Glucocorticoids and especially dexamethasone are increasingly used to prevent post-operative nausea and vomiting (PONV). Successful use of corticosteroids as antiemetics in antineoplastic chemotherapy was first reported in the early 1980s.1 More recently, prophylactic administration of dexamethasone has emerged as an antiemetic option in surgical patients and is also increasingly used as an analgetic adjuvant.2 In 2003, the drug was officially licensed in Germany and other European countries for prevention of PONV and is implemented in consensus guidelines for the management of PONV.3
So far there is no evidence to suggest that a single dose of 8 mg dexamethasone exerts any significant adverse effects after surgery, including the impairment of wound healing or an increased infection rate.4 However, the majority of studies were small in sample size and conducted on patients undergoing surgical procedures, which are associated with minimal tissue trauma, that is, ear, nose and throat (ENT) surgery, dentistry and neurosurgery.
One particular concern of routine corticosteroid intake is its impact on glucose and lipid metabolism, that is, the potential alteration of insulin sensitivity resulting in hyperglycaemia and stimulated lipolysis. Considering the effects of the catabolic responses to surgery, particularly the potentially negative influence of hyperglycaemia on surgical outcome and in-hospital mortality, any disturbance of perioperative glucose5–7 and lipid homeostasis8 by pharmacological interventions assumes clinical importance.
Although the effects of short-term and long-term use of corticosteroids have been widely studied in the non-surgical patient population, the metabolic consequences of dexamethasone administration in surgical patients undergoing abdominal surgery have received only modest attention. This randomised double-blind controlled trial was designed to study the effect of a single prophylactic oral dose of 8 mg dexamethasone on blood glucose and plasma levels of non-esterified fatty acid (NEFA) concentrations during and after abdominal hysterectomy. The oral dose of 8 mg dexamethasone was chosen to compensate the lower bioavailability of the drug (80 ± 12%) compared to the intravenous route.
Materials and methods
After approval by the local Ethics Committee of the Philipps University of Marburg (Az:03/02) and written informed consent by the participating patients, 90 non-diabetic American Society of Anesthesiologist's classification I and II women (class I: healthy patients; class II: patients with mild, controlled, functionally non-limiting systemic disease) undergoing elective non-cancer abdominal hysterectomy were included in this randomised placebo-controlled trial. All patients were screened pre-operatively and excluded from the study if blood glucose was higher than 5.5 mmol l−1 after a 6 h fasting period, if they had a previous pathological glucose tolerance test, intake of oral antidiabetics, insulin, β-blockers or an arbitrarily chosen level of a BMI more than 35 kg m−2.
The patients fasted from midnight the day before surgery and received 20–40 mg clorazepate at 10 p.m. The usual medication (e.g. thyroxin, iodine, antihypertensive drugs but no diuretics) was continued throughout the peri-operative period. Two hours before scheduled surgery, patients received oral premedication with 7.5 mg midazolam and the study drug ingested with a small amount of clear non-sweetened liquid. The latter was stored in a sealed box prepared by the hospital pharmacy prior to the beginning of the study according to a computer-generated random list in order to ensure allocation concealment. The box contained a tablet with 8 mg dexamethasone or an undistinguishable placebo tablet.
In the operating room, the routine monitoring devices were attached and an i.v. line was inserted. Ringer solution was infused. Routine general anaesthesia was then administered. Fentanyl (0.2–0.5 mg), propofol (2–3 mg kg−1) and cisatracurium (0.1 mg kg−1) were titrated according to body weight and individual demands of the patient, the latter according to results of a qualitative relaxometry. The airway was secured by an endotracheal tube and anaesthesia was then maintained using 3–5% of desflurane in oxygen-air to keep depth of anaesthesia within a range of 40–60 according to the bispectral index (BIS). Additional doses of fentanyl (up to a maximum cumulative dose of 1 mg) were administered if signs of stress-response occurred (increase in heart rate or blood pressure above baseline values). Cisatracurium was repeated according to qualitative relaxometry. Remifentanil was continuously infused, at least during the last 60 min of surgery to facilitate post-operative recovery. Routine antiemetic prophylaxis with 12.5 mg dolasetrone was given to all patients at the end of surgery. In all patients, a total hysterectomy was performed by a transverse (Pfannenstiel) incision. Most patients were undergoing surgery because of severe hypermenorrhoe and/or myomas. They were extubated at the end of surgery without reversal of neuromuscular block. Post-operative analgesia was performed with 2.5 g metamizole (dipyridone) and repeated i.v. boluses of 0.05–0.1 mg kg−1 piritramide, a pure μ-opioid agonist with approximately 75% of the potency of morphine, until patients reported pain levels of 3 or lower on a 0–10 numeric rating scale. All patients received 40 mg of enoxaparine at 6 p.m. the evening before surgery and on the day of surgery (after completion of the study).
Patients were monitored until the second post-operative day for the occurrence of PONV and until discharge from hospital for signs of delayed wound healing or wound infections using the ASEPSIS scoring system.9
Peripheral venous blood samples were drawn the day before surgery (patients were fasted for at least 4 h), immediately before induction of anaesthesia, after removal of the uterus, and 2, 6 and 10 h after the end of anaesthesia while the patients were still fasting (only pure water was allowed as oral refreshment). During the first 24 h after surgery, patients received 125 ml h−1 lactated Ringer solution. The oral dose of 8 mg dexamethasone was chosen to compensate the lower bioavailability of the drug (80 ± 12%) compared to the intravenous route. Fig. 1 shows the time course of the study protocol.
Blood glucose was analysed using a blood gas analyser (ABL 700 Series'; Radiometer GmbH, Willich, Germany) and verified using standard laboratory measurement. For analysis of NEFA levels, the samples were immediately centrifuged and stored at −30°C. NEFAs were analysed using the NEFA-C enzymatic colorimetric method (Wako Chemicals GmbH, Neuss, Germany). The correlation between this method and a reference method is excellent (r = 0.992; y = 1.078x−0.024). The error of measurement in the expected range of measurements (precision) was not more than ± 2.7% of the actual concentration.
Assuming a clinically relevant mean group difference of 2 mmol l−1 with an average standard deviation of measurements of 40 patients per group provide a power of more than 80% to detect this between group difference and a more than 90% power to detect within group variations (effects over time) using an analysis of variance (ANOVA) for repeated measures. There were no changes in methods or trial outcomes after commencement of the trial.
The flow of patients from screening to analysis of the obtained data is shown in Fig. 2. The biometric data of all patients (‘intention to treat sample’) are shown in Table 1. There were no statistically significant or clinically relevant differences between the two groups. PONV occurred in 53% (20 of 38; 95% confidence interval: 36–69%) of the patients in the control group that received dolasetrone alone and in 34% (15 of 44; 95% confidence interval: 20–50%) of the dexamethasone group that also received dolasetrone. There was one case of wound infection in the placebo group. This complication occurred in a woman with a BMI of 34.2 kb m−2 who developed hyperglycaemia in the post-operative period. Maximum blood glucose was 11.2 mmol l−1. She received treatment with cefuroxime and recovered within 4 days. No other relevant side-effects occurred.
The time course of the blood glucose levels is shown in Fig. 3. Compared with pre-operative values, intraoperative and all post-operative measurements were increased irrespective of whether or not dexamethasone was administered. This increase was more pronounced in patients receiving dexamethasone (interaction term of the ANOVA: P = 0.02) indicating an additional hyperglycaemic effect of the glucocorticoid apart from the known effects of surgical stress on blood glucose. Elevated blood glucose (>7 mmol l−1) during the post-operative observation period while the patients were still fasting was observed in 36 participants (placebo: 9 = 24% and dexamethasone: 27 = 61%). Eleven patients showed hyperglycaemia (defined as blood glucose level of 8.5 mmol l−1) during at least one of the measurements. Of these, nine patients had received dexamethasone, that is, a 20% rate in this group of patients. In nine of these 11 patients, blood glucose decreased below 8 mmol l−1 at the final measurement 10 h after the end of anaesthesia or showed a marked trend towards normal values (Fig. 4). In most of the patients (77 out of 82; 94%), glucose levels were decreasing at the end of the observation period in both groups, indicating that dexamethasone had only short-lasting effects in these fasting patients.
Concerning the measurements of the free NEFA, two additional patients were excluded from the analysis because they inadvertently received more than an induction dose of propofol. This hypnotic is formulated in an oil-in-water emulsion and contains soybean oil (100 mg ml−1). Thus, the effects of the administration of larger doses on NEFA plasma levels are not known. The effects on the concentrations of NEFA were minimal. Figure 5 gives an overview on these measurements that did not show a statistically significant change.
Surgical trauma is followed by metabolic and endocrine changes commonly described as the catabolic stress response. Typical features are increased circulating concentrations of glucose and NEFA. Anti-insulinergic hormones such as cortisol are thought to be major mediators of catabolism. Therefore, the therapeutic use of exogenous corticosteroids may accentuate the metabolic alterations induced by surgical tissue trauma.
This assumption is supported by the results of the present study showing that PONV prophylaxis with 8 mg oral dexamethasone augments the hyperglycaemic response to abdominal hysterectomy.
Chronic intake of corticosteroids in metabolically healthy patients causes so-called ‘steroid diabetes’, which is characterised by increased hepatic gluconeogenesis, impaired peripheral glucose utilisation, abnormal glucose tolerance, and, to a varying degree also hyperglycaemia, hypertriglyceridaemia, hypercholesterolaemia, and increased plasma NEFA levels.10 Conversely, short-term administration of corticosteroids together with glucose in normal volunteers produced less hyperglycaemia than glucose alone11 indicating that glucocorticoids are metabolically inert when used once and at a low dose in non-surgical patients.
Persistent hyperglycaemia has been shown to be an independent risk factor for death, major cardiovascular, respiratory, infectious and renal complications after surgery.5,6 According to the results of a study in 20 patients after craniotomy, 10 mg of dexamethasone increased blood glucose levels to a significantly greater extent than placebo (placebo: from 4.9 to 5.8 mmol l−1; dexamethasone: 5.4 to 8.3 mmol l−1).12 Similar results were obtained in an observational (non-randomised) trial in patients undergoing neurosurgical procedures.13 Single dose administration of a high dose of dexamethasone in a small number of cardiac surgery patients (1 mg kg−1 before anaesthesia induction followed by 0.5 mg kg−1 8 h later) induced a significantly greater hyperglycaemic response than placebo (10.7 ± 0.6 vs. 7.4 ± 0.5 mmol l−1).14 More recently, the metabolic effect of 10 mg i.v. dexamethasone was studied in a mixed group of diabetic and non-diabetic patients undergoing bariatric and non-bariatric procedures.15 Blood glucose increased post-operatively by 1.5–2 mmol l−1 in both groups. Values returned to baseline level 4 h later. This study was flawed because of the heterogeneity of surgical trauma and the lack of a control group.16
Although the metabolic impact of prophylactic dexamethasone administration has been studied after craniotomies,12,13 open heart surgery,14 and strabismus surgery,17,18 its effect on glucose metabolism during and after abdominal surgery is still unclear.
PONV prophylaxis with i.v. dexamethasone 0.25, 0.5, or 1 mg kg−1, given immediately after induction of anaesthesia in 168 children aged 2–15 years undergoing paediatric strabismus surgery17 slightly increased blood glucose 4 h after steroid administration in all patients with mean blood glucose values remaining within the normal range.
In the dexamethasone group of the present study, mean peak blood glucose values at 7 mmol l−1 were observed 2 h after surgery (approximately 6 h after administration of dexamethasone) followed by a slight decline afterwards. Fifty-six percent of our patients were normoglycaemic (defined as glucose concentrations of below <7.0 mmol l−1) throughout the entire observation period. Despite fasting, a greater number of patients receiving dexamethasone showed transiently elevated glucose levels (glucose 7.0–8.5 mmol l−1: dexamethasone: 61% vs. placebo: 24%) and some (20 vs. 5%) even developed hyperglycaemia (glucose >8.5 mmol l−1). In light of the recent finding that fasting blood glucose concentrations greater than 7 mmol l−1 were associated with a relevant increased in-hospital mortality7, our results gain clinical relevance. When comparing patients with and without episodes of hyperglycaemia, hyperglycaemic patients were older (48 vs. 39 years) and heavier (BMI 31 vs. 25 kg m−2), indicating that these factors may have aggravated, or contributed to some extent, to the hyperglycaemic action of dexamethasone.
In the present study, only patients who were not treated for diabetes mellitus and showed normal fasting blood glucose levels were included. As glucose tolerance tests were not routinely performed pre-operatively, it cannot be excluded that patients with pre-existing, undiagnosed impaired glucose tolerance or prediabetes entered the study. Obese patients with an impaired glucose tolerance undergoing Roux-en-Y gastric bypass surgery had significantly higher maximum glucose values (10.4 ± 1.6 mmol l−1) after treatment with 8 mg dexamethasone than placebo-treated patients (8.8 ± 1.7 mmol l−1).19 These results indicate that hyperglycaemic response after is more pronounced in patients with impaired glucose tolerance, but not substantially different when dexamethasone is added in antiemetic doses.
One limitation of our study is that the sample size is much too small to detect clinically relevant endpoints (i.e. mortality, wound infection, etc.) that might be associated with elevated blood glucose levels. One may criticise our decision to draw blood samples not dependent on the time of dexamethasone administration, but relatively to the end of surgery resulting in inconstant time intervals between drug administration and glucose and NEFA measurements due to variations in duration of surgery. However, we assumed that surgery itself will have more impact on the post-operative glycaemic response than the administration of dexamethasone, a drug known to have a long biological duration of action. The decision to administer the drug by the oral route 2 h before induction of anaesthesia was based on the intention to detect metabolic effects of the drug during surgery. The additional variability in plasma concentration by inconstant resorption from the gastrointestinal tract is low because dexamethasone has a very high bioavailability after oral intake (78 ± 12%).
As a consequence of stimulated lipolysis in the peripheral fat tissue and the counter-regulatory endocrine response to stress, NEFA concentrations typically increase during and after abdominal surgery. Increased circulating levels of free fatty acids have been shown to inhibit peripheral glucose uptake and, thus, can aggravate post-operative hyperglycaemia, a mechanism known as the Randle cycle.20 Clinically, increased circulating concentrations of NEFA are associated with depressed myocardial contractility. Accumulation of toxic fatty acid derivatives causes membrane damage, arrhythmias and increased myocardial oxygen consumption.8,21 In our study, plasma NEFA levels were not affected by dexamethasone prophylaxis suggesting that low-dose dexamethasone is devoid of relevant effects on lipolysis. As the actual NEFA concentration is a function of the relationship between fatty acid release by adipose tissue and fatty acid re-uptake by adipocytes and the liver, NEFA plasma levels are only poor indicators of whole body lipolysis. Only direct measurement of NEFA kinetics, for example, using isotope tracer techniques, would allow any insight into changes of the absolute lipolysis rate.
Even a small dose of 8 mg dexamethasone increases the risk for hyperglycaemia in the perioperative period in non-diabetic patients, but was not associated with changes in the perioperative NEFA levels. Thus, the benefits of administering corticosteroids should be weighed against the potential side-effects of hyperglycaemia. Hyperglycaemia seems to be short lasting, but increased caution should be exercised with patients with risk factors for impaired glucose tolerance until studies on the effect of glucocorticoid administration in diabetic patients or those with an impaired glucose tolerance are available.
The authors have made substantive contribution to design of the study, its conduction and the preparation of the article. Only institutional funding has been obtained. None of the authors has personal or financial interests in drugs or methods used in this trial.
1 Rich MW, Abdulhayoglu G, DiSaia PJ. Methylprednisolone as an antiemetic drug during cancer chemotherapy: a pilot study. Gynecol Oncol 1980; 9:193–198.
2 Rasmussen ML, Mathiesen O, Dierking G, et al
. Multimodal analgesia with gabapentin, ketamin and dexamethasone
in combination with paracetamol and ketorolac after hip arthroplasty: a preliminary study. Eur J Anaesthesiol 2010; 27:324–330.
3 Gan TJ, Meyer TA, Apfel CC, et al
. Society for Ambulatory Anesthesia guidelines for the management of postoperative nausea and vomiting. Anesth Analg 2007; 105:1615–1628.
4 Holte K, Kehlet H. Perioperative single-dose glucocorticoid administration: pathophysiologic effects and clinical implications. J Am Coll Surg 2002; 195:694–712.
5 Anderson RE, Klerdal K, Ivert T, et al
. Are even impaired fasting blood glucose levels preoperatively associated with increased mortality after CABG surgery? Eur Heart J 2005; 26:1513–1518.
6 Doenst T, Wijeysundera D, Karkouti K, et al
. Hyperglycemia during cardiopulmonary bypass is an independent risk factor for mortality in patients undergoing cardiac surgery. J Thorac Cardiovasc Surg 2005; 130:1144–1150.
7 Umpierrez GE, Isaacs SD, Bazargan H, et al
. Hyperglycemia: an independent marker of in-hospital mortality in patients with undiagnosed diabetes. J Clin Endocrinol Metab 2002; 87:978–982.
8 Oliver MF, Opie LH. Effects of glucose and fatty acids on myocardial ischaemia and arrhythmias. Lancet 1994; 343:155–158.
9 Wilson AP, Treasure T, Sturridge MF, Gruneberg RN. A scoring method (ASEPSIS) for postoperative wound infections for use in clinical trials of antibiotic prophylaxis. Lancet 1968; 311:e313.
10 Eisenstein AB, Tiglao C. Effect of endocrine therapy on nutritional status. Am J Publ Health 1964; 54:2068–2074.
11 Perkoff GT, Parker V, McCall J, Taylor FH. Effect of cortisol on glucose metabolism in man. J Lab Clin Med 1963; 62:431–436.
12 Pasternak JJ, McGregor DG, Lanier WL. Effect of single-dose dexamethasone
on blood glucose concentration in patients undergoing craniotomy. J Neurosurg Anesthesiol 2004; 16:122–125.
13 Lukins MB, Manninen PH. Hyperglycemia in patients administered dexamethasone
for craniotomy. Anesth Analg 2005; 100:1129–1133.
14 Morariu AM, Loef BG, Aarts LPHJ, et al
: benefit and prejudice for patients undergoing on-pump coronary artery bypass grafting. Chest 2005; 128:2677–2687.
15 Hans P, Vanthuyne A, Dewandre PY, et al
. Blood glucose concentration profile after 10 mg dexamethasone
in nondiabetic and type 2 diabetic patients undergoing abdominal surgery. Br J Anaesth 2006; 97:164–170.
16 Schricker T, Carvalho G. Abdominal surgery, morbid obesity, age, dexamethasone
, diabetes mellitus and glucose metabolism. Br J Anaesth 2006; 97:751–752.
17 Madan R, Bhatia A, Chakithandy S, et al
. Prophylactic dexamethasone
for postoperative nausea and vomiting in pediatric strabismus surgery: a dose ranging and safety evaluation study. Anesth Analg 2005; 100:1622–1626.
18 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.
19 Nazar CE, Lacassie HJ, López RA, Muñoz HR. Dexamethasone
for postoperative nausea and vomiting prophylaxis: effect on glycaemia in obese patients with impaired glucose tolerance. Eur J Anaesthesiol 2009; 26:318–321.
20 Bonadonna RC, Groop LC, Simonson DC, Defronzo RA. Free fatty acid and glucose metabolism in human aging: evidence for operation of the Randle cycle. Am J Physiol 1994; 266:E501–E509.
21 Liedtke AJ. Lipid burden in ischemic myocardium. J Mol Cell Cardiol 1988; 20(Suppl 2):65–74.