postoperative Renal Dysfunction
Other patient factors, including previous cardiac surgery, peripheral vascular disease, hypertension, and chronic obstructive pulmonary disease, have been identified as potential risk factors for postoperative AKI. These factors, which have not been uniformly accepted as risk factors, are discussed in the following section.
Surgical Interventions Associated With Substantial Intraoperative Renal Ischemia
Cardiopulmonary Bypass
The incidence of postoperative AKI after cardiac surgery ranges from 2% to 50%, with approximately 0.4%–4.7% of patients who develop postoperative AKI requiring dialysis ( Table 8.2 ). In a landmark study of 42,773 patients by Chertow and colleagues, dialysis-dependent postoperative AKI occurred in 1.1% of patients with a 63.7% associated mortality—striking when compared with the 4.3% mortality among those who did not require dialysis after cardiac surgery. Postoperative AKI heralds a poor prognosis with increasing complications and mortality, especially for patients with postoperative respiratory failure, hypotensive episodes, bleeding, atrial fibrillation, or other end-organ dysfunction.
| Author (ref. no.) | Number of patients | Study design | Definition of renal dysfunction | Nondialysis outcome | Dialysis outcome | ||
|---|---|---|---|---|---|---|---|
| Incidence | Mortality | Incidence | Mortality | ||||
| (%) | (%) | (%) | (%) | ||||
| Yeboah et al. | 428 | — | — | 26 | 38 | 4.7 | 70 |
| Abel et al. | 500 | Prospective | Cr > 1.5 mmol/L | 21.6 | 13.8 | 3 | 100 |
| Bhat et al. | 490 | Retrospective | Cr ≥ 1.6 mmol/L a | 28.1 | 10.9 | 2.2 | 45 |
| McLeish et al. | 1542 | Retrospective | — | Not reported | Not reported | 1.3 | 35 |
| Hilberman et al. | 204 | Case control | BUN ≥ 30 mmol/L b | 2.5 | 60 | 2.5 | 60 c |
| Gailiunas et al. | 752 | Retrospective | Cr > 1.5 mmol/L | 17 | Not reported | 1.5 | 27 |
| Lange et al. | 2959 | — | — | Not reported | Not reported | 1.2 | 53 |
| Corwin et al. | 572 | Case control | Cr postop ≥ 1.5 Cr preop | 6.3 | Not reported | 1 | 33 |
| Slogoff et al. | 504 | — | Cr ≥ 1.5 mmol/L d | 2.4 | 0.2 | 0.4 | 0 |
| Zanardo et al. | 775 | Prospective | Cr ≥ 1.5 mmol/L | 15.1 | 9.5 | 0.5 | 44 |
| Mangano et al. | 2417 | Prospective | Cr ≥ 2.0 mmol/L | 7.7 | 19 | 1.4 | 40 |
| Abrahamov et al. | 2214 | Retrospective | CrCl < 40 mL/min/1.73 m 2 e | 2.1 | Not reported | 1% | 30 |
| Grayson et al. | 5132 | Retrospective | Cr > 2.0 mmol/L | 2 | Not reported | 0.9 | 32.9 or 46.2 f |
| Antunes et al. | 2455 | Prospective | Cr ≥ 2.1 mmol/L g | 5.6 | 5.8 | 0.6 | 33.3 |
| Oprea et al. | 6130 | Retrospective | Cr ≥ 0.3 mmol/dL | 58 | 30.7 | Not reported | Not reported |
a Also, Cr postop – Cr preop ≥ 0.4 mmol/L (mg/dL).
b Or inulin or creatinine clearance ≤ 50 mL/min/1.73 m 2 .
c Mortality is combined for dialysis and nondialysis patients.
d And fractional excretion of sodium > 1% or total urinary sodium > 20 μg/L or urine with casts, epithelial cells, or cellular debris.
e Creatinine clearance at least a 15-mL/min decline from preoperative level.
f Depending on the type of surgery (32.9% after coronary artery bypass grafting [CABG]; 46.2% after valve operation with or without CABG).
g Plus an increased creatinine level of ≥ 0.9 mmol/L from preoperative to maximum postoperative values.
These patients remain in the intensive care unit and hospital for greater durations and are more likely to require specialized long-term care. The overall mortality for patients who developed postoperative AKI requiring renal replacement therapy ranges from 27% to 100%. The economic impact of postoperative AKI in cardiac surgery patients is considerable. In fact, it is estimated that nationally, the direct hospital cost of caring for patients with post-CPB renal failure is approaching $645 million.
Postoperative AKI after cardiac surgery with CPB in patients with advanced age can sometimes be the result of acute renal ischemia superimposed on preexisting limited renal reserve. Predictive risk factors indicative of compromised renal perfusion, such as advanced age (> 70 years old), preoperative left ventricular dysfunction, atherosclerotic vascular disease, decreased renal reserve, DM, prior myocardial revascularization, a longer duration of CPB time (> 3 hours), type of operation (e.g., valvular surgery), perioperative use of nephrotoxic agents (e.g., radiocontrast dyes), and postoperative low CO, all have been associated with an increased risk for postoperative AKI.
Historically, extracorporeal circulation was thought to be associated with multiple perturbations in renal physiology and function. During CPB, there are substantive decreases (25%–75%) in RBF and GFR and increases in renal vascular resistance. These physiologic perturbations are likely sequelae of the loss of pulsatile blood flow, increases in circulating catecholamines and inflammatory mediators, macroembolic and microembolic insults to the kidneys (organic and inorganic debris), release of free hemoglobin from traumatized red blood cells, and decrease in flow rates and mean arterial pressure during CPB. In addition, extreme hemodilution and deep HCA have been associated with a significantly greater likelihood of postoperative AKI. However, several studies have found no effects of hypothermia without circulatory arrest, pulsatile perfusion, pH-stat management, or type of membrane oxygenators on renal function after CPB. Lema and colleagues studied the GFR and effective renal plasma flow of patients with a Cr level of less than 1.5 mg/dL undergoing hypothermic CPB and found that renal function was not adversely affected by CPB.
The off-pump coronary artery bypass approach to coronary revascularization was expected to substantially reduce the incidence of end-organ dysfunction (e.g., renal, cerebral) observed in patients undergoing CABG surgery with CPB. However, a substantial number of retrospective and prospective studies comparing the incidence of adverse outcomes associated with off- versus on-pump CABG surgery have failed to indicate that off-pump CABG surgery was associated with a lower risk for postoperative complications. Although studies of patients undergoing off-pump CABG surgery revealed significantly fewer changes in microalbuminuria, fractional extraction of sodium, free water clearance, free hemoglobin, and N -acetyl-β- d -glucosaminidase, this was not associated with a significant reduction in postoperative AKI. In patients with preoperative non–dialysis-dependent AKI, however, off-pump CABG surgery appears to lower the incidence of postoperative renal failure, need for renal replacement therapy, and mortality. A recent meta-analysis indicated some potential benefit of off-pump CABG surgery for the reduction in the incidence of postoperative AKI; however, there was no difference found in need for dialysis or mortality. In contrast, Lamy and colleagues failed to show a difference between new-onset renal failure requiring dialysis at either 30 days, 1 year, or 5 years after on- or off-pump CABG surgery. Additional studies have found similar results with respect to postoperative AKI. Finally, McCreath and colleagues suggested that a minimally invasive cardiac surgical approach such as the port access minithoracotomy approach to mitral valve surgery may confer a reduction in the incidence of postoperative AKI when compared with a standard median sternotomy approach.
Pharmacologic Management of Post–Cardiopulmonary Bypass AKI
Pharmacologic approaches to reduce postoperative AKI have been studied extensively in recent years. Although most of these pharmacologic agents appeared to promote UO in the perioperative period, intraoperative UO had no correlation with postoperative renal function, especially when diuretics were used intraoperatively. Dopamine (DA) at low doses activates the DA-1 receptor and has the theoretical benefits of renal artery dilation, natriuresis, and diuresis. Although DA was once shown to increase renal plasma flow, GFR, and urinary sodium excretion, these dopaminergic effects were not observed in patients with impaired renal function (GFR < 50 mL/min/1.73 m 2 ), probably because of a lack of renal reserve capacity in response to the effects of DA. Randomized controlled trials with cardiac surgery patients have not demonstrated that prophylactic low-dose DA can preserve renal function, reduce the development of AKI, and decrease mortality. In addition, the positive inotropic and chronotropic effects of DA could lead to the development of perioperative arrhythmias and an increase in myocardial oxygen consumption that are potentially deleterious to cardiac surgery patients. Furthermore, the use of DA to promote diuresis in patients who are hypovolemic is likely to exacerbate renal failure. In view of the lack of proven benefits and the potential harm, routine use of DA to promote diuresis in the perioperative setting is not recommended.
Mannitol is a hyperosmotic agent that increases GFR during periods of renal hypoperfusion, augments renal cortical and medullary blood flow, and promotes scavenging of reactive hydroxyl free radicals. It reduces renal oxygen consumption during periods of ischemia and enhances diuresis of intraluminal debris. However, the prophylactic use of mannitol in patients undergoing CPB has not produced convincing improvement in renal function and mortality outcome.
Loop diuretics such as furosemide offer benefits similar to those of mannitol in reducing oxygen consumption and improving diuresis by preventing the accumulation of obstructive casts. In two separate prospective randomized studies, the use of furosemide during and after CPB was found to have no clinical benefits and to be potentially detrimental to renal function. Although DA, mannitol, and furosemide can convert oliguric to nonoliguric renal failure and facilitate management of fluid balance and electrolytes, in the absence of evidence that forced diuresis translates to survival benefit, the routine use of these medications is not encouraged.
A selective DA-1 receptor agonist, fenoldopam, has the theoretical advantages of decreasing renal vascular resistance and increasing RBF and GFR. Caimmi and colleagues studied patients with a preoperative Cr level of greater than 1.5 mg/dL and found that infusion of fenoldopam (0.1–0.3 μg/kg/min) during CPB and in the early postoperative period was associated with an improvement in postoperative Cr level and CrCl. In a prospective multicenter cohort study of high-risk cardiac surgery patients with renal failure, prophylactic infusion of fenoldopam was associated with a 50% reduction of AKI and a decrease in mortality from 15.7% to 6.5%. These clinical benefits were based on a preliminary experience with fenoldopam and will have to be confirmed by larger-scale prospective randomized clinical trials before it is used routinely.
Vasodilators such as calcium channel blockers (e.g., nifedipine, diltiazem) have been shown to improve GFR in patients undergoing cardiac surgery with CPB. However, clinically significant decreases in morbidity and mortality outcomes are lacking to advocate their routine use for the prevention and treatment of postoperative AKI.
Clonidine is a nonselective α-adrenergic agonist that may prevent renal hypoperfusion by inhibiting stress-induced catecholamine–mediated vasoconstriction. Kulka and colleagues found that in patients without an elevated risk for postoperative AKI, clonidine at 4 μg/kg prevented the deterioration of CrCl after cardiac surgery compared with placebo. However, the long-term benefit of clonidine infusion for reduction of mortality in high-risk patients has not been fully evaluated.
Atrial natriuretic peptide (ANP) is important in intravascular fluid and circulatory regulation. It promotes diuresis and natriuresis, increases GFR, reverses afferent renal vasoconstriction and efferent renal dilation, and inhibits sodium reabsorption. Intraoperative volume loading has been shown to regulate ANP release, and its concentrations may predict long-term outcomes after CABG surgery. Intraoperative infusion of ANP has been shown to decrease central venous pressure, pulmonary capillary wedge pressure, pulmonary vascular resistance, peripheral vascular resistance, and the renin-angiotensin-aldosterone system, and its use was associated with increasing RBF, GFR, and diuresis. The precise effects of ANP on post-CPB renal function and their potential role in curtailing renal failure and its associated mortality will have to be determined by future clinical trials.
Two studies that examined the institution of early renal replacement therapy in the form of continuous venovenous hemodiafiltration in patients with established postoperative AKI found a statistically significant reduction in the overall rate of mortality. Even outside the setting of the perioperative period, a debate exists on the optimal timing of renal replacement therapy for AKI in critically ill patients.
In summary, the effects of CPB on the kidneys have significant impacts on patient management and outcomes. However, despite intensive investigation into the pathogenesis and prevention of postoperative AKI, there has been only limited progress in the development of effective protective strategies. As intravascular volume depletion and hypoperfusion can lead to exacerbation of renal ischemia and accentuate the risk for postoperative AKI, avoidance of nephrotoxic agents and close attention to maintenance of intravascular volume, blood pressure, and CO are key in the effort to reduce the occurrence of postoperative AKI after cardiac surgery. Identification of patients with genetic variants of specific inflammatory markers (interleukin-6 [IL-6] gene promoter polymorphism, apolipoprotein E) to determine predisposition for postoperative AKI may be an additional means of identifying patients at risk for postoperative AKI. In recent years, kidney-specific biomarkers measured perioperatively have been correlated with prolonged CPB time and may help predict AKI after CPB.
Major Aortic Surgery
Patients undergoing major aortic surgery with or without HCA are particularly vulnerable to postoperative AKI. These patients are typically of advanced age and have atherosclerotic disease of the large arterial vessels and end-organ vascular beds (e.g., of the heart, brain, and kidney). Thus vascular surgery patients have many of the characteristics that are prognostic for postoperative AKI in cardiac surgery patients. Review of the literature suggests that the presence of preoperative AKI is the most consistent predictor for postoperative AKI following major vascular surgery. However, because each type of major vascular surgery procedure is associated with a significant mechanical perturbation that interferes with RBF—cross-clamping of the aorta, left-heart bypass, renal artery reimplantation or surgery, or CPB with or without HCA—the type and duration of the particular mechanical intervention render other potential variables less important predictors of postoperative AKI.
The reported incidences of AKI and dialysis after major aortic surgery vary substantially ( Table 8.3 ). The reported incidences of postoperative AKI range between 5% and 29% and between 1.6% and 22.2%, respectively. This reflects the nonuniform definitions of preoperative and postoperative AKI but also, and perhaps more important, the underlying potential for significant, potentially life-threatening differences in patient characteristics and management strategies in this group of patients.
