TRICC TRIAL AND OTHER TRANSFUSION TRIGGERS.pptx

To determine whether a restrictive strategy of red-cell transfusion and a liberal strategy produced equivalent results in critically ill patients, we compared the rates of death from all causes at 30 days and the severity of organ dysfunction.

1. TRICC TRIAL AND OTHER TRANSFUSION TRIGGERS Dr. ASHISH NAIR Mbbs , md anaesthesia, fra , idccm , dm critical care

2. BACKGROUND Red cell transfusions are used to augment the delivery of O 2 to avoid harmful effects of O 2 debt. Until recently, Hb - 10g/ dL was considered are threshold for RBC transfusion in critically ill patients. Critically ill patients, however, are at increased risk for immunosuppressive and microcirculatory complications of red-cell transfusions. Hence multiple RCTs have evaluated restrictive v/s liberal transfusion strategies and recommend accepting lower Hb values as threshold for transfusions.

3. OPTIMAL TRANSFUSION TRIGGER It should: Avoid delay in transfusion Avoid premature transfusions Take in to account different degrees of anaemia tolerance of different organs Be determined continuously Have a high sensitivity and specificity

4. Tomic Mahecic T, Dünser M, Meier J: RBC Transfusion Triggers: Is There Anything New? Transfus Med Hemother 2020;47:361-369. doi : 10.1159/000511229

5. RESEARCH QUESTION In patients admitted to ICU, does a restrictive v/s liberal blood transfusion strategy alter all cause mortality at 30 days?

6. POPULATION: Patients admitted to 1 of 22 tertiary level ICUs and 3 community ICUs in Canada between Nov 1994- Nov 1997. 6541 adults were assessed out of which 838 were randomised. Inclusion criteria: Euvolemic patients after initial treatment as assessed by attending physician. Expected ICU stay > 24 hours Acute anemia with Hb ≤ 9g/dl within 72 hours of admission

7. POPULATION Exclusion criteria: Age < 16 years Inability to receive blood products Active blood loss at the time of enrolment Defined as evidence of on-going blood loss accompanied by a decrease in Hb concentration of 3g/dl in preceding 12 hours or a requirement for at least 3 units of packed red cells during same period. Chronic anemia Defined as Hb concentration < 9g/dl on at least one occasion > 1 month before admission Pregnancy Brain death or imminent death (within 24 hours) or possibility of witholding or withdrawing of on-going treatment by the physician Admission after routine cardiac surgical procedure

8. STUDY DESIGN Prospective randomised controlled trial Informed consent obtained from patient or closest family member before enrolment. Consecutive patients with normovolemia were assigned to one of the 2 treatment groups stratified according to centre and disease severity (APACHE II) of ≤15 or > 15 and balanced using permuted blocks of 4 or 6. Further randomised using sealed opaque envelops arranged in a computer generated random order.

9. STUDY DESIGN

10. INTERVENTION AND CONTROL RESTRICTIVE STRATEGY: (n= 418) Hb concentration maintained in range of 7-9g/dl with transfusions given when Hb < 7g/dl LIBERAL STRATEGY: (n=420) Hb concentration maintained in range of 10-12g/dl with transfusion threshold of 10g/dl. PCV were transfused 1 at a time and Hb concentration was measured after each transfusion.

11. BASELINE CHARACTERISTICS Noted at the time if randomization Demographic, diagnostic therapeutic information along with disease severity (APACHE II and multiple organ dysfunction score), primary reason for ICU admission and perioperative diagnosis were noted.

12. OUTCOMES Primary outcome: All cause mortality in 30 days after randomization Secondary outcomes: 60 day rates of death from all causes Mortality rates during the stay in ICU and during hospitalization Survival times in first 30 days Measures of organ failure and dysfunction ICU and hospital length of stay

13. SAMPLE SIZE They estimated that 2300 patients would be needed to rule out absolute difference of 4% in 30-day mortality assuming combined mortality of 18%. Interim analysis conducted once 404 patients were enrolled. Revealed actual combined mortality of 23% which changed the detectable difference to 4.5% . Because of this increase, they decided to decrease the sample size to 1620. The executive committee decided to terminate the study prematurely because of decrease in enrolment to below 20% of predicted level over a period of several months.

14. RESULTS No differences in baseline characteristics. Average daily Hb concentration in restrictive strategy group was 8.5 ± 0.7g/dl and 10.7± 0.7g/dl in liberal strategy group (p<0.01). Average of 2.6 ± 4.1 red cell units per patient were administered in restrictive strategy group as compared to 5.6 ± 5.3 units per patient in liberal strategy group. (p<0.01) This equals in relative decrease of 54% in no. of transfusions when lower threshold was used. 33% of patients in restrictive strategy group did not receive any transfusion after randomization as compared to 0% in liberal strategy group. (p<0.01)

15. RESULTS Non-compliance of physicians (finding of Hb concentrations outside of pre-specified range for at least 48 hours): 1.4% v/s 4.3% (p=0.02) Cross over rate (physicians either administered or with-held the transfusion) was 1.8%. 1% v/s 2.6% (p=0.12) No significant difference found in: Use of medications including vasoactive agents Fluids and daily fluid balance Pulmonary artery catheterizations Dialysis Mechanical ventilation Surgical procedures

16. OUTCOMES All cause 30 day mortality after admission: 18.7% v/s 23.3% (p=0.11) Mortality rates during hospitalization: 22.2% v/s 28.1% (p=0.05) Mortality rate during stay in ICU: 13.4% v/s 16.2% (p=0.29) 60-day mortality rate: 22.7% v/s 26.5% (p=0.23)

17. OUTCOMES

18. OUTCOMES Number of patients with multiple organ failure (>3 organs) was not significant: 5.3% v/s 4.3% (p=0.36) Mean multiple organ dysfunction score : 8.3±4.6 v/s 8.8±4.4 (p=0.10) When all the patients who died within 30 days were given multiple organ failure score of 7 and multiple organ dysfunction score of 24, the results were marginally better in restrictive strategy group.

19. OUTCOMES

20. OUTCOMES: SUB-GROUP ANALYSIS Sub groups: Age : < 55 years v/s ≥ 55 years APACHE II score : ≤ 20 v/s > 20 No difference in baseline characteristics. All outcomes similar for patients ≥ 55 years and with APACHE II score >20 (p>0.36). 30-day mortality was significantly lower in restrictive strategy group for patients < 55 years (5.7% v/s 13%; p=0.02) and APACHE II score ≤ 20 (8.7% v/s 16.1%; p=0.03)

21. KAPLAN-MEIER SURVIVAL CURVES

22. OUTCOMES: SUB-GROUP ANALYSIS No significant difference in 30-day mortality in subgroups with 1˚ or 2˚ diagnosis of cardiac disease (20.5% v/s 22.9%; p=0.69) Severe infections and septic shock (22.8% v/s 29.7%; p=0.36) Trauma (10% v/s 8.8%; p=0.81)

23. OUTCOMES: SUB-GROUP ANALYSIS ADJUSTED MULTIPLE ORGAN DYSFUNCTION SCORES: For patients with APACHE II score > 20, age ≥ 55years and specific diagnoses like cardiac disease, trauma and severe infections with septic shock: No significant difference (p>0.30) For patients with APACHE II ≤ 20 and age < 55 years: APACHE II ≤ 20- 8.3 ± 6.2 v/s 10.0 ± 7.2 (p=0.01) Age < 55years- 8.8 ± 5.7 v/s 10.3 ± 6.6 (p=0.03)

24. COMPLICATIONS Cardiac events – pulmonary oedema and MI were more frequent with liberal strategy group. (p<0.01)

25. STRENGTHS Multi-centre, randomized Average APACHE II score was 21 indicating reasonably sick population 33% did not receive transfusion in restrictive group: All patients in liberal group received transfusion identifying high transfusion rates thus making the findings more impressive. A lower hospital mortality with an average transfusion difference of 2 units is a powerful challenge to previously held belief that high Hb levels were always necessary in critically ill patients.

26. LIMITATIONS Large selection bias. Only 13% were included. INADEQUATE sample size, randomized 838 patients v/s targeted 1620; potential for type 2 error. Changed subgroups post-hoc from APACHE II < 15 to < 20. Included patients with anaemia within 3 days of admission.

27. CONCLUSION: A restrictive strategy of red-cell transfusion is at least as effective as and possibly superior to a liberal transfusion strategy in critically ill patients with possible exception of patients with acute myocardial infarction and unstable angina.

28. ARE RESULTS VALID?? Was assignment of patients to the treatment truly randomized? Yes Were all the patients who entered the study accounted for appropriately in the end? Yes Were all the patients, physicians and those doing the assessment blind to the treatment? No Was similarity between the groups documented Yes Aside from the intervention, were the groups treated in the same way? Yes

29. APPLICATION OF RESULTS IN ROUTINE PRACTICE Can the results be applied to my patients? Yes Were all clinically important outcomes reported? Yes Are the likely treatment benefits greater than potential harm and costs? Yes

30. EVIDENCE FOR TRANSFUSION IN OTHER CLINICAL SENARIOS

31. VARIOUS GUIDELINES AND RECOMMENDATIONS FOR RBC TRANSFUSIONS: National Blood Authority (Australia, 2012): Restrictive strategy to be used in critically ill patients Transfusion likely to be appropriate with H<7g/ dL Hb - 7-9g/ dL : transfusions to be decided based on clinical senario Transfusion appropriate in ACS with Hb < 8g/ dL British Committee for standards in Haematology (2013): Transfusion threshold ≤ 7g/ dL should be default in critically ill patients unless specific comorbidities ( strong recommendation ) ASA task force on peri -operative blood management (2015): Restrictive strategy ( Hb <8g/ dL ) may be used to reduce transfusion requirements

32. VARIOUS GUIDELINES AND RECOMMENDATIONS FOR RBC TRANSFUSIONS: American association of Blood banks (2016): Restrictive strategy in hospitalized, haemodynamically stable, critically ill patients. ( Strong recommendation ) Transfuse when Hb <7g/ dL Restrictive strategy for patients undergoing orthopedic or cardiac surgey ( Strong recommendation ) Transfuse when Hb <8g/ dL Restrictive strategy in hospitalized, haemodynamically stable patients with CVS disease ( weak recommendation ) Transfuse when Hb <8g/ dL Insufficient evidence for specific transfusion strategy in patients with ACS, severe thrombocytopenia and chronic transfusion dependent anaemia

33. RECENT EVIDENCE FOR TRANSFUSION IN CRITICALLY ILL PATIENTS Restrictive red-cell transfusion strategy potentially reduced in-hospital mortality in critically ill patients with anaemia compared to liberal strategy

34. EVIDENCE FOR BLOOD TRANSFUSION IN CARDIAC PATIENTS TRICS-III (2017) : in patients undergoing cardiac surgery with moderate to high risk of death, does restrictive transfusion strategy compared with liberal impact composite outcome of death, MI and AKI requiring HD? RCT, un-blinded, international, multi-centre POPULATION INTERVENTION CONTROL OUTCOME > 18yrs with EuroSCORE 1 of 6 or higher 5243 randomized, 5092included Restrictive group: received red cell transfusion if Hb <7.5g/dl intra- or post-operatively Liberal group: received red cell transfusion if Hb <9.5g/dl intra- or post operatively or < 8.5g/dl in non-ICU ward. 1˚: restrictive strategy was non-inferior to liberal for composite outcome of death, new onset Aki requiring HD , MI during hospitalization

35. EVIDENCE FOR BLOOD TRANSFUSION IN CARDIAC PATIENTS Risks of Restrictive Red blood cell transfusion strategies ( Hb 7-8g/dl) in patients with cardiovascular disease (CVD): a meta-analysis (2018) In patients hopsitalized for non-cardiac indications, liberal transfusion strategies are associated with a decrease risk of MACE in both with and without known (CVD). However, this provides a survival benefit to CVD patients not admitted for CVD corrective procedures.

36. EVIDENCE OF TRANSFUSION IN CARDIAC PATIENTS REALITY trial (2021): in patients with acute MI and anaemia is restrictive compared with liberal strategy, non-inferior with respect to major adverse cardiovascular events at 30 days? RCT, un-blinded, international, multi-centre MACE: all cause mortality at 30 days, recurrent MI, emergency revascularization prompted by ischemia, stroke POPULATION (668) INTERVENTION CONTROL OUTCOME Age >18years with Acute MI (STEMI or NSTEMI), with symptoms in last 48 hours and elevated biomarkers of injury and Hb between 7-10 g/dl during admission Liberal transfusion (n=324): transfusion for Hb <10g/dl with target post – transfusion Hb >11g/dl Restrictive transfusion (n=342): transfusion if Hb <8g/dl with target 8-10g/dl. (previous threshold Hb <7g/dl- modified to maximise investigator adherence to protocol. At 30 days MACE occurred in 14% v/s 11% (CI= -8.4% to 2.4%). This met the non-inferiority criteria. In patients with AMI with anaemia, restrictive transfusion strategy was non-inferior in regards to MACE at 30 days.

37. EVIDENCE FOR TRANSFUSION IN UGI BLEED Compared to liberal transfusion strategy, a restrictive strategy significantly improved outcomes 51% in restrictive strategy group did not require transfusion as compared to 14% in liberal strategy (p<0.001) Probability of survival at 6 weeks was higher in restrictive strategy group (95%) v/s (91% in liberal strategy group (p=0.02) Further bleeding in 10% v/s 16% (p=0.01) Adverse events 40% v/s 48% (p=0.02) Portal pressure gradient within first 5 days increased significantly in patients assigned to liberal strategy (p=0.03) Probability of survival higher in restrictive strategy for patients with peptic ulcer and cirrhosis (Child-Pugh A or B disease) but not for Child-Pugh C disease.

38. RED CELL TRANSFUSION THRESHOLD IN ADULTS CONDITION HAEMOGLOBIN THRESHOLD FOR TRANSFUSION Symptomatic patient ( eg . Myocardial ischemia, haemodynamic instability) 10g/ dL Hospitalized patients Pre-existing coronary artery disease 8g/ dL Acute coronary syndromes, including AMI 8-10g/ dL ICU ( haemodynamically stable) 7g/ dL Gastrointestinal bleeding ( haemodynamically stable) 7g/ dL Orthopedic surgery 8g/ dL Cardiac surgery 7.5g/ dL Ambulatory out- patients Oncology patient in treatment 7-8g/ dL Palliative care setting As needed for symptoms, hospice benefits may vary

39. EVIDENCE FOR TRANSFUSION IN SEPSIS TRISS (2014): in patients with septic shock, does blood transfusion at lower or higher Hb threshold improve mortality? RCT, multi-centre Surviving Sepsis guidelines (2016): recommend restrictive RBC trabsfusions when Hb <7g/ dL in absence of myocardial ischemia, severe hypoxemia or acute haemorrhage. POPULATION INTERVENTION CONTROL OUTCOME Adults with septic shock with Hb ≤ 9g/dl. Data obtained from 998 patients out of randomized 1005 Transfusion threshold of ≤7g/dl Transfusion threshold of ≤9g/dl 1˚- no difference in 90 day mortality. In patients with septic shock, mortality and rates of ischemic events were similar in those assigned to blood transfusion at a higher v/s lower threshold.

40. EVIDENCE FOR TRANSFUSION IN SEPSIS Systematic review and meta-analysis (2019): 3 RCTs: TRICOP, TRISS and TRICC Could not show any difference in 28 or 30 day mortality between liberal (threshold 9-10g/ dL ) and restrictive transfusion (threshold 7g/ dL ) strategies in patients with sepsis or septic shock.

41. EVIDENCE FOR TRANSFUSION IN TRAUMA PROMMTT study (2013): Higher plasma and platelet ratios in early resuscitation were associated with decreased mortality in patients who received at least 3 units of blood products during first 24 hours. Among survivors at 24 hours, subsequent risk of death by 30 days was not associated with plasma or platelet ratios. PROPPR trial (2015): Multicentre RCT comparing massive transfusion protocols based on plasma: platelets: RBCs ratios of 1:1:1 v/s 1:1:2 No statically significant difference in 24 hour and 30day mortality. Death from exsanguination reduced in first 24 hours significantly decreased in 1:1:1 group (p=0.03). Achieved haemostasis was significantly greater in 1:1:1 group (p=0.006)

42. EVIDENCE FOR TRANSFUSION IN TRAUMA COMBAT (July 2018): Single centre, placebo controlled, semi-blinded RCT Compared pre-hospital infusion of plasma to normal saline in patients with haemorrhagic shock. No difference in 28 day mortality (15% v/s10%, p- not significant) PAMPer trial (August 2018): Pragmatic, phase 3, multicentre, cluster-randomized trial. Randomized air medical services 2 units of plasma versus goal-directed crystalloid bases resuscitation targeting SBP≥ 90mmHg. 30day mortality significantly lower in in plasma group (23% v/s 33%, p=0.03)

43. EVIDENCE FOR TRANSFUSION IN TRAUMA RePHILL trial (April 2022): Multi-centre, open label, phase 3 RCT 2 units of packed red cells + 2 units of lyophilised plasma v/s normal saline (up to 2 litres) Trial did not demonstrate benefit for composite outcome of mortality and lactate clearance in trauma patients with haemorrhagic shock receiving PRBC- LyoPlas compared with NS. Effect of liberal v/s restrictive transfusion strategy on mortality in moderate to severe head injury (2006): Unable to detect significant difference with liberal and restrictive strategies.

44. EVIDENCE FOR TRANSFUSION IN PREGNANCY Transfusion is necessary if Hb < 6g/ dL and there are < 4weeks for delivery. Hb -7g/ dL : for women with continued bleeding or at risk of haemorrhage or cardiac decompensation . Hb < 7g/ dL in labour or in immediate postpartum period: blood transfusion is only indicated if there is previous history of bleeding or patient prone for bleeding. For sickle cell disease or thalassemia: should only be reserved for severe cases. Prophylactic transfusions are associated with increased costs, number of hospitalizations and risk of alloimmunisation Jadon A, Bagai R. Blood transfusion practices in obstetric anaesthesia. Indian J Anaesth . 2014 Sep;58(5):629-36. doi : 10.4103/0019-5049.144674. PMID: 25535427; PMCID: PMC4260311.

45. GOALS FOR TRANSFUSION IN OBSTETRIC PATIENT: Hb > 8g/ dL Platelet count >75 x 10 9 /L Prothrombin time and activated PT < 1.5 x mean control Fibrinogen > 1g/L Jadon A, Bagai R. Blood transfusion practices in obstetric anaesthesia. Indian J Anaesth . 2014 Sep;58(5):629-36. doi : 10.4103/0019-5049.144674. PMID: 25535427; PMCID: PMC4260311.

46. ONGOING TRIALS HEMOTION trial: HEMOglobin Transfusion Threshold in Traumatic Brain Injury Optimization TRICS IV: Transfusion Requirements in Younger Patients Undergoing Cardiac Surgery

47. THANKYOU…