In this cross-sectional, single centre, retrospective study conducted in the Department of Anaesthesia, Fortis Hospital, Delhi, India (November 2016 to May 2017). Adult patients between 18 to 65 years of age who had under gone an elective surgery for more than two hours in the preceding six months by using low-flow sevoflurane anaesthesia (fresh gas flow less than one liter per minute) were included. All patients had normal preoperative renal functions. Patients with past history of dialysis, renal transplantation or surgery for any renal disorder, those who had undergone an emergency surgery, evidence of use of any contrast media during the surgery, known hypersensitivity to sevoflurane or any other condition that precluded use of sevoflurane were not included. Similarly, pregnant and/or lactating women were also excluded from the study. Medical records of all patients admitted for surgery were obtained from the Medical Records Department of the hospital and evaluated for the eligibility for inclusion in the study. The administration of general anaesthesia was individualised based on the patient’s response. Change in serum creatinine and BUN values from baseline to postsurgery were evaluated for all enrolled patients. Number of patients experiencing postoperative glucosuria and proteinuria were also measured. In addition, the fresh gas flow rate, carrier gases used (O₂ plus air), end tidal concentration of sevoflurane, type of CO₂ absorber used, duration of surgery, type of breathing circuit used (open/closed), inducing agents (other than sevoflurane) used and other anaesthetics, muscle relaxants used were recorded. Approval from Institutional Ethics Committee was obtained before initiation of the study.

Discrete data are summarised using numbers and percentages whereas continuous data are summarised using mean and standard deviation. Statistical analysis was performed using SAS^® version 9.2 (SAS Institute Inc., USA). Statistical difference in the preoperative values of serum creatinine and BUN was evaluated using paired t-test at 5% level of significance.

Medical records of 169 patients were screened and 100 patients were included in the analysis. Other 69 (40.83%) patients did not meet the enrollement criteria, hence were considered as screen failures and not included in the study.

The study included 44 (44%) males and 56 (56%) females. Overall mean age of the population was 41.3±12.59 years. Mean weight and height was 83.9±27.81 kg and 162.7±10.48 cm, respectively.

History of hypertension, diabetes, hypothyroidism, musculoskeletal or connective tissue disorder and asthma was present in 32 (32%), 19 (19%), 16 (16%), 4 (4%) and 3 (3%) patients respectively. History of coronary artery disease, goitre, dyslipidemia, chronic obstructive pulmonary disease, coronary artery bypass and paraplegia after cervical spine surgery was present in one patient each whereas sleep apnoea syndrome, depression and seizures were present in two patients each. Only 2 (2%) patients had significant family medical history, both of hypertension.

A total of 22 (22%) patients were receiving antihypertensive medicines which included amlodipine 6 (6%), telmisartan 6 (6%), metoprolol 3 (3%), olmesartan 2 (2%), valsartan 2 (2%), atenolol 2 (2%) and bisoprolol 1 (1%). A total of 28 (28%) patients were on endocrine medicines which included metformin 8 (8%), levothyroxine sodium 8 (8%), glimepiride 5 (5%), human soluble insulin 3 (3%), vildagliptin 3 (3%), and sitagliptin 1 (1%). Five patients were taking medicines for cardiac disorders which included clopidogrel and acetylsalicylic acid (n=2 each) and atorvastatin (n=1).

The mean duration of surgery was 163.3±51.78 minutes ranging from 122 to 365 minutes. All patients received sevoflurane with dial setting of two or more and fresh gas flow less than one liter per minute in a closed breathing circuit. The mean fresh gas flow rate was 0.9±0.06 liter/minute whereas mean end tidal concentration was 1.8±0.11. A total of 91 (91%) patients received midazolam whereas all patients were administered fentanyl and propofol as inducing agents. Atacurium was used in 99 (99.00%) patients. Sofnolime was used as CO₂ absorber in all patients.

The mean temperature before surgery was 98.34±0.31°F whereas after surgery it was 98.37±0.16°F. Values of mean pulse rate, respiratory rate, systolic blood pressure and diastolic blood pressure before surgery and after surgery are shown in [Table/Fig-1]. There was no clinically significant difference in the temperature, pulse rate, respiratory rate or blood pressure after surgery.

There was no clinically significant change in serum creatinine values from first visit to second visit. Also, there was no statistical significance in serum creatinine from preoperative visit to postoperative visit (p=0.2415) [Table/Fig-2].

The mean change in BUN from first visit (preoperative) to second visit (postoperative) observed was 0.5 mg/dL. Similarly, no clinically significant change in BUN values from preoperative visit to postoperative visit was observed (p=0.2624) [Table/Fig-2].

There were no reports of change in the dose or anaesthetic drug due to intolerance or adverse event due to sevoflurane. No clinically significant changes in vital signs, physical findings, or other observations related to the safety were recorded in patients in this study. No adverse events were reported in the study. Overall, no clinically relevant significant changes in serum creatinine, BUN, proteinuria or glucosuria were observed post-surgery when compared to pre-surgery laboratory parameters.

Although, sevoflurane is a well accepted inhalational anaesthestic agent among the Indian anaesthesiologists, there are some concerns about renal toxicity due to fluoride ions and compound A generated from sevoflurane [14]. In this retrospective study, we evaluated the effect of low-flow sevoflurane on renal functions of patients exposed to wide variety of surgeries lasting for more than two hours. Sevoflurane was administered with dial setting of two or more and fresh gas flow less than 1 L/minute in a closed breathing circuit. Muscle relaxant and inducing agent were used to obtain adequate anaesthesia. The dial concentration of sevoflurane was monitored by vaporizer calibrated specifically for its delivery. The delivered sevoflurane was monitored with anaesthesia gas monitor (Philips IntelliVue GS-M1019A). The administration of general anaesthesia was individualised based on the patient’s response.

Overall, low-flow sevoflurane was well tolerated by patients in this study. No significant or new safety signals were detected. A study had reported no change in plasma creatinine level, however with significant reduction in plasma BUN with low-flow desflurane, sevoflurane and propofol [15]. In another study, the percentage of patients with creatinine increase was not significantly different with sevoflurane compared to isoflurane or propofol in patients with elective coronary artery surgery [16]. In the present study, we did not find clinically significant change in serum creatinine or BUN after low-flow sevoflurane.

The present study observations are different from that of a study conducted by Ebert TJ and Arain SR [15]. In their study, significant increase in urine glucose and protein was observed in all study groups i.e., low-flow desflurane, sevoflurane and propofol and changes were similar in all three groups without the association of anaesthetic concerntation with abnormal renal findings. As changes were observed in all the study groups, the authors suggested role of nonanaesthetic factors such as surgical or postoperative factors for the same. Higher changes after surgeries in central regions indicate role of glomerular capillary haemodynamic effects or surgical stress in postoperative renal changes [15]. In another study, Higuchi H et al., [17] also showed mild and transient proteinuria associated with low-flow sevoflurane, without change in BUN and creatinine in patients without preexisting kidney disease. Renal tubular toxicity and disturbed proximal tubular reabsorption are the described reasons for proteinuria and glycosuria [17]. None of the patients in the present study developed glucosuria and proteinuria after administration of low-flow sevoflurane.

A study comparing renal function after administration of low-flow sevoflurane and isoflurane using biomarkers of tubular damage showed similar effects with both the agents [18]. The results of the study showed that moderate duration low-flow sevoflurane maybe as safe as using low-flow isoflurane. Expsoure to compound A during surgeries of short period (i.e., two to three hours) with sevoflurane anaesthesia do not affect the kidney. Similarly, in surgeries lasting for 10 or more hours, prolonged low-flow sevoflurane has shown to have similar effect on renal parameters as that of high-flow sevoflurane and low-flow isoflurane anaesthesia [19]. In the present study results also showed renal safety of sevoflurane exposure for surgeries lasting for over two hours (mean duration of surgery 163.3 minutes).

Low-flow sevoflurane has also been used in patients with preexisting stable renal insufficiency. In such patients, renal functions are similar with low-flow isoflurane [20]. We enrolled patients with normal renal functions, but effect of low-flow sevoflurane in Indian patients with stable preexisting renal insufficiency can be evaluated in future studies.

Because of interactions of carbon dioxide absorbent and sevoflurane resulting in production of compound A and differences in carbon dioxide absorbents, its choice is also a topic of debate. A study comparing four carbon dioxide absorbents did not show significant differences in renal safety profile [21]. In the present study, sofnolime was used in all patients. A Russian study showed no significant renal injury in healthy patients with compound A level 275 ppm/hour during minimal flow anaesthesia [22]. In the present study, patients had normal renal functions before undergoing surgery. These observations reaffirm the use of low-flow sevoflurane in patients undergoing surgery.

Overall, observations from present study suggest no adverse effect on renal parameters with sevoflurane in Indian patients’ undergoing surgery for more than two hours. In humans, the renal cysteine conjugate β-lyase pathway responsible for metabolism of compound A is much less active compared to that in rats. Postulation that if this mechanism is responsible for renal damage, humans may have a lesser potential for risk, is supported by the observations in the present study [6]. We did not evaluate the concentration of compound A, but no significant change in renal functions in patients from present study supports this theory.

In the present study has some limitations. We did not compare the renal function changes with other inhalational anaesthetic agents. The present study was evaluation parameters were restricted to traditional renal function tests without any biomarkers. Significantly reduced compound A formation is possible with some carbon dioxide absorbants. We did not evaluate or compare the levels of compound A with different carbon dioxide absorbants; hence, it is difficult to conclude whether lack of renal toxicity seen in the present study is result of carbon dioxide absorbant, low-flow sevoflurane itself or both. Use of only one carbon dioxide absorbent in all patients restricted us from comparing effects with other agents. Single centre, retrospective study design and short term follow up limits generalization of findings to all patients. Neverthless, the observations affirm the renal safety profile of sevoflurane and suggests no new safety signal when used in patients with normal renal function tests. Larger, prospective studies are required to confirm these findings.

In the present study, administration of low-flow sevoflurane was not associated with any major concerns related to the renal safety. The results of present study suggest that low-flow sevoflurane anaesthesia in patients with elective surgeries lasting for more than two hours is found to be renal safe and can begiven for maintainace of anaesthesia in patients with normal renal functions.

The research funding support was provided by Abbott Healthcare Pvt Ltd. The authors have declared and confirmed that there is no conflict of interest with respect to this authored publication.