Role of preoperative carbohydrate loading: a systematic review

INTRODUCTION Surgical stress in the presence of fasting worsens the catabolic state, causes insulin resistance and may delay recovery. Carbohydrate rich drinks given preoperatively may ameliorate these deleterious effects. A systematic review was undertaken to analyse the effect of preoperative carbohydrate loading on insulin resistance, gastric emptying, gastric acidity, patient wellbeing, immunity and nutrition following surgery. METHODS All studies identified through PubMed until September 2011 were included. References were cross-checked to ensure capture of cited pertinent articles. RESULTS Overall, 17 randomised controlled trials with a total of 1,445 patients who met the inclusion criteria were identified. Preoperative carbohydrate drinks significantly improved insulin resistance and indices of patient comfort following surgery, especially hunger, thirst, malaise, anxiety and nausea. No definite conclusions could be made regarding preservation of muscle mass. Following ingestion of carbohydrate drinks, no adverse events such as apparent or proven aspiration during or after surgery were reported. CONCLUSIONS Administration of oral carbohydrate drinks before surgery is probably safe and may have a positive influence on a wide range of perioperative markers of clinical outcome. Further studies are required to determine its cost effectiveness.

Insulin resistance is a central metabolic change during surgical stress that is directly proportional to the magnitude of the operation.It causes hyperglycaemia in non-diabetic patients.As a consequence, various endocrine and infl ammatory systems are stimulated.This results in an exacerbation of the existing postoperative catabolic state with marked loss of body fat and protein stores. 1,2Aggressive treatment with insulin to maintain glycaemic control has resulted in reduced organ dysfunction and mortality. 3,4Additionally, insulin resistance has been shown to be an independent factor infl uencing length of stay in hospital postoperatively. 1 The aim of this review is to systematically appraise the available data regarding the safety and benefi cial role of preoperative carbohydrate loading in patients undergoing surgery and, where possible, make comparison with placebo or traditional practice.

Methods
A PubMed literature search was undertaken using the keywords 'carbohydrate loading', 'preoperative', 'surgery' and 'insulin resistance'.Search limits consisted of any  article published up until September 2011, studies involving adults undergoing general surgical operations and English language manuscripts.The references of all articles were cross-checked to include all pertinent articles (Fig 1).The primary outcome measure was effect of preoperative carbohydrate loading on insulin resistance.Secondary outcome measures were the effect of carbohydrate treatment on gastric emptying, gastric acidity, wellbeing of patient (assessed qualitatively), immunity, clinical outcome and nutrition.

Effect of preoperative carbohydrate on insulin resistance
][7][8][9][10][11] Various methods were used to analyse insulin resistance (Table 1).[10][11] One article assessed insulin resistance using the HOMA-IR (homeostatic model assessment -insulin resistance) equation. 7One study used an artifi cial pancreas with a closed loop system (STG-22). 5One study used the quantitative insulin sensitivity check index. 6The time of assessment var-ied from one week preoperatively to up until three days after surgery.Frequency varied from being assessed once postoperatively or on two separate occasions (preoperative and after surgery).0][11] The maximum improvement in insulin action observed was by a factor of 50% (p<0.01)after the morning dose of carbohydrate on the day of surgery. 12n contrast, only one study demonstrated no effect of carbohydrate on postoperative peripheral insulin sensitivity (borderline signifi cance, p=0.049).This may be due to a type II error, the small sample size (12 patients) or the timing of surgery and diverse fasting durations. 8

Effect of preoperative carbohydrate on gastric emptying
][14][15] The protocol used in the trials varied with comparisons made between oral carbohydrate drinks and fasting from midnight or water or oral nutritional supplement or mixture of carbohydrate and soy protein or intravenous glucose and electrolytes.The time and number of drinks administered was also variable.
3][14][15] However, three patients in the study conducted by Hausel et al had large residual gastric fl uid volumes. 15It was noted that one of these patients had a history of previous intestinal obstruction, one had a short interval between intake of drink and premedication, and the third patient had abnormal fasting plasma glucose.

Effect of preoperative carbohydrate on gastric acidity
Three randomised trials examined the effect of carbohydrate drinks given preoperatively on gastric acidity in patients undergoing surgery. 6,13,15Modes of assessing gastric acidity varied between the studies (Table 4).The study by Hausel et al assessed acidity by automatic back titration with sodium hydroxide to pH 7. 15 A study by Yagci et al used a urine pH meter. 13The third trial used biochemical indicator paper. 6All the studies demonstrated that there was no difference in gastric acidity following a carbohydrate drink compared with placebo or intravenous glucose or, more importantly, fasting.

Effect of preoperative carbohydrate on patient wellbeing
16][17]19 The variables measured by the VAS were thirst, hunger, anxiety, depression, pain, tiredness, weakness, inability to concentrate, mouth dryness and nausea.The numbers of variables studied in each trial were different.In the study by Kaska et al, the psychosomatic status of patients was assessed by the modifi ed Beck questionnaire, 6 which consists of 21 questions addressing symptoms such as fatigue and irritability.Hausel et al used two methods: VAS and objective analysis by nursing staff. 18reparation with carbohydrate led to a signifi cant reduction in thirst, hunger, anxiety and malaise in two trials com-  pared with fasting and placebo (fl avoured water). 15,18The improvement in thirst was similar for the placebo and carbohydrate groups.Two trials compared intravenous glucose with fasting from midnight and oral carbohydrate drinks. 6,17hese studies demonstrated that the fasted patients had increased thirst, hunger, tiredness, anxiety and mouth dryness scores.On the contrary, both intravenous and oral carbohydrates alleviated feelings of tiredness and weakness compared with fasting.However, intravenous glucose infusion did not decrease the sense of thirst and hunger as effectively as oral carbohydrates.
Hausel et al investigated the effect of carbohydrate on postoperative nausea and vomiting in 172 patients undergoing an elective laparoscopic cholecystectomy. 18Between 12 and 24 hours after surgery, there was a signifi cantly lower incidence of nausea and vomiting in the carbohydrate group than in the fasted group.Three studies demonstrated no benefi cial effect of carbohydrate drinks on similar variables measuring general wellbeing of the patient prior to surgery. 14,16,19

Effect of preoperative carbohydrate on immunity and clinical outcome
Two trials examined the impact of carbohydrate drinks on postoperative immunity and clinical outcomes. 16,20Mathur et al conducted the largest double blind placebo controlled trial in 2009 to study the effect of preoperative carbohydrate drinks on a number of clinical outcomes after colorectal surgery and hepatic resection. 16There was no difference in the incidence of postoperative infections in the carbohydrate group compared with the placebo group.Furthermore, no signifi cant benefi t was observed in the carbohydrate group with regard to length of stay and time to intake of oral diet.
In contrast to this, Noblett et al demonstrated that preoperative treatment with carbohydrate drinks reduced the length of hospital stay compared with placebo or water. 20arlier return of gut function was also noticed in the carbohydrate group although this was not statistically signifi cant.The early return of gut function could be a contributory factor for reduced postoperative hospital stay.

Effect of preoperative carbohydrate on nutrition
Five studies examined the effect of preoperative carbohydrate on the postoperative nutritional status of the patient. 6,14,16,20,21Varied methods were employed for measurement of nutrition.Four trials used anthropometric measurements. 14,16,20,21A dynamometer measured grip strength in the dominant hand.Other measurements included triceps skinfold thickness and mid-arm circumference.One study measured muscle power in hand grip with a digital tension meter. 6n the study conducted by Henriksen et al, no signifi cant difference was observed between the groups when analysed per se. 14However, when the results of the two intervention groups (carbohydrate only, and carbohydrate and peptide) were pooled together, they had a signifi cantly better muscle strength in the quadriceps muscles than the control (water) group after one month (p<0.05).Despite this, no difference was observed between the three groups in voluntary isometric hand grip strength.Noblett et al demonstrated a signifi cant reduction in grip strength on discharge in the fasted patients compared with their preoperative values (p<0.05). 20n contrast, both the carbohydrate and placebo groups did not show a signifi cant reduction in their postoperative grip strength.Similar results were noted by Kaska et al but the values were not signifi cant. 6uill et al found no signifi cant difference in the body mass index between the carbohydrate and control groups or loss of fat mass from baseline to discharge. 21Nevertheless, preoperative oral glucose improved preservation of muscle mass compared with placebo.In contrast to the above fi ndings, Mathur et al did not notice greater preservation of muscle mass in the carbohydrate group. 16Furthermore, carbohydrate treatment did not ameliorate postoperative nitrogen loss although it did increase the levels of insulin-like growth factor 1 postoperatively.

Effect of preoperative carbohydrate in diabetic patients
Only one trial investigated the effect of carbohydrate drink in diabetic patients. 22The effect of carbohydrate drink was compared in 25 type 2 diabetic patients with 10 healthy con-

Discussion
The traditional practice of fasting patients before surgery results in depletion of hepatic glycogen, enhanced gluconeogenesis and insulin resistance. 23,24This is further aggravated by the insulin resistance caused by surgery. 25The practice of overnight fasting was fi rst challenged in 1994 by Ljungqvist et al in patients undergoing an open cholecystectomy. 11Postoperative insulin resistance was reduced by 50% in patients receiving overnight intravenous glucose infusion.Moreover, in patients receiving glucose infusion, hepatic glycogen content was increased by 65% during surgery compared with fasting patients. 26reoperative thirst has been suggested to be the main contributory factor of patient discomfort, followed by hunger and anxiety. 27Clear drinks alleviate thirst but their effect on hunger is inconclusive. 28,29Use of high carbohydrate drinks preoperatively was pioneered by Nygren et al in 1995. 12It was specially designed, consisting mainly of polymers to reduce the osmotic effect of the drink on gastric emptying.They demonstrated that the carbohydrate drink left the stomach in 90 minutes in patients undergoing a laparoscopic cholecystectomy after ingestion on the morning of surgery.None of the studies analysed in this review reported any adverse events following ingestion of carbohydrate drinks such as apparent or proven aspiration during or after surgery.
Preoperative carbohydrate drinks improved patient wellbeing after surgery signifi cantly, especially hunger, thirst, malaise, anxiety and nausea.However, no benefi t was noted by Bisgaard et al although the values were not statistically signifi cant. 19A combination of heterogeneous surgical procedures, surgical access and anaesthetic protocols introduces a number of variables that could diminish the possibility of detecting any clinical benefi t of carbohydrate drinks. 16A longer fasting time due to delay in the start of surgery and lower carbohydrate dose can also alleviate the effects of carbohydrate.
The review of trials examining the effect of carbohydrate on preservation of muscle mass presents a mixed picture and no conclusion could be drawn regarding the role of preoperative carbohydrate treatment.The varied methodology and outcome measures used could be a contributory factor.Future studies need to be carried out to investigate this further.
Factors that increase the risk of gastric aspiration are pregnancy, obesity, history of metabolic disorders including diabetes, ASA grade >2 and gastrointestinal disorders.They were excluded from all the trials owing to fear of gastric aspiration resulting in pulmonary complications.No evidence was available with regard to the safety of use of carbohydrate drinks in these patients.Diabetic patients are particularly at risk of poor glycaemic control after surgery. 30These patients have been excluded from the majority of the studies because of fear of delayed gastric emptying. 31,32In order to use preoperative carbohydrate loading in diabetic patients, it would be helpful to recognise patients with delayed gastric emptying.4][35] The only study that examined the effects of carbohydrate drinks in diabetic patients was small (35 patients). 22The results cannot therefore be generalised to all diabetic patients.Furthermore, it needs to be explored whether carbohydrate loading has a similar benefi cial effect on the metabolism as in non-diabetic patients.
Various oral carbohydrate preparations have been analysed and compared with placebo or overnight fasting (Table 6).The most commonly used oral formula for preoperative carbohydrate loading in the trials was a 12.5% carbohydrate drink (preOp ® , Nutricia, Trowbridge, UK) in quantities of 400ml or 800ml.It has been shown to be iso-osmolar and found to leave the stomach in 90 minutes with no reported adverse effects.The commercial preparation is available in a 200ml carton.The cost of one carton is £3.50 so it will cost £21 per patient (4 x 200ml in the evening before surgery and 2 x 200ml on the morning of the surgery) per procedure.One should evaluate whether this additional cost is worth the advantageous effects of the carbohydrate drink on clinical outcome.

Conclusions
Administration of oral carbohydrate drinks before surgery is probably safe as it leaves the stomach in 90 minutes and does not affect gastric acidity.It may have a positive infl uence on a wide range of perioperative markers of clinical outcome.There has been considerable focus in improving the recovery times and therefore shortening postoperative stay after both major and minor elective surgical procedures.This ethos has formed the basis of the enhanced recovery programme.Preoperative carbohydrate loading may be a useful adjunct in improving postoperative recovery.Further studies are required, however, to assess its cost effectiveness.

Figure 1
Figure 1 Flow diagram of study selection