Impact of different intravenous fluids on blood glucose levels in nondiabetic patients undergoing elective major noncardiac surgeries.

BACKGROUND: Intravenous (IV) fluids are an integral part of perioperative management. Intraoperative hyperglycemia is associated with poor clinical outcomes in patients undergoing major surgeries even in nondiabetics. AIM: This study was conducted to observe the effect of different maintenance fluid regimens on intraoperative blood glucose levels in nondiabetic patients undergoing major surgeries under general anesthesia. SETTINGS AND
DESIGN: Randomized double-blind study.
MATERIALS AND METHODS: One hundred nondiabetic patients of either sex were divided randomly into two Groups I and II of 50 each undergoing elective major surgeries of more than 90 min duration under general anesthesia. Both groups were given calculated dosage of IV fluids accordingly 4-2-1 formula while Group I was given Ringer lactate (RL) and Group II was given 0.45% dextrose normal saline and potassium chloride 20 mmol/L. Changes in vital parameters, % oxygen saturation, and urine output were monitored at regular intervals. Capillary blood glucose (CBG) was measured half-hourly until end of surgery. If CBG level was more than 150 mg%, then calculated dose of human insulin (CBG/100) was given as IV bolus dose. STATISTICAL ANALYSIS: Statistical analysis was done using SPSS 22.0 software (IBM Corporation, Armonk, New York, USA), paired t-test and Chi-square test.
RESULTS: A significant increase of CBG level and was observed during intraoperative and immediate postoperative period (P < 0.001) in Group II.
CONCLUSION: RL solution is probably the alternative choice of IV fluid for perioperative maintenance and can be used as replacement fluid in nondiabetic patients undergoing major surgeries.

RCT Entities:   Population Interventions Outcomes

INTRODUCTION

Intravenous (IV) salt solutions were first used in the 1830s for the treatment of fluid loss due to cholera[12] and IV saline was administered to surgical patients in the late 19th century. The administration of IV fluids is nearly a universal practice for the patients undergoing major surgeries, where they fulfill the requirement to maintain intravascular volume at a time when that volume may be depleted due to preoperative fasting, surgical blood loss, evaporation, urinary excretion, vasodilatation caused by anesthesia, loss of fluid into the third space, and transcapillary leak of albumin caused by inflammatory mediators.[3] The concept of fluid resuscitation aiming for supranormal hemodynamic parameters was introduced by Shoemaker et al. in the 1970s and 1980s.[4]

The aim of perioperative fluid therapy is to maintain the circulating plasma volume and thus ensure organ perfusion and oxygen delivery to the tissues.[5] This has traditionally been achieved by infusion of large volumes of fluids,[5] resulting in a positive intraoperative fluid balance, which can, however, be associated with significant morbidity postoperatively.[6] Generally, fluid therapy includes replacement of basal fluid requirement and fluid losses such as by perspiration, blood loss, and exudation through the surgical wound, as well as maintenance of physiological functions and to avoid hypotension.[7] Surgeries are considered to be the combination of multiple factors including tissue damage, fasting, blood loss, effects of medication, and temperature changes from a metabolic point of view.[8] Combinations of all these factors gives rise to stress response. The stress response to surgery is characterized by increased secretion of pituitary hormones and activation of the sympathetic nervous system.[9] Stress response to surgery includes a wide range of hormonal and metabolic changes, among which two main systemic effects the neuroendocrine and the hemato-immunological effects can be distinguished.[10]

The ultimate effect of these various endocrine changes is increased catabolic activity by increased secretion of catabolic hormones such as cortisol and glucagon. The effect of these endocrine and metabolic changes ultimately leads to increased neoglucogenesis and hyperglycemia. Hence, this stress response may be quantified by the incidence of hyperglycemia. The metabolic changes appear to be proportional to the severity of the surgical trauma with plasma cortisol and blood glucose concentration rising slightly during minor surgical procedures but significantly during major intra-abdominal operations.[11] Acute hyperglycemia has many harmful effects such as impaired endothelial nitric oxide generation with decreased vasodilatation, increased expression of endothelial and leukocyte adhesion molecules, reduced complement function, impaired neutrophil function, and increased cytokine synthesis.[12] Together these changes enhance the inflammatory response to injury and likelihood of infection. Many of these responses are shown at glucose concentrations of 180–200 mg/dl. The use of insulin to reduce hyperglycemia has been shown to decrease endothelial activation, stimulate glucose uptake, protect hepatic mitochondria, improve the circulating lipid profile, and decrease circulatory inflammatory markers.[13] Tight perioperative glucose control is associated with a reduction in renal impairment and renal failure in nondiabetic cardiac surgical patients.[14]

The type of anesthesia may influence the hyperglycemic response during surgery. General anesthesia has been shown to result in higher blood glucose concentration than local and epidural analgesia.[1516] Circulating catecholamines, cortisol and glucagon concentration are higher in patients undergoing general anesthesia.[17]

Intraoperative glycemic control has an important role in improving surgical outcomes even in nondiabetic patients. Patient outcomes including length of stay, stroke, renal insufficiency, and mortality have been shown to improve with tight glycemic control, so it has an economic impact. There were many studies on intraoperative blood glucose levels in diabetics, and studies on nondiabetic patients were conspicuously lacking especially noncardiac surgeries.

In this context, we have studied the effect of various maintenance fluids on glucose concentration in nondiabetic patients undergoing major surgeries under general anesthesia.

Aims and objectives of the study

This study was conducted to observe the effect of Ringer lactate (RL) and 0.45% normal saline in 5% dextrose with 20 mmol/L potassium on intraoperative blood glucose levels in nondiabetic patients undergoing elective major noncardiac surgery under general anesthesia.

Intraoperative blood glucose levels and incidence of perioperative hyperglycemia (capillary blood glucose [CBG] ≥150 mg%) in two groups of patients receiving different maintenance fluids was studied.

MATERIALS AND METHODS

One hundred nondiabetic patients (50 in each group) of American Society of Anesthesiologist (ASA) physical status I and II of either sex, aged between 18 and 60 years undergoing elective major abdominal (duration of surgery >90 min) surgery were enrolled for this prospective, randomized, double-blind study. The sample size was based on the power analysis calculated by the previous study so that 50 patients in each group would provide a power of >0.8 (α =0.05) to detect an increase of intraoperative and postoperative blood glucose levels. The patients were randomized into the two study groups. Method of blinding was done by covering the fluid bottles with a paper. Group I patients received RL solution as maintenance fluid and Group II patients received 0.45% sodium chloride with 5% dextrose and 20 mmol/L potassium chloride, as per calculated hourly infusion rate according to their body weight. Fluid deficit arising from overnight fasting was corrected by the maintenance fluid according to 4-2-1 formula (Holliday–Segar's formula) Fifty percent of the total deficit was corrected in the 1st h, and the remaining 50% was corrected in next 2 h. Blood loss and other plasma losses were approximately calculated from mops and suction drain bottle. These losses up to the transfusion threshold (blood loss up to 15% of body weight) were replaced volume per volume with haemaccel (3.5% gelatin-based colloidal IV solution). We targeted a urine output of at least 1 ml/kg/h intraoperatively, and urine output was measured after every 30 min. If urine output was found to be below desired target, 200 ml bolus of 3.5% gelatin-based colloidal solution was given over 10 min. We also maintained systolic blood pressure between 100 and 130 mm Hg. If at any point during surgery patients became hypotensive (systolic blood pressure <100 mm Hg), additional bolus of 200 ml colloid was administered. When the hypotension was not corrected by fluid bolus alone, 5 mg of ephedrine was administered IV.

All patients were premedicated with 0.2 mg glycopyrrolate IV, 1 mg midazolam IV, pantoprazole 8 mg IV, butorphanol 1 mg IV, and ondansetron 4 mg IV 5 min before induction of anesthesia. Following preoxygenation, induction of anesthesia was done with 2 mg/kg propofol IV and tracheal intubation was facilitated by 1 mg/kg IV succinylcholine. Anesthesia was maintained with 66% nitrous oxide and sevoflurane in 33% oxygen, on controlled ventilation. Vecuronium 0.1 mg/kg followed by 0.05 mg/kg at 30 min intervals or as clinically judged. Patients were on controlled ventilation.

All patients had baseline CBG measured immediately before the start of IV infusion of any fluid. Thereafter, CBG levels were measured half-hourly until the end of surgery. Glucose concentration was determined in fresh capillary blood by reflectance photometry using an Accu-Chek® active blood glucose monitor. If at any time during the study, CBG was found to be more than or equal to 150 mg/dl, calculated dose of human soluble insulin (=CBG/100 units) was given as IV bolus.

We considered CBG value of 150 mg/dl as treatment initiating point because a recent analysis of the cohort found a positive correlation between the average postoperative glucose level and mortality, with the lowest mortality in patients with an average postoperative blood glucose of 150 mg/dl.[18]

All the data were compiled and statistically analyzed using SPSS software version 22.0, Chi-square test and paired t-test. SPSS Statistics for Windows, version 22.0 was made by IBM Corporation, Armonk, New York, USA and released in 2013.

RESULTS

One hundred nondiabetic patients (50 in each group) of ASA physical status I and II of either sex, aged between 18 and 60 years were enrolled for this prospective, randomized, double-blind study.

There was no statistically significant difference between the groups in terms of the demographic characteristics of the patients, namely, age, weight, sex and ASA physical status distribution, and duration of surgery [Table 1].

Table 1

Demographic profile

There was statistically no significant difference in fasting blood sugar and immediate preoperative blood glucose levels. The preoperative serum electrolytes such as serum sodium, serum potassium, and serum calcium levels were comparable in both groups [Table 2].

Table 2

Baseline clinical profile

CBG was significantly high at baseline, 15, 30, 60, 90, 120, and 150 min intraoperatively in Group II (patients receiving 0.45% dextrose normal saline). We found that CBG level increased with progression of surgery and the mean CBG values at 15, 30, 60, 90, 120, and 150 min were significantly higher in patients who received 0.45% sodium chloride with 5% dextrose and 20 mmol/L potassium chloride as maintenance fluid in contrast to those receiving RL solution [Graph 1].

Graph 1

Comparison of blood glucose levels in both groups at different intervals

Percentage of hyperglycemic episodes in Group II patients was also significantly higher. Patients were considered hyperglycemic when CBG >150 mg%. In Group I, no CBG value was more than 150 mg until end of surgery. There were 16, 38, 60, 28, and 4 hyperglycemic events at intervals 30, 60, 90, 120, and 150 min [Graph 2].

Graph 2

Episodes of hyperglycemia in Group II. Number of patients showing hyperglycemic events at various intervals in Group II

Exogenous insulin requirement to maintain normoglycemia was significantly higher in patients receiving dextrose containing solution than those receiving RL solution. Rescue insulin dosage was given IV when CBG was >150 mg% in dosage of (CBG/100 = insulin IU). In Group I, there was no rescue insulin required as no patient in this group had hyperglycemia. In Group II, the number of rescue insulin doses increased as the intraoperative time increased.

Postoperative serum sodium levels in Groups I and II patients were 149.65 ± 4.92 and 139.04 ± 3.81 mEq/L, respectively, postoperative serum potassium levels in Groups I and II were 4.10 ± 0.98 and 4.91 ± 1.65 mEq/L, respectively, and serum calcium levels in Groups I and II were 5.12 ± 0.35 and 4.42 ± 0.16 mEq/L which were statistically significant (P < 0.05) [Graph 3].

Graph 3

Postoperative serum electrolytes (mEq/L) in Groups I and II

DISCUSSION

Hyperglycemia is an independent risk factor for morbidity and mortality in the perioperative period, during which the patient care is provided by anesthesiologists or intensive care physicians. Hyperglycemia, a common response to intraoperative stress of surgery and critical illnesses[1920] is associated with poorer outcomes in critically ill patients regardless of their diabetic status.[2122232425] It has been studied that degree of rise of blood sugar levels due to surgical stress in controlled diabetics is less significant as compared to nondiabetics.[14]

Glycemic control is usually maintained by IV infusion of insulin even in nondiabetics. A glycemic target of 140–180 mg/dl appears safe and acceptable for the majority of critically ill patients in perioperative settings.

This study was conducted to compare the effect of RL versus 0.45% dextrose normal saline on intraoperative blood glucose levels in nondiabetic patients undergoing elective major abdominal surgery under general anesthesia.

Many studies have been conducted on diabetic patients to study the effect of intraoperative hyperglycemia on perioperative complications, but very few studies are available in nondiabetic patients.

Our aim was to improve the outcome and decrease the incidence of perioperative complications arising due to intraoperative hyperglycemia in nondiabetic patients, thereby reducing the length of hospital stay and financial burden on both patient as well as hospital.

All parameters such as age, sex, weight, and preoperative blood glucose (measured by Accu-Chek glucometer) in Groups I and II were found comparable and nonsignificant [Tables 1 and 2].

CBG was measured by Accu-Chek glucometer at 0, 15 min, and then after every 30 min intervals until the completion of surgery and 30 min after the surgery. The baseline blood glucose levels in both the groups were nonsignificant (P > 0.05). In Group I, a persistent 10–15% decline (but nonsignificant) in value of CBG levels was noted at 15 min and thereafter there was a persistent increase in levels at all intervals while in Group II, there was increase in CBG levels at all intervals as the surgery progressed and found to be highly significant. These findings were in concordance with a study conducted in October 2015 by Rao and Indra[14] to compare blood glucose levels under general anesthesia in nondiabetic and controlled diabetics. There was a decline in blood sugar level at 5 min postintubation; in our study, we noted blood sugar level at 15 min. In their study, postintubation blood sugar levels showed a decrease of 1.064% (nonsignificant) that was attributed to suppression of stress response of intubation by fentanyl and propofol. A study conducted by Saringcarinkul and Kotrawera[26] in 2009 found a progressive increase in blood glucose levels in patients receiving 5% dextrose during surgery. They considered blood glucose levels of more than 150 mg/dl as significant. A prospective, randomized, parallel study was conducted in 2013 by Maitra et al.[11] on 200 nondiabetic patients to study intraoperative blood glucose levels in patients undergoing elective major surgery receiving different crystalloid solutions, as maintenance fluids. It was found that CBG levels increased with progression of surgery and mean CBG values at 1st, 2nd, and 3rd h were significantly higher in patients receiving 0.45% sodium chloride with 5% dextrose. We also observed that patients receiving RL remained normoglycemic throughout the study period, whereas patients who received 0.45% dextrose saline had significantly elevated plasma glucose levels; similar findings were observed by Chin et al. in 2006.[27] They conducted randomized controlled trial in 50 nondiabetic adult patients undergoing elective surgery and infused 500 ml of either 5% dextrose in 0.9% normal saline, lactated Ringer's solution, or 0.9% normal saline over 45–60 min. Plasma glucose, electrolytes, and osmolarity were measured before infusion and at 15 min and 1 h after completion of infusion. None of the patients had preoperative hypoglycemia despite average fasting times of almost 13 h. Patients receiving lactated Ringer's and normal saline remained normoglycemic throughout the study period. Patients receiving dextrose normal saline had significantly elevated plasma glucose 15 min after completion of infusion (11.1 [9.9–12.2, 95% confidence interval] mmol/L). They concluded that initiation of IV fluid replacement with dextrose-containing solutions was not required to prevent hypoglycemia in elective surgery. On the contrary, a relatively small volume of 500 ml causes significant, albeit transient, hyperglycemia, even in nondiabetic patients.

The postoperative blood glucose levels in Group I was 138 ± 9.34 mg/dl and in Group II as 145.00 ± 8.42 mg/dl which was statistically significant (P < 0.05). Pili-Floury et al.[28] in 2010 conducted a study on glycemic dysregulation in nondiabetic patients after major lower limb prosthetic surgery. They assessed blood glucose from preoperative period to the end of the postoperative period day 2 and considered blood glucose levels 126 mg% as hyperglycemia. They observed a significant increase in blood sugar levels (P < 0.001) in 74% of patients.

In our study, there was no incidence of hyperglycemia, i.e., CBG ≥150 mg% in Group I (patients receiving RL as maintenance fluid), while 60% of patients of Group II had hyperglycemic event. Every single incidence of hyperglycemia was taken into account during calculation of hyperglycemic events. These values were highly significant (P < 0.01). A study was conducted by Azarfarin and Alizadeh Asl[29] to know the prevalence of hyperglycemia in nondiabetic patients undergoing coronary artery bypass grafting which concluded that 54.6% patients had hyperglycemia. We also observed the similar finding in our study where CBG was found to be higher than 150 mg%, and calculated dose of human soluble insulin was given as IV bolus and found that requirement of rescue insulin was significantly higher in a group of patients receiving 0.45% dextrose normal saline than patients receiving RL infusion. Our observations are similar to a study conducted by Maitra et al.[11] in 2013, who found that requirement of rescue insulin to maintain glycemic homeostasis in patients receiving dextrose as maintenance fluid during surgery was significantly higher than those who received RL.

Preoperative serum sodium, serum potassium, and serum calcium were found to be nonsignificant, but postoperative values of serum electrolytes were found to be statistically significant (P < 0.05).

Preoperative serum sodium in Groups I and II were 139.94 ± 3.45 and 141.50 ± 4.04 mEq/L, respectively (P > 0.05), whereas postoperative serum sodium in Group I and II was 149.65 ± 4.92 and 139.04 ± 3.81 mEq/L, respectively, which was statistically significant (P < 0.05).

Preoperative serum potassium in Groups I and II were 4.06 ± 1.21 and 4.12 ± 1.38 mEq/L, respectively (P > 0.05), whereas postoperative serum potassium in Group I and II was 4.10 ± 0.98 and 4.91 ± 1.65 mEq/L, respectively, which was statistically significant (P < 0.05).

Preoperative serum calcium in Groups I and II were 4.76 ± 0.34 and 4.74 ± 0.31 mEq/L, respectively (P > 0.05), whereas postoperative serum calcium in Group I and II was 5.12 ± 0.35 and 4.42 ± 0.16 mEq/L, respectively, which was statistically significant (P < 0.05).

A study was conducted by Hadimioglu et al.[30] to know the effect of different crystalloid solutions on acid-base balance and early kidney transplant and found 19% of patients who received normal saline developed hyperkalemia, whereas none of the patients who received RL had hyperkalemia.

Limitation

Long-term follow-up of the patients should have been done to know the impact of perioperative hyperglycemia on morbidity and mortality.

During surgeries of long duration in patients receiving infusion of 0.45% dextrose normal saline, hyperglycemia becomes a matter of concern.

Serum chloride level: Chloride is one of most important electrolyte in the blood, and it helps to keep the amount of intracellular and extracellular fluid in balance. It helps to maintain blood volume, blood pressure, and pH of body fluids. However, due to some technical issues and financial constraints in developing country like ours, it could not be done.

CONCLUSION

We conclude that stress-induced hyperglycemic response in patients undergoing major noncardiac surgery is common in nondiabetic population. Maintenance fluid therapy by dextrose containing infusions as opposed to RL infusion increases the incidence of perioperative hyperglycemia in nondiabetic patients. To achieve normal CBG levels by IV bolus dose of soluble human regular insulin, significantly higher doses are required in patients receiving dextrose containing saline as maintenance fluid.

Therefore, RL solution is probably the alternative choice of IV fluid for perioperative maintenance and can be used as replacement in healthy patients undergoing elective surgeries.

For nondiabetic patients undergoing major surgical procedures, a significant stress response can be mounted with elevations in blood glucose. In these patients, we support the practice of targeting blood glucose levels to <150 mg/dl with frequent blood glucose measurements.

Recommendations

We feel that follow-up of the patients should be done to know the incidence of surgical site infection, length of stay in the hospital, and to know if patients with hyperglycemia had higher morbidity and mortality in the long run so as to make tight glycemic control mandatory to maintain euglycemia.

Future research should be done on the large patient population to establish criteria for intraoperative glucose monitoring and threshold for treatment. Establishing these standards may improve the outcomes in the patients.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.