Dialytic therapy in Acute Renal Failure

1
Dialytic therapy in Acute Renal Failure
starting from square one
Dr Chan Ching Kit Medical Officer Renal
Unit, Department of Medicine, PYNEH
2
Outline

• Background and definitions
• Modalities of dialytic therapy
• Practical problems in dialytic therapy in ARF -5w
• Why to start?
• When to start?
• What to start? IHD / CRRT dilemma
• How to start? anticoagulation , choice of
  dialyser
• How much to wash? solute and fluid clearance
  and adequacy
• Conclusion

3
Background

• Mortality in patients with ARF is surprisingly
  high, and has not changed significantly despite
  advances in medical technology / introduction of
  dialysis for more than 30 years
• 10-23 of ICU patient developed ARF
• Brivet et al. Crit Care Med 24192-198, 1996
• Groneveld et al. Nephron 59602-610, 1991
• 70 required renal replacement therapy (RRT) to
  sustain life.
• McCullough et al. Am J Med 103368-375, 1997
• Period prevalence of ARF on RRT in ICU was 5-6
• Uchino et al. JAMA 294813-818, 2005

4
Background

• Sepsis /- MODS as leading cause for ARF
• In patient mortality from 30 in nephrotoxic
  drug induced ARF to gt90 when ARF is associated
  with multiple organ failure.
• Turney et al. JAMA 2751516-1517
• Chertow et al. Arch Intern Med 1551505-1511, 1995

5
Background

• ARF is an independent risk factor for morbidity
  and mortality
• Metnitz et al. Crit Care Med 302051-8, 2002
• Uraemic state and the need for RRT among
  critically ill patients frequently results in
  therapy-related complications, which may further
  aggravate the underlying condition
• Managing ARF in ICU is a significant ongoing
  challenge to Intensivists and Nephrologists.

6
(No Transcript)
7
AC Fry et al PostgradMedJ 200682106-111
8
(No Transcript)
9
(No Transcript)
10
Definition of ARF

• Wide variation in quoted figures due to no
  consensus definitions for ARF!
• More than 35 different definitions have been used
  in the literature, creating confusion and
  difficulties in comparison among different
  studies.
• Kellum JA et al. Curr Opin Crit Care 8509-514,
  2002
• The Acute Dialysis Quality Initiative (ADQI)
• Develop consensus and evidence-based statements
  in the field of ARF

11
RIFLE Criteria
12
Linear increase in odds ratio from Risk to
Failure (Odds ratios, Risk 2.5, Injury 5.4,
Failure 10.1)
Crit Care Med 2006 Vol 34 No71913-1917
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
(No Transcript)
20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
Practical problems in dialytic therapy in ARF
26
Why to start?

• Indications for renal replacement therapy in ICU
• Renal
• Acid-base disturbance mainly metabolic acidosis
• Electrolytes disturbance e.g hyperkalaemia
• Intoxication
• Overload of fluid e.g. pulmonary oedema
• Uraemia
• Non-renal
• Allowing administration of fluids and nutrition
• Elimination of inflammatory mediators (?)
• Stefan and Eckardt. Seminar in Dialysis vol 19,
  No.6 (Nov-Dec)2006, p455-464

27
When to start?

• Need for RRT in critically ill patients with ARF
  depends on numerous factors
• Remaining diuresis
• Accumulation of uraemic solutes
• Hypercatabolic state
• Patient size
• Desired level of metabolic control

28
When to start? Ur / Cr level ?

• Urea generation is not constant between patients,
  or even for the same patient over time
• Volume of distribution (V) of urea may change
  over time
• Ur / Cr levels depends on
• Production
• Volume of distribution
• Renal elimination
• Blood urea nitrogen (BUN) or serum creatinine
  levels are not good indicators of severity of ARF
• So far no biomarkers / clinical predictors of the
  course of AKI available .

29
Benefits and risks

• Early initiation of RRT may avoid severe
  derangements in metabolic control, and subsequent
  adverse effects of ARF
• ? Improvement in survival

• RRT may have negative consequences e.g.
  influences on immune system e.g activation of
  neutrophils or the complement system
• Possible complications from RRT e.g bleeding
  complication

30
Early initiation of RRT

• Nonrandomized studies suggested that both early
  initiation of RRT and the use of higher
  ultrafiltration rates improve survival and renal
  recovery
• In post-traumatic AKI, early initiation of CVVH
  (BUN 4313mg/dL, D1015) was associated with a
  39 survival when compared to 20.3 survival in
  late RRT group (BUN 9428, D1927) (plt0.05)
• Gettings et al. Intensive Care Med 25805-813,
  1999

31
Early initiation of RRT

• Early CVVHDF (Kgt5.5, SCr gt 5mg/dL regardless of
  urine output) was associated with fever days of
  mechanical ventilation, ICU stay, as well as
  lower ICU and hospital mortality (17.6 and 23.5
  vs. 48.1 and 55.5), when compared to historical
  control (urine output lt 100ml/8hr).
• Demirkilic et al. J Card Surg 2004 1917-20
• Early CVVH (urine output lt 100ml/8 hrs despite
  frusemide infusion) post Cardiac surgery was
  associated with lower hospital mortality, when
  compared to late CVVH (BUN gt84mg/dL, SCr gt3mg/dL,
  K gt6 regardless of urine output).
• Elahi et al Eur J Card Surg 2004261027-1031

32
Early initiation of RRT

• Higher doses of RRT and therefore better uraemic
  control led to an improvement of survival Mean
  starting BUN in patients who survived was lower
  than in non-survivors in all study groups
• Ronco et al. Lancet 35626-30, 2000

33
Early initiation of RRT

• No improvement in D28 survival and renal
  recovery, with the use of high ultrafiltration
  rate or early hemofiltration in oliguric ARF
  patients.
• Bouman et al. Crit Care Med 302205-2211,2002
• Disease severity too low in this study to
  demonstrate significant difference between early
  vs. late approach.

34
Early initiation of RRT

• prophylactic hemofiltration performed in 24
  trauma patients
• Positive effects on hemodynamic parameters
• No benefit with respect to the severity and
  duration of illness or patient outcome.
• Bauer et al. Intensive Care Med 27376-383,2001.

35
Valerie et al. Seminar in dialysis Vol 17, No 1
(Jan Feb) 2004, p30-36
36
Confounding factors

• Indications for renal support are likely to be
  different
• Early initiation may be drived by volume overload
  or electrolyte disturbance, as opposed to
  azotemia in patients in late initiation group
• Insignificant trend toward greater duration of
  therapy in late initiation group
• More severe renal injury in late initiation
  group, leading to an increased time to recovery
  of renal function and contributing to mortality
  difference.
• Selection bias
• Patients who developed early AKI but did not have
  renal support initiated early and who either
  recovered renal function or died without ever
  receiving RRT.

37
Timing of initiation - Conclusion

• No clear cut recommendation at this moment
• Decision should be based on individual basis
• As ARF and its associated metabolic alternations
  appears to increase the risk of severe extrarenal
  complications, initiation of RRT should be
  started early in patients with severe, rapidly
  developing oliguric ARF.

38
What to start? IHD vs. CRRT
39

• Prior to the development of CRRT, IHD and PD were
  the only two modalities of RRT.
• With improved technology, CRRT has gained
  increasing popularity, and developed into a whole
  family of related therapies to provide
  uninterrupted renal support to critically ill
  patients over period of days

40
Why CRRT ?

• Slow gradual removal of fluid and solute
• Enhance hemodynamic stability
• Permit better fluid and solute control
• Allow more aggressive nutritional management
• Enhanced clearance of inflammatory mediators,
  particularly using hemofiltration in patients
  with concomitant sepsis

41
Is CRRT more superior ?

• Majority of studies comparing IHD and CRRT are
  non-randomized observational studies or
  retrospective case studies.
• Confounding factors
• variation in disease severity between treatment
  groups
• unfair randomization due to intolerance to IHD
• significant crossover between both groups.
• So far no consensus on this issue yet.

42
Raymond Vanholder et al. J Am Soc Nephrol
12S40-43,2001
Teehan GS et al. J Intensive Care Med 2003 18130
43
Mehta et al

• Prospective RCT involving 166 patients with ARF
• Study period 56 months
• Either CVVH/DF or IHD
• Baseline characteristics higher males, higher
  APACHE III scores, higher prevalence of liver
  failure among patients randomized to CRRT group
• Univariate intention-to-treat analysis revealed a
  higher mortality among patients receiving CRRT
  (66 vs. 48, plt0.02)

44
Mehta et al

• On multivariate analysis, the RRT mortality had
  no impact on all cause mortality nor renal
  recovery, while being replaced by more
  traditional risk factors, mainly APACHE III score
  and numbers of failed organs
• Problems
• Uneven distribution of severity of illness
  between 2 groups
• Allow patients to cross over therapies for
  medical reasons.

45
Comment from Claudio RoncoClin J Am Soc Nephrol
2597-600

• Patients were allowed to cross over, making true
  comparison impossible
• Patients with hemodynamic instability (MAP lt
  70mmHg) were excluded
• If patients received a sufficient trial of CRRT
  and survived, renal recovery was dramatically
  increased (92.3 VS 59.4, plt0.01), and therefore
  IHD delayed or impeded renal recovery
• CRRT delivered superior control of uraemia

46
Mehta et al Kidney Int 60 1154-1163,2001
47
Vinsonneau et al Hemoleaf study Group

• Largest, best powered prospective randomized
  multicentre study to compare the results of CRRT
  with IHD
• 360 patients
• Intention to treat
• No difference in D60 survival (32 in IHD vs. 33
  in CRRT)

48
Vinsonneau et al Lancet 368379-385, 2006
49
Vinsonneau et al Hemoleaf study Group

• Unexpected progressive and significant increase
  in survival rates in the IHD group over time
  (relative risk 0.67/year 95 CI 0.56-0.80,
  plt0.001)
• Learning curve for optimizing IHD therapy in the
  study environment
• An increase in the frequency of IHD during the
  first 8 days over the course of the study,
  without corresponding increase in dialytic dose
  in CRRT group
• Helbert Rondon-Berrios et al. Curr opinion
  Nephrol Hypertens 1664-70, 2007

50
Tonelli et al. AJKD 40875-885,2002
51
Modality of RRT - conclusion

• Remained unanswered
• ? Another examples to illustrate that sound
  technology / devices / drugs may not necessarily
  be associated with benefits or good outcome
• E.g. liberal use of blood transfusion, COX2
  inhibitors, Swan Ganz catheter.

• Jonathan Himmelfarb. Continuous renal
  replacement therapy in the treatment of acute
  renal failure Critical assessment is required.
  Clin J Am Soc Nephrol 2385-389,2007

52
Choice of dialysers
53
Biocompatibility

• Blood membrane interaction activate cellular and
  humoral components of blood, leading to
  generation of several biological responses
• complement activation
• coagulation cascade activation
• monocytes activation
• neutrophil degranulation
• release of reactive oxygen species

54
Bio-incompatible membranes (e.g. Cuprophane,
hemophane)

• May worsen the catabolic state of ARF
• Aggravate the pro-inflammatory state of sepsis
• Activation and subsequent exhaustion of
  mononuclear and polymorphonuclear cells may
  predispose patients to bacterial infections
• Higher observed mortality rate due to sepsis
• Delay recovery of ARF, due to leucocyte
  activation and infiltration of renal parenchyma,
  esp. following ischemic reperfusion injury

55
Membrane biocompatibility
56
(No Transcript)
57
With the cost differential between
bioincompatible and biocompatible dialysers used
in ARF settling rapidly diminishing, there
remains no persuasive reason to use unsubstituted
cellulose dialysers. Hemodialysis in ARF Does
the membrane matter? Modi GS et al. Seminars in
Dialysis Vol 14 No 5 (Sept-Oct) 2001 p318-332
58
Dialytic dose in ARF
59
GFR (100ml/min) x60x24 x7 days 1008 L/week
60

• No RRT is ever as efficient as native kidneys
• There is some indication in ESRF patients that up
  to certain level, delivered dose of RRT is
  inversely proportional to morbidity and
  mortality.
• Urea kinetic modeling (UKM)
• Dialysis adequacy is measured by urea reduction
  ratio (URR) and Kt/V.
• K dialyser urea clearance
• t duration of dialysis treatment
• V volume of distribution of urea total body
  water

61
(No Transcript)
62
(No Transcript)
63
URM in ARF

• There is no current guidelines for measurement of
  solute clearance in the setting of ARF
• Application of UKM in ARF is not valid because
• Whether urea is a surrogate marker for the toxic
  metabolites in ARF is not established, esp. in
  the setting of multiple organ failure.
• Hypercatabolic state negative nitrogen balance
  in ARF, therefore steady state assumption of
  kinetic models does not apply
• In multiple organ failure, ARF is characterized
  by instability of hemodynamic parameters,
  increased permeability of vasculature, and the
  use of vasoactive substances, which all produce
  disequilibrium in urea distribution.
• Stefan and Eckardt. Seminar in Dialysis Vol 19,
  No 6 (Nov-Dec) 2006, p 455-464
• Despite these limitations, URR remains the most
  widely used markers of dialysis adequacy in ARF
  treated with intermittent therapy

64

• There is a marked discrepancy between prescribed
  and delivered dose of dialysis.
• observed Kt/V in ARF patients have been showed to
  be 30 lower than prescribed.
• Jaber et al. Blood Purif. 200220154-160
• Schiffl et al. NEJM 2002 346305-310
• Early discontinuation of dialysis due to
  hypotension / clotted circuit / catheter
  dysfunction / access recirculation.
• Lack of steady state
• High urea rebound after dialysis
• Uncertainty about true total body water (TBW) and
  Volume of distribution of urea
• Presence or absence of residual renal function

65
Only 15-32 of treatment sessions achieved Kt/V gt
1.2. Teehan GS et al. J Intensive Care Med
2003 18 130
66
UKM and clinical outcomes in ARF

• URR gt 58 was associated with significant
  reduction in mortality, although patients with
  very low and very high severity scores , their
  survival rates were not altered with dialytic
  doses manipulation (78 and 0 respectively)
• Paganini et al. Am J Kidney Dis 1996 28S81-S89.
• This finding suggest the presence of interplay
  between severity of illness and delivered dose of
  dialysis, and not necessarily cause-effect
  relationship.

67
(No Transcript)
68
(No Transcript)
69
Schiffl et al NEJM 2002346305-310

• 160 patients with ARF, assigned in alternate
  order, to 6x/week IHD or alternate day IHD
• High flux dialysis
• Baseline characteristics / APACHE III score
  similar in both groups
• Treatment time / blood flow / intradialytic
  weight loss similar

70
Schiffl et al NEJM 2002346305-310

• Delivered Kt/V higher in daily IHD (5.8 vs. 3.0)
• 14 days all cause mortality significantly lower
  in daily IHD when compared to alternate day IHD
  (28 vs. 46, p0.01)
• Patients with clinical deterioration were allowed
  to switch over to CRRT, arguing for the need for
  a true efficacy rather than intention-to-treat
  analysis
• Nutritional intake not reported (expected to be
  more liberal in daily IHD)

71
UKM in CRRT

• Dialysers used in CRRT usually with high UF
  coefficient
• Remove urea and middle to large molecules (0.3-5
  Kda) including cytokines and other inflammatory
  mediators
• Using a computer-based model, Clark et al
  demonstrated that in a 50kg patient, a steady
  state blood urea nitrogen of 60mg/dL by CRRT
  would require 4.4 sessions of IHD per week
• Similar degree of metabolic control using IHD was
  not achievable in patient of gt 90kg.

72
(No Transcript)
73
Ronco et alLancet 35626-30,2000

• Prospectively compared outcomes in patients with
  ARF receiving different doses of CVVH
• Patients receiving UF of 35ml/kg/hr had
  significantly better outcomes than those
  receiving 20ml/kg/hr (survival 57 vs. 41)
• No statistically significant difference in
  survival between patients receiving 35 and
  45ml/kg/hr.

74
(No Transcript)
75
How to compare dialytic doses in different
modalities?
76
Artificial organs 30178-185
77
(No Transcript)
78
(No Transcript)
79
(No Transcript)
80
(No Transcript)
81
Dialytic dose - Conclusion

• UKM in ESRF not validated in ARF setting
• Dialytic dose estimation in ARF will be difficult
  due to deviation in t and V, and delivered doses
  tend to be lower than estimated doses
• Dialytic doses more easily achieved by CRRT
• Severity of illness and dialytic dose determine
  outcomes, but not in cause effect relation.

82
(No Transcript)
83
(No Transcript)
84
AC Fry et al Postgrad Med J 2006 82106-116
85
(No Transcript)
86
Conclusion

• There is no consensus in the modality of choice
  for dialytic therapy in ARF.
• Biocompatible dialysers give potential benefit of
  high chance of renal recovery in nonoliguric
  patients.
• Potential benefit of CRRT on clearance of
  inflammatory mediators in ARF with sepsis
• There is tendency of underdialysis (in term of
  solute clearance) in ARF, esp. if using IHD.
• Dialysis adequacy more easily achieved by CRRT
• Dialysis regimen should be tailored according to
  patient need.

87
Future directions

• hybrid therapy (EDD)
• extracorporeal blood treatment (EBT) e.g. HVHF,
  HPHF, CPFA for mediators removal and improve
  systemic hemodynamics and organ perfusion.

88
Kidney disease beyond Nephrology Intensive
CareZaccaria Ricci and Claudio RoncoNephrol
Dial Transplant (2007) 22708-711

• Crude mortality assessment shows that the overall
  hospital outcome of ARF has remained high today,
  and has not changed in the past 30 years
  nevertheless such analysis is profoundly
  misleading.
• Patients with ARF in hospitals 30 years ago were
  mostly treated outside ICU, did not require or
  receive mechanical ventilation or vasopressor
  drugs, were 20-30 years younger in age and their
  outcome was typically assessed retrospectively
  and in academic centres only...

89
Kidney disease beyond Nephrology Intensive
CareZaccaria Ricci and Claudio RoncoNephrol
Dial Transplant (2007) 22708-711

• much greater illness severity for patients
  treated in 2005, the mortality of ARF has not
  increased, the duration of treatment has markedly
  decreased in terms of need for dialysis, time in
  ICU and time in hospital and the techniques of
  artificial renal support have also changed
  markedly
• as the therapeutic capability improves and the
  system continues to accept a mortality of 50 as
  reasonable for these very sick patients, the
  healthcare system will progressively admit and
  treat sicker and sicker patients with ARF.

90
- END -
91
Dialysis related thrombogenicity
92
Anticoagulation in RRT
93
(No Transcript)
94
Hemodialysis adequacy
95
Replacement fluid and dialysate
96
Predilutional vs postdilutional
97
(No Transcript)
98
(No Transcript)
99
(No Transcript)
100
(No Transcript)
101
(No Transcript)
102
(No Transcript)
103
(No Transcript)
104
(No Transcript)