DKA

DIABETIC KETOACIDOSIS

Andrew J. BauerPediatric Endocrinology

WRAMC

GOALS

• REVIEW TYPE 1 DIABETES AND METABOLISM AS THEY RELATES TO DKA

• CLINICAL DIAGNOSIS and MISLEADING LABS

• TREATMENT and CONTROVERSIES

• TREATMENT GUIDELINES

Type 1 DM

• Autoimmune destruction of the pancreatic islet cell

• Hallmark = lymphocytic infiltration of islets

• Progresses over years• Leads to insulin

deficiency• Later may be associated

with glucagon deficiency as well

Progression to Type 1 DM

Autoimmune destruction

“Diabetes threshold”

Honeymoon

100% Islet loss

Typical Presentation

• Polyuria, polydypsia, weight loss

• Vomiting • Rapid-deep respiration• CNS depression – coma• Precipitating event

“Typical” Setting…..

• 9 yo boy presents to clinic with CC “ 6 day history of stomach pain and diarrhea.” “Vomiting started 2 days ago and has persisted.” – (+) weight loss– PE: HR 140, RR 28, T97.8 Weight: 27 Kg

• Tachy mucous membranes• Abd - soft, (+)BS, mild left CVA tenderness

– DX: viral gastroenteritis with mild dehydration

• Returned to ER 24 hours later– PE: cachectic, quiet, tired, cooperative, (+) ketotic breath

Background

• 15-30% of new diabetics present in DKA– < 4 yrs of age = 40% with DKA @ diagnosis

• Most common cause of death in diabetics less than 20 years of age– 70% of related deaths in diabetics less than 10

yrs of age

• Mortality: 5-15% (1-2% at MEDCEN)• Preventable

Diagnostic Criteria

• Blood glucose > 250 mg/dl• pH < 7.35

• HCO3 < 20 mEq/L

• Anion Gap > 12• ketonemia

Etiology

• Results from inadequate insulin– Accidental or intentional omission– Inappropriate intervention when stressed

Etiology

➨ DKA violates rules of common sense• Increased insulin requirement despite decreased

food intake• Marked urine output in setting of dehydration• Catabolic state in setting of hyperglycemia and

hyperlipidemia

• Insulin Deficiency is the Primary defect• Stress hormones accelerate and exaggerate

the rate and magnitude of metabolic decompensation

Pathophysiology Counter-Regulatory Hormones

Pathophysiology Hormone• Impaired insulin secretion Epi• Anti-insulin action Epi, cortisol, GH• Promoting catabolism All• Dec glucose utilization Epi, cortisol, GH

Islets of Langerhans

β-cell destruction Insulin Deficiency

Adipo-cytes

Muscle

Liver

Decreased Glucose Utilization &Increased Production

GlucagonIncreasedProtein Catabolism Increased

KetogenesisGluconeogenesis,GlycogenolysisIncreasedLipolysis

HyperglycemiaKetoacidosis

HyperTG

PolyuriaVolume Depletion

Ketonuria

AminoAcids

FattyAcids

StressEpi,Corti

sol

GH

Threshold180 mg/dl

Pathophysiology

Insulin

GlucagonEpinephrine

CortisolGrowth Hormone

Pathophysiology

Dec Glucose UtilizationLipolysis

Insulin

GlucagonEpinephrine


DKA - Early• Relative Insulin Deficiency ➨ Glycogenolysis & gluconeogenesis restrained

Peripheral glucose uptakeElevates blood glucose

Decreased Utilization

➨ post-prandial and Stress-Induced

hyperglycemia

Pathophysiology

GluconeogenesisGlycogenolysis

LipolysisKetogenesis

Insulin

GlucagonEpinephrine


• Insulin Deficiency GlycogenolysisGluconeogenesisHepatic glucose outputPeripheral glucose uptakeElevates blood glucoseLipolysisRelease FFA -> liverVLDL & ketones Ketonemia and hyperTG ➨ Acidosis & Diuresis

DKA - Late Increased Production &Decreased Utilization

➨ Fasting hyperglycemia

DKA

Initial Evaluation

• Hx and PE - – Duration of onset – Level of dehydration– Evidence of infection

• Labs - STAT – Electrolytes – Venous blood gas– Serum Osmolality– U/a

Osmolality= 2 x (Na + K) + Glucose/18

+ BUN/3

9 yo lab Evaluation

• 148| 109| 325.6 | <5 | 1.4

• Blood Gas - pH 7.0 5/1.020Glu >1000, (+) Ketones

700 16.847.5

51824.4

9 yo lab Evaluation

• 148| 109| 325.6 | <5 | 1.4

• Blood Gas - pH 7.0 5/1.020Glu >1000, (+) Ketones

700 16.847.5

51824.4

Misleading Labs

• Sodium• Potassium• Ketones• WBC

Misleading Labs

Sodium

• Na+ depressed 1.6 mEq/L per 100 mg% glucose• Corrected Na+ = measured Na +

1.6 meq/L x (glucose-100)/100))• Example:

– Na+ = 123 meq/L and Glucose = 1,250 mg/dl– 1,250 – 100 = 1,150 / 100 = 11.5 x 1.6 = 18 meq/L – Corrected Na+ = 123 + 18 = 141 meq/L

Misleading Labs

Sodium

• Triglycerides also artificially lower Na

Lipid LipidNa Na Na Na Na NaNa Na Na

Na Na Gluc NaNa Gluc

Serum

• Acidosis leads to flux of K+ out of cells as H+ enters cells to buffer

• Dehydration and volume depletion– Aldosterone ➨ Na reabsorption and K+

wasting➨ Serum K+ usually normal or high, but total

body K+ is low

Misleading Labs

Potassium

DKA- Risks of Therapy

Hypokalemia/Hyperkalemia

• With insulin therapy– K+ moves into cells (1 meq/L / 0.1 unit pH )

• Even with K+ you must– Give large doses (40 meq/L) K+

– Monitor K+ levels and EKG • High K - tall peaked T, long PR, wide QRS• Low K - depressed ST, diphasic T, Prom U-wave

– Cardiac dysrythmia

• Acetyl-CoA condenses to acetoacetate• Insulin prevents utilization of acetoacetate • so levels and shunt to ß-hydroxybutyrate and acetone

Misleading Labs

Ketones

• In the absence of insulin, FFA go to the liver, and into mitochondria via carnitine

• ß-oxidation excess acetylCoA

Nitroprusside reaction

Misleading Labs

Screening for Ketonemia

• Urine Dip stick vs. anion gap/serum bicarbSensitivity Specificity

DKA 99 % 69 %➨ Diabetic with minor signs and symptoms

and negative urine ketone dip stick is unlikely to have acidosis= high negative predictive value for excluding DKA

Am J Emer Med 34: 1999

Misleading Labs

WBC count

• N = 247 DKA admissions over 6 years– Mean WBC = 17,519/mm3 (+/- 9,582)– 69% without infection– 17.8% presumed viral infection– 12.9% bacterial infection - more common in

children < 3 years of age

➨ All need to be evaluated and re-evaluated if persistent acidosis

Am J Emer Med 19: 270-3, 2001

Let’s start treatment…..

Controversies and Risks of Therapy

• Fluids - composition, bolus amount and total fluids/day• Use of Bicarbonate• Phosphate replacement

CerebralEdema

DKA – Controversy

Cerebral Edema - Truths ?

• Idiogenic osmoles in CNS accumulate fluid

• Cerebral edema – present in 100% of patients prior to therapy

• Treatment exacerbates cerebral edema– Vigorous fluid

administration– Hypotonic fluids– Bicarbonate

Acute

LateSequelae

DKA – Cerebral Edema

Actualities

• Etiology is not known• Occurs exclusively in pediatric patients• Mortality Rate = 21%• Morbidity Rate = 27% (permanent neurologic

sequelae)➨ Difficulty is relatively rare occurrence (1-3 %)

with subsequent small numbers of patients in retrospective or prospective studies


Actualities

• NEJM - Jan 2001– N = 6977 DKA patients from 10 centers over 15

years– 61 developed cerebral edema (0.9%)

• Pediatrics - Sep 2001– N = 520 DKA patients over 5 1/2 years– 2 developed cerebral edema


Total Fluids

• > 4 L/m2/day, or > 50 ml/kg in first 4 hrs α hyponatremia α herniation– May occur in patients that receive less– Of 52 patients with neurologic

complications 21 had either a rise of serum Na or fall less than 4 mmol/L

JCE

M 8

5:50

9-51

3, 2

000

➨ Attention to fluid rate and tonicity is essential, but may not be sufficient to predict subset that will

develop neurologic complications

J P

eds

113:

10-1

4, 1

988


Total Fluids

• > 4 L/m2/day, or > 50 ml/kg in first 4 hrs α hyponatremia α herniation– May occur in patients that receive less– Of 52 patients with neurologic

complications 21 had either a rise of serum Na or fall less than 4 mmol/L

JCE

M 8

5:50

9-51

3, 2

000

➨ Attention to fluid rate and tonicity is essential, but may not be sufficient to predict subset that will

develop neurologic complications

J P

eds

113:

10-1

4, 1

988

DKA – Cerebral Edema Variable Time of Onset

0

1

2

3

4

5

6

7

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 21 25

Hours after Initiation of Therapy

# of

Ch

ild

ren

wit

h N

euro

logi

c D

eter

iora

tion

NE

JM 3

44:2

64-6

9, 2

001

Prior to therapy; longer durationsymptoms before diagnosis


Other

• Hypoxemia– Children’s brains have higher oxygen

requirement, 5.1 mL/100g vs. 3.3 mL/100g– Hypophosphatemia with resultant decreased

2,3-DPG decreases O2 delivery to brain cells

– Mannitol - earliest effects are related to decreased viscosity, not to shift of fluid from extravascular space

JCEM 85: 509-13, 2000Neurosurg 21: 147-156, 1987


Signs and Symptoms

1. Sudden and persistent drop in heart rate- not bradychardia - not assoc with HTN

- not related to hydration status2. Change in sensorium 7. Fall in serum3. Headache Na, or failure4. Emesis to rise5. Incontinence6. Unexplained tachypnea

JCEM 85:509-513, 2000


Evaluation

• CT may be non-diagnostic at time of symptoms– 9 of 30 - no edema, 6 read as normal– 5 of 9 - 2.5 to 8 hours after onset of coma, read as

normal

JCEM 85:509-513, 2000

➨ Cerebral Edema is a clinical diagnosis.Need to treat BEFORE imaging.

DKA – Risks of Therapy

Bicarbonate Administration

• Administration to acidotic patient generates rapid rise in CO2

• CO2 enters CNS rapidly

• HCO3- is delayed by blood-brain barrier

• Increased CNS CO2 exacerbates cerebral acidosis

CO2 + H2O H2CO3 H+ + HCO3-

• May also reduce partial pressure of O2 in CSF ➨ vasoconstriction ➨ brain hypoxia/ischemia



NEJM 344:264-269, 2001

• Multi-center study from 10 pediatric centers, USA and Melbourne, Australia over 15 yr period– 6977 DKA hospitalizations: 61 cases cerebral edema

(0.9%)

• Presentation: PaCO 2 BUN Glucose Bicarb Cerebral Edema 11.3 27 758 23/61 (32%) Controls 15.1 21 700 43/174 (23%)

∀≠ fluid, insulin, or sodium administration, nor rate of fall in glucose was associated



• Variations in treatment exacerbate an on-going pathologic process

• Brain ischemia is major underline etiology– Hyperglycemia increases extent of neurologic damage– Extreme dehydration, hypocapnia – Concept of idiogenic osmotically active substances not

supported (no relationship to change in glucose, rate of fluid or Na administration)

➨ Risk related to duration and severity of DKA

NEJM 344:264-269, 2001

**** ******** ****

**** ******** ****

DKA- Controversy

Phosphate

• Essential phosphate deficit• W/treatment serum phosphate and 2,3-DPG fall

• Shift oxyhemoglobin curve reducing O2 deliver

Theoretical

Practical

• No evidence of direct benefit, but less Cl-

• Give ½ K+ replacement as K-phos x 8 hours

• Limit to 2 mEq/kg/day to avoid hypocalcemiaEndo Met Clin 29:Dec 2000

Elements of Therapy

Elements of Therapy

• Fluids – treat shock, then sufficient to reverse dehydration and replace ongoing losses (will correct hyperglycemia)

• Insulin – sufficient to suppress ketosis, reverse acidosis, promote glucose uptake and utilization (will stop ketosis)

• Electrolytes – replace profound Na+ and K+ losses

Typical Therapy - Fluids

• 10% dehydration is standard estimate (use ▲weight if known)– Bolus: treat shock, usual 20-30cc/kg

given 10cc/kg at a time– Replace deficit over 48-72 hours– ie. 10 % in 20 Kg pt = 2000ml over 48hrs

= maintenance + 42cc/hr x 48 hours


• Use ½ NS to NS • Average = 2 x maintenance

– 4:2:1 cc/kg/hr or 100:50:20 cc/kg/day – ie. 25 kg patient

• (4 x 10) + (2 x 10) + (1 x 5) = 65 cc/hr• (100 x 10) + (50 x 10) + (20 x 5)/24 hours

= 66.7 cc/hr


Insulin

Insulin Level

Biological effect

100%

100 uU/ml

Current therapy usescontinuous insulin drip➨ Drop glucose 50-100 mg/dl/hr

0.1 units/kg/hr

Typical Therapy - Insulin

• 0.1 unit/kg/hr continuous drip (regular)– Flush tubing with 50 ml– 250 units regular in 250 cc NS (1.0 units/

ml)= 0.1 u/kg/hr = 0.1 ml/kg/hr

Typical Therapy Glucose - 2 Bag Method

• Goal - decrease blood glucose by 50-100 mg/dl/hr

• Must continue insulin therapy to correct acidosis

• Order D10 NS to bedside– when serum glucose < 300: add D5NS ( = 1/2

D10NS + maintenance bag)– when serum glucose < 200: Change to D10NS

Typical Therapy

• K+ 40 meq/L (split between KCl and Kphos)

• Reverse insulin resistance– Treat infection– Treat underlying illness - stress

• Bicarb - only if severe circulatory failure and high risk of cardiac decompensation from profound acidosis

Monitor

• ICU - pH < 7.3 and/or HCO3 < 15

• Available staff• Strict I/O (NPO)

– Fluid calculations must account for ongoing losses – vomiting, diarrhea, excessive urine

– ? If > 4 L/m2/day

• CNS activity - headache, change in sensorium

Monitor

• Vitals - sudden drop in HR, tachypnea• Neurologic checks - q30-60 minutes• Weight - bid• Labs

– dstick q1 hour– Urine dip q void - resolution of ketonuria may

lag behind clinical improvement

Monitor

• Labs – Lytes, VBG q 2-4 hours

➨ Drop in Na - increase risk of cerebral edema, ? SIADH vs. cerebral salt wasting

➨ HCO3- / pH in first 2-3 hours may drop further due

to re-perfusion of tissue, lactic acidosis

DKA

Guidelines

• Common ground to start from• Does not eliminate need to individualize

therapy• Large deviations should be an opportunity

to critically review clinical and therapeutic course

DKA

Flowsheet

• CIS is not a flow sheet, but rather a database

• Inability to review all data at one time decreases ability to make sound decisions

• Maintenance of flowsheet is the first step in critical analysis of response to therapy

9 yo lab Evaluation

• 27 Kg - assume 10% dehydrated• 148| 109| 32

5.6 | <5 | 1.4 70016.847.5

51824.4

• Anion Gap = • Osm = • Corrected Na =

• Fluid Def = • Maintenance = • IV rate (24hrs) =

Transport of Patient with DKA

• 2 large bore PIV• Must have documentation of previous

treatments– PE with vitals and notes on mental status – Fluids - bolus and current– ? SQ Insulin given - time and amount– Contact phone number for labs/cultures

• Must have glucagon, mannitol and IV glucose with patient at ALL times

DKA

Prevention

• 50% DKA admissions are in known diabetics

• Failure of Physician-Patient relationship– non-compliance– Inappropriate intervention– Sick day rules need to be understood and

followed– Availability is essential


Improved Management ?

• All patients given 20 cc/kg NS bolus over 30-45 minutes

• Started on 0.1 units/kg/hour Insulin without bolus

• Fluids - 2.5 x maintenance of 3/4 NS regardless of degree of dehydration

• Glucose used to maintain insulin rate

Pediatrics 108: 735-740; 2001


Improved Management ?

• Outcome– 23 % fewer fluid changes = decreased error risk– Mean total fluids in first 24 hours lower (5 vs 4

l/m2/day)– Dec time to resolve acidosis shorter (16 vs 12

hours) ??– Reduced fluid cost ($1060 to $776)

Pediatrics 108: 735-740; 2001

“Typical” Setting…..

• 7 yo boy with 24 hour history of n/v/d. Diagnosed with IDDM 2 yrs ago. Woke up with moderate ketones and dstick of 350 mg/dl.– Is this DKA ?– What is your responsibility ?

• 12 yo patient on CSII. Last 4 hours dsticks increasing from 120 to 450 mg/dl. Now complaining of headache and nausea. Large ketones on dip-stick.

DKA

• Acidosis– Primary buffer is intracellular protein– K+ moves out of cells and H+ moves in– In association with aldosterone (induced from

hypovolemia)• Potentiates K+ wasting

➨➨ Hypokalemia

EKG Changes During DKA

Normal

Hi K Lo K


Other

• Insulin associated activation of Na+/H+ pump– Not commonly found during initial treatment– As acidosis resolves, H+ diffuses out of brain

cells and Na+ enters (along with H20)

– Rabbit model - drop in glucose secondary to insulin administration vs. peritoneal dialysis results in cerebral edema

JCEM 85: 509-13, 2000

0

200

400

600

800

1000

1200

1400

12 1 2 3 4 5 6 7 8

Glucose

6

6.5

7

7.5

8

12 1 2 3 4 5 6 7 8

pH

3

3.5

4

4.5

5

5.5

6

6.5

7

12 1 2 3 4 5 6 7 8

K

DKA

Documents

glucose utilizationepi

insulin deficiencylater

glucagon deficiency

setting of hyperglycemia

stressedetiology dka

weight losspe

utilization postprandial

new diabetics present