• 
• 

Three Part Question

In [children] is [prolonged preoperative fasting] associated with [perioperative/postoperative hypoglycaemia]?

Clinical Scenario

A child requires a surgical procedure and therefore must be kept nil-by-mouth for several hours prior to the operation. This precaution reduces the volume and acidity of gastric contents, thereby reducing the risk of aspiration. Although surgery may increase blood glucose via a systemic stress response, the period of pre-operative starvation may cause children to become profoundly hypoglycaemic(as they are less able to regulate their blood sugars effectively than adults). This problem is contributed to by the fact that the fasting period is often longer than planned because of bigger surgical lists or poor planning. Are current fasting practices getting the correct balance between prevention of complications and safe blod sugars?

Search Strategy

MEDLINE 1950 – June week 1 2010
EMBASE 1980 to 2010 Week 23
CINAHL 1982 to June week 1 2010
Using the OVID interface.

((exp Preoperative Period/ or exp Preoperative Care/ or preoperative.mp.) OR (exp Anesthesia, General/ or exp Anesthesia/ or Anesthesia, Intravenous/ or an*esthesia.mp. or exp Anesthesia, Inhalation/) OR (an*esth$.mp.)) AND (Fasting.mp. or exp Fasting/) AND ((exp Hypoglycemia/) OR (hypoglyc$.mp.)) (limit 9 to (english language and humans and ("newborn infant (birth to 1 month)" or "infant (1 to 23 months)" or "preschool child (2 to 5 years)" or "child (6 to 12 years)" or "adolescent (13 to 18 years)"))

Search Outcome

MEDLINE found 45 results, of which 23 were relevant and of sufficient quality to be included. Three of these papers were review articles and as such not suitable for full critical appraisal but will be discussed in a later section.
One of the papers could not be obtained using the John Rylands University Library’s Inter-library loans service, and could therefore not be appraised.
EMBASE and CINAHL found no further papers.
Searching of the references and hand searching found 3 more relevant papers.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Gupta. P et al 2004 India	56 children aged 12 and under admitted for elective surgery. Divided into 6 age groups (0-1month, 1-6 months, 6months-2yrs, 2-6yrs, 6-14 yrs).	Cohort Study (level 3a)	Possible correlation between age, weight and fasting plasma glucose	Correlation not significant (p=0.828 for age and p=0.523 for weight)	Lack of Blinding. Method of randomisation not specified. Small sample size of 56 children. Lack of sample size estimates.
			Correlation between length of fast and plasma glucose level	Correlation not significant except in 2-6yrs age group where p=0.027
			Five patients showed signs of hypoglycaemia, only one of whom had a BM within the hypoglycaemic range (defined as <40mg/dl for neonates and <60mg/dl for infants and children). One child was hypoglycaemic but asymptomatic.	Child 1= Headache, age 1.5, fasted for 16 hrs, fasting plasma glucose of 100mg/dl (5.55mmol/L).
Bush. G et al 1996 United Kingdom	One four year old male child admitted for elective circumcision. He was fasted from midnight, and anaesthesia induced at 08.45am.	Case Study (level 4)	Severe Hypoglycaemia (with long term sequalae of epilepsy and cerebral damage)	Blood Glucose level of 1.7mmol/L measured at 21.25pm. Had probably dipped even lower prior to this reading.	Unknown whether hypoglycaemia would still have occurred in the absence of propranolol. Extremely small sample size (one child only).
Somboonviboon. W et al 1996 Thailand	84 healthy children aged 1 month to 13 years admitted for outpatient surgery. Divided into three groups according to age (Group 1 = <1year, Group 2= 1-5 years, Group 3 = 5+). Variation in length of time fasted according to age. Hypoglycaemia defined as BM <60mg/dL (3.33 mmol/L).	Cohort Study (level 3a)	Variations in length of time fasted according to age.	(Time in hours, Mean ±S.D) Group 1= 8.52±2.69 Group 2= 10.92±2.18 Group 3= 11.89±2.76
			Preoperative Blood Glucose level	(Time in minutes, Mean ±S.D) Group 1= 40.53±18.46 Group 2= 33.11±16.89 Group 3= 32.03±16.94
Somboonviboon. W et al 1996 Thailand	84 healthy children aged 1 month to 13 years admitted for outpatient surgery. Divided into three groups according to age (Group 1 = <1year, Group 2= 1-5 years, Group 3 = 5+). Variation in length of time fasted according to age. Hypoglycaemia defined as BM <60mg/dL (3.33 mmol/L).	Cohort Study (level 3a)	Variations in length of time fasted according to age.	(Time in hours, Mean ±S.D) Group 1= 8.52±2.69 Group 2= 10.92±2.18 Group 3= 11.89±2.76	Lack of blinding. Lack of randomisation. No sample size estimates. Big variation in ages, and therefore fasting times, may cause inaccuracies. Variation in length of time under anaesthetic meant that postoperative readings were taken at different times between children.
			Variation in length of time under anaesthetic	(Time in minutes, Mean ±S.D) Group 1= 40.53±18.46 Group 2= 33.11±16.89 Group 3= 32.03±16.94
			Postoperative Blood Glucose level	(mg/dl ±S.d) Group 1= 129.07±37.90 Group 2= 115.62±29.63 Group 3= 111.53±23.07, No Significant correlation between length of fast and post-op BM (p >0.05)
Amponsah. G et al 1993 Ghana	100 children admitted for surgery (60 major operations and 40 minor), divided into four groups according to age: Group 1= < 1 year (17 children) Group 2= 1-5 years (49 children) Group 3= 6-10 years (26 children) Group 4= >10 years (8 children) Hypoglycaemia was defined as blood sugar <4.4mmol/L. All patients had ‘overnight fast’	Cohort Study (level 3a)	Variation in mean body weight between groups (Kg)	Group 1= 6.81 Group 2= 11.43 Group 3= 21.71 Group 4= 28.49	Lack of Blinding Lack of randomisation No sample size estimates. Did the health of the children affect their physiological response to fasting? Variation in length of time fasted. 4.4mmol/L and below considered hypoglycaemic- too high?
			Variation in duration of starvation (hours)	Group 1= 14.1 ± 4.65, Group 2 = 14.98 ± 4.61, Group 3= 17.3 ± 2.73, Group 4= 17.2 ± 1.94
			Blood Sugar Concentrations (mmol/L) immediately post-induction	Group 1= 4.14± 1.1, Group 2= 4.06 ±0.96, Group 3= 4.47± 0.79, Group 4= 4.7 ± 0.83
			Percentage of patients who were defined as hypoglycaemic ( BM <4.4 mmol/L)	Group 1= 65%, Group 2= 61%, Group 3= 50%, Group 4= 25%
Shah. M et al 1990 Pakistan	104 children aged 6 months – 10 years, undergoing inpatient anaesthesia for minor procedures (e.g. tonsillectomy). All were fasted from midnight the night before until induction of anaesthetic (mean 10.87hrs ±2.68hrs).	Cohort Study (level 3a)	Pre- induction hypoglycaemia (mmol/L) (defined as blood sugar of 2.7-3.5 mmol/L)	Detected in 4 children (5.8%), distributed as follows: Child 1- 1 year old girl, fasted for 9hrs total, Pre induction BM=3.3.
			Increase in Blood Sugar seen in samples taken 30 minutes after induction in hypoglycaemic children (except in one case) (mmol/L)
			Possible correlation between pre-induction BM and age
Shah. M et al 1990 Pakistan	104 children aged 6 months – 10 years, undergoing inpatient anaesthesia for minor procedures (e.g. tonsillectomy). All were fasted from midnight the night before until induction of anaesthetic (mean 10.87hrs ±2.68hrs).	Cohort Study (level 3a)	Pre- induction hypoglycaemia (mmol/L) (defined as blood sugar of 2.7-3.5 mmol/L)	Detected in 4 children (5.8%), distributed as follows: Child 1- 1 year old girl, fasted for 9hrs total, Pre induction BM=3.3, Child 2- 2 year old boy, fasted for 13hrs total, Pre induction BM= 3.3, Child 3- 3 year old boy, fasted for 10hrs total, Pre induction BM= 3.2, Child 4- 4 year old boy, fasted for 15hrs total, Pre-induction BM= 2.7	Lack of Blinding No randomisation Lack of sample size estimates No information given about contents of pre-medication Children used may respond differently to fasting than those living in the developed world.
			Increase in Blood Sugar seen in samples taken 30 minutes after induction in hypoglycaemic children (except in one case) (mmol/L)	Increase between pre to post-induction BM- r= 0.466
			Possible correlation between pre-induction BM and age	No significant corellation (r=0.068)
Aun. C et al 1990 United Kingdom	20 healthy children aged 1-5years presenting for minor surgery. All were given a milk-based drink at midnight, then nothing further was given. Then randomly allocated into two groups according to feed given 4hrs before surgery. Group 1 were given nothing/water Group 2 were given 5% dextrose solution.	Randomised control trial (level 1b)	No preoperative hypoglycaemia was found in either group	Average Blood Sugar (mmol/L) Group 1= 4.64, Group 2= 4.58	No sample size estimates. Lack of blinding. Method of randomisation not stated. Time when last solid food given is not specified. The pre-med given to all children was sugar based.
			Plasma glucose found to be increased intra-operatively and post operatively when compared to pre-op values. Peak at 30 mins post op.	Increase statistically significant (p <0.01)
			Mean plasma glucose thought to be higher in fasted group.	No significant difference between groups.
			Increased concentrations of glucagon, cortisol and growth hormone postoperatively.	Increase from pre-operative values statistically significant (p<0.01)
			Correlation between volume of gastric contents (aspirated immediately after induction) and age.	Significant correlation in the fasted group (p=0.03)
			Correlation between volume of gastric contents (aspirated immediately after induction) and body weight.	Significant correlation in the fasted group (p=0.02)
Long. T et al 1989 United Kingdom	One 6 year old boy admitted for adenotonsillectomy. He was fasted from midnight until anaesthesia was induced	Case Report (level 4)	Symptoms of profound hypoglycaemia were experienced when attempting to induce anaesthetic	At 10.45 am, fingerprick glucose measurement estimated a blood sugar reading of 2mmol/L. Analysis of plasma glucose showed blood glucose to be 3.2mmol/L.	Small sample size of 1 child. Conducted retrospectively.
O'Flynn. P et al 1989 United Kingdom	34 healthy children (aged between 11 and 99 months) admitted for daycase grommet insertion. All were instructed not to eat after midnight the night before surgery.	Cohort Study (level 3a)	Large variation in length of time between last meal and induction of anaesthesia	Parents advised to fast children from midnight but many were fasted much longer	No sample size estimates. Small sample size of 34 children. Lack of blinding and randomisation.
			Preoperative hypoglycaemia	Found in 3 of the 34 children (8.8%) at induction. Blood sugars (mmol/L) were as follows: Child 1= 2.3 Child 2= 2.7 Child 3= 2.9
Van der Walt et al 1990 Australia	62 infants aged less than three months presenting for routine surgery. Divided into two groups according to which feed they were given 4 hours pre-operatively. Group 1- Breast Milk (30 infants) Group 2- Formula Milk (32 infants) After this feed, all infants were fasted.	Randomised Control Trial (level 1b)	Variation in age between groups	Average age for groups (days): Group 1= 48 ±24, Group 2= 72± 22	Lack of Sample size estimates. No blinding. No randomisation.
			Blood Sugar measured at induction of anaesthesia.	No hypoglycaemia in either group
			Blood sugar measured intraoperatively, in some cases following the administration of IV fluids	No hypoglycaemia in either group
Welborn. L et al 1986 United States of America	446 healthy children aged 1 month- 6 years, presenting for minor surgery. Divided into two groups according to intraoperative fluids given: Group A – Lactated Ringer’s solution Group B- 5% Dextrose and Lactated Ringer’s solution. All fasted from midnight, but allowed clear fluids thereafter until 4-6hrs before surgery.	Randomised Control Trial (level 1b)	Variation in mean fasting time (hours), dependent on age	0-1 years= 6.1, 1-2 years= 11.0 , 2-4 years= 12.2, 4-6 years= 12.7	Lack of blinding Method of randomisation not specified No sample size estimates Wide range in time initial blood sample was taken
			Two cases of hypoglycaemia at induction	Blood sugars of 2.61mmol/L and 2.2mmol/L at induction of anaesthesia. Both occuring in children who had a prolonged fast.
			Increase in pre-operative-post operative blood glucose values (mg/dl)	Increase from pre-op to post-op BM was statistically significant across all groups (p<0.0001)
Van der walt. J et al 1986 Australia	123 infants aged between 5 days and 12 months presenting for routine surgery. Fasted for 6hrs pre-op, then randomly allocated into one of four groups according to feed given 3-4 hrs pre-op. Group 1-20% Poly Joule Group 2- 5% Dextrose Group 3- Cow’s Milk Group 4- Control Group Hypoglycaemia defined as plasma glucose <2.2mmol/L	Randomised Control Trial (level 2a)	Plasma glucose (mmol/L) measured immediately before induction of anaesthesia	No hypoglycaemia in any group	Lack of blinding. Method of randomisation not specified. Lack of sample size estimates. Large variation in length of fast.
			Plasma glucose (mmol/ L) measured 5 minutes after induction	No hypoglycaemia in any group
			Difference between pre and post-induction plasma glucose measured (mmol/L)	Increase statistically significant (p<0.001)
Redfern. N et al 1986 United Kingdom	54 healthy children children aged between 1 and 5 years. Divided into two groups according to scheduling of surgery. Group 1- fasted from midnight the night before. Morning surgery. Group 2- given cereal at breakfast time them fasted for afternoon surgery.	Randomised Control Trial (level 1b)	Mean fasting time	Significantly longer for patients fasted overnight for morning surgery (mean 14.9 hrs) than those in afternoon (mean 8.8hrs) (p<0.001)	No sample size estimates. Lack of blinding. Method of randomisation not specified. Pre-med contained sugar so may have affected blood glucose concentrations.
			Mean post-induction glucose level	Slightly lower in afternoon group (mean 4.4mmol/L) than morning group (4.8mmol/L) (despite shorter fast) (P<0.05)
			Possible correlation between length of individual fast and post-induction blood glucose level	No significant corellation between length of fast and blood sugar
			Intraoperative increase in glucose level compared to pre-op	Significant increase in both groups (p<0.001)
			Hypoglycamia (defined as BM< 2.6mmol/L)	No hypoglycaemia in either group
Payne et al 1984 United Kingdom	100 healthy children under 5 years and under 20Kg weight. Divided into two groups- those under and those over 1 year old. They were further subdivided according to weight- those below the 3rd centile, those between the 3rd and 25th centiles and those over the 25th centile. Children fasted from midnight, those under 1 given milk at 02.00am. All children recived 5% dextrose solution according to weight 4 hours before surgery.	Randomised control trial (level 1b)	Those children below the 3rd centile for weight are more likely to become hypoglycaemic (defined as BM < 3mmol/L)	In those children over 1 year of age (group A), there was a statistically significant tendency to develop hypoglycaemia (8 out of 22 children were shown to have a BM <3mmol/L before or immediately after induction) (p<0.05)	Lack of blinding. Method of randomisation not specified. No sample size estimates. Confusing grouping of patients.
			Two cases of extremely low plasma glucose in children below the 3rd centile.	Seen in two two-year old marasmic children weighing 9kg and 8.3kg. Post induction their BMs were 1.8mmol/L (child 1) and 2.8mmol/L (child 2). Child 1’s BM increased slightly post-op to 3.3mmol/L, but child 2’s BM decreased post-operatively to 1.9mmol/L.
			Two cases of hypoglycaemia in children above the 3rd centile.	Both aged <15 months, with pre-op fasts of 7.5 and 10hours.
Srinivasan. G et al 1986 United States of America	16 infants and neonates scheduled for elective surgery with baseline glucose levels <150mg/dl (8.33mmol/L)	Cohort Study (level 2b)	Possible correlation between blood glucose concentration and weight	Postsurgical plasma glucose levels showed negative correlation with weight (p<0.01)	Length of fast not specified. Very small sample of 16 children. No sample size estimates. Lack of blinding or randomisation.
			Some post operative hyperglycaemia	Seven infants had blood sugars higher than 13.8mmol/L. On average these children were lighter than those children who did not develop hyperglycaemia (p<0.01)
			Significant inrease in blood glucose post-operatively from pre-op value	p<0.01. Non hypoglycaemia in any child.
Nilsson. K et al 1984 United Kingdom	70 healthy children under 2 years. All were given breast milk 4 hours before start of surgery. Then divided into two groups: Group A received Ringer acetate solution intraoperatively Group B received ringer acetate with 2% glucose	Randomised control trial – level 1b.	Group given glucose had larger increase when comparind pre and post-operative BMs	Difference significant (p<0.001)	No definition of hypoglycaemic range Lack of Blinding Method of randomisation not specified Small sample of 70 No sample size estimates
			No post-operative hypoglycaemia	Lowest recorded BM= 2.9
Jensen et al 1982 United Kingdom	134 healthy children admitted for minor surgery. Divided into inpatient and outpatient groups. Further subdivided according to age: Group I- children <4yrs Group II- children >4yrs All outpatients were fasted from bedtime. Inpatients were divided into two groups: Group A were fasted from bedtime (46 children, of whom 24 were under 4 years(Group I)). Group B were given a fruit syrup in amounts according to body weight (36 children, of whom 15 were under 4 years (group I).	Randomised control trial- 1b.	Potential difference in blood glucose concentrations between inpatients and outpatients	No significant difference between groups.	Lack of blinding. Method of randomisation not specified. No sample size estimates. Confusing grouping of patients.
			Possible correlation bertween length of fast and post-operative blood sugar	No correlation in any group (p>0.1)
			Children over 4 years given fruit syrup had significantly higher blood sigars than those fasted from midnight	(0.02
Allison et al 1982 United Kingdom	92 children aged between 8 months and 8 years, fasted overnight prior to anaesthesia.	Cohort Study (level 2b)	Symptomatic hypoglycaemia	None of those children becoming hypoglycaemic were symptomatic	No sample size estimates. Lack of blinding or randomisation.
			Hypoglycaemia (defined as blood sugar <3.3mmol/L)	Overall (n=92), 11% of children became hypoglycaemic. In the group of children under 4 years old (n=30), 23% became hypoglycaemic.
			Severe hypoglycaemia (blood sugar < 2.2mmol/L)	One child (3%) had a blood sugar <2.2mmol/L
			Risk of hypoglycaemia related to weight of the child	Those children under the 25th centile for weight (n= 28) had a mean plasma glucose of 4mmol/L. Those children above the 25th centile had an average BM of 4.6mmol/L.
Bevan et al 1978 United Kingdom	243 healthy children aged 5 months – 12 years undergoing inpatient procedures. Divided into two groups: ‘Starved Group’- 142 children, fasted overnight if morning surgery, or for at least 8hrs if in afternoon. ‘Fed Group’ – 101 patients. Given ‘gastro-caloreen’ (a glucose polymer dissolved in water) 3-5 hrs after the previous night’s evening meal. Given more at 06.00 on day of surgery and every 4 hrs thereafter until 3-4 hours before surgery.	Randomised control trial (level 1b)	Mean blood glucose higher in fed patients than starved.	In the ‘starved’ group twenty-four	Does not taken into account protein requirement (only carbohydrate) Lack of blinding. Method of randomisation not specified. No sample size estimates. Does not mention that giving sugar may affect insulin secretion.
Ffoulkes. D et al 1976 Nigeria	28 healthy children between 2 months and 15 years. BM was measured 15 minutes before anaesthesia, 15 minutes after induction and 15 minutes after start of surgery.	Randomised control trial- level 1b.	Two cases of hypoglycaemia (defined as BM <45mg/dl)	Children had pre-anaesthetic BMs of 45mg/dl (2.77mmol/L) and 27mg/dl (1.5mmol/L) respectively.	Lack of blinding. Method of randomisation not specified. Small sample size of 28. No sample size estimates. No p values given therefore difficult to assess validity.
			Cases of hypoglycaemia occurred in children under 2 years weighing 3.2-12.7Kg.	Children under two with lower body weights had lower blood sugars than heavier children, following a comparative fast. ?statistically significant
Ware et al 1976 United Kingdom	57 children under 4 years old undergoing surgery.	Cohort study (level 3a)	Six children were found to be hypoglycaemic post operatively overall, many of whom were below the 10th centile for weight	Mean pre-op glucose in children below the 10th centile= 52.6±16.1, Mean pre-op glucose in children above the 10th centile= 64.5±19.7, (0.2>P>0.1, said to be statistically significant using students t test)	Dextrostix used to measure blood glucose and may have been inaccurate in some cases. No sample size estimates. Relatively small sample of 57 children. Lack of blinding and randomisation.
			2 hours postoperatively, one child was hypoglycaemic	Blood sugar of 37mg/dl (2.05mmol/L)
			4 hours post –op, three were hypoglycaemic	Blood sugars of 1.83, 1.55 and 2.05 (mmol/L)
			6 hours post-op, two children were hypoglycaemic	Blood sugars of 1.88 and 2.05 (mmol/L)
Thomas et al 1974 United Kingdom	62 children aged between 19 months and 166 months, undergoing afternoon surgery for correction of strabismus. All children were fasted overnight then received a bowl of cereal at 06.00. Divided into two groups: Starvation group (SG) received no further food. Milk Group (MG) were given milk to drink 4 hrs before surgery.	Randomised control trial (level 2a)	Group given glucose had higher mean blood glucose	Difference between groups significant (p<0.0005)	No sample size estimates. Realtively small sample size of 62. Lack of blinding. Method of randomisation not specified.
			Cases of hypoglycaemia different between groups	In the milk group, no hypoglycaemia occurred. In the starved group, 15.2% were hypoglycaemic (BG<40mg/dl (2.22mmol/L)). Difference between groups significant (p=0.05)
			Within the starvation group, the smaller children were the ones who became hypoglycaemic	Children in SG <47months old and <15.5Kg (18 children)= 5 became hypoglycaemic (28%), Children in SG >47months old and >15.5Kg (15 children)= none were hypoglycaemic.
			In children <47 months old and <15.5Kg in weight, there were differences in blood sugar between the MG and the SG	Starved group mean blood sugar level = 46.4mg/dl (2.57mmol/L), Milk Group mean blood sugar level = 60.0 (3.33mmol/L), Difference significant between groups (p<0.0005)
Watson et al 1972 United Kingdom	80 children between 22 months and 15 years presenting for surgery (combination of major and minor.	Randomised control trial (level 1b)	Overall increase in blood glucose concentrations in response to surgery	Increase from pre-operative to post-operative values were statistically significant (p<0.001)	Length of pre-operative fast not specified. Lack of blinding No sample size estimates Method of randomisation not specified
			Eight patients (10%) were hypoglycaemic pre-operatively (defined as blood sugar < 40mg/dl)	The following blood sugars were recorded at induction (mmol): 28, 40, 38, 30, 30, 40, 30, 25. No significant correlation between type of surgery (major/minor), or length of fast and risk of hypoglycaemia.

Comment(s)

Being hypoglycaemic is problematic during surgery as it can affect the way in which a child behaves physiologically, both when under anaesthetic and when regaining consciousness afterwards. It is therefore important to prevent this occuring, whilst balancing this with the prevention of aspiration pneumonitis. The findings of the papers studied are extremely variable. Some of the older papers (Thomas et al, Watson et al, Bevan et al and Allison et al) found worryingly high occurance of severe hypoglycaemia. Several of the papers found that hypoglycaemia was affecting a smaller but significant minority of children (Shah et al, O'flynn et al, Welborn et al and Jensen et al). Despite variation in definitions of hypoglycaemia and results, there is a common trend running throughout many of the papers. This is that the youngest, and therefore smallest, children are less able to regulate their blood sugars effectively following a period of starvation. This was the case in many of the papers studied, as demonstrated by Payne et al (16), who found that 36% of those children over 1 year but under the 3rd centile for weight were hypoglycaemic when anaesthetic was induced, and found hypoglycaemia at some point during the anaesthetic (BM< 3mmol/L) in 11% overall. However, several of the papers read demonstrated that the children studied were able to readily regulate their blood sugars without becoming hypoglycaemic (Somboonviboon et al, Van der Walt et al, Shah et al, Aun et al, Nilsson et al and Srinivasan at al). All papers that measured blood glucose postoperatively confirmed the expected hypothesis that blood glucose would increase after surgery due to a physiological stress response. One or two children did not fit this pattern, but many of these were hypoglycaemic before surgery and may have behaved differently because of this. Many of the papers studied were several decades old and only obtained data from a small cohort; therefore perhaps more detailed research into the area is required in the form of large randomised control trials comparing starved children to those allowed a source of glucose pre-operatively. Further evidence may add weight to one side of the argument and enable clinicians to decide on a definite ‘safe’ fasting time, or on whether to provide pre-operative glucose.

Clinical Bottom Line

It is difficult to draw a definite conclusion on the issue with such varied results; however the research and this paper have highlighted the issue as an important one which must be considered further due to the potentially disastrous consequences of prolonged hypoglycaemia. Many of the papers suggest that children are able to regulate their blood sugar without food, as long as fasting is not protracted. An important trend in many papers is that younger or lighter children are the ones becoming hypoglycaemic, perhaps indicating that specific care must be taken when considering the length of pre-operative fast in this group of patients. Almost all of the papers showed that children remained asymptomatic despite having low blood sugars, perhaps suggesting that much hypoglycaemia goes undetected in clinical practice where blood sugars are not measured as routine at induction of anaesthetic.

References

1. Murat. I et al Perioperative Fluid Therapy in Pediatrics Pediatric Anaesthesia 2008, (18); 363-370
2. Gupta. P et al Preoperative Fasting and Plasma Glucose levels in Children Journal of Anaesthetics and clinical Pharmacology 2004, (20)2; 161-163
3. Bush. G et al Severe hypoglycaemia associated with preoperative fasting and intraoperative propranolol. A case report and discussion Paediatric Anaesthesia 1996 6: 415–417
4. Somboonviboon. W et al Blood Glucose Concentration in Pediatric Outpatient Surgery Journal of the Medical Association of Thailand 1996 Apr;79(4):236-9.
5. Somboonviboon. W et al Blood Glucose Concentration in Pediatric Outpatient Surgery Journal of the Medical Association of Thailand 1996 Apr;79(4):236-9.
6. Phillips. S et al Preoperative fasting for paediatric anaesthesia- Review Article British Journal of Anaesthesia 1994, (73); 529-536
7. Aun. C et al Preoperative Fasting in Children- Review Article Annals of the Academy of Medicine Singapore 1994, (23); 572-578
8. Huang SJ et al Prolonged fasting in pediatric outpatients does not cause hypoglycemia Anaesthesiologica Sinica 1993 Dec;31(4):249-52
9. Amponsah. G et al Blood sugar concentrations in children undergoing surgery under general anaesthesia West African Journal of Medicine Medicine1993 Apr-Jun;12(2):89-92
10. Shah. M et al The effect of Pre-anaesthetic fasting on blood glucose level in children undergoing surgery Journal of the Pakistani Medical Association 1990, Oct;40(10):243-5.
11. Shah. M et al The effect of Pre-anaesthetic fasting on blood glucose level in children undergoing surgery Journal of the Pakistani Medical Association 1990, Oct;40(10):243-5.
12. Aun. C et al Paediatric glucose homeostasis during anaesthesia British Journal of Anaesthesia 1990 1990, (64); 413-418
13. Long. T et al An unusual case of symptomatic hypoglycaemia in a child Anaesthesia 1989, (44); 765-766
14. O'Flynn. P et al Fasting in children for day case surgery Annals of the Royal College of Surgeons or England 1989, (71); 218-219
15. Van der Walt et al A Study of Preoperative Fasting in Infants Aged Less Than Three Months Anaesthesia and Intensive Care 1990, (18); 527-531
16. Welborn. L et al Perioperative blood glucose concentrations in pediatric outpatients Anesthesiology 1986, (65); 543-547
17. Van der walt. J et al The effect of different Pre-operative feeding regimens on plasma glucose and gastric volume and pH in Infancy Anaesthetics and Intensive Care 1986, (14); 352-359
18. Redfern. N et al Blood Glucose in anaesthetised children Anaesthesia 1986, (41); 272-275
19. Payne et al Plasma glucose levels in the peri-operative period in children Anaesthesia 1984, (39); 868-872
20. Srinivasan. G et al Glucose homeostasis during Anaesthesia and Surgery in Infants Journal of Pediatric Surgery 1986, (21)8; 718-721
21. Nilsson. K et al Blood-Glucose Concentrations During anaesthesia in Children British Journal of Anaesthesia 1984 1984, (56); 375
22. Jensen et al Preoperative starvation and blood glucose concentrations in children undergoing inpatient and outpatient anaesthesia British Journal of Anaesthesia 1982, (54); 1071
23. Allison et al Pre-operative starvation in children -The role of alanine in blood glucose homeostasis Anaesthesia 1982. (31); 274-271
24. Bevan et al Acid base changes and Anaesthesia-The influence of pre-operative starvation and feeding in paediatric surgical patients Anaesthesia 1978, (28); 415-422
25. Ffoulkes. D et al Effect of anaesthesia and surgery on blood sugar and carbohydrate tolerance in African Children Canadian Journal of Anaesthesia 1976, Sep;23(5):486-91
26. Ware et al Postoperative Hypoglycaemia in small children The British Medical Journal 1976, (2); 499-501
27. Thomas et al Hypoglycaemia in Children before operation: Its incidence and prevention British Journal of Anaesthesia 1974, (46); 66-68
28. Watson et al Blood glucose levels in children during surgery British Journal of Anaesthesia 1972, (44); 712