Good overview of acute renal failure but this was written before the most recent ATN data which negates one of the premises of the lectuer that higher doses of dialysis are beneficial in patients in ARF.

1. acute renal failure
…from basics to the latest advances
Joel M. Topf, MD
Clinical Nephrologist

2. the house
moment

3. Dr. Haas invented the first dialysis machine designed
for humans and in 1928 he treated 6 patients.
All of them died.

4. In 1943, Willem Kolff’s, working in the
Nazi occupied Netherlands created
the second human dialysis machine.
In 1943 he dialyzed his first patient, a
young man with acute nephritis.

 Dr. Haas
In 1945, a 67-year-old woman in
uremic coma presented to Dr Kolff.
Regained consciousness after 11
hours of hemodialysis.

5.  Commonly quoted
mortality of 70% is
for dialysis requiring
ICU patients
 For hospital acquired
ARF: 20%

8.  37 year old AA female
 Multiple GSW
 Prolonged hypotension
 Aorta was cross
clamped during
exploratory laparotomy
 Anuric x 18 hours
 Cr from 0.8 to 2.2

9.  36 y.o. African American
women with menorrhagia.
 Has prolonged bleeding
following fibroidectomy
 Contrasted CT scan used to
determine source of
bleeding.
 Cr rises from 0.8 to 2.2
 Patient is non-oliguric

10. Two women.
Same age.
Same race.
Same rise in creatinine.
Same diagnosis: acute renal failure.
Two completely different diseases.

11. definition of acute renal failure
“Acute and sustained reduction in renal function.”
35 definitions

12. biochemical
definitions
Contrast nephropathy
ARF is defined by a 0.5
mg/dL or 25% increase
in serum creatinine

13. event driven
definitions
Dialysis dependent
ARF is often used in
retrospective cohorts
 Easy to capture
 Unambiguous
 Important end-point

16. rifle criteria for
stratifying arf
R isk
I njury
F ailure
L oss of function
E nd-Stage Renal disease

17. R isk
 Increase in Cr of 1.5-2.0 X baseline or
 urine output < 0.5 mL/kg/hr for more than 6 hours.
I njury
F ailure
L oss of function
E nd-Stage Renal disease

18. R isk: Inc Cr 50-100% or U.O. < 0.5 mL/kg/hr for > 6 hrs
I njury
 increase in Cr 2-3 X baseline (loss of 50% of GFR) or
 urine output < 0.5 mL/kg/hr for more than 12 hours.
F ailure
L oss of function
E nd-Stage Renal disease

19. R isk: Inc Cr 50-100% or U.O. < 0.5 mL/kg/hr for > 6 hrs
I njury: Inc Cr 100-200% or U.O. < 0.5 mL/kg/hr > 12 hrs
F ailure
 increase in Cr rises > 3X baseline Cr (loss of 75% of GFR) or
 an increase in serum creatinine greater than 4 mg/dL, or
 urine output < 0.3 mL/kg/hr for more than 24 hours or
anuria for more than 12 hours.
L oss of function
E nd-Stage Renal disease

20. R isk: Inc Cr 50-100% or U.O. < 0.5 mL/kg/hr for > 6 hrs
I njury: Inc Cr 100-200% or U.O. < 0.5 mL/kg/hr > 12 hrs
F ailure: Inc Cr > 200% or > 4 mg/dL or U.O. < 0.3 mL/kg/hr >
24 hrs or anuria for more than 12 hours
L oss of function
 persistent renal failure (i.e. need for dialysis) for more than 4
weeks.
E nd-Stage Renal disease

21. R isk: Inc Cr 50-100% or U.O. < 0.5 mL/kg/hr for > 6 hrs
I njury: Inc Cr 100-200% or U.O. < 0.5 mL/kg/hr > 12 hrs
F ailure: Inc Cr > 200% or > 4 mg/dL or U.O. < 0.3 mL/kg/hr >
24 hrs or anuria for more than 12 hours
L oss of function: Need for dialysis for more than 4 weeks
E nd-Stage Renal disease
 persistent renal failure (i.e. need for dialysis) for more than 3
months.

22. R isk: Inc Cr 50-100% or U.O. < 0.5 mL/kg/hr for > 6 hrs
I njury: Inc Cr 100-200% or U.O. < 0.5 mL/kg/hr > 12 hrs
F ailure: Inc Cr > 200% or > 4 mg/dL or U.O. < 0.3 mL/kg/hr >
24 hrs or anuria for more than 12 hours
L oss of function: Need for dialysis for more than 4 weeks
E nd-Stage Renal disease : Need for dialysis for more than 3
months

23. nice criteria. do they work?
 20,126 consecutive
admissions to a
university hospital
 Excluded kids
 Kidney transplant and
dialysis patients
 Patients admitted for <
24 hours
 Using RIFLE:
 Risk 9.1%
 Injury 5.2%
 Failure 3.7%
Uchino S, Bellomo R, Goldsmith D. Crit Care Med 2006 Vol 34 1913-1917.

24. >3x BL Cr
Cr > 4
Hospital Mortality

25. nice criteria. do they work in the icu?
 University of Pittsburgh
has 7 ICUs
 5,383 patients
 Excluded dialysis
 Subsequent admissions
 Frequency of acute
Kidney failure:
 No AKD 1,766
 Risk 670
 Injury 1,436
 Failure 1,511
Hoste E, Clermont G, Kersten A. Crit Care 2006 Vol 310

27. when Hoste looked at markers of
severity of illness excluding the renal
system:
No survival difference between the
4 groups:
• Lack of renal failure
• Risk
• Injury
• Failure

28. RIFLE is dependent on creatinine.
creatine is a functional marker of
organ damage
Functional
markers: old
and busted

29. biomarkers are foot prints of actual
organ damage
Biomarkers,
new hotness

30. functional versus biomarkers
Functional
Marker Biomarker
SGOT
Hypoalbuminemia
Liver damage Coagulopathy
SGPT
GGT

31. functional versus biomarkers
Functional
Marker Biomarker
SGOT
Hypoalbuminemia
Liver damage Coagulopathy
SGPT
GGT
Troponin I
Hypotension
Heart damage Arrhythmia
Troponin T
CK-MB

32. functional versus biomarkers
Functional
Marker Biomarker
SGOT
Hypoalbuminemia
Liver damage Coagulopathy
SGPT
GGT
Troponin I
Hypotension
Heart damage Arrhythmia
Troponin T
CK-MB
Creatinine
KIM-1
Kidney damage BUN
NGAL
Cystatin C

33. creatinine as a lagging indicator
 4,118 Cardiac surgery patients
 Prospectively looked at changes of creatinine
48 hours post-op on 30-day mortality
 All odds ratios were controlled for 26
variables found to be significant predictors of
mortality in univariate analysis

34. Creatinine falls Creatinine rises
<0.5 0.4 0.2 0.1 0.3 0.5 0.7 0.9
Delta Creatinine (mg/dL)

35. candidates for a renal troponin:

36. candidates for a renal troponin:
kidney injury molecule-1 (kim-1)
 Transmembrane
protein expressed 2.00
in the proximal
tubule. 0.69
 Expression is 0.34
increased 0.13
following ischemic
damage
 Can be found 12
hours after renal
insult
Han WH, Bailly V, Abichandani. Kidney Int 2002 62, 237–244.
Liangos O, Han WK, Wald R. Abstract J Am Soc Nephrol 16: 318A, 2005.

37. candidates for a renal troponin:
kidney injury molecule-1 (kim-1)
 Transmembrane Time starts at aorta cross
protein expressed clamp. Cr rose to 2.1.
in the proximal
tubule.
 Expression is inc-
reased following
ischemic damage
 Can be found 12
hours after renal
insult
Han WH, Bailly V, Abichandani. Kidney Int 2002 62, 237–244.
Liangos O, Han WK, Wald R. Abstract J Am Soc Nephrol 16: 318A, 2005.

38. urinary neutrophil gelatinase-
associated lipocalin (ngal)
 Protein that is secreted  Prospective
by the kidney in res- observational trial
ponse to ischemic injury  81 adults going for
 Early data in children Cardiac surgery
showed nearly perfect  65 No AKI
sensitivity and  1 died of MOF
specificity  16 AKI (Risk or higher)
 5 required CVVH
 False positives with UTI
 5 died of MOF
Mishra J, Ma Q, Prada A. J Am Soc Nephrol 2003; 14: 2534-43.
Wagener G, Jan M, K M. Anesthesia 2006; 105: 485-91.

40. differential diagnosis

41. etiologies of arf
 Seventy percent have concurrent oliguria
 < 400 mL/day
 < 0.5 mL/kg/hr in children
 < 1 mL/kg/hr in infants
 Complicates 5-7% of hospitalizations

42. Hou SH, Bushinsky DA, Wish JB. Am J Med 1983; 74: 243-8.
Nash K, Hafeez A, Hou S. Am J Kidney Dis. 2002; 39: 930-6.
Kaufman J, Dhakal M, Patel B, Et al. Am J Kidney Dis 1991; 17: 191-8.

43. Hou SH, Bushinsky DA, Wish JB. Am J Med 1983; 74: 243-8.
Nash K, Hafeez A, Hou S. Am J Kidney Dis. 2002; 39: 930-6.
Kaufman J, Dhakal M, Patel B, Et al. Am J Kidney Dis 1991; 17: 191-8.

44. N=389 N=256 N=103
Pascual J, Liano F. J Am Geriatr Soc 1998, 46: 1-5.

45. hospital acquired acute renal failure

46. hospital acquired acute renal failure

49. Pre-renal azotemia
No BP, no pee pee

50. no bp, no ARF pee
Pre-renal pee
 Properly functioning kidney, properly
responding to a drop in systemic and renal
perfusion
RPF
ulat ion
au toreg
GFR

51. differentiation of prerenal from
intrinsic renal disease
 Use of FENa
 Fraction of filtered sodium which is excreted in the
urine.
 Patients with prerenal azotemia will be sodium
avid and minimize renal excretion of sodium
lowering the FENa below 1%

52. Fractional excretion of
sodium:
Excreted Na
Filtered Na

53. Calculating the Numerator
Excreted Na = Urine Na x Urine Volume

54. Calculating the Denominator
Filtered Na = Serum Na x GFR
GFR = Urine Cr x Urine Volume
Serum Cr
Filtered Na = Serum Na x UrCr x UrVol
Serum Cr

55. Excreted Na
FENa =
Filtered Na
Urine Na x Urine Volume
FENa =
Serum Na x UrCr x Urine Volume
Serum Cr
Urine Na
FENa =
Serum Na x UrCr
Serum Cr
Urine Na x Serum Cr
FENa =
Serum Na x UrCr

56. FENa the easy way
 FENa is a small number 0.1% to 3%
 So the calculation will be 0.001-0.03 prior to
converting to percent by X 100
 So make the fraction small by putting the small
numbers over the big numbers
 Sr Na
 Sr Cr Sr Cr x Ur Na
FENa =
 Ur Na Sr Na x Ur Cr
 Ur Cr

57. FeNa. what is it good for?
 The discriminator for differentiating between prerenal azotemia
and ATN is 1%:
 FENa < 1 indicates pre-renal  FENa > 1 indicates ATN
azotemia
Pre-renal ATN (oliguric and Pre-renal ATN (oliguric and
azotemia non-oliguric) azotemia non-oliguric)
FENa < 1 27 4 FENa > 1 3 51
FENa > 1 3 51 FENa < 1 27 4
 Sensitivity: 90%  Sensitivity: 93%
 Specificity: 93%  Specificity: 90%
Miller, Schrier, Et al. Annals Int Med, 1978 Vol 89. p 47-50

58. Low FENa, Not pre-renal
FENa False Positive
 Pre-renal Azotemia  ATN tested too early
 Contrast Nephropathy  ATN with CHF
 Hemoglobinuric  ATN with cirrhosis
nephropathy  ATN with severe burns
 Myoglobinuric nephropathy  Non-oliguric acute renal
 Acute rejection failure
 Cyclosporin and Tacrolimus  Acute Glomerulonephritis
toxicity*  ACEi in bilateral RAS or in
 Hepatorenal syndrome RAS with solitary kidney
 Acute interstitial nephritis  NSAID induced ARF

59. FeNa false negatives
 Diuretics  Metabolic alkalosis

60. Low fractional excretion of sodium in acute
renal failure
Role of timing of the test and ischemia
 Patients with a decreased FENa were tested
earlier than those with an elevated FENa
 1.7 days for the low FENa group
 3.4 days for the high FENa group
 70% of patients in the low FENa group had a
subsequent FENa > 1%

61. fractional excretion of urea
 Based on the physiologic increase in urea
reabsorption with pre-renal azotemia
 Normal FE Urea is 50-65% in well hydrated
individuals
 In prerenal azotemia this falls below 35%
 Not affected by diuretics
Sr Cr x Ur Urea
Na
FEurea =
FENa =
Sr Na x UrUr Cr
Urea x Cr
Kaplan, Kohn. American J Nephrol, 1992; 12: 49-54.

63. FEurea in the differential diagnosis
of atn
 102 patients with ARF
 Gold standard was consultants full analysis
and retrospective analysis of response to
treatment.
 Divided the cases into:
 ATN
 Prerenal without diuretic
 Prerenal treated with diuretics
Carvounis, Sabeeha, Nisar, Et al. Kidney Int, 2002 Vol 62. p 2223-2229

64. FEUrea
FENa

65. therapy
Renal replacement therapy
Furosemide
Dopamine
Fenoldapam
hANP (Anaritide)

66. renal replacement therapy

67. Conventional Dialysis
Diffusive Clearance 67
136 108
5.8 17
3.8
0
145 110
2 35
0
Dialysate

68. Isolated Ultrafiltration: CHF Solutions 80 mmol K
= 13.8 liters
5.8 mmol/L
Minimal clearance 67
136 108
5.8 17
3.8
67
136 108
5.8 17
3.8

69. CVVH
Convective clearance 67
136 108
5.8 17
3.8
Ultrafilter 3+
liters/hour
Replace all ultrafiltrate 0
with sterile fluid at ideal 140 108
plasma concentrations 4
2 30
0

70. CVVH
Convective clearance
Post-filter replacement fluid

71. CVVH
Convective clearance Pre-filter replacement fluid

72. CVVHDF
Convective and Diffusive

73. high dose dialysis
High dose
survival
Low dose
Severity of illness (CCARF Score)

74. Ronco’s landmark dialysis dose
study
 425 patients with dialysis dependent acute
renal failure were randomized to one of three
doses of CVVH
 20 mL/kg/hr of effluent
 35 mL/kg/hr
 45 mL/kg/hr

75. 45 mL/kg/hr
35 mL/kg/hr
20 mL/kg/hr
Ronco C, Bellomo R, Hormea P, Et al. Lancet 2000; 355: 26-30.

76. Schiffl: daily dialysis versus three
days/wk dialysis
 160 patients
P=0.01 P=0.001
Schiffl, H. et al. N Engl J Med 2002;346:305-310

77. odds ratio of death
P=0.002
P=0.005
P=0.007
P=0.02
Schiffl, H. et al. N Engl J Med 2002;346:305-310

78. adding dialysis to CVVH
 206 dialysis patients randomized to
 CVVH 1-2.5 L/hr
 CVVH plus 1-1.5 liters of dialysate (CVVHDF)
P=0.03 P=0.008
Saudin P, Niederberger S, De Seigneux S, Et al. Kidney Int 2006; 70: 1312-7.

79. Study n treatment groups
Ronco 425 CVVH 20/h vs. 35-45 ml/kg/h*
Bouman 106 CVVH 20ml/kg/h* vs. 48 ml/kg/h
Schiffl 160 Alternate day vs. daily hemodialysis
Saudan 206 CVVH 25 ml/kg/h vs. CVVHDF 42 ml/kg/h
Total (fixed effects)
Total (random effects)
1 10
Odds ratio
*For purposes of analysis the two high-dose arms in Ronco were combined, as were the two low-dose arms in
Bouman. If these groups are removed the odds ratio is unchanged (1.94; P <0.001).
Kellum J. Nature Clin Practice Nephrol 2007 3: 128-9.

80. future data
 US trial: ATN
 Primarily veterans hospital
 Prospective randomized, multi-center trial
 Dose finding study
Conventional daily dialysis
SLED
CVVH
CVVHD
CVVHDF
 Australian trial: RENAL

81. furosemide
 Decreased activity of the ascending loop of
Henle decreases renal oxygen demand by the
kidney
 Better align demand and supply in ischemia

82. Mehta’s trial of furosemide in arf
 Retrospective review of
ICU patients
 Diuretic responsiveness
determined survival
Mehta, R. L. et al. JAMA 2002;288:2547-2553.

83. furosemide the rct
 338 with dialysis dependent ARF
 Randomized to high dose furosemide (2,000
mg/day) vs placebo
 End-point length of dialysis
 No improvement of survival, length of
dialysis, number of dialysis sessions
 Shorter time to 2 liters/day of urine output
Cantarovich F, Rangoonwala B, Et al. Am J Kidney Dis 2004; 44: 402-9.

84. dopamine: still doesn’t work
 In healthy volunteers low
dose dopamine increases
renal blood flow and
induces diuresis Increased RBF
 Patients in the intensive
care unit do not respond
this way.
Increased urine

85. dopamine: still doesn’t work
 In healthy volunteers low
dose dopamine increases
renal blood flow and
induces diuresis
 Patients in the intensive
care unit do not respond
this way.
 RCT of 380 ICU patients
with early renal failure
ANZICS Clinical Trials Group. Lancet 2000;356:2139-47.
Kellum JA, Decker JM.Crit Care 2001; 29:1526-31.

86. dopamine: still doesn’t work
 In healthy volunteers low
dose dopamine increases
renal blood flow and
induces diuresis
 Patients in the intensive
care unit do not respond
this way.
 RCT of 380 ICU patients
with early renal failure
 Meta-analysis of 58 studies
and 2,149 patients
ANZICS Clinical Trials Group. Lancet 2000;356:2139-47.
Kellum JA, Decker JM.Crit Care 2001; 29:1526-31.

87. dopamine: the randomized
controlled trial
 328 ICU patients with SIRS
 Early signs of renal failure
 < 0.5 cc/kg/hr
 Cr > 1.7 mg/dL without a prior history of renal
disease
 A rise in serum Cr of 0.9 mg/dL in less than 24
hours
 The primary outcome was peak serum
creatinine
ANZICS Clinical Trials Group. Lancet 2000;356:2139-47.

88.  Secondary end points:
 Furosemide dose 192 mg vs 268 mg p=0.39
 Duration of mechanical ventilation 10 vs 11 p=0.63
 Duration of ICU stay 13 vs 14 p=0.67
 Survival to hospital discharge 92 vs 97 p=0.66

89. meta-analysis
 Kellum and Decker searched MedLine
(English and non-English literature) for every
article on human trials with dopamine for the
treatment or prevention of ARF from 1966 to
1999.
 They included 58 studies with 2149 patients
Kellum JA, Decker JM.Crit Care 2001; 29:1526-31.

90. A. Exclude radiocontrast
studies
B. Limited to heart studies
C. Excludes studies in
which had abnormal
control groups or
increased variance

91.  Dopamine increases cortical blood flow more
than medullary blood flow
 Cortical blood flow increases GFR
 Cortical blood flow increases renal oxygen demand

92. complications of low-dose
dopamine
 Increase arrhythmias
 Increased myocardial oxygen demand
 Gut ischemia
 Suppressed respiratory drive
 Increased sensitivity to radiocontrast agents
 Decreases in T-cell activity

93. dopamine 2.0: fenoldapam
 Isolated DA-1 activity
 Licensed as an IV anti-hypertensive
 Increases medullary blood flow more than
cortical blood flow
 Improved oxygenation
 Does not increase renal work

94. RCT of fenoldapam
 155 patients randomized within 24 hours of
50% increase in Cr
 Primary end-point incidence of need-for-
dialysis and/or survival at 21 days
 Fenoldapam or half normal saline for 72
hours
 Protocolized definition of need-for-dialysis
Tumlin JA, Finkel KW, Murray PT, Et al. Am J Kidney Dis. 2005; 46:26-34.

95. P=0.235 P=0.163 P=0.068
P=0.048 P=0.015
P=0.036 P=0.022
Tumlin JA, Finkel KW, Murray PT, Et al. Am J Kidney Dis. 2005; 46:26-34.

96. Tumlin JA, Finkel KW, Murray PT, Et al. Am J Kidney Dis. 2005; 46:26-34.

97. prophylactic fenoldapam in sepsis
 300 patients with sepsis and no signs of AKI
 Non-oliguric
 Cr < 1.7
 Randomized to prophylactic fenoldapam vs
placebo

98. P=0.006
P=0.056
Fenoldapam
Placebo

99. atrial natriuretic peptide
 Recombinant Anaritide is therapeutic form
 Dilates afferent arterioles
 Improves GFR and urine output in animal
models of ATN
 Three high profile studies looked at using
ANP in human AKI.

100. radiocontrast nephropathy
 30 minutes of ANP
before contrast
 30 minutes of ANP after
contrast
 Cr > 1.8
 Randomized to placebo
or 1 of 3 doses of
anaritide
 Creatinine increase of
0.5 or 25% defined RCN
Kurnik B, Allgren RL, Genter FC. Am J Kid Dis 1998; 31: 674-80.

101.  504 critically ill patients
 Creatinine at randomization
was 4.6
 75% had a normal BL
creatinine
 24-hour infusion of Anaritide
p=0.008
Allgren R, Manbury T, Rahman SN. N Eng J Med 1997; 336: 828-34.

102. oliguric follow-up. strict EBM.
 222 oliguric patients  24-hour infusion of ANP
P=0.51 P<0.001
P=0.22
Lewis J, Salem M, Chertow G. Am J Kid Dis 2000; 36: 767-74.

103. fixing everything that was wrong
 Early treatment
 50% increase in creatinine
 Low dose anaritide
 50 ng/kg/min vs 200 ng/kg/min
 Anaritide run continuously until renal recovery or
dialysis.
 Previous studies used 24 hour infusion
 Protocol defined indication for dialysis
 UO < 0.5 cc/kg/hr  Pulmonary edema and
for 3 hours FiO2 >0.8
 Cr > 4.5  K>6.0
Swärd K, Valsson F, Odencrants P, Et al. Crit Care Med 2004; 32: 1310-5.

104.  N=61
 Average Cr 2.3
Swärd K, Valsson F, Odencrants P, Et al. Crit Care Med 2004; 32: 1310-5.

105. summary
 Prognosis is grim
 We now have a validated, consensus definition
 R isk
 I njury
 F ailure
 L oss of function
 E srd
 Outpatient and inpatient acquired ARF differ in
etiology
 Hospital acquired disease is your fault

106. summary
 FE of Urea is a validated way to separate pre-renal
from AKI even in the presence of diuretics
 Use of high dose dialysis regardless of methodology
offers a survival benefit
 There is no proven benefit of one modality over
another
 Except peritoneal dialysis which has been proven to be
inferior to CVVH
 Dopamine doesn’t work
 Fenoldapam and anaritide may have a role in
reducing mortality from ARF.

108. Done