2. GOALS
• REVIEW TYPE 1
DIABETES AND
METABOLISM AS
THEY RELATES TO
DKA
• CLINICAL DIAGNOSIS and MISLEADING
LABS
• TREATMENT and CONTROVERSIES
• TREATMENT GUIDELINES
3. 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
4. Progression to Type 1 DM
Autoimmune destruction
Honeymoon
100% Islet loss
“Diabetes threshold”
6. “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
7. 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
9. Etiology
• Results from inadequate insulin
– Accidental or intentional omission
– Inappropriate intervention when stressed
10. Etiology
² 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
11. Pathophysiology
Counter-Regulatory Hormones
• Insulin Deficiency is the Primary defect
• Stress hormones accelerate and exaggerate
the rate and magnitude of metabolic
decompensation
Pathophysiology Hormone
• Impaired insulin secretion Epi
• Anti-insulin action Epi, cortisol, GH
• Promoting catabolism All
• Dec glucose utilization Epi, cortisol, GH
12. Islets of β-cell destruction Insulin Deficiency
Langerhans
l
Ep i,Cortiso Decreased Glucose Utilization &
Stress GH Increased Production
Muscle
Adipo- Amino Glucagon
Increased Liver
cytes Acids Protein
Catabolism Increased
Ketogenesis
FattyAcids Gluconeogenesis,
IncreasedLipolysis Glycogenolysis
Polyuria Threshold
180 mg/dl Hyperglycemia
Volume Depletion
Ketoacidosis
Ketonuria
HyperTG
23. Misleading Labs
Sodium
• Triglycerides also artificially lower Na
Lipid Lipid
Na Na Na
Na Na Na
Serum Na Na
Na Na Na Gluc Na
Na Gluc
24. Misleading Labs
Potassium
• Acidosis leads to flux of K+ out of cells as H+
enters cells to buffer
• Dehydration and volume depletion
– Aldosterone ² Na reabsorption and K+
wasting
² Serum K+ usually normal or high, but total
body K+ is low
25. 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
26. Misleading Labs
Ketones
• In the absence of insulin,
FFA go to the liver, and Nitroprusside
reaction
into mitochondria via
carnitine
• ß-oxidation excess
acetylCoA
• Acetyl-CoA condenses to acetoacetate
• Insulin prevents utilization of acetoacetate
• so levels and shunt to ß-hydroxybutyrate and acetone
27. Misleading Labs
Screening for Ketonemia
• Urine Dip stick vs. anion gap/serum bicarb
Sensitivity Specificity
DKA 99 % 69 %
² 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
28. 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
² All need to be evaluated and re-evaluated if
persistent acidosis
Am J Emer Med 19: 270-3, 2001
30. Controversies and Risks of Therapy
• Fluids - composition, bolus
amount and total fluids/day
Cerebral
• Use of Bicarbonate
Edema
• Phosphate replacement
31. DKA – Controversy
Cerebral Edema - Truths ?
Acute
• Idiogenic osmoles in
CNS accumulate fluid
• Cerebral edema –
present in 100% of
patients prior to therapy
• Treatment exacerbates
cerebral edema
– Vigorous fluid
administration Late
Sequelae
– Hypotonic fluids
– Bicarbonate
32. DKA – Cerebral Edema
Actualities
•Etiology is not known
•Occurs exclusively in pediatric patients
•Mortality Rate = 21%
•Morbidity Rate = 27% (permanent neurologic
sequelae)
² Difficulty is relatively rare occurrence (1-3 %)
with subsequent small numbers of patients in
retrospective or prospective studies
33. DKA – Cerebral Edema
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
34. DKA – Cerebral Edema
Total Fluids
• > 4 L/m2/day, or > 50 ml/kg in first 4 hrs α
JCEM 85:509-513, 2000 J Peds 113:10-14, 1988
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
² Attention to fluid rate and tonicity is essential, but
may not be sufficient to predict subset that will
develop neurologic complications
35. DKA – Cerebral Edema
Total Fluids
• > 4 L/m2/day, or > 50 ml/kg in first 4 hrs α
JCEM 85:509-513, 2000 J Peds 113:10-14, 1988
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
² Attention to fluid rate and tonicity is essential, but
may not be sufficient to predict subset that will
develop neurologic complications
36. DKA – Cerebral Edema
Variable Time of Onset
# of Children with Neurologic Deterioration
7
Prior to therapy; longer duration
6 symptoms before diagnosis
5
4
3
2
NEJM 344:264-69, 2001
1
0
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
37. DKA – Cerebral Edema
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
Neurosurg 21: 147-156, 1987
38. DKA – Cerebral Edema
Signs and Symptoms
1. Sudden and persistent drop in heart rate
- not bradychardia - not assoc with HTN
- not related to hydration status
2. Change in sensorium 7. Fall in serum
3. Headache Na, or failure
4. Emesis to rise
5. Incontinence
6. Unexplained tachypnea
JCEM 85:509-513, 2000
39. DKA – Cerebral Edema
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
² Cerebral Edema is a clinical diagnosis.
Need to treat BEFORE imaging.
JCEM 85:509-513, 2000
40. 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
41. DKA – Risks of Therapy
Bicarbonate Administration
• 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: PaCO2 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
NEJM 344:264-269, 2001
42. **** ****
****
DKA – Risks of Therapy ****
**** ****
********
Bicarbonate Administration
• 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)
² Risk related to duration and severity of DKA
NEJM 344:264-269, 2001
43. DKA- Controversy
Phosphate
Theoretical
• Essential phosphate deficit
• W/treatment serum phosphate and 2,3-DPG fall
• Shift oxyhemoglobin curve reducing O2 deliver
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 hypocalcemia
Endo Met Clin 29:Dec 2000
45. 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
46. Typical Therapy - Fluids
• 10% dehydration is standard estimate (use
vweight 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
47. Typical Therapy - Fluids
• 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
48. DKA – Risks of Therapy
Insulin
100%
Biological
0.1 units/kg/hr
effect
Current therapy uses
continuous insulin drip 100 uU/ml
² Drop glucose Insulin Level
50-100 mg/dl/hr
49. 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
50. 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
51. 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
52. 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
53. 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
54. Monitor
• Labs
– Lytes, VBG q 2-4 hours
² Drop in Na - increase risk of cerebral edema, ?
SIADH vs. cerebral salt wasting
² HCO3- / pH in first 2-3 hours may drop further due
to re-perfusion of tissue, lactic acidosis
55. 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
56. 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
57. 9 yo lab Evaluation
• 27 Kg - assume 10% dehydrated
• 148| 109| 32 16.8 518
700 24.4
5.6 | <5 | 1.4 47.5
• Anion Gap = • Fluid Def =
• Osm = • Maintenance =
• Corrected Na = • IV rate (24hrs) =
58. 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
59. 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
60. Typical Therapy - Fluids
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
61. Typical Therapy - Fluids
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
62. “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.
63. 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
²² Hypokalemia
65. DKA – Cerebral Edema
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