Perfusion - Pressure - Creatinine Trial

1. Introduction: Renal function and cardiac surgery Impaired kidney function is a well known complication to cardiac surgery and is observed in up to 30% of the patients, depending on the definition of acute kidney injury (AKI). The precise pathophysiological mechanisms are unknown, it is, however, quite obvious that AKI is not initiated by a single harmful factor but seems to be the result of different combined factors. Known risk factors for the development of postoperative AKI are cardiac incompensation, poor ejection fraction, gender, chronic obstructive lung disease, insulin dependent diabetes mellitus (IDDM), previous heart surgery, acute surgery, complexity of surgery, preoperative creatinine levels and prolonged duration of extracorporeal perfusion. These associations are not necessarily causal. It is not surprising that a subacute operation for a severe aortic stenosis in a patient with preoperatively poor circulation and a marginally increased S-creatinine, results in impaired renal function postoperatively. However, to conclude that cardiac surgery is the reason for AKI is not justified. Several peroperative factors are likely to contribute in the development of AKI. Renal perfusion and oxygen delivery might be reduced (hypotension, lack of pulsatile flow during extracorporeal circulation (ECC), vasoactive pharmacological agents, anaesthesia), risk for emboli, general inflammatory response, direct nephrotoxins (free haemoglobin, free radicals, gentamycin etc.) and haemodilution. Current knowledge is limited and the general insight into regional perfusion during extracorporeal perfusion (heart-lung-machine) is very sparse. There do, however, exist guidelines regarding flow and blood pressure during ECC, but these are based upon empiric data, which are controversial and vary from centre to centre. The primary focus has been on regional cerebral perfusion and the possibilities to reduce the frequency of cerebral complications after cardiac surgery. This interest can easily be explained by the obvious catastrophic consequences for the patient due to coma, hemiparesis or serious cognitive dysfunction. The consequences of AKI are less evident, however there is a direct and strong association between a moderately increased postoperative creatinine level, reduced glomerular filtration rate (GFR) and 30 day mortality as long term survival. A recent analysis of the kidney function of lung transplanted patients in our department after use of ECC showed that GFR was dramatically reduced by 40% 14 days after surgery. This impairment has furthermore been shown to be irreversible. Cyclosporine A is known to be nephrotoxic in the long term, but this can hardly explain the observed GFR reduction after 14 days, which must result from the combined influence of surgical/anaesthesiological trauma, cyclosporine, and side effects of antibiotic, antiviral, antifungal and immunosuppressive treatment. Data from our internal prospective registry shows an increasing incidence of patients with renal complications. In 2008 temporary dialysis was necessary in 7% of the patients (76/1130) and in 19% of the patients a postoperative increase of S-creatinine > 200 μmol/l was observed, mirroring a reduction in kidney function of 60-70%. We assume that the kidney function is completely normalized over time, but this is actually not known because these patients are not routinely controlled on a cardiac surgery outpatient basis. Currently, we are gathering 1-year outcome data on patients, who have undergone postoperative dialysis in 2008. One of the reasons for the increasing frequency of AKI might be increasing patient age, and thereby, an increase in comorbidity. At the same time the administration of ECC has changed. Ten years ago patients were routinely cooled during ECC. This is not the case anymore and results in a reduced degree of peripheral vessel constriction and consequently reduced blood pressure. The question regarding which blood pressure and which blood flow is optimal during ECC remains controversial, but generally a much lower blood pressure is now accepted, as long as the calculated blood flow is maintained and cerebral saturation values are sufficient (not measured routinely). The literature on the subject is sparse but two randomized trials exists. One had a primary focus on kidney autoregulation at different perfusion pressures, the other recent trial could not demonstrate significant differences in kidney function measured by s-creatinine at 3 different perfusion pressures. However, S-creatinine is a very crude method for measuring kidney function. In addition, the trial included patients with low risk of postoperative AKI. Postoperative increase in S-creatinine typically occurs on day 2 or 3 after surgery, and S-creatinine does not increase before GFR is reduced below 50%. This implies that the possible initial 50% reduction in kidney function cannot be detected by routine biochemistry. It is fair to assume that kidney injury occurs during surgery or within the first postoperative 24 hours. From a biochemical point of view there is an apparent lack of robust and reliable early markers of tubular kidney damage, which would allow a relevant early clinical intervention, and possible prevention of AKI aggravation. Creatinine kinases MB and troponins are very reliable markers for cardiac damage and can reveal acute myocardial infarction within hours after the occlusion of coronary vessels. A corresponding early diagnosis of AKI does not exist. At the same time it seems likely that specific enzymatic or cell membrane components could be detectable in the urine a few hours after an AKI. There has been some focus on possible candidates for early detection of AKI, but none have shown to be sufficiently robust for clinical use, yet. During the last 20 years there have been a few trials with focus on prevention and prophylactic intervention to avoid AKI with anti-inflammatory drugs, natriuresis inducing agents (ANP) or renal blood flow increasing pharmacological intervention. However, none of these interventions have had sufficiently convincing effects for further clinical testing. A recent Japanese study seems to indicate that prophylactic infusion of ANP can reduce postoperative kidney impairment, but these results need confirmation. Our department is currently performing a phase 2 trial in cooperation with Action Pharma (a Danish biotech company) in which we are testing AP214, a new small anti-inflammatory protein, with potential kidney protective characteristics (clinicaltrials.gov: 00903604). At the same time we are investigating if AP214 can reduce the nephrotoxic effect of cyclosporine in a pig kidney transplant model. Summary / Conclusion:

• An increasing incidence of renal complications is seen after cardiac surgery.
• The aetiology is unknown.
• It is unclear if the reduction in renal function is reversible.
• Renal complications are associated with increased short- and long term mortality and morbidity after cardiac surgery.
• Renal regional perfusion during extracorporeal circulation is poorly understood, and the impact of blood pressure and perfusion on postoperative kidney function in patients with increased risk for AKI after cardiac surgery is unknown.
• There are several potential biochemical markers for the early detection of acute tubular kidney damage, however there is a substantial lack of clinical data.
• Several interventions might be kidney protective, however there are no robust clinical data.

1. Hypothesis for the PPC trial The glomerular filtration rate (GFR) is better preserved after cardiac surgery with ECC with an intended increased periprocedural arterial blood pressure compared to current practice.
2. Methods
3. Design: Phase 1: Randomized proof of concept trial, n = 12. Simple 1:1 randomisation. Phase 2: If the proof of concept trial turns out positive and logistically feasible (which we have shown, see later) the trial will be extended to include 2 x 50 patients. There are no similar trials or data in the literature on which to base an estimate of the necessary trial size. N = 100 is a qualified assumption based on what is clinically possible to include within a 1-1½ year period. As earlier shown in the lung transplant population GFR was reduced with 40%. A clinically relevant improvement of an average reduction in GFR would be a reduction of 20%. However, GFR data are quite heterogeneous and some patients have perfectly normal GFR's throughout surgery and the postoperative course, while others develop pronounced AKI. Phase 3: Test of different preventive and prophylactic interventions (AP214, proANP, reduced haemodilution, pulsatile flow during ECC). Trial size and outcome measures will be dependent on results from phase 2. There are no earlier data on continuous saturation measurement in renal venous blood or sequential mass spectrometric analysis of urine in a cardiac surgery population. Current results will therefore decide on future primary outcome measures.
4. Inclusion- and exclusion criteria: Patients with increased risk of postoperative AKI, who give written consent after appropriate information, are included. Inclusion criteria are:

• Age > 70
• Combined procedures (heart valve-, bypass surgery and aortic aneurysm) Exclusion criteria are:
• Age < 70
• S-creatinine > 200 µmol/L
• Acute operation
• Endocarditis

1. Intervention: Control group: Operation procedure and ECC run are performed according to current guidelines at the Dept. of Thoracic and Cardiovascular Surgery / Rigshospitalet. Intervention group: During ECC run a MAP > 60 mmHg is maintained. In both groups GFR measurements are performed using the Cr-EDTA clearance method. GFR is measured the day before surgery, day 4-6 after surgery and finally after 4 months (+/- 14 days). A catheter is inserted in the renal vein through the femoral vein. The position is controlled by fluoroscopy. Repeated saturation measurements, by blood gas analysis, will be performed during and in the first 24 hours after surgery. Blood samples are taken from the same access. The catheter is removed before discharge from the intensive care unit.
2. Outcome measures: Primary outcome measure: Cr-EDTA clearance, the day before surgery, at day 4 - 6 and 4 months (+/- 14 days) after surgery Secondary outcome measures:

• Oxygen consumption on-line in renal venous blood
• Renal renin and proANP excretion
• U-NGAL, U-Apolipoprotein M, U-IL-18, U-KIM-1 and S-cystatin C
• Urine proteomics (mass spectrometry) Tertiary outcome measures: Standard clinical data collection (haemodynamics, diuresis, routine biochemistry) and clinical outcome in terms of time in intensive care, time to discharge, incidence rates of adverse- and serious adverse events.

1. Economy:

This investigation is a non-profit trial. None of the project members have economic interests in the trial. The involved clinical departments cover direct and indirect expenses regarding the prolonged admission, GFR-measurements and additional biochemical analyses estimated to be 15.800 DKK per patient.

The Danish Heart Association covers the salary for Ph.D. student Kristian Kandler.

5: Perspectives

This project combines expert knowledge from different medical fields on a highly scientific and clinical level (advanced cardiac surgery in high risk patients, advanced postoperative intensive care, innovative monitoring (invasive, biochemical, clinical physiology). We expect to provide original, innovative and scientifically relevant results that will contribute to a better understanding of the renal pathophysiology and autoregulation during and after cardiac surgery. As a clinical randomised trial this study may show new possibilities to reduce the incidence of postoperative renal complications in a group of patients with significant mortality and morbidity.