Diabetic KetoAcidosis As easy as D, K, A? - PowerPoint Presentation

Slide1

Diabetic KetoAcidosis

As easy as D, K, A?

Clinical Case

25-year-old male with type 1 diabetes

lantus

26 units ON and

novorapid

12 units

TDS

recurrent attendances with

DKA

admitted with 24h history of vomiting and abdominal

pain

c

apillary blood glucose (CBG) 18mmol/l

<11.1mmol/L

blood ketones

6.7mmol/l

<0.3mmol/L

pH

7.11

7.35-7.45

bicarbonate

9mmol/l

18-23mmol/L

Is this DKA?

DKA — Definitions

D

iabetes

CBG

>11.1

mmol

/l OR known diabetes

K

etosis

capillary

blood ketones ≥3.0mmol/l OR urine ketones ≥3+

A

cidosis

pH

<7.30 OR bicarbonate ≤15mmol/l

have to have all 3 (high BMs plus ketones without acidosis ≠ DKA

)

CBG may be relatively normal e.g. pregnancy (

euglycaemic

DKA

)

many unwell patients with T2DM have acidosis and ketosis

e.g

. from vomiting, sepsis and renal

failure

≠

DKA

DKA — Management Priorities

Fluid Resuscitation

Clearance of ketones and resolution of acidosis - insulin

Prevention and treatment of low K

+

(hypokalaemia)

Identification and treatment of precipitating cause

Monitoring (ketones, glucose, K

+

, other complications)

Patient education and referral to diabetes specialist team

NOT

getting the BM down as quickly as

possible…

DKA — Fluid Resuscitation

fluid deficit in DKA up to

100mL/kg

=

7L

for 70kg person

more

if

sepsis

deficits in sodium

10mmol/kg

,

chloride and potassium

3-5mmol/kg

aim to correct over 24h using 0.9%

saline:

1L

stat

2L

over next 4h (500ml/h)

2L over

next 8h (250ml/h)

2L

over next 16h (125ml/h)

start 10% glucose infusion at

125mL/h

when CBG <

14mmol/L

0.9% saline should usually continue

alongside

the glucose

infusion

DO

NOT

stop 0.9% saline when starting 10% glucose within

the first

24h as

glucose

will not replace intravascular volume

DKA — Resolution of KetoAcidosis

50 units

actrapid

in 50ml 0.9% saline giving a

1unit/mL

solution

FIXED RATE IV INSULIN INFUSION (FRIVII)

start at

0.1units/kg/h

=

7units/h

=

7mL/h

for a 70kg

person

NOT

variable rate IV insulin infusion (

VRIVII or

sliding scale

)

rate of insulin does

NOT

go

up or down

depending on

CBG

aim for fall in ketones of ≥

0.5mmol/L/h

fall

in

glucose

of

3mmol/L/h

if

this is

NOT

being achieved i.e. ketones not falling by

0.5mmol/L/h

check insulin infusion (rate, lines,

venflon

, volume TBI)

increase insulin infusion rate by 1unit (

1mL)/

h and

monitor

DKA — Insulin

if patient takes long-acting insulin

e.g. Lantus

®

or

Levemir

®

ensure this is

prescribed and given

ALONGSIDE

the IV insulin

infusion

if patient

normally

on

SC

insulin infusion

pump:

ensure this is set to basal rate only

BEFORE

starting the DKA

protocol

FRIVII can be stopped when

BOTH

ketosis and acidosis have resolved

i.e. pH >7.30 AND ketones <0.3mmol/l

if patient eating and drinking

switch to

usual SC

insulin

overlap

IV insulin for

30-60min AFTER

first SC insulin

dose

if patient not E&D, vomiting or septic

switch to VRIVII protocol

sliding scale

DKA — Prevention of Hypokalaemia

although K

+

is frequently normal on admission, total body stores of K

+

are usually low and serum potassium level will fall rapidly after starting IV insulin

first litre of IV fluid usually given

STAT

without adding any

KCl

potassium should be added to

all

subsequent

bags depending on repeated measurements of serum K

+

check VBG +/- U&E at least every 4h during first 24h

Serum Potassium (

mmol

/L)

Amount of

KCl

to add/L of 0.9%

NaCl

(

mmol

/L)

>5.5

0

4.5-5.5

20

<4.5

40

DKA — Identify & Treat Cause

infection / sepsis

poor compliance with

insulin

intentional

or accidental

new diagnosis of T1DM

myocardial infarction or other critical illness

pregnancy

drugs e.g. steroids

history

routine bloods

ECG

urinalysis (including ß-HCG)

CXR

blood cultures

stool cultures

DKA — Monitoring

hourly

obs

including

urine

output

consider catheter

and GCS

if

severe

DKA

hourly CBG

and

blood ketones

document religiously

maintain accurate fluid balance chartkeep

checking that…

insulin

is running

at correct rate and no problems with pump etc.

fluids are being given in time with prescribed

schedule

gently

remind doctors to…

repeat

U&E plus VBG

at

least every

4h

increase

insulin rate if ketones not falling

adequately

add sufficient

potassium to fluids

Clinical Case cont.

diagnosed with DKA

3L of 0.9% saline written up over first 4h

started on FRIVII at

6units(6mL)/

h

body weight 60kg

26units

of

lantus

prescribed on drug chart

after 2h blood ketones are 6.3mmol/L

6.7mmol/L on admission

insulin being delivered at correct rate

insulin rate increased to

7units(7mL)/h

2h later ketones 4.2mmol/L and CBG

12mmol/L

Clinical Case cont.

second cannula sited and 10% glucose started at

100mL/h

in addition to 0.9% saline continued

nurse repeats VBG and informs doctor that K

+

is

3.8mmol/L

40mmol

KCl

added to next

2L of

0.9%

NaCl

CXR shows R basal

pneumonia —IV abx prescribedafter 12h VBG is repeated and shows pH 7.33 and bicarbonate 18mmol/l —blood ketones now 0.2mmol/L

patient still not eating and drinking

VRIVII (‘sliding scale’)

commenced, patient referred to DNS and transferred

to

Avon Ward for ongoing

care

Does this ever occur in the real world?

‘Doctor

the

ketones

are not falling

…’

insulin infusion running at

0.1mL/h

rather than

7mL/h

0.1units/kg/h prescribed

line

from insulin pump clamped

off after trip to x-ray

venflon

removed and trickling insulin into patient’s mattresspump constantly alarming so only 14units given over 6h

42 units should have been infused

patient collapsed in toilet due to severe

hypoglycaemia

SC

insulin pump still running despite being started on FRIVII

no IV insulin

for

4h as nurse wanted to ‘wait until

Consultant

ward

round’

patient

rebounded back into DKA

long-acting insulin not written up on drug chart

0.9% saline stopped when 10% glucose commenced before

patient rehydrated

VRIVII (‘sliding scale’) being used instead of FRIVII

ketones not regularly

checked

with BMs

Audit of DKA Management on AMU

100

consecutive cases of DKA admitted to

AMU in 2015

Not a single case was managed

completely in accordance with

the 2010

JBDS DKA

guidelines…

inadequate initial fluid resuscitation in 45%

10% glucose not commenced when BM <14mmol/l in 50%

FRIVII not delivered or delivered incorrectly in 30%

CBG and ketones not

checked / documented

hourly in 35%

failure to adjust FRIVII appropriately in 55%

long acting SC insulin not prescribed in 50%

inadequate

potassium replacement in 70%

no septic screen in 65%

inadequate repeat U&E/VBG in 70%

DKA —Key Messages

we need to

improve:

nurse-led

management

fluid resuscitation

is the most important part of initial treatment

fixed rate IV insulin infusion (FRIVII)

not

sliding scale (VRIVII)

check CBG (BM) and blood ketones

hourly

if ketones not falling by 0.5mmol/h

check pump and lines

before

increasing insulin rate by

1unit/mL/h

give long acting SC insulin

as well as IV

insulin

remember to check patients for SC insulin infusion pumps

start 10% glucose when CBG <

14mmol/L but

continue 0.9

% saline

frequent monitoring /

obs

and repeat

U&E / VBG essential

adequate

potassium replacement

identify and treat cause, monitor for complications, refer to DNS

DKA — Complications

hypokalaemia

risk

of life threatening arrhythmias

cerebral oedema

adolescents

and

children

pneumonia

including aspiration

MI

venous thromboembolism

ensure

tinzaparin

is prescribed

hypoglycaemia

hypomagnesaemia and

hypophosphataemia

rarely require treatment

hypoxaemia and ARDS

Despite improvements in management MORTALITY remains at

2-5%

HHS - Hyperosmolar Hyperglycaemic State (HONK)

tends to affect

older / Type

2 diabetics

glucose often much higher than in DKA

usually

>

30mmol/L

more profound dehydration

sodium usually very

high and fluid deficit >

10L

acidosis

is not

required for diagnosis

often

acidotic due to sepsis or

AKI

ketosis may be present due to starvation/acute illness

rarely

>

3.0mmol/L

fluid deficit should be corrected more slowly

e.g. over 72h

FRIVII is not required

use VRIVII at 50

% of usual dose if insulin

naïve

10%

glucose

is rarely

required to maintain

euglycaemia

invariably associated with severe precipitating illness

e.g. MI, sepsis

prognosis much worse

mortality

up to 50

%