• 
• 

Three Part Question

In [diabetic children] should [sliding scale insulin be used perioperatively] to [control blood glucose more tightly and minimise complications]?

Clinical Scenario

A child with diabetes requires a surgical procedure. Most recent hospital guidelines recommend the use of a ‘sliding scale’ for insulin delivery only in those children undergoing major or emergency operations. Administration of insulin intravenously as opposed to subcutaneously may facilitate tighter control of blood sugars throughout a procedure. Would this delivery method be beneficial for all diabetic children having surgery?

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 Diabetes Mellitus, Type 1/ or Diabetes.mp. or exp Diabetes Mellitus/) OR (diabet$.mp.)) AND ((exp Anesthesia/ or exp Anesthesia, General/ or anaesthesia.mp.) OR (an*esth$.mp.)) AND ((exp Insulin Infusion Systems/ or insulin.mp. or exp Insulin/) OR (sliding scale.mp.)) (limit 10 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 40 papers, of which 5 were relevant. Four of these were review articles, and as such only one paper was suitable for full critical appraisal.
EMBASE and CINAHL found no further papers.
Searching of the references and hand searching found no further relevant papers.
Searching the internet for recent recommendations found two major guidelines (one from the Australasian Paediatric Endocrine Group and one from the Canadian diabetes association).

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses
Kaufman et al 1996 United States of America	19 children with type I diabetes mellitus admitted for surgery between July 1989 and June 1992. Divided into two groups according to whether insulin was delivered IV (Group 1) or SC (Group 2). Average age 12.6 ±3.9 years in Group 1 and 9.6 ± 3.9 years in Group 2).	Cohort Study (Level 2b)	Baseline blood glucose level at 2hr before surgery	No significant difference between groups	Data collected retrospectively. Variation in type of surgery being performed. Some children had relatively minor procedures such as tonsillectomy while others underwent more major ones e.g. cardiac surgery. Very small sample size of 19 children.
			Bedside blood glucose 1hr before surgery	Blood glucose levels significantly higher in Group 2 (p<0.05)
			Intraoperative blood sugar levels measured at 0hrs (start of surgery), 0.5 and 1hr into procedure, and every 4 hours thereafter	At 0hrs, 4hrs, 16hrs, 20hrs, 24hrs, 30hrs, 36hrs 54hrs and 72hrs blood glucose was significantly higher in Group 2 than Group 1 (p<0.05)
			Goal to keep all patients blood sugar between 5.5 and 8.3 mmol/L at all times was not achieved	Blood Glucose levels > 11.1mmol/L were recorded immediately post-op and on days 2 and 3 in multiple patients in Group 2. This did not occur at all in Group 1
			Insulin dose (units/kg/24hr) required to maintain blood glucose within target range differed between Groups 1 and 2	Group 1 were required significantly more insulin on day of surgery (p <0.01), and for three days postoperatively (p<0.025)0

Comment(s)

The physiological stress caused by surgery can significantly affect blood glucose concentrations in diabetic children. Surgery increases the secretion of counter-regulatory hormones which work antagonistically against insulin, thereby increasing the child's insulin requirements peri-operatively. The effects of this physiological response, combined with the implications of pre-operative fasting mean that this period can be extremely disruptive for diabetic control in these patients. Effective insulin delivery is necessary during this time to minimise the complications of the blood sugar straying outside an optimum range. Current guidelines at the Royal Manchester Children's Hospital suggest that subcutaneous insulin will be adequate to control the blood sugars of those children undergoing minor, elective surgery lasting less than one hour. It states that for major or longer operations a sliding scale of insulin should be commenced one hour prior to surgery to maintain blood sugars between 7 and 13mmol/L. This also applies to operations where post-operative fasting is required. Insulin is not administered routinely via a sliding scale in shorter or more simple operations, despite the tighter blood sugar control it could possibly provide. The available literature concerning this topic (in terms of papers suitable for critical appraisal) was very limited, as current practice in this area is governed largely by expert opinion. Kaufman et al demonstrated that blood sugars were more effectively kept within a defined range when using inravenous insulin as opposed to subcutaneous. However, this paper does not look specifically at the use of a sliding scale for more basic operations. The current international guidelines of Australia and Canada are the leading recommendations upon which most local guidelines are based. Expert groups from both countries suggest that intravenous insulin is helpful when attempting to control the hypoglycaemia which can result from pre-operative fasting in diabetic children. They do not routinely advocate sliding scale insulin as the 'gold standard' in diabetic children undergoing operations. This is in line with the opinions of several review articles on the topic by Betts et al, Rhodes et al, Chadwick et al and Holvey et al.

Clinical Bottom Line

Although the research into the use of sliding scale insulin for diabetic children undergoing surgery is very limited, the one randomised control trial specific to the area reports that IV insulin is more effective for controlling blood sugars within a tighter range than subcutaneous injection. However, this paper does not look at how the method of delivery should be tailored to the length and nature of the surgery being performed. The evidence must be considered alongside considerable expert opinion in the area which suggests that subcutaneous insulin may be sufficient for planned, short operations. As the data is so limited, it would seem that more randomised control studies should be carried out to ascertain the efficacy of IV insulin both in shorter operations and as an overall method of perioperative insulin delivery in paediatrics.

References

1. Kaufman F, Devgan S, Roe T et al. Perioperative management with Prolonged Intravenous infusion verus Subcutaneous Insulin in children with Type I Diabetes Mellitus Journal of Diabetes and its Complications 1986, (10); 6-11
2. Australasian Paediatric Endocrine Group Clinical practice guidelines: Type 1 diabetes in children and adolescents
3. Canadian Diabetes Association Clinical Practice Guidelines for the Prevention and Management of Diabetes in Canada Canadian Journal of Diabetes September 2008, Volume 32, Supplement 1.
4. Holvey. SM et al Surgery in the Child with Diabetes Pediatric Clinics of North America 1969 Aug;16(3):671-9.
5. Chadwick et al Diabetes Mellitus and the Pediatric Anaesthetist Paediatric anaesthesia September 2004: (9):716-23
6. Rhodes et al Perioperative Management of Pediatric Surgical Patients with Diabetes Mellitus Anesthesia & Analgesia October 2005, 101(4); 986-999
7. Betts et al, International Society for Paediatric and Adolescent Diabetes Management of children and adolescents with diabetes requiring surgery Paediatric Diabetes 2009, (10); 169-174