Sw eetand sour: Intraoperativeglycaem iccontrol D aveStory •H ead ofResearch, D epartm ent ofA naesthesia, A ustin H ealth; •A /Prof, D epartm ent ofSurgery, U niversity of M elbourne •C hair, A N ZCA TrialsG roup
Sweet and sour: Intraoperative glycaemic control
Dave Story•Head of Research, Department of Anaesthesia, Austin Health;•A/Prof, Department of Surgery, University of Melbourne•Chair, ANZCA Trials Group
Conflict of interestConflict of interestType 1 diabetes >20 years
- age 23
Pens:
-Aspartate insulin (Novorapid): GM human; 3 X 12 units
-Glargine (Lantus): GM human: 36 units nocte
-BSL – Glucometer 5 seconds: digital + log
Gerich, Am J Med, 2002
Frequency perioperative diabetes Frequency perioperative diabetes
REASON study
4,150 older inpatients
23 hospitals ANZ
22% diabetes, 30-day mortality 5% (OR 1.0)
-27% IHD (20% all)
-26% CRI (16% all)Story et al, Anaesthesia, 2010
Diabetes diagnosisDiabetes diagnosis
• Random BSL >11mmol/L• Fasting BSL ≥ 7 mmol/L• OGTT 2hr BSL >11mmol/L
– Diabetes likely – Diabetes unlikely – Impaired glucose tolerance, >5.5 mmol/L
Diabetes Australia + RACGP,
Diabetes Management, 2009
Diabetes “severity”Diabetes “severity”
Using haemoglobin A1C: HbA1c “A1C”
Hb + glucose irreversibly attached to beta chain
A1C - 3 months; <30 days 50%, 60 to 120 days 25%
A1C Mean BSL
6% 7.5 mmol/L
7% 9.5 mmol/L
8% 11.5 mmol/L
9% 13.5 mmol/L
10% 15.5 mmol/L
Burtis et al, Tietz Textbook Clinical Chemistry, 2006
Not acute Type 1RBG > 11 mmol/L diagnostic
Diabetes Care, 2009
Endorsed by Diabetes Society of Australia
Medical Journal of Australia 2011
Flinders 200911% (262/2360) undiagnosed666 surgical patients52 (8%) known diabetes54 (8%) unknown diabetes
Surgery, ASurgery, A1C1C and infection and infection
490 diabetic patients, non-cardiac
VA Conneticut
Median age 72, median A1C 7.3%
A1C < 7.0%, n= 197, infection 12%
A1C ≥ 7.0%, n= 293, infection 20%
Adjusted OR
A1C ≥ 7.0%, infection OR 2.1 (1.2 to 3.7)
Dronge et al, Arch Surg, 2006
Insulin PumpsInsulin Pumps
• Basal rapid acting infusion + boluses
Killen et al, Anaesth Intensive Care, 2010
HypoglycemiaHypoglycemiaVariation between and withinUS mg/dl = mmol/L X 18 approx 20
200mg/dl = 11.1 mmol/L (approx 10 mmol/L)
Reference Range: 4 mmol/L to 6 mmol/L
3 mmol/L – sympathetic – sweating, hunger
2.5 mmol/L – altered CNS: confusion, diplopia
Eventually coma, deathService, NEJM, 1995
Perioperative diabetesPerioperative diabetes• Limited evidence
Glister + Vigersky, Endocrinol Metab Clin N Am, 2003
• Ask patient:“What and when” - sugar and insulin
• Three parts:-Basal: glargine – avoids ketosis (cf GIK)-Nutritional – rapid (W/H)-Corrective –rapid s/c or IV regular
Assundi + Calles-Escandon, J Hosp Med, 2007Killen et al, Anaesth Intensive Care, 2010
Dumb things with insulinDumb things with insulin
• Forget to take it• Take twice
– short – long
• Take wrong one• Take too much• Eat too much for usual dose• Eat too little for usual dose
Unusual and/or stressful situations
Insulin pumpsInsulin pumps
Beware: Technology + no underlying long acting
Options:• IV regular infusion at basal rate (day surgery)• Continue with pump if confident• Convert to s/c rapid + glargine
Assundi + Calles-Escandon, J Hosp Med, 2007
Killen et al, Anaesth Intensive Care, 2010
What BSL?What BSL?
Aim: 8 mmol/ L (5 to 10 mmol/L)
NICE-SUGAR: 6000 ICU patients
4.5 to 6.0 mmol/L (tight) vs <10 mmol/L (usual)
90 day mortality, Tight control surgical OR: 1.31
Hypos 6.8% vs 0.5%
NICE-SUGAR Investigators NEJM, 2009
Hypo…how much dextrose?Hypo…how much dextrose?• Mild to mod: 3 to 5 mmol/L; severe < 3 mmol/L
• Don’t over treat: target 8 mmol/L (5 to 10 mmol/L)
– Dextrose 5% = 5 g/100ml
– Dextrose 50% = 50 g/100 ml = 5g/10 ml
IV push
• BSL 3-5 Dose = 0.1 g/kg 2ml/kg 5% dextrose
• BSL < 3 Dose = 0.15 g/kg 3ml/kg 5% dextrose
Can’t remember = 150 ml 5% Dextrose (7.5g)
= 15 ml 50%Assundi + Calles-Escandon, J Hosp Med, 2007
Why is the patient hypo?Why is the patient hypo?
• Poor management: eg delay• Mistake in insulin or intake?• Is the problem fixed?
– beware duration too much long-acting• Beware insulin infusions
Most likely…hyperglycemiaMost likely…hyperglycemia
My glucometer on Christmas day…My glucometer on Christmas day…
Most likely…hyperglycemiaMost likely…hyperglycemia• BSL > 10 mmol/L • Hours: unpleasant hyperosmolar, dehydrated• Don’t over treat; target 8 mmol/L (5 to 10 mmol/L)• 80/total daily insulin = 1 unit effect mmol/L BSL• Me: 72 units / day
80/72 = 1.1 mmol/L for 1 unitAdult rule of thumb: BSL - 8 = IV regular insulin
OR S/C rapid
Glister + Vigersky, Endocrinol Metab Clin N Am, 2003
PostopPostop
• Physicians
• Three parts (alternative to insulin infusion) • 0.5 units / kg / day (conservative start)
-Basal: glargine 0.25 units / kg / day-Nutritional –rapid s/c 0.25 units / kg / day-Corrective – rapid s/c
• RABBIT 2, Diabetes Care 2011
Assundi + Calles-Escandon, J Hosp Med, 2007
Higher A1C less tolerant of lower glucoseHigher A1C less tolerant of lower glucose
Egi et al, Crit Care Med, 2010
Concluding thoughts…Concluding thoughts…Balance of probabilities:• A1C in all coronary + vascular patients• ?A1C in others eg 70+
• No DM + A1C >6% - med review – risks• A1C >8% +/- DM - med review
Research: A1C in ANZ populations: complications :RCT usual care vs A1C < 7.0 preop
Measure the blood sugarMeasure the blood sugar
Concluding thoughts…Concluding thoughts…• No evidence for very tight control in OR• Aim: 8 mmol/L (Range: 5 to 10 mmol/L)• Give basal• W/H rapid • Don’t overreact• Use IV regular or s/c rapid to correct• Beware pumps• Antiemetics• D5W is our friend• Endocrine involvement for O/N stay
Measure the blood sugarMeasure the blood sugarAhmed et al, Anaesth Analg, 2005