Proposed Study of Tranexamic Acid

1
Proposed Study of Tranexamic Acid

• Philip Arnold

2
Questions

• Efficacy of tranexamic acid
• Appropriate dose

3
Fibrinolysis
4
Evidence Efficacy

• 1. Zonis Z, Seear M, Reichert C, Sett S, Allen C.
  The effect of preoperative tranexamic acid on
  blood loss after cardiac operations in children.
  J Thorac Cardiovasc Surg 1996111(5)982-7.
• 2. Reid RW, Zimmerman AA, Laussen PC, Mayer JE,
  Gorlin JB, Burrows FA. The efficacy of tranexamic
  acid versus placebo in decreasing blood loss in
  pediatric patients undergoing repeat cardiac
  surgery. Anesth Analg 199784(5)990-6.
• 3. Chauhan S, Bisoi A, Modi R, Gharde P, Rajesh
  MR. Tranexamic acid in paediatric cardiac
  surgery. Indian J Med Res 200311886-9.
• 4. Chauhan S, Bisoi A, Kumar N, Mittal D, Kale S,
  Kiran U, et al. Dose comparison of tranexamic
  acid in pediatric cardiac surgery. Asian
  Cardiovasc Thorac Ann 200412(2)121-4.
• 5. Chauhan S, Das SN, Bisoi A, Kale S, Kiran U.
  Comparison of epsilon aminocaproic acid and
  tranexamic acid in pediatric cardiac surgery. J
  Cardiothorac Vasc Anesth 200418(2)141-3.
• 6. Bulutcu FS, Ozbek U, Polat B, Yalcin Y, Karaci
  AR, Bayindir O. Which may be effective to reduce
  blood loss after cardiac operations in cyanotic
  children tranexamic acid, aprotinin or a
  combination? Paediatr Anaesth 200515(1)41-6.

5
Efficacy
6
Efficacy
7
Efficacy
8
Efficacy

• Cochrane analysis (updated 2007)
• Patients over 18years only
• 53 trials 3836 patients.
• Reduced risk of exposure to allogenic blood by
  39 (RR 0.61, 95 CI 0.54 to 0.70)
• 29 trials 2488 patients in cardiac surgery
• Reduced risk of exposure by 31 (RR 0.69, 95 CI
  0.60 to 0.79
• 17 trials reported post operative blood loss,
  mean reduction 263mls/kg
• there is no need for further placebo controlled
  trials of lysine analogues in cardiac surgery

9
Efficacy
10
Efficacy

• Meta analysis efficacy of tranexamic acid during
  major paediatric surgery.
• In cardiac surgery tranexamic acid reduced blood
  loss by a mean of 11ml/kg.
• In scoliosis surgery in children/ adolescents.
  Use of tranexamic acid reduced blood loss by 682
  ml (95 CI 1149 214 ml).

11
Efficacy

• Conclude that a large multicenter placebo
  controlled trial should be performed in children.
  powered to detect clinically important outcomes
  such as mortality, use of blood products,
  reoperation and side effects

12
Dose Finding
13
Dose finding
14
Dose Finding

• Chauhan et al. Dose comparison of tranexamic acid
  in pediatric cardiac surgery. Asian Cardiovasc
  Thorac Ann 200412(2)121-4
• compared 4 dosing regimes against placebo
• single bolus of 50mg/kg
• 10mg/kg followed by an infusion of 1mg/kg/hr
• two doses of 20mg/kg
• three doses of 10mg/kg.

15
Dose finding

• The single dose regime was no better than placebo
  and significantly inferior to other regimes
• No conclusive difference between other regimes.
• Recommended 10mg/kg 3
• Comparison of different dosage regimes rather
  than cumulative dose. Dosing regime more
  important than dose
• 30 patients per group. Underpowered to detect
  small differences between regimes

16
Dose finding

• Vacharaksa et al Tranexamic acid as a means of
  reducing the need for blood in children
  undergoing open heart surgery for congenital
  cyanotic heart disease. J Med Assoc Thai 200285
  Suppl 3S904-9.
• Small study 67 patients
• Two doses of 15mg/kg no better than one

17
Dose finding (adults)

• Karski single dose of 10g (approx 150mg/kg)
  10g over 5 hours. No difference
• Katoh 100mg.kg1 vs 100mg/kg 50mg/kg. Higher
  dose superior.
• Armellin 230mg/kg vs 15mg/kg1mg/kg/hr. No
  difference
• Horrow compared five dosage 2.5mg/kg
  (0.25mg/kg/hr) to 40mg/kg (4mg/kg/hr.) They
  found no advantage of dosages above 10mg/kg (with
  1mg/kg/hr),

18
Dose finding

• Grant et al. Perioperative blood transfusion
  requirements in pediatric scoliosis surgery the
  efficacy of tranexamic Acid. J Pediatr Orthop
  200929(3)300-4
• 10mgkg and 1mg/kg/hr vs 20mg/kg and 10mg/kg/hr
• Nonrandomized, retrospective, poorly powered
• Trend to reduced blood usage with higher dose

19
Dose Finding

• Trials in adults and children largely
  inconclusive
• Dosage schedules that maintain plasma
  concentrations appear superior to single dose
  regimes
• Dosages mostly not designed to achieve
  predetermined plasma concentrations

20
PK of tranexamic acid

• No PK data available in children
• Single iv dose 1g
• three exponential decays
• Initially rapid, then T1/2 of 2 hours and 9-18
  hours
• Volume of distribution is 9-12litres
• The drug is distributed widely
• Little protein binding
• Eliminated unchanged in the urine
• clearance similar to GFR
• Prolonged in renal dysfunction

21
PK during adult CPB

• Dowd NP Pharmacokinetics of tranexamic acid
  during cardiopulmonary bypass. Anesthesiology
  200297(2)390-9.
• Bypass increased volume of distribution and
  reduced clearance
• two compartment model modified for the effects of
  bypass
• existing dosing regimes produce unreliable and
  unstable plasma activity

22
Simulation of Paediatric PK

• NONMEM used to perform simulation (user-written
  code to allow for changes in PK estimates pre-,
  during- and post-CPB)
• Method of TA administration stipulated as
• Bolus Dose pre-CPB (given over 30mins)
• Maintenance continuous infusion to begin at the
  end of bolus dose
• A bolus dose in the pump prime
• Target concentrations stipulated as 10mcg/ml,
  30mcg/ml and 100mcg/ml.
• PK Parameters used to write model

23
Why 10,30,100mcg/ml

• 10mcg/ml level able to produce 80 inhibition of
  fibrinolysis
• 30mcg/ml possibly able to cause direct inhibition
  of plasmin and platelet inhibition. Required in
  some in vitro experiments to produce inhibition
  of fibrinolysis
• 100mcg/ml consistent with currently used
  paediatric dosing regimes. Should produce 98
  inhibition of plasmin activation.

24
Dose regimes
25
Toxicity

• Inhibition of fibrinolysis could theoretically
  lead to increased thrombosis
• In fact little/ no evidence to support this
• Most high quality data is as comparison to
  aprotinin in adults (lower risk)
• Not associated with graft occlusion in adults
• Not associated with early fenestration closure
  post Fontan (71 patients)
• Possibly increased risk of thrombotic
  complications on ECMO with epsilon aminocaproic
  acid

26
Statistical Methods

• Unlikely to be possible to identify different
  response between doses using a traditional dose
  comparison study
• Alternative approach is PK-PD modeling
• Robust PK model is built using established
  methods (population PK, sparse sampling theory)
• Can be linked to response such as validated
  biomarker or clinical outcome (PD)
• In the absence of adequate PD data dose can be
  derived to achieve plasma level consistent with
  previous in vivo or in vitro work

27
PK-PD modeling

• Is possible to estimate an effect for dosages
  between the dosages actually studied
• The PK modeling allows optimization of the dosing
  regime as well as total dosage.
• Much of the variability in response to the drug
  may be due to PK variability. By mapping effect
  to plasma concentration, rather than dose, some
  of this variability is removed.
• Depending on the robustness of the model it may
  be possible to extrapolate outside of the dosage
  range studied.
• Depending on the robustness of the model and of
  data collection it is possible to test the impact
  of covariates on the response.

28
Genetics PAI-1

• PAI-1 endogenous inhibitor of fibrinolysis
• PAI-1 has been extensively investigated as risk
  factor for cardiovascular disease
• G5/G5 polymorphism (16-23) has been implicated
  in increased fibrinolysis and bleeding and
  increased response to Tx Acid
• G5/G5 polymorphism has been associated with lower
  levels of PAI-1
• Has not been investigated in children

29
Fibrinolysis
30
Proposed study design

• Patients aged 0-16 years undergoing heart surgery
  utilising cardiopulmonary bypass. Anticipated to
  involve cardiopulmonary bypass of greater than 90
  minutes duration?
• Neonates (defined as any patient aged less than
  50 weeks post conceptual age). ngt40 to establish
  PK in neonates
• Patients undergoing a repeat Stenotomy
• Patients with cyanotic heart disease and baseline
  saturations less than 83 or baseline haematocrit
  greater than 55

31
Study Design

• 50 patients in each dosage group (total 200)
• Placebo and three dosage regimes
• Based on power to establish efficacy
• Stratified to hospital and neonates

32
Dose regimes
33
Primary outcome

• Chest drain loss (as recorded on ICU charts by
  bedside nurse) between the 2nd and 12th
  postoperative hour.

34
Important Secondary Outcomes

• Total Allogenic Red Cell Usage during five day
  period from day of surgery (d0 d5)
• Evidence of life endangering bleeding
• Red cell replacement in theatre post bypass
  (including autologous blood) of gt50 estimated
  blood volume
• Chest drain loss in ICU of gt50 blood volume in
  first 12 hours
• Re-exploration for bleeding
• Diagnosis of tamponade within 24 hours of end of
  surgery (haemodynamically and/or
  echocardiographically)
• Use of rFVIIa

35
Other Secondary clinical outcomes related to
efficacy

• Total use of allogenic platelets (in five days
  from surgery)
• Total use of allogenic FFP (in five days from
  surgery)
• Total use of allogenic cryoprecipitate (in five
  days from surgery)
• Total use of rFVIIa
• ?Time to chest closure after completed protamine
  administration

36
Hematological Markers
37
Other clinical data including evidence toxicity

• Taken from clinical records and clinically
  indicated tests
• Major negative post operative sequelae during
  hospital admission
• Death within 30 days
• Mechanical circulatory or respiratory support
• New episode of renal failure requiring dialysis
• New focal neurological deficit
• New seizures
• Thrombotic event including shunt occlusion
  vascular occlusion echo evidence of intracardiac
  or extracardiac thrombosis coronary occlusion
• Laboratory evidence of elevated creatinine

38
Summary of Blood Samples
39
Conclusion

• Evidence of efficacy in children is weak and
  further investigation is appropriate
• PK (PD) modeling is likely to give the best
  evidence to support dosing regimes
• Collection of data on PAI-1 will help to explain
  response to this drug
• This is worth doing with potential wide
  application