An Update in Diabetic Ketoacidosis (DKA) and Hyperglycemic - PowerPoint PPT Presentation

An Update in Diabetic Ketoacidosis (DKA) and Hyperglycemic Hyperosmolar State (HHS) Dr Ketan Dhatariya MBBS MSc MD MS FRCP Consultant in Diabetes and Endocrinology Norfolk and Norwich University Hospitals

Financial Disclosures • In the last 12 months I have attended advisory board meetings for – Eli Lilly – Sanofi Aventis – Roche Diagnostics • I have received travel grants from – Genentech – Roche Diagnostics – Eli Lilly – Novo Nordisk • All of my conflicts of interest are available on www.norfolkdiabetes.com

Acknowledgements • Ian Nunney • JBDS • Gillian Iceton • ABCD • Kath Higgins • DISN UK • Mike Sampson • Diabetes UK

Outline • A bit of history • Where we ended up • Where we are now • What might help us get better

DKA

A Bit of History – Pre Insulin • DKA was universally fatal • After the siege of Paris in 1870, it was realised that strict diets could prolong life for a year or two – As promoted by Fredrick Allen in the US and Bernard Naunyn in Germany (he was also amongst the first to advocated the use of alkali’s in ‘diabetic coma’) Bliss M The discovery or insulin – 25 th anniversary edition. University of Toronto Press. 2000

A Bit of History – The Early Years • Between 1923 and 1925, Elliot Joslin reported that 31 out of 33 people he treated for DKA had survived – But with very gentle parenteral fluid replacement • In the UK, RD Lawrence advocated aggressive fluid replacement – 3-6 pints over 1-2 hours – Given with adrenaline, stropanthin and a caffeine enema Joslin EP et al Med Clin N Amer 1925;8:1873-1919 Lawrence RD Br Med J 1930;i(3614):690 Lawrence RD Br Med J 1936;ii(3940):81-82

Fluids • Hence the start of the discussion about rate of fluid replacement

A Bit of History – The Early Years • In 1945, Howard Root in Boston reported how mortality had dropped from 12% between January 1923 and August 1940 to 1.6% between August 1940 and May 1944 – 1923 to 1940, average insulin dose in the 1 st 24 hours was 237 units – 1940 to 1944, average insulin dose in the 1 st 24 hours was 287 units (range 50 to 1770) Root HF JAMA 1945;127(10):557-564

A Bit of History – The Early Years • In 1948, Micks in Dublin advocated – 100 units for those presenting in ‘pre - coma’ – 500 units for those in true coma and then 100 units intravenously every 15 minutes until there was signs of improvement (‘doses of over 2000 units have sometimes been required’) Micks RH Br Med J 1948;ii(4568):200-203

A Bit of History – The Early Years • In 1949, Black and Malins reported a case series of 170 consecutive cases treated with an average of 265 units – Mean of 265 units (140 to 500) of intravenous insulin for those ‘drowsy but rousable’ – Mean of 726 units (250 to 1400) for those ‘rousable with difficulty’ – Mean of 870 units (500 to 1400) for those ‘unconscious on admission’ Black AB & Malins JM Lancet 1949;253(6541):56-59

Insulin • Hence the introduction of high dose insulin infusions – With ‘young and enthusiastic’ junior medical staff constantly by the patients’ bedside – A well equipped laboratory available 24 hours a day

Low Dose Insulin Rate of glucose and β -OHB drop with low dose (5-6 units/hr) insulin infusion Page MM et al Br Med J 1974;ii(921):687-690 Semple PF et al Br Med J 1974;ii(921):694-698

People Were Still Dying • Given 0.1u/kg/hr and 1-2 L of fluid on admission then 1 L every 3-4 hours, and giving potassium 20-40mmol/hour Carroll P et al Diabetes Care 1983;6(6):579-585

So… • DKA was treated with – Fluid – Intravenous insulin – Potassium – ± bicarbonate & phosphate But how much and how fast?

Danger • How do the order sets for DKA vary amongst your local hospitals? • Are they the same or so they have minor differences? • What about across Counties? Across States? • What effect on outcomes do these (minor) differences have?

Where Are We Now? • In 2010 the JBDS produced a guideline on the management of DKA • With > 20,000 hard copies given out or downloaded • An updated guideline was published in late 2013 • A national survey was conducted in Autumn 2014 http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm

Overall Quality of JBDS Guidelines Overall quality of JBDS – IP guidelines in 104 UK Trusts DKA Over 90% of respondents rated the guidelines as ‘Excellent’ or ‘Good’ Data from Mike Sampson. Feedback from 104 hospitals - first shown at Diabetes UK March 2013

Diagnostic Criteria - ADA and JBDS (200mg/dl) Kitabchi AE et al Diabetes Care 2009;32(7):1335-1343 http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm

Markers of Severity – ADA and JBDS Marker of severity Value Marker of severity Value Mental status GCS<12 or abnormal Pulse <60 or >100 bpm APVU score Oxygen saturation <92% on air (assuming Urine output <0.5ml/kg/hr or AKI normal baseline) pH <7.1 Blood ketones >6.0mmol/l Potassium <3.5 or >6.0 mmol/l Bicarbonate <5.0mmol/l Systolic BP <90mmHg Anion gap >16.0mmol/l Kitabchi AE et al Diabetes Care 2009;32(7):1335-1343 http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm

A Question • How do we know that what we are doing is correct?

Where Are We Going?

Results • 283 forms were received from 72 hospitals between May and November 2014 • There are hundreds of messages in the data! • A few of the main ones are:

Precipitants (%) Dhatariya K et al Submitted for publication

Fixed Rate Intravenous Insulin • The use of 0.1units/kg/hr led to excellent rises in pH and bicarbonate – so DKA resolved by 18.77 hours Dhatariya K et al Submitted for publication

Potassium • But despite an aggressive potassium replacement regimen – more than 50% of patients became hypokalaemic Dhatariya K et al Submitted for publication

Take Home Message • Despite the existence of widely adopted national guidance – there are areas that need addressing • Has the slow evolution of the ‘evidence’ resulted in ‘complacency’? • We need to make sure the guidance that we give has a robust evidence base

HHS

‘ Diabetic 1st Reports of HHS Treatment pre- Coma’ and post-insulin Julius De Graeff y Lips August W. Theodor Von Von Stosch Frerichs Dreschfeld Sament y Schwartz 1828 1880 1880 1957 Pasquel FJ, Umpierrez GE. Diabetes Care 2014;37(11):3124-3131

First Mention in English? • On the 18 th August 1886 by Dreschfeld in the Bradshawe Lecture at the Royal College of Physicians of London – Diabetic coma “ though of small compass, is yet full of interest both to the physician and to the pathologist” • He described 3 types of coma – Drowsiness, passing onto coma – An excited nervous system (resembling alcohol intoxication), then drowsiness and coma – Dyspnoea with acetone (the most frequent sort) Dreschfeld J Br Med J 1886;ii(1338):358-363

Prescience • He described the coma in those who did not have acetone – As occurring mainly in those over 40 – Who were stout and well nourished at the time of death – Fatty liver on necropsy Dreschfeld J Br Med J 1886;ii(1338):358-363

Early Mentions of Non Ketotic Diabetes • RD Lawrence in 1951 – Described ‘lipo - plethoric’ or ‘fat diabetics’ – And the rarer ‘lipo - atrophic’ or ‘thin diabetics’ • This was associated with ‘intense lipidaemia’ • Sament and Schwartz in 1957 describe a case where 270 units of insulin reduced glucose from 1568mg/dl to 700mg/dl – describing much greater insulin sensitivity compared to DKA Lawrence RD BMJ 1951;1:373 Sament S S Afr Med J 1957;31(36):893-894

Continuing Mortality • Given 0.1u/kg/hr and 1-2 L of fluid on admission then 1 L every 3-4 hours, and giving potassium 20-40mmol/hour Carroll P et al Diabetes Care 1983;6(6):579-585

Evolution Kitabchi AE et al Diabetes Care 2009;32(7):1335-1343

Joint British Diabetes Societies for Inpatient Care • In August 2012 JBDS published a national guideline on the management of HHS http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm

Overall Quality of JBDS Guidelines Overall quality of JBDS – IP guidelines in 104 UK Trusts HHS Over 90% of respondents rated the guidelines as ‘Excellent’ or ‘Good’ Data from Mike Sampson. Feedback from 104 hospitals - first shown at Diabetes UK March 2013

ADA and JBDS HHS Definitions ADA (2009) JBDS (2012) Plasma glucose >600mg/dl (33.3mmol/l) >540mg/dl (30mmol/l) Arterial pH >7.3 >7.3 Serum bicarbonate >18mEq/l >15mmol/l Urine ketones Small Not referenced Serum ketones Small <3.0mmol/l Effective serum osmolality >320mOsm/Kg >320mOsm/Kg Anion gap Variable Not referenced Mental status Stupor / coma Not referenced Kitabchi AE et al Diabetes Care 2009;32(7):1335-1343 http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm

Do These Regimens Work? • No idea • Another national survey is needed! http://www.diabetologists-abcd.org.uk/JBDS/JBDS.htm Kitabchi AE et al Diabetes Care 2009;32(7):1335-1343