Perfusionist techniques of reducing acute kidney injury following cardiopulmonary bypass: an evidence-based review

Introduction

Over the last several decades, there has been an upsurge in the use of medical technology. These advances have revolutionized the practice of medicine and have resulted in a marked improvement in outcomes. The practice of cardiac surgery has been changed by the growth of these new technologies and, by the late 1950s and early 1960s, with the advent of the heart-lung machine (HLM), it became possible to perform complex cardiac surgery procedures without the high risks associated with these procedures. ¹ Despite these advances, studies have shown that there are detrimental effects to body systems when the HLM is used during these procedures. As early as the 1960s, it was noted that there was a change that occurs with kidney function and cardiopulmonary bypass (CPB). ² As materials for CPB have decreased complications, there remains significant morbidity that can occur during these procedures. These include a decrease in renal function as measured by postoperative changes in glomerular filtration rate (GFR), renal blood flow, urine output, renal plasma clearance and serum creatinine. ³ These complications have been associated with an increased incidence of postoperative morbidity, mortality and length of hospital stay. ⁴

CPB using the HLM and postoperative renal dysfunction are correlated. ⁵ The length of time on CPB is an independent risk factor for AKI in the postoperative setting. ⁶ What occurs during the time on CPB may have an effect on clinical outcomes. Perfusionists can alter blood flow and mean arterial pressure (MAP), control temperature, give fluids and administer medications via the CPB circuit. They also are responsible for the selection and preparation of the CPB circuit. Few studies clearly define the relationships between perfusion-specific factors and AKI in the adult population of patients undergoing CPB. This review paper will examine the relationships between CPB and AKI in order to determine the roles that these relationships have on outcomes. Studies were included based upon the relationship between CPB and AKI with normal kidney function. Studies were excluded in patients with pre-existing kidney disease or if there was no clear CPB-related cause to AKI. The review focuses on adult CPB-related management that requires perfusionist intervention, including hemodilution, systemic pump flow, pulsatile perfusion, MAP, mannitol, furosemide, ultrafiltration, hemolysis and mini-circuits.

Defining the need for evidence-based guidelines

Every year there are 800,000 cardiac surgical procedures performed worldwide. ⁷ It is estimated that AKI occurs in as many as 40% of these patients. ⁴ Patients who develop AKI have twice the average intensive care unit stay. ⁸ Having to undergo dialysis after a heart operation can increase a patient’s mortality rate by as much as 45%. ⁹ Significant healthcare resources are spent on patients who have AKI. ¹⁰ Patients who develop AKI after cardiac surgery are increasingly being discharged to skilled nursing facilities, causing even more resource consumption. ⁹ If practitioners are to engage in strategies that will help reduce the number of patients who develop this complication after surgery, then healthcare costs will decrease. ⁸ The impetus to ensure healthcare dollars are spent wisely is being felt throughout the industry. Exceptional patient care with as little morbidity as possible should be the goal. There is currently a movement in medicine today towards evidence-based practice guidelines. ¹¹ Perfusionists continue to be slow at translating evidence-based research knowledge into the clinical environment where it will affect patients most. ¹² The use of evidence-based guidelines can encourage research and allow practitioners to rely less on instinct.

Renal function & physiology

The kidney is a complex organ based upon its design and habitus in the body. It receives approximately 20% of the cardiac output. ¹³ It has unique oxygen requirements due to how blood flows to the renal medulla. ¹⁴ The renal medulla, because of its design, functions at low oxygen levels, but, during periods of ischemia, does not have the capability to survive without some injury. ¹⁵ The oxygen tension inside the medulla is low and any alterations in blood flow related to decreased hematocrit and pressure may damage the kidney. ¹⁶ The common problem post CPB is acute tubular necrosis (ATN) and granular casts are visible in damaged kidneys. ⁴ The renal cortex is sensitive to injury during the rewarming phase of a cardiac surgical procedure due to hypoperfusion. ¹⁷ The kidney responds if the mean arterial blood pressures fall below 80 mmHg by reducing the glomerular filtration rate (GFR). ¹⁸

Classification systems for acute kidney injury

The definition of AKI was developed to ensure a standardized nomenclature for comparison. Thirty-five definitions of renal failure are mentioned in the literature. ¹⁹ The Acute Dialysis Quality Initiative (ADQI), in 2004, developed the RIFLE classification, which stands for risk, injury, failure, loss of kidney function and end stage renal failure. ²⁰ The first three classes are stages while the final two classes are outcomes. This classification compares the effect of serum creatinine values along with glomerular filtration rate (GFR) and compares them to decreases in urine output to grade the severity of renal dysfunction as well as outcomes. ²⁰ Loss of kidney function is graded when a patient has been in the state for 4 weeks or more and end-stage kidney disease is observed after three months. ²⁰ The term AKI was derived from the ADQI and replaced the former use of the term acute renal failure. ²⁰

The Acute Kidney Injury Network, in 2007, issued the AKIN classification to allow for a rapid diagnosis. ²¹ The changes in creatinine over a shorter timeframe are believed to be prognostic of AKI. The first three stages of the RIFLE classification correlate with three stages of kidney injury. ²¹ The AKIN classification measures renal dysfunction in a 48-hour timeframe while the RIFLE classification takes one week. ²¹ The AKIN classification also does away with the outcome-based final two measures of the RIFLE system. Instead of a patient winding up on renal replacement therapy (RRT) or dialysis, they are automatically assigned to the highest class in the AKIN system, but that is not necessarily the case with the RIFLE classification. ²²

The literature demonstrates the difficulty in coming to a consensus before the AKIN classification. Research studies did not have similar endpoints as to what defines kidney injury. ⁷ Some studies have examined the role of creatinine and transient increases from preoperative levels as a marker of kidney injury while other studies look at whether patients end up on dialysis or RRT. ³ Recent research has focused on two new biomarkers which are sensitive to kidney injury: neutrophil gelatinase-associated lipocalin (NGAL) and cystatin C. These might be new and faster ways to determine AKI following cardiac surgery. ²³ These two markers have been found to be independent predictors of AKI quicker than serum creatinine levels as well as being predictive of the length of stay in the intensive care unit. ²³

Risk factors

A number of preoperative risk factors that predispose patients to AKI include advanced age, preoperative anemia, diabetes mellitus, increased body mass index (BMI) and use of the intra-aortic balloon pump (IABP). ²⁴ The duration of CPB correlates strongly with AKI. ²⁵ Also, patients undergoing valve procedures as well as combined procedures of coronary artery bypass grafting (CABG) and valve surgery are at greater risk. ²⁵

The use of CPB has been discovered to be an independent risk factor for AKI. ²⁶ A retrospective study of 2199 patients by Stallwood et al. of isolated CABG patients without renal dysfunction found that surgery with CPB was an independent risk factor for AKI in the postoperative setting. ²⁶ Patients who had gone on CPB had a 2.9% incidence of AKI versus 1.4% in the off-CPB group. ²⁶ Patients in both groups were similar in terms of comorbid conditions and the off-CPB group trended for reoperation. ²⁶ Previous work at establishing CPB as an independent risk factor in smaller cohorts did not show the same effect. ²⁶

The evidence indicating CPB as a risk factor exists, but the exact mechanism for AKI is complex. The exact role of CPB-related AKI is hard to define and, while there is an increase in mortality, it is hard to pinpoint one factor that causes this. The systemic inflammatory response syndrome (SIRS) is a potential problem of CPB. ²⁷ The whole body inflammation that occurs during SIRS subjects the kidneys to a potential insult from the large foreign surface area of the bypass circuit. ²⁸ SIRS is a risk factor for the development of AKI postoperatively. ⁶ Patients on CPB are at risk for receiving various forms of emboli during surgery. Perfusionist interventions, such as injections of medications into the circuit, have been listed as a source of microemboli generated during CPB. ²⁹ Increased embolic load from perfusionist interventions during CPB has been correlated to rising postoperative creatinine levels. ³⁰ Focus on decreasing interventions through the sample manifold and using larger bore syringes has been shown to decrease, but not eliminate, the amount of emboli generated during CPB. ³¹

Anemia & hemodilution

Anemia has been shown to be an independent risk factor for AKI during cardiac surgery. ³² Anemia potentially primarily contributes to AKI by decreasing oxygen delivery, which can also lead to patients receiving homologous blood transfusions. The foreign surface of the CPB circuit can cause coagulopathy to develop, leading to further blood product administration. As the number of red blood cell (RBC) transfusions increases, there is a tendency for higher AKI rates. ¹⁶ Swaminathan et al. made the association between nadir hematocrit (Hct) and renal outcomes in a retrospective study of isolated coronary artery bypass graft (CABG) patients. ³³ As the nadir Hct fell below 24%, renal failure increased. ³³ Patients who received RBC transfusions to treat a low Hct had increased renal injury as well. ³³ Habib et al. reported this association in a retrospective study of isolated CABG patients. ³² A nadir Hct of less than 24%, RBC transfusions and increased CPB time all played a role in renal injury. ³² The avoidance of hemodilution and maintaining an Hct equal to or greater than 24% may be beneficial as it can lead to less RBC transfusion during CPB.

Loor et al. described the multiple hit phenomena, which looked at anemia, transfusions or both. This retrospective study of more than 8000 patients reported that patients who experienced a hemoglobin level less than 8 g/dl exhibited an increased tendency for AKI. ³⁴ If RBCs were administered before this time, there was a decreased likelihood of AKI. ³⁴ If the patient had received both an RBC transfusion and experienced a hemoglobin less than 8 g/dl during surgery, there was an increased tendency for AKI to develop. ³⁴

Mehta et al. retrospectively analyzed data on more than 13,000 patients and found that, as Hct levels decreased in both men and women, AKI increased. ³⁵ Higher mortality was found in men at higher nadir Hct levels than women. ³⁵ The implication is that different genders may be able to tolerate different nadir Hct levels during CPB. ³⁵ It is, however, difficult to make this association without prospective, randomized trials to delineate what nadir Hct may injure patients.

The patient becomes hemodiluted in cardiac surgery unless measures are taken to limit the volume of fluid the patient receives. ³⁶ The perfusionist can limit the degree of hemodilution during CPB by using retrograde autologous priming (RAP). ³⁴ RAP is a technique in which the patient’s own blood is used to displace the crystalloid prime volume in the CPB circuit. RAP has been shown to be a valuable technique to limit the amount of fluid given during CPB and may cause patients to receive less homologous blood transfusions. ³⁷ Other techniques of CPB circuit modification may decrease prime volume and further reduce hemodilition. Among them are vacuum-assisted venous drainage, mini-circuits and decreased circuit surface area. ³⁷ The degree of hemodilution and nadir hematocrit on bypass are independent risk factors for postoperative AKI.^32,33

Mini-circuits

Mini-circuits for CPB are closed systems without a venous reservoir. ³⁸ These circuits dramatically decrease the foreign surface area the blood is exposed to during CPB. ³⁸ The use of mini-circuits has been shown to decrease the incidence of AKI following CABG surgery. ³⁸ Less hemodilution and concomitantly fewer RBC transfusions may have contributed to the mechanism that lowered the incidence of AKI. ³⁸ A single-center, retrospective study of more than 4600 patients looked specifically at the role mini-circuits might play in renal outcomes. ³⁹ No difference was found between mini and conventional CPB circuits in preventing AKI. ³⁹ Asteriou et al. found, in high-risk isolated CABG surgery, that mini-circuits decreased adverse events, including renal failure. ⁴⁰ The limited application of mini-circuits to use in isolated CABG surgery makes it difficult to assess their role in decreasing AKI.

Ultrafiltration

The use of conventional ultrafiltration (CUF) during CPB to remove excessive fluid is believed by some to have an effect on kidney function. If the hemoconcentrator removes fluid the kidneys are supposed to remove, then it is believed that the kidneys may not function as efficiently. A prospective, observational study looked at the effects of CUF during CPB on kidney function postoperatively and found no difference between the two groups. Non-CUF patients tended to receive more fluid, but only produced slightly more urine on CPB. ⁴¹ During the 24-hour period following surgery, the urine output of the non-CUF patients increased; however, they still maintained a higher fluid balance. ⁴¹ Perez-Vela et al. also found no increase in AKI in patients randomized to receive CUF during their surgery. ⁴² The limited evidence suggests that CUF does not protect against AKI nor does it cause injury.

Pump flow

The role of the perfusionist during CPB is to ensure that the physiologic and metabolic needs of the patients are being met. ¹ The CPB circuit provides the cardiac output during surgery and is an area where perfusionist management can cause harm. ⁴³ The typical pump flow for a patient undergoing normothermic bypass is from 2.2 to 2.5 L/min/m², but many factors play into whether this pump flow is achieved: ⁴⁴ the type of surgery being performed, collateral blood flow returning to the heart as well as surgeon preference are all factors which can cause the perfusionist to pump less than the targeted flow. ¹

The oxygen delivery (DO₂) to the patient can be calculated with information that is readily available during the procedure. Ranucci et al. prospectively studied the relationship between DO₂, systemic pump flow and Hct in a cohort of over 1400 patients. ³⁶ As a patient’s DO₂ decreased below the 272 ml/min/m² threshold, the tendency for AKI to develop increased. ³⁶ The perfusionist can augment oxygen delivery by increasing the pump flow, even in the presence of anemia. The DO₂ should be considered another option to be examined other than hematocrit alone. ³⁶ De Somer et al. retrospectively studied the idea that DO₂ values influence AKI. ⁴⁵ Increased carbon dioxide production (VCO₂) and DO₂ were studied and a ratio comparing their values was calculated (DO₂/VCO₂). ⁴⁵ As nadir DO₂ values decreased below 262 ml/min/m² and DO₂/VCO₂ ratios fell below 5.3 the odds of AKI developing increased. ⁴⁵ Additionally, when the hematocrit fell below 23.5%, the incidence of AKI increased. The predictive value of the DO₂ and DO₂/VCO₂ models are more than 90% effective, with the nadir DO₂ being the most predictive in this setting. Although this study is retrospective, maintaining higher DO₂ values may very well mitigate the development of AKI.

Mean arterial pressure

Kidneys are autoregulated over a range of mean arterial blood pressures (MAP). It has been asserted that autoregulation is lost below a MAP of less than 80 mmHg. ¹⁸ During CPB, the patient is exposed to extremes in terms of MAP. ⁴⁶ Whether or not a low MAP at a normal cardiac index on CPB can cause AKI is unknown. Similarly, what the ideal MAP is during CPB to ensure adequate renal protection is much debated. One study looked at different MAP ranges during bypass and found no correlation to AKI. ⁴⁷ Kanji et al., on the other hand, in a prospective observational study, found that a MAP variance of more than 26 mmHg between preoperative and CPB values is an independent risk factor for AKI development. ⁴⁶ This so called delta MAP observation was noted in the presence of low systemic pump flows on CPB, further demonstrating the need for adequate systemic pump flow during CPB to protect the kidneys. ⁴⁶ In a retrospective study with a small cohort, Fischer et al. found the development of AKI was associated with a MAP of less than 60 mmHg, low systemic pump flow and long CPB time. ⁴⁸ Haase et al. reported that severe hypotension (MAP less than 60 mmHg) occurring with a hemoglobin less than 8 g/dl predicted AKI. ⁴⁹ While no consensus exists regarding the ideal MAP on CPB, a target MAP of at least 60 mmHg may reduce the risk of AKI. ¹

Temperature management

The link between temperature and CPB has been studied with a particular focus on neurologic events. Boodwani et al. were the first to show the relationship between extreme hypothermia and hyperthermia causing problems while on CPB. ⁵⁰ The study demonstrated that increased AKI was with patients warmed to 37°C or cooled below 32°C. ⁵⁰ Newland et al. retrospectively found that perfusate exiting the oxygenator at 37.5°C correlated to an increased incidence of AKI when compared to perfusate that did not exceed 37.0°C. ⁵¹ Additionally, the length of time at higher temperatures during rewarming increased the risk of AKI postoperatively. ⁵¹

Mannitol, furosemide, and sodium bicarbonate

Urine output is one of the simplest ways to measure kidney function in the operating room during surgery. ²⁵ If a patient has an abrupt decrease in urine production, it can be indicative of problems that can be addressed. In many institutions, the perfusionist administers one of two diuretics to promote urine output. Mannitol is an osmotic diuretic, which increases renal blood flow. ⁵² Typically, mannitol is given in doses from 10 to 50 grams. While it does increase urine output, mannitol has not been shown to prevent AKI following surgery. ⁵³ A prospective, randomized, double-blinded, clinical trial of patients with normal kidney function found no evidence that mannitol protected against AKI. ⁵² Carcona et al. found that mannitol use increased the excretion of β₂-microglobulin following open-heart surgery, which indicated potential damage to the kidney tubules. ⁵⁴ If mannitol was administered along with dopamine during surgery, kidney damage was increased. ⁵⁴ The routine use of mannitol for cardiac surgery cannot be recommended for preventing AKI because no evidence has demonstrated decreased damage. ¹⁴ Mannitol use has also been examined prospectively in patients with existing kidney dysfunction. Urine output increased in patients who received mannitol, but did not provide any renal protection against further damage. ⁵³

Furosemide is a diuretic that exerts its effect on the loop of Henle by preventing sodium reabsorption, but does not increase renal blood flow. Dosage ranges from 10 mg to 100 mg. ⁵⁵ Furosemide also increases urine output, but, in higher doses, can cause renal dysfunction. ⁵⁵ Parolari et al. found that administration of furosemide along with low urine output during CPB were independent risk factors for developing AKI postoperatively. The use of any drug to augment urine output during CPB is to treat a symptom not a cause. ³ Caution should be exercised whenever mannitol or furosemide is administered because their relationship to AKI has not been established.

The use of sodium bicarbonate to alkalinize the urine of patients at risk for AKI is routine following cardiac catheterization. It is believed to help prevent damage to the tubules by slowing free radical oxygen species generation. The use of sodium bicarbonate is believed to be the best available drug to decrease the incidence of AKI. ⁵⁶ A pilot study showed that sodium bicarbonate may mitigate AKI following CPB. ⁵⁷ A prospective, randomized, double-blinded, multicenter trial involving 350 patients at high risk of developing AKI showed no decrease in AKI development between groups, but an increased mortality in the sodium bicarbonate group. ⁵⁸ This caused early termination of the study as it was deemed that the use of sodium bicarbonate was causing harm. ⁵⁸ The neutrophil gelatinase-associated lipocalin (NGAL) biomarker was increased in the sodium bicarbonate group and was predictive between 6- 24 hours at diagnosing AKI. ⁵⁸ Alkalinizing a patient’s urine with sodium bicarbonate as a routine measure in order to prevent AKI cannot be recommended at this time in light of this study.

Hemolysis

The use of any extracorporeal circuit is associated with some degree of hemolysis. ⁵⁹ Subjecting red blood cells to shear stress in any part of the CPB circuit can cause hemolysis. ³⁹ This occurs when haptoglobin can no longer bind free hemoglobin in the circulation, leading to the formation of plasma free hemoglobin. This can be seen in the urine when it turns to a pink color. There has been a prevalent belief that pink-colored urine is a normal by-product of CPB and may not lead to AKI. ⁶⁰ The association between CPB-induced hemolysis was recently examined in aortic surgery. Vermeuelen Windsant et al. found an association between the formation of plasma free hemoglobin from CPB and injury to the proximal kidney tubules. ⁵⁹

Cardiotomy suction is the primary source of plasma free hemoglobin generated by the extracorporeal circuit. ⁶¹ The air-blood interface caused by excessive cardiotomy suction increases red cell fragility. CPB time and degree of hemolysis are also directly related. Vanek et al. observed increased hemolysis during heart valve surgery which was associated with length of time on CPB. ⁶² The type of systemic pump head has little effect on the generation of hemolysis. A meta-analysis done by Sachzowzki et al. found no significant difference between roller and centrifugal pumps in any variable measured, including plasma free hemoglobin. ⁶³

Mechanisms to reduce CPB hemolysis and AKI are not clear due to the lack of direct research. The choice of centrifugal or roller pump does not make a difference. The normal treatment for hemolysis is clear fluid and alkalinizing the urine. A recent study questions the practice of urine alkalization and the potential for kidney damage. ⁵⁸ Since there is little that can be done in the treatment for hemolysis, it should be the operative team’s goal to reduce the amount of intraoperative hemolysis that occurs. Areas that will decrease hemolysis include the avoidance of over-occlusion of roller pumps and the avoidance of unnecessary cardiotomy suction.

Pulsatile flow

The kidneys function physiologically with pulsatile flow, but, in most institutions, CPB is carried out in a non-pulsatile manner. A recent meta-analysis by Sievert et al. found that, of the studies examined, there was increased creatinine clearance and decreased lactate levels with pulsatile CPB. ⁶⁴ A prospective, randomized trial by Mohammadezah et al. found that pulsatile CPB was associated with better kidney protection. ⁶⁵ Adademir et al. prospectively examined the effect of pulsatile CPB and found low levels of interleukin-18 (IL-18) and NGAL in the pulsatile group, which indicates kidney protection. ⁶⁶ Pulsatile perfusion is not without controversy. The means by which to generate a pulse with the CPB equipment may result in increased hemolysis.

Conclusion

In recent years, researchers have become increasingly focused on the relationship between CPB and AKI. The development of evidence-based practice guidelines in the perfusion community has been an area that has been slow to be adopted. ¹² Bartels et al. noted that the entire field of CPB is practiced more on experience than evidence. ⁶⁷ Studies have been slow to define the independent effect of particular techniques employed during CPB. ¹ The growth of medical databases collecting procedurally specific data may delineate causative relationships to define the relationship between CPB and AKI. Additional research through randomized clinical trials will also allow for evidence-based practice to grow in CPB ¹

AKI following cardiac surgery requiring CPB may be reduced by the application of a coordinated prevention plan by multidisciplinary teams preoperatively, intraoperatively and postoperatively. Techniques that avoid a nadir Hct of less than 24% and RBC transfusion should be applied throughout a patient’s hospital stay. In particular, elective operations on anemic patients should be avoided until the condition is resolved. Any technique that minimizes hemodilution during surgery should also decrease the need for RBC transfusion, including the avoidance of unnecessary preoperative crystalloid fluid administration, the reduction of extracorporeal circuit prime volume, the application of RAP and the assiduous and appropriate usage of a cell saver and pericardial suction. The perfusionist should select equipment designed to cause less shear stress, thereby, producing less plasma free hemoglobin. The literature indicates that maximizing pump flow, as well as maintaining a MAP of greater than 60 mmHg during the procedure, may reduce the incidence of postoperative AKI. Further research needs to be done in new biomarkers that can earlier identify those at risk of developing AKI.

The limitations of this review stem largely from its reliance upon an all-adult patient population while examining perfusionist interventions related to AKI. Any analyses or conclusions cannot be applied to children. Some studies that met inclusion criteria were retrospective or lacked a control group. Several studies had a very low number of patients enrolled and researchers may have not been blinded in their analyses. This review is wide in scope and, while every effort to ensure relevant studies were included, some may have been excluded based upon how they were cataloged in databases.

In summary, AKI following cardiac surgery is a complication that causes significant morbidity and mortality. It increases the resources used to care for these patients by doubling their stay in the ICU. The long-term implications of patients ending up on hemodialysis or renal replacement therapy should guide practitioners to new and better ways to care for these patients. Knowledge is still growing in this area and prospective studies need to be done to establish a direct relationship between techniques and outcomes.