The Effect of Isoflurane or Sevoflurane on Kidney Function

Study Protocol Post-Operative Creatinine Changes In Patients With Pre-Existing Renal Impairment After Low-Flow Isoflurane Or Sevoflurane: A Randomised Clinical Trial

Introduction

Sevoflurane is a volatile anaesthetic agent available for clinical use in most countries including Australia, the United States, the United Kingdom, and Canada. Since the first clinical use of sevoflurane in Japan during the 1980s, millions of anaesthetics have been given with sevoflurane. Sevoflurane has a number of advantages over other agents for general anaesthesia, particularly isoflurane. The benefits to patients and the environment include:

• A lower blood and tissue solubility allows patients to wake up faster after anaesthesia 1, 2.
• The lower blood solubility allows faster titration during surgery.
• Heart rate is more stable during increases in inspired concentration.
• Airway resistance is lower than with other volatile agents.
• A pleasant smell allows gas induction
• There is less nausea and vomiting than with isoflurane
• It is a fluorocarbon rather than a chlorofluorocarbon and does not worsen the atmospheric ozone hole.

The most important concern about sevoflurane is its effect on renal function due to breakdown products. When sevoflurane is exposed to the carbon dioxide absorbents soda lime and baralyme a vinyl breakdown product called Compound-A is formed. High dose exposure to Compound-A alters kidney histology in rats. Some, but not all, studies have found that Compound-A produces increases urinary secretion of renal tubule enzymes in humans. The clinical importance of enzyme secretion is unclear.

The most widely used indicator of renal function that has established clinical importance is plasma creatinine concentration. No study of sevoflurane has shown that sevoflurane produces greater increases in plasma creatinine concentration than other anaesthetic drugs. The Austin Health Department of Anaesthesia has recently completed a study in cardiac anaesthesia that included sevoflurane. We found that, comparing isoflurane, sevoflurane, and propofol, there was no difference in creatinine change after surgery between the three groups. This study will be published in Anesthesiology, the manuscript is attached.

The fresh gas flow into an anaesthetic machine is the amount of oxygen, or oxygen and air, or oxygen and nitrous oxide that flows into the anaesthetic circuit per minute. Moderate fresh gas flows are around 3 l/min. Low fresh gas flows are around 1 l/min. Low fresh gas flows have benefits for patients, the environment, and the hospital:

• Patients lose less heat
• Patients lose less water
• There is less environmental contamination with chlorofluorocarbons and nitrous oxide (a green house gas).
• Lower anaesthetic costs.

At low fresh gas flows, patients are exposed to higher concentrations of Compound A.

A recent multicentre study looked at creatinine changes after major surgery. This study found no difference between isoflurane and sevoflurane with low fresh gas flows of 1 l/min. Patients with pre-existing renal impairment were excluded.

Pre-operative renal impairment increases the risk of clinically important deterioration of renal function after surgery. Studies of changes in creatinine after sevoflurane have included only a small number of patients with pre-existing renal impairment 10. None of these small studies has shown that sevoflurane aggravates pre-existing renal impairment more than other anaesthetic drugs. This includes a subgroup analysis in our Cardiac Anaesthesia study. No study, however, has examined creatinine changes in those with pre-existing renal impairment and with low fresh gas flow. This would expose those with increased risk of further renal impairment to greater doses of Compound-A. From other research and our own, we believe that sevoflurane would not affect renal function more than isoflurane in patients with preexisting renal impairment receiving low flow anesthesia.

Research Question What are the post operative changes in serum creatinine after anaesthesia with sevoflurane or isoflurane using fresh gas flows of 1 l/min in patients with pre-operative renal impairment?

Methods The study methods draw heavily on our previous study 7, copy attached. The study design conforms to the Consolidated Standard of Reporting Trials (CONSORT) 11.

Study type Randomized, clinical trial.

Study Site Austin and Repatriation Medical Centre

Eligible patients

Adult patients for elective surgery who will:

1. need to stay at least one night
2. need general anaesthesia
3. have a pre-operative serum creatinine concentration greater than 130 umol/l.

Exclusions

1. The patient refuses.
2. The patient is less than 18 years of age.
3. The patient is on pre-operative dialysis
4. The anaesthetist plans to use only regional or total intravenous anaesthesia.
5. Specific types of surgery: craniotomy, carotid, and cardiac.

Recruiting One of the investigators will recruit patients as soon as possible after the patients are booked for surgery. We expect to enroll at least 3 patients per week.

Treatment arms

1. Isoflurane OR 2. Sevoflurane Anaesthesia maintenance will be with fresh gas flows of 1 l/min and inspired oxygen of 30% to 100% in nitrogen (oxygen/air) or nitrous oxide. Individual anaesthetists will determine all other aspects of the peri-operative care including fluid therapy.

Random allocation Patients will be randomly allocated to receive isoflurane or sevoflurane. Before the trial ARMC pharmacy staff will provide the allocation sequence using a table of random numbers. Allocations will be kept in opaque envelopes by the trial coordinator. Anaesthetists will be notified of the allocation to isoflurane or sevoflurane on the day of surgery.

Primary end point The primary end point will be the highest change in serum creatinine from before operation to the first three days postoperatively. Supporting data will be from the highest creatinine and urea in the first three postoperative days. Patients leaving hospital before 3 postoperative days will be included. Good clinical care for patients with renal impairment includes a creatinine measurement on admission and at least one postoperative creatinine measurement. The trial coordinator will oversee this. The data for this study will be from clinically indicated tests. Specific blood tests for this study will not be needed. Data will be collected from the ARMC clinical chemistry results. Data will be stored on computer in a secure are of the ARMC Department of Anaesthesia in line with the Therapeutic Goods Administration Guidelines for Good Clinical Research Practice in Australia.

Clinically important changes

When analysing the differences between the isoflurane and sevoflurane groups, two criteria will need to be met for the changes in creatinine to be clinically important:

1. at least one mean increase in creatinine will be greater than 45 umol/l above pre-operative concentration AND
2. the difference between the means will be greater than 20 umol/l. For example: The group 1 mean rises by 50 umol/l and the group 2 mean rises by 25 umol/l.

Power analysis Using a standard deviation of 50 umol/l from our Cardiac Anaesthesia trial data, for a power of 0.90, alpha < 0.05, to detect a 20 umol/l difference between 2 groups we would need 270 patients. We estimate we would need two years to complete this study.

Data Collection and Statistical analysis Data will be collected and analysed by investigators blinded to the drug allocation. Statistical analysis will be by parametric tests if data are normally distributed. Otherwise equivalent non-parametric tests will be used.

The statistics will be the point estimate for the difference between the isoflurane and sevoflurane increases in creatinine, the 95% confidence interval for the point estimate, and the p value from an unpaired Student's t-test.

An additional analysis will be an estimate of the proportions of patients in the isoflurane and sevoflurane groups who have a postoperative creatinine change greater than 45 umol/l. The statistics will be the point estimate for the difference in the proportions, the 95% confidence interval for the point estimate, and the p value from a Chi square test.

Supporting statistics will be from the point estimates for the difference between the peak creatinine and urea, the 95% confidence interval for the point estimates, and the p values from unpaired Student's t-tests.

Adverse events. In addition to usual clinical care, patients in this study will have added surveillance of their clinical chemistry results. Important changes will be passed on to the surgical units caring for the patient. The investigators will notify the ARMC Human Research Ethics Committee of any death or serious morbidity of patients enrolled in this study.

Significance of the project For this large trial, with a simple design, our hypothesis is that sevoflurane will not produce greater increases in creatinine than isoflurane in patients with preexisting renal impairment. For this project an unequivocal finding would be that the 95% confidence intervals for the differences between the isoflurane and sevoflurane groups did not include clinically important differences and that any differences that were found had a p value of greater than 0.05. Unequivocal findings would lead us to two conclusions: (1) that it is highly unlikely that sevoflurane has clinically important adverse renal effects, and (2) that sevoflurane is safe to use with low fresh gas flows in patients with preexisting renal impairment.

The benefits of sevoflurane and low fresh gas flows are discussed in the introduction. A finding of no adverse renal effects would provide high quality evidence supporting the safety of using sevoflurane and low fresh gas flows in patients with preexisting renal impairment.