A comparison of 5% dextrose in 0.9% normal saline versus non-dextrose-containing crystalloids as the initial intravenous replacement fluid in elective surgery.
Intravenous fluid replacement in adult elective surgery is often initiated with dextrose-containing fluids. We sought to determine if this practice resulted in significant hyperglycaemia and if there was a risk of hypoglycaemia if non-dextrose-containing crystalloids were used instead.
We conducted a randomized controlled trial in 50 non-diabetic adult patients undergoing elective surgery which did not involve entry into major body cavities, large fluid shifts, or require administration of >500 ml of intravenous fluid in the first two hours of peri-operative care. Patients received 500 ml of either 5% dextrose in 0.9% normal saline, lactated Ringer's solution, or 0.9% normal saline over 45 to 60 minutes. Plasma glucose, electrolytes and osmolarity were measured prior to infusion, and at 15 minutes and one hour after completion of infusion.
None of the patients had preoperative hypoglycaemia despite average fasting times of almost 13 hours. Patients receiving lactated Ringer's and normal saline remained normoglycaemic throughout the study period. Patients receiving dextrose saline had significantly elevated plasma glucose 15 minutes after completion of infusion (11.1 (9.9-12.2, 95% CI) mmol/l). Plasma glucose exceeded 10 mmol/l in 72% of patients receiving dextrose saline. There was no significant difference in plasma glucose between the groups at one hour after infusion, but 33% of patients receiving DS had plasma glucose [greater than or equal to]8 mmol/l.
We conclude that initiation of intravenous fluid replacement with dextrose-containing solutions is not required to prevent hypoglycaemia in elective surgery. On the contrary, a relatively small volume of 500 ml causes significant, albeit transient, hyperglycaemia, even in non-diabetic patients.
Key Words: intravenous fluid, dextrose, saline, hypoglycaemia, hyperglycaemia, surgery, stress
Concern has been expressed that prolonged preoperative fasts can lead to occult hypoglycaemia and it has been recommended that dextrose-containing solutions be administered during surgery to prevent this (1). This rationale has led to elective surgical patients at our institution receiving 500 ml of intravenous 5% dextrose in 0.9% normal saline at the start of anaesthesia, followed by further fluid replacement with non-dextrose-containing crystalloids. However, this practice may not be justified, given the capacity for endogenous glucose production in a healthy adult'. Coupled with the metabolic response to surgical stress, intravenous dextrose infusion may in fact cause significant hyperglycaemia.
At present though, there is insufficient evidence to support the use of either dextrose-containing or non-dextrose-containing solutions when initiating intravenous fluid replacement in the adult elective surgical population. We therefore conducted a prospective randomized trial to compare the effect of infusing 500 ml of 5% dextrose in 0.9% normal saline with that of two other commonly used intravenous crystalloids (lactated Ringer's solution, and 0.9% normal saline) on plasma glucose levels in non-diabetic adults presenting for elective surgery, within a period of one hour following completion of the infusion.
MATERIALS AND METHODS
Following approval by the institutional ethics committee and written informed consent, we recruited 60 adult patients of American Society of Anesthesiologists' (ASA) physical status 1 or 2 who were presenting for elective surgery between January 2002 and August 2002. Patients were excluded from the study if they had a preoperative diagnosis of diabetes mellitus, renal or hepatic dysfunction; were receiving perioperative corticosteroid therapy; were undergoing surgery involving entry into major body cavities or large fluid shifts; or had anticipated intravenous fluid requirements exceeding 500 ml in the first two hours of perioperative care.
Patients were randomized into three groups using a computer-generated random number table and sealed opaque envelopes. Group DS received 500 ml of dextrose 5% in 0.9% normal saline (dextrose 5 g/dl, sodium and chloride 150 mmol/l each), Group LR received 500 ml of lactated Ringer's solution (sodium 130 mmol/l, potassium 4 mmol/l, calcium 3 mmol/l, chloride 109 mmol/l, lactate 28 mmol/l), and Group NS received 500 ml of 0.9% normal saline (all solutions manufactured by BBraun Medical Industries S.B., Penang, Malaysia). Infusion was commenced in the operating theatre once an intravenous cannula had been inserted, at a rate titrated to ensure completion within 45 to 60 minutes.
Blood was drawn from the patient prior to commencing the infusion (baseline), 15 minutes after completion of the infusion and one hour after completion of the infusion. The blood drawn was analysed in the hospital's central laboratory for plasma glucose, sodium, potassium, chloride, urea and osmolarity. Plasma glucose levels were measured using an enzymatic colorimetric assay in an automated analyser. No further intravenous fluid was administered until the last blood sample had been taken. No other restrictions were placed on the conduct of anaesthesia.
Differences between groups were analysed using one-way analysis of variance. Within-groups comparison of data across time was made using analysis of variance for repeated measures. A post hoc Bonferroni correction for multiple testing was performed for all pairwise comparisons. A P value of <0.05 was considered significant. All results are reported as mean (95% confidence intervals) or mean[+ or -]standard deviation.
Sample size was determined by a power analysis based on the following assumptions: 1) a mean baseline plasma glucose of 6 mmol/l, with a standard deviation of 2.5 mmol/l; 2) a definition of a clinically significant increase in plasma glucose levels as 50% or more; 3) an error of 0.05 and power of 80%. This produced a sample size estimate of 48 patients in total (16 per group). This was increased to 60 patients (20 per group) to allow for patient dropouts.
Sixty patients were recruited into the study. Fifty patients completed the study protocol: 18 in Group DS and 16 each in Group LR and NS. Ten patients were excluded from analysis because of protocol violations involving the infusion of additional fluid during the study period. There were no significant differences in baseline characteristics between the three groups except for age (Table 1).
Mean duration of the preoperative fast was almost 13 hours in all three groups. The distribution of cases by type of surgery and anaesthetic technique are also shown in Table 1. Fentanyl and propofol were used to induce general anaesthesia, which was then maintained with isoflurane and nitrous oxide. Morphine for analgesia and atracurium for muscle relaxation were administered when required. All spinal anaesthetics were for total knee replacement surgery; patients received 10 [mu]kg of hyperbaric bupivacaine and 10 mg of fentanyl. None of the patients required ephedrine or other sympathomimetics during the study period.
Plasma glucose at the various sampling times is illustrated in Figure 1. Mean plasma glucose at baseline was similar in all three groups (4.6 (4.4-4.9), 4.9 (4.4-5.3),4.8 (4.4-5.2) mmol/l in Groups DS, LR and NS respectively). The lowest recorded preoperative value was 3.3 mmol/l. Plasma glucose was significantly higher than baseline in all three groups at 15 minutes and one hour after completion of infusion.
[FIGURE 1 OMITTED]
There was no significant difference between Groups LR and NS 15 minutes and one hour after completion of infusion. Both groups remained normoglycaemic throughout the study period. In Group DS however, plasma glucose 15 minutes after completion of infusion was significantly higher compared to the other two groups. The mean value was 11.1 (9.9-12.2) mmol/l versus 5.8 (5.3-6.3) mmol/l and 5.8 (5.4-6.2) mmol/ lin Groups LR and NS respectively (P<0.0001). The highest recorded value in Group DS was 15.8 mmol/l. Plasma glucose exceeded 12 mmol/l in 28% (5/18), 10 mmol/l in 72% (13/18), and 8 mmol/l in 94% (17/18) of patients in Group DS. None of the patients in the other two groups had plasma glucose levels exceeding 8 mmol/l.
At one hour after completion of infusion, mean plasma glucose was 6.9 (6.1-7.7) mmol/l in Group DS, compared to 6.2 (5.4-7.0) mmol/l and 5.8 (5.4-6.3) mmol/l in Groups LR and NS respectively; this difference was not statistically significant (P=0.081). Plasma glucose levels remained [greater than or equal to]8 mmol/l in 33% (6/18) of patients in Group DS.
There were no significant differences between groups or across time with respect to plasma urea, sodium, potassium, or osmolarity; all values were within normal limits.
When patients undergoing spinal anaesthesia were excluded in a post hoc analysis, plasma glucose at 15 minutes after completion of infusion was still significantly higher in Group DS compared to Groups LR and NS [10.8 (9.3-12.3), 5.7 (5.2-6.2), 5.9 (5.5, 5.3) respectively, P<0.0001].
Current ASA guidelines recommend a preoperative fast of at least six hours following intake of solids or non-human milk and two hours following clear fluids (3). Excessive fasting periods are discouraged as it results in significant patient discomfort (4). Nevertheless it is still common practice for hospitals to enforce a "nothing by mouth after midnight" rule in adult patients scheduled for elective surgery in a morning list. This simplifies preoperative instructions and allows flexibility in rescheduling should the need arise; however it also translates into a fasting period of 8 to 12 hours or longer. Our institution is no exception; the average fasting time in our patients was more than 12 hours.
The evidence for hypoglycaemia resulting from prolonged preoperative fasts is conflicting. In one study, 14% of healthy young women had serum glucose of 2.5 mmol/l or less after a preoperative fast exceeding 12 hours; none of them, however, were symptomatic (2). On the other hand, the absence of preoperative and intraoperative hypoglycaemia, even when surgery is prolonged, has also been demonstrated (3). Our results support the view that the risk of perioperative hypoglycaemia in the healthy adult undergoing elective surgery is minimal.
Instead, a statistically significant rise in plasma glucose from baseline was observed in all three groups of patients, which may be attributed to the catabolic response to surgery. The surgical stress response increases endogenous glucose production (from glycogenolysis and gluconeogenesis) whilst decreasing clearance (5). Despite this, normoglycaemia was maintained throughout the study period in the groups receiving non-dextrose-containing crystalloid solutions. On the other hand, the administration of 500 ml of dextrose 5% in 0.9% normal saline resulted in significant hyperglyacaemia despite the modest caloric load. This was transient, however, and its clinical significance remains uncertain. Nevertheless, acute hyperglycaemia has been associated with impaired phagocytosis by polymorphonuclear leukocytes (2), dysfunction of the complement system (7) and stimulated sympathoadrenergic activity (8). Perioperative hyperglycaemia has also been associated with glycosuria (1); and in the context of coronary artery bypass graft surgery, longer hospital stays, more recurrent wound infections and ischaemic episodes, and decreased survival over a two-year period (9).
What about the choice between lactated Ringer's solution and normal saline? It has been suggested that the metabolism of lactate may contribute to an increase in plasma glucose (10); however this may be pertinent only in the diabetic population. We did not observe any significant differences between the two with respect to plasma glucose, electrolytes or osmolarity, although it should be noted that only 500 ml of each was administered and that our study was not designed to examine the impact on biochemical variables other than plasma glucose.
Central neuraxial anaesthesia attenuates the hyperglycaemic response to surgery (11) and the unequal distribution of spinal anaesthetics between study groups may have exaggerated the observed difference in plasma glucose. However a significant difference in plasma glucose was still observed in a post hoc analysis that excluded the patients who received spinal anaesthesia. The mean age of the patients, although greater in Group LR compared to the others, is unlikely to have been a significant confounder. Older patients might be expected to have a greater tendency to hyperglycaemia because of an increased incidence of impaired glucose tolerance; however this was not observed in Group LR. Another limitation of our study was the fluid infusion rate of 500 to 750 ml/h. A range rather than a fixed infusion rate was chosen to allow flexibility in managing individual patients. It is also our clinical impression that the first 500 ml of fluid are usually infused even faster (over <30 minutes) during anaesthesia. We deliberately extended the duration of infusion to ensure our results were relevant to the majority of anaesthetic practitioners. It is likely that we have underestimated, rather than overestimated, the degree of hyperglycaemia induced by the use of 5% dextrose-containing solutions in practice. Finally we would like to point out that while dextrose 5% in 0.9% normal saline is a hyperosmolar solution (560 mOsm/1) compared to the other crystalloids as well as plasma, it is nevertheless still isotonic. Once infused, the dextrose is rapidly metabolized, leaving isotonic normal saline (12). Dextrose 5% in water, on the other hand, is iso-osmolar but hypotonic and distributes throughout total body fluid, rather than confining itself to the extracellular fluid compartment. Dextrose-saline is therefore a much more effective intravascular volume expander and better suited to perioperative use if intravenous administration of glucose is desired.
In conclusion, it does not appear necessary to administer 5% dextrose-containing solutions to prevent hypoglycaemia in adult patients fasted for elective surgery. On the contrary, even a relatively small volume of 500 ml causes hyperglycaemia, albeit transient. Given the absence of evidence for benefit and the possibility for harm, we recommend that the routine use of 5% dextrose-containing solutions as the initial intravenous replacement fluid in elective adult surgery should be avoided altogether.
(1.) Doze VA, White PE Effects of fluid therapy on serum glucose levels in fasted outpatients. Anesthesiology 1987; 66:223-226.
(2.) Chambrier C, Aouifi A, Bon C, Saudin F, Paturel B, Bouletreau P Effects of intraoperative glucose administration on circulating metabolites and nitrogen balance during prolonged surgery. J Clin Anesth 1999; 11:646-651.
(3.) Report by the American Society of Anesthesiologists Task Force on Preoperative Fasting. Practice guidelines for preoperative fasting and the use of pharmacologic agents to reduce the risk of pulmonary aspiration: Application to healthy patients undergoing elective procedures. Anesthesiology 1999; 90:896-905.
(4.) Sutherland AD, Stock JG, Davies JM. Effects of preoperative fasting on morbidity and gastric contents in patients undergoing day-stay surgery. Br J Anaesth 1986; 58:876-878.
(5.) Geisser W, Schreiber M, Hofbauer H et al. Sevoflurane versus isoflurane--anaesthesia for lower abdominal surgery. Effects on perioperative glucose metabolism. Acta Anaesthesiol Scand 2003; 47:174-179.
(6.) Rassias AJ, Marrin CAS, Arruda J et al. Insulin infusion improves neutrophil function in diabetic cardiac surgery patients. Anesth Analg 1999; 88:1011-1016.
(7.) Kwoun MO, Ling PR, Lydon E et al. Immunologic effects of acute hyperglycaemia in nondiabetic rats. J Parenteral Enteral Nutr 1997; 21:91-95.
(8.) Nordenstrom J, Jeevanandam M, Elwyn DH, Carpentier YA. Increasing glucose intake during total parenteral nutrition increases norepinephrine excretion in trauma and sepsis. Clin Physiol 1981; 1:525-534.
(9.) Lazar HL, Chipkin SR, Fitzgerald CA, Bao Y, Cabral H, Apstein CS. Tight glycemic control in diabetic coronary artery bypass graft patients improves perioperative outcomes and decreases recurrent ischemic events. Circulation 2004; 109:1497-1502.
(10.) Thomas JG, Alberti KG. Hyperglycaemic effects of Hartmann's solution during surgery in patients with maturity onset diabetes. Br J Anaesth 1978; 50:185-188.
(11.) Lattermann R, Carli F, Wykes L, Schricker T Epidural blockade modifies perioperative glucose production without affecting protein catabolism. Anesthesiology 2002; 97:374-381.
(12.) Sterns RH, Silver SM. Salt and water: read the package insert. QJM 2003; 96:549-552.
K. J. CHIN *, J. MACACHOR [dagger], K. C. ONG [double dagger], B. C. ONG [section]
Department of Anaesthesia, Singapore General Hospital, Singapore
* F.A.N.Z.C.A., Registrar, Department of Anaesthesiology, Tan Tock Seng Hospital.
[dagger] M.D., Medical Officer, Department of Anaesthesia, National University Hospital.
[double dagger] M.Med, Registrar, Department of Anaesthesiology, Tan Tock Seng Hospital.
[section] M.Med, Senior Consultant.
Address for reprints: Dr Ki-Jinn Chin, Registrar, Department of Anaesthesiology, Tan Tock Seng Hospital, 11 Jalan Tan Tock Seng, 308433, Singapore.
Accepted for publication on June 15, 2006.
TABLE 1 Baseline characteristics of sample population Dextrose Lactated 5% in 0.9% Ringer's normal saline solution No. of patients 18 16 Age (years)* 38[+ or -]18 50[+ or -]15 Weight (kg) 63[+ or -]12 62[+ or -]15 Male: female ratio 9 : 9 8 : 8 Preoperative fast 12.7[+ or -]2.6 12.7[+ or -]3.9 (hours) Duration of 88[+ or -]48 93[+ or -]42 surgery (min) No. of spinal 4 2 anaesthetics Type of surgery Orthopaedic -- 9 5 joint replacement Orthopaedic 2 1 -- other surgery ENT surgery 4 8 Breast surgery 2 0 General surgery 1 2 -- minor Co-morbid conditions Hypertension 3 1 Ischaemic heart 1 1 disease Hyper-/ 1 2 hypothyroidism 0.9% normal saline No. of patients 16 Age (years)* 35[+ or -]15 Weight (kg) 72[+ or -]15 Male: female ratio 13: 3 Preoperative fast 13.0[+ or -]3.0 (hours) Duration of 72[+ or -]46 surgery (min) No. of spinal 1 anaesthetics Type of surgery Orthopaedic -- 5 joint replacement Orthopaedic 2 -- other surgery ENT surgery 7 Breast surgery 1 General surgery 1 -- minor Co-morbid conditions Hypertension 3 Ischaemic heart 0 disease Hyper-/ 0 hypothyroidism *P=0.023.
|Printer friendly Cite/link Email Feedback|
|Author:||Chin, K.J.; Macachor, J.; Ong, K.C.; Ong, B.C.|
|Publication:||Anaesthesia and Intensive Care|
|Date:||Oct 1, 2006|
|Previous Article:||Pulmonary atelectasis following spinal anaesthesia for caesarean section.|
|Next Article:||Narrative Pain and Suffering--Progress in Pain Research and Management.|