Transfusion practice in patients undergoing cardiac surgery in New Zealand—impact of the TRICS III study (the TRICS TRIPS study)
View Article PDF
Cardiac surgery is the largest perioperative user of donated blood products accounting for 14–20% of total blood product consumption and consuming 50% of all red blood cell (RBC) transfusions given to surgical patients.[[1–3]] Perioperative anaemia and administration of RBC transfusions have been shown to be independent predictors of poor outcomes.[[4–10]] While restrictive strategies have reduced transfusion in other populations,[[11–13]] it may be inappropriate and unsafe to extrapolate these findings to cardiac surgery. Previously, inter-physician and inter-hospital variability in transfusion practice has been shown, often unrelated to patient risk factors.[[14,15]]
The Transfusion Requirements in Cardiac Surgery III (TRICS III) study was an international, open label, randomised-controlled, noninferiority study in moderate- to high-risk patients undergoing cardiac surgery conducted 2014–2017.[[16]] The study enrolled 5,243 participants (633 in New Zealand) and showed that a restrictive transfusion strategy (transfuse if haemoglobin level was <75g/L starting from induction of anaesthesia) reduced transfusion of allogeneic red blood cells and was non-inferior to a liberal strategy (transfuse if haemoglobin level was <95g/L in the operating room or Intensive Care Unit (ICU) or was <85g/L in the non-ICU ward) for a composite outcome of mortality and major morbidity.[[16,17]] Participants were enrolled between 14 November 2014 and 17 March 2017 in New Zealand and the study published 12 November 2017.
Of equal importance to the generation of evidence is the translation of findings into clinical practice. We wished to understand how the dissemination of the results of the TRICS III study impacted real-world clinical practice.
Methods
This study of the impact of knowledge translation involved two components:
1. Multi-time period, prospective observational study.
2. Cross-sectional, self-administered online questionnaire of clinicians ordering blood transfusions.
Prospective observational study of RBC administration
Setting
A prospective, observational study at all five publicly funded cardiac surgical centres in New Zealand.
Aim
To benchmark current RBC transfusion practice and determine the proportion of transfusions compliant with findings of the TRICS III study.
Conducted for four weeks at the following time points:
• October 2017 after completion of TRICS III study enrolment and before results were known (T1).
• May 2019 after the publication of TRICS III study results (T2).
Participants
Consecutive patients admitted to the ICU following cardiac surgery using cardiopulmonary bypass during the study periods. Each site collected data on eligible patients over a four-week period capped at 50 patients at each time point.
Ethical approval
Ethics (13/CEN/189/AM07) and local site governance approval were obtained before study commencement. The requirement for written informed consent was waived.
Data collection
Data was collected by trained research coordinators at each site and managed using REDCap (Research Electronic Data Capture) hosted at the Medical Research Institute of New Zealand.
Data included:
• Demographic data (ethnicity, age, gender), type of surgery.
• At the time of each transfusion episode, the following data was recorded by the clinical staff administering the transfusion: pre- and post-transfusion haemoglobin and haematocrit values (routinely collected values measured before transfusion and first value measured after transfusion), place of transfusion (theatre, ICU or ward) and the primary reason for transfusion from a pre-specified list.
Clinician survey
Aim:
To describe self-reported practice regarding the administration of RBCs.
The survey was conducted:
3. September 2018 after completion of study enrolment and before results were known.
4. May 2018 two months after the results of the TRICS III study were published.
5. May 2019 12 months after the results of the TRICS III study were published.
Participants
All cardiac surgeons, anaesthetists, perfusionists and ICU consultants in New Zealand that cared for patients undergoing cardiac surgery were invited to participate. An invitation to participate containing a link to the questionnaire was distributed by email to all participants. Completion of the survey implied consent to participate. Participants received a reminder, weekly for four weeks, to complete the survey. As clinicians could choose to participate in all surveys the sample at each time point was not identical, however, there was likely significant overlap, but information to link them together was not collected.
Data collection
An online questionnaire was administered using REDCAP. The survey consisted of demographic questions regarding participants including specialty, length of practice, number of cases performed per year and location. The respondent was asked to provide an answer regarding the administration of RBCs to hypothetical patients. Finally, respondents were asked if they had participated in the TRICS III study or not; if they knew the results of the study, and whether this knowledge had impacted their practice.
Data analysis
Data were summarised separately by time point and transfusion status for the prospective observational study of RBC administration, and by survey (time point) for the clinician surveys. All data were described using means (± standard deviation [SD]) for continuous variables or number (n) and percentage (%) for categorical variables. For categorical variables, comparisons between time points/groups/surveys were performed using chi-squared tests or Fisher’s exact test. To compare time points/groups/surveys by continuous variables, a t-Test or Kruskal–Wallis test (for measures with non-normal distribution) was applied. For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were documented as occurring for a major bleeding episode. Major bleeding was defined as clinical evidence of ongoing blood loss and a decrease in the haemoglobin concentration of 3 g/L in the preceding 12 hours or a requirement for at least three units of RBCs during the same period. For the question that asked clinicians to rank additional considerations when ordering RBCs, responses were analysed as follows: each factor ranked first was given three points; second-ranked received two points; third-ranked considerations received one point. Points were then summed for each factor.
All tests were independent and two-sided, with P-values <0.05 considered statistically significant. All statistical analyses were performed using Stata 16 (StataCorp LP, College Station, Texas).
Results
View Tables 1–4.
Prospective observational study of RBC administration
At baseline, (T1) 209 patients were admitted following cardiac surgery over a four-week period in October 2017. At T2, 181 patients were admitted following cardiac surgery over a four-week period in May 2019.
Participant characteristics.
Characteristics of participants were comparable between timepoints and typical of the patient population that undergoes cardiac surgery in New Zealand (Table 1).
Transfusion characteristics
There was no significant difference in the number of units transfused at each transfusion episode between time points (Table 2). A similar proportion of patients received a RBC transfusion in T1 (45.9%) and T2 (43.6%; Table 1). Participants who were transfused received a median (IQR, range) of 3 units (1–4; 1–17) in T1 and 4 units (2–4; 1–25) in T2.
This study showed a significant increase from T1 to T2 in the proportion of RBC transfusions that were compliant with the restrictive arm of the TRICS III study (T1=44% T2=56.7%, p=0.01). The increase in compliance in ICU was more marked (41.5% of transfusions compliant at T1 increasing to 60% at T2, p=0.004).
RBC transfusions were mostly ordered and administered in the ICU with most patients requiring only one unit (Table 2).
The mean pre-transfusion haemoglobin and haematocrit did not differ significantly between T1 and T2, respectively. However, the proportion of transfusions given when the pre-transfusion haemoglobin was <75g/L was significantly higher at T2 (Table 2, Figure 1).
Clinical indication for transfusion differed between T1 and T2 (p<0.001). The most common reason recorded for RBC transfusion was low haemoglobin followed by major bleeding. There was a higher proportion of major bleeding and shock/altered tissue perfusion in T1 vs T2 and higher reporting of minor bleeding and disseminated intravascular coagulation in T2 (Table 2).
For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were clinician-reported as a major bleeding episode (n=65). Findings remained unchanged.
Clinician survey
Survey 1 (S1) received 97 responses of which 90 were complete and included in the analysis (response rate=97/15; 62%). Survey 2 (S2) received 67 responses of which 63 were complete and included in the analysis (response rate=67/161; 41.6%). Survey 3 (S3) received 72 responses of which 64 were complete and included in the analysis (response rate=72/15;, 45.9%).
Respondent characteristics and existing guidelines
Most responses were received from anaesthetists followed by ICU Consultants (Table 3). Close to half of the respondents reported having no formal written guidelines for the administration of RBCs but where they did exist, a significant increase in perceived adherence was seen at later time points (p=0.001).
Responses to scenarios and transfusion decision making
When asked what modifying factors would allow acceptance of a higher transfusion threshold there was broad agreement between intra-operative and post-operative clinicians. Both groups rated emergency surgery, valve surgery and age over 70 years as the most important factors (Figure 2).
Participation in and knowledge of TRICS III study
Most respondents had participated in the TRICS III study, with the majority reporting that participation had not resulted in them changing their practice. After the publication of TRICS III, 75.4% of respondents felt that there should be New Zealand-specific guidelines developed for the administration of RBCs, and an additional 9.8% thought we should wait for international guidelines to be developed (Table 4).
View Figures 1–2.
Discussion
We conducted a national, multi-timepoint prospective observational study and clinician survey of transfusion practice after cardiac surgery to assess the impact of the dissemination of the findings of the TRICS III study.[[16]]
Our key findings were of a significant change in transfusion practice, with clinicians being more restrictive in their use of RBCs and strong support for the development of guidelines for transfusion practice.
There was a 29% relative increase in the proportion of RBC transfusions compliant with the findings of the TRICS III study. This occurred in theatres, ICUs and the postoperative wards, driven by a significant change in practice in the ICUs, which recorded an increase in compliance of 45%. One reason for this may be that the TRICS III study was run out of the ICUs in four of the five centres: therefore, ICU clinicians may have been more aware and engaged in disseminating and translating the results of the study. Another reason might be that ICUs had the highest reported rates of guidelines for RBC transfusion.
The findings of the TRICS III study appear to have been translated into practice more rapidly than may have been expected. Translating evidence and knowledge into practice is a challenging process due to the complexity of healthcare systems.[[18]] It has been demonstrated previously that change in healthcare can be slow with reports of it taking up to 17 years.[[19]] Our study was undertaken 12 months following the dissemination of the TRICS III study results. It would be interesting to undertake the observational study again, perhaps through regular clinical audit, to see if there was a sustained change.
While there was a significant improvement in compliance with the restrictive arm of the TRICS III study, this only reached 56.7% at best. There may be several factors for this, perhaps the most important one being that in the TRICS III study protocol a higher transfusion trigger was allowed in patients who were bleeding. In our study approximately 50% of noncompliant transfusion episodes were attributed to bleeding and, in a further 25%, the reason was unknown. This is consistent with findings of the clinician survey where ongoing active bleeding ranked highest across all three clinical areas as additional consideration for RBC transfusion.
There was some variability in transfusion practice between centres however this is less than previously reported and did not reach significance at T2, which perhaps adds weight to any plan for national transfusion guidelines.
We also found that reasons considered before prescribing an RBC transfusion did not differ across the surveys and relate to previously published work. When asked what additional factors were considered important, clinicians cited age and low body mass index. TRICS IV—a trial of restrictive versus liberal transfusion in younger patients undergoing cardiac surgery—is a large new study attempting to clarify the relationship between patient age and transfusion thresholds (ClinicalTrials.gov Identifier: NCT04754022).[[20]]
We found that the proportion of clinicians who reported that they would aim for a higher Hb reduced significantly from S1 to S3 which may reflect a change in self-reported practice based on the results of the TRICS III study.
These results also concur with the recommended haemoglobin threshold of 75g/L for cardiac surgery patients reported in the 2018 Patient Blood Management International Consensus.[[21]]
We undertook both the observational study and the clinician-reported survey to demonstrate both what had actually occurred in practice and what clinicians thought had changed. This adds strength to the study findings and impetus for the development of future practice guidelines.
As with all observational work this study does have several limitations. We did not link individuals across time points for the clinician surveys, so could not assess individual changes in practice but only those at an overall level. We also cannot say whether the findings of this study relate specifically to the trial results or whether other factors may have also influenced over the study periods including background temporal changes in practice. It should also be recognised that this study was undertaken in one country only and therefore may not be generalisable to other healthcare systems and practices. In addition, because all five cardiac surgical hospitals participated in this study, we did not have available control sites that did not participate to compare.
Our study suggests a rapid translation of the results of the TRICS III study into clinical practice in New Zealand. Participation of all five cardiac surgical centres in the study, rapid and targeted presentation of the findings in multi-disciplinary fora, and the use of local guidelines at each hospital may have contributed to this. It is anticipated that there may be savings to the public hospital system through a more consistent approach to blood transfusion.
In conclusion, following the publication of the findings of the TRICS III trial, clinicians involved in the care of cardiac surgery patients in New Zealand were more restrictive in their administration of red blood cell transfusions than before the trial findings were published. There was strong support for the development of clinical guidelines for transfusion of RBCs.
Summary
Abstract
Aim
Cardiac surgery is the largest perioperative user of donated blood products. There is significant uncertainty as to the optimal threshold for RBC transfusion in patients undergoing cardiac surgery with little evidence to guide practice. We wished to determine whether the results of a large randomised controlled trial had changed practice.
Method
A prospective observational study of red blood cell (RBC) transfusions of patients undergoing cardiac surgery utilising cardiopulmonary bypass was undertaken as well as a cross-sectional self-administered online practice survey of clinicians ordering red blood cell transfusions in all publicly funded cardiac centres in New Zealand.
Results
Significantly more transfusions were administered to a pre-transfusion haemoglobin <75g/L and thus considered in agreement with the restrictive arm of the TRICS III study after completion of TRICS III study enrolment and before results were known (T1)=44% when compared to after results were known (T2=56.7%, p=0.01). Most respondents in the clinician survey had participated in the TRICS III study.
Conclusion
After the publication of the findings of a large multi-national clinical trial, clinicians involved in the care of cardiac surgery patients were more restrictive in their administration of red blood cell transfusions than before the trial findings were published.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Josephson CD, Glynn SA, Kleinman SH, Blajchman MA. State-of-the-Science Symposium Transfusion Medicine Committee. A multidisciplinary “think tank”: the top 10 clinical trial opportunities in transfusion medicine from the National Heart, Lung, and Blood Institute–sponsored 2009 state-of-the-science symposium. Transfusion. 2011 Apr;51(4):828-41.
2) Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion. 1996 Aug;36(8):699-706.
3) Covin R, O'Brien M, Grunwald G, et al. Factors affecting transfusion of fresh frozen plasma, platelets, and red blood cells during elective coronary artery bypass graft surgery. Arch Pathol Lab Med. 2003 Apr;127(4):415-23.
4) Karkouti K, Wijeysundera DN, Beattie WS. Reducing Bleeding in Cardiac Surgery Investigators. Risk associated with preoperative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008 Jan 29;117(4):478-84.
5) Corwin H, Gettinger A, Pearl R, et al. The CRIT Study: Anemia and blood transfusion in the critically ill--current clinical practice in the United States. Crit Care Med. 2004 Jan;32(1):39-52.
6) Karkouti K, Wijeysundera DN, Yau TM, et al. The influence of baseline hemoglobin concentration on tolerance of anemia in cardiac surgery. Transfusion. 2008 Apr;48(4):666-72.
7) Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006 Jun;34(6):1608-16.
8) Scott B, Seifert F, Grimson R. Blood transfusion is associated with increased resource utilisation, morbidity and mortality in cardiac surgery. Ann Card Anaesth. Jan-Jun 2008;11(1):15-9.
9) Surgenor SD, Kramer RS, Olmstead EM, et al. The association of perioperative red blood cell transfusions and decreased long-term survival after cardiac surgery. Anesth Analg. 2009 Jun;108(6):1741-6.
10) Vincent J, Baron J, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA. 2002 Sep 25;288(12):1499-507.
11) Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999 Feb 11;340(6):409-17.
12) Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011 Dec 29;365(26):2453-62.
13) Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013 Jan 3;368(1):11-21.
14) Goodnough L, Soegiarso R, Birkmeyer J, Welch H. Economic impact of inappropriate blood transfusions in coronary artery bypass graft surgery. Am J Med. 1993 May;94(5):509-514.
15) Yap C-H, Lau L, Krishnaswamy M, et al. Age of Transfused Red Cells and Early Outcomes After Cardiac Surgery. Ann Thorac Surg. 2008 Aug;86(2):554-9.
16) Mazer CD, Whitlock RP, Fergusson DA, et al. Restrictive or liberal red-cell transfusion for cardiac surgery. N Engl J Med. 2017 Nov 30;377(22):2133-2144.
17) Mazer CD, Whitlock RP, Fergusson DA, et al. Six-Month Outcomes after Restrictive or Liberal Transfusion for Cardiac Surgery. N Engl J Med. 2018 Sep 27;379(13):1224-1233.
18) Greenhalgh T, Papoutsi C. Studying complexity in health services research: desperately seeking an overdue paradigm shift. BMC Med. 2018 Jun 20;16(1):95.
19) Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011 Dec;104(12):510-20.
20) U.S. National Library of Medicine. Transfusion Requirements in Younger Patients Undergoing Cardiac Surgery (TRICS-IV). 2022. Last updated January 24, 2022. Cited February 13, 2022. Available from https://clinicaltrials.gov/ct2/show/NCT04754022.
21) Mueller MM, Van Remoortel H, Meybohm P, et al. Patient Blood Management: Recommendations From the 2018 Frankfurt Consensus Conference. 2019 Mar 12;321(10):983-997.
For the PDF of this article,
contact nzmj@nzma.org.nz
Transfusion practice in patients undergoing cardiac surgery in New Zealand—impact of the TRICS III study (the TRICS TRIPS study)
View Article PDF
Cardiac surgery is the largest perioperative user of donated blood products accounting for 14–20% of total blood product consumption and consuming 50% of all red blood cell (RBC) transfusions given to surgical patients.[[1–3]] Perioperative anaemia and administration of RBC transfusions have been shown to be independent predictors of poor outcomes.[[4–10]] While restrictive strategies have reduced transfusion in other populations,[[11–13]] it may be inappropriate and unsafe to extrapolate these findings to cardiac surgery. Previously, inter-physician and inter-hospital variability in transfusion practice has been shown, often unrelated to patient risk factors.[[14,15]]
The Transfusion Requirements in Cardiac Surgery III (TRICS III) study was an international, open label, randomised-controlled, noninferiority study in moderate- to high-risk patients undergoing cardiac surgery conducted 2014–2017.[[16]] The study enrolled 5,243 participants (633 in New Zealand) and showed that a restrictive transfusion strategy (transfuse if haemoglobin level was <75g/L starting from induction of anaesthesia) reduced transfusion of allogeneic red blood cells and was non-inferior to a liberal strategy (transfuse if haemoglobin level was <95g/L in the operating room or Intensive Care Unit (ICU) or was <85g/L in the non-ICU ward) for a composite outcome of mortality and major morbidity.[[16,17]] Participants were enrolled between 14 November 2014 and 17 March 2017 in New Zealand and the study published 12 November 2017.
Of equal importance to the generation of evidence is the translation of findings into clinical practice. We wished to understand how the dissemination of the results of the TRICS III study impacted real-world clinical practice.
Methods
This study of the impact of knowledge translation involved two components:
1. Multi-time period, prospective observational study.
2. Cross-sectional, self-administered online questionnaire of clinicians ordering blood transfusions.
Prospective observational study of RBC administration
Setting
A prospective, observational study at all five publicly funded cardiac surgical centres in New Zealand.
Aim
To benchmark current RBC transfusion practice and determine the proportion of transfusions compliant with findings of the TRICS III study.
Conducted for four weeks at the following time points:
• October 2017 after completion of TRICS III study enrolment and before results were known (T1).
• May 2019 after the publication of TRICS III study results (T2).
Participants
Consecutive patients admitted to the ICU following cardiac surgery using cardiopulmonary bypass during the study periods. Each site collected data on eligible patients over a four-week period capped at 50 patients at each time point.
Ethical approval
Ethics (13/CEN/189/AM07) and local site governance approval were obtained before study commencement. The requirement for written informed consent was waived.
Data collection
Data was collected by trained research coordinators at each site and managed using REDCap (Research Electronic Data Capture) hosted at the Medical Research Institute of New Zealand.
Data included:
• Demographic data (ethnicity, age, gender), type of surgery.
• At the time of each transfusion episode, the following data was recorded by the clinical staff administering the transfusion: pre- and post-transfusion haemoglobin and haematocrit values (routinely collected values measured before transfusion and first value measured after transfusion), place of transfusion (theatre, ICU or ward) and the primary reason for transfusion from a pre-specified list.
Clinician survey
Aim:
To describe self-reported practice regarding the administration of RBCs.
The survey was conducted:
3. September 2018 after completion of study enrolment and before results were known.
4. May 2018 two months after the results of the TRICS III study were published.
5. May 2019 12 months after the results of the TRICS III study were published.
Participants
All cardiac surgeons, anaesthetists, perfusionists and ICU consultants in New Zealand that cared for patients undergoing cardiac surgery were invited to participate. An invitation to participate containing a link to the questionnaire was distributed by email to all participants. Completion of the survey implied consent to participate. Participants received a reminder, weekly for four weeks, to complete the survey. As clinicians could choose to participate in all surveys the sample at each time point was not identical, however, there was likely significant overlap, but information to link them together was not collected.
Data collection
An online questionnaire was administered using REDCAP. The survey consisted of demographic questions regarding participants including specialty, length of practice, number of cases performed per year and location. The respondent was asked to provide an answer regarding the administration of RBCs to hypothetical patients. Finally, respondents were asked if they had participated in the TRICS III study or not; if they knew the results of the study, and whether this knowledge had impacted their practice.
Data analysis
Data were summarised separately by time point and transfusion status for the prospective observational study of RBC administration, and by survey (time point) for the clinician surveys. All data were described using means (± standard deviation [SD]) for continuous variables or number (n) and percentage (%) for categorical variables. For categorical variables, comparisons between time points/groups/surveys were performed using chi-squared tests or Fisher’s exact test. To compare time points/groups/surveys by continuous variables, a t-Test or Kruskal–Wallis test (for measures with non-normal distribution) was applied. For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were documented as occurring for a major bleeding episode. Major bleeding was defined as clinical evidence of ongoing blood loss and a decrease in the haemoglobin concentration of 3 g/L in the preceding 12 hours or a requirement for at least three units of RBCs during the same period. For the question that asked clinicians to rank additional considerations when ordering RBCs, responses were analysed as follows: each factor ranked first was given three points; second-ranked received two points; third-ranked considerations received one point. Points were then summed for each factor.
All tests were independent and two-sided, with P-values <0.05 considered statistically significant. All statistical analyses were performed using Stata 16 (StataCorp LP, College Station, Texas).
Results
View Tables 1–4.
Prospective observational study of RBC administration
At baseline, (T1) 209 patients were admitted following cardiac surgery over a four-week period in October 2017. At T2, 181 patients were admitted following cardiac surgery over a four-week period in May 2019.
Participant characteristics.
Characteristics of participants were comparable between timepoints and typical of the patient population that undergoes cardiac surgery in New Zealand (Table 1).
Transfusion characteristics
There was no significant difference in the number of units transfused at each transfusion episode between time points (Table 2). A similar proportion of patients received a RBC transfusion in T1 (45.9%) and T2 (43.6%; Table 1). Participants who were transfused received a median (IQR, range) of 3 units (1–4; 1–17) in T1 and 4 units (2–4; 1–25) in T2.
This study showed a significant increase from T1 to T2 in the proportion of RBC transfusions that were compliant with the restrictive arm of the TRICS III study (T1=44% T2=56.7%, p=0.01). The increase in compliance in ICU was more marked (41.5% of transfusions compliant at T1 increasing to 60% at T2, p=0.004).
RBC transfusions were mostly ordered and administered in the ICU with most patients requiring only one unit (Table 2).
The mean pre-transfusion haemoglobin and haematocrit did not differ significantly between T1 and T2, respectively. However, the proportion of transfusions given when the pre-transfusion haemoglobin was <75g/L was significantly higher at T2 (Table 2, Figure 1).
Clinical indication for transfusion differed between T1 and T2 (p<0.001). The most common reason recorded for RBC transfusion was low haemoglobin followed by major bleeding. There was a higher proportion of major bleeding and shock/altered tissue perfusion in T1 vs T2 and higher reporting of minor bleeding and disseminated intravascular coagulation in T2 (Table 2).
For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were clinician-reported as a major bleeding episode (n=65). Findings remained unchanged.
Clinician survey
Survey 1 (S1) received 97 responses of which 90 were complete and included in the analysis (response rate=97/15; 62%). Survey 2 (S2) received 67 responses of which 63 were complete and included in the analysis (response rate=67/161; 41.6%). Survey 3 (S3) received 72 responses of which 64 were complete and included in the analysis (response rate=72/15;, 45.9%).
Respondent characteristics and existing guidelines
Most responses were received from anaesthetists followed by ICU Consultants (Table 3). Close to half of the respondents reported having no formal written guidelines for the administration of RBCs but where they did exist, a significant increase in perceived adherence was seen at later time points (p=0.001).
Responses to scenarios and transfusion decision making
When asked what modifying factors would allow acceptance of a higher transfusion threshold there was broad agreement between intra-operative and post-operative clinicians. Both groups rated emergency surgery, valve surgery and age over 70 years as the most important factors (Figure 2).
Participation in and knowledge of TRICS III study
Most respondents had participated in the TRICS III study, with the majority reporting that participation had not resulted in them changing their practice. After the publication of TRICS III, 75.4% of respondents felt that there should be New Zealand-specific guidelines developed for the administration of RBCs, and an additional 9.8% thought we should wait for international guidelines to be developed (Table 4).
View Figures 1–2.
Discussion
We conducted a national, multi-timepoint prospective observational study and clinician survey of transfusion practice after cardiac surgery to assess the impact of the dissemination of the findings of the TRICS III study.[[16]]
Our key findings were of a significant change in transfusion practice, with clinicians being more restrictive in their use of RBCs and strong support for the development of guidelines for transfusion practice.
There was a 29% relative increase in the proportion of RBC transfusions compliant with the findings of the TRICS III study. This occurred in theatres, ICUs and the postoperative wards, driven by a significant change in practice in the ICUs, which recorded an increase in compliance of 45%. One reason for this may be that the TRICS III study was run out of the ICUs in four of the five centres: therefore, ICU clinicians may have been more aware and engaged in disseminating and translating the results of the study. Another reason might be that ICUs had the highest reported rates of guidelines for RBC transfusion.
The findings of the TRICS III study appear to have been translated into practice more rapidly than may have been expected. Translating evidence and knowledge into practice is a challenging process due to the complexity of healthcare systems.[[18]] It has been demonstrated previously that change in healthcare can be slow with reports of it taking up to 17 years.[[19]] Our study was undertaken 12 months following the dissemination of the TRICS III study results. It would be interesting to undertake the observational study again, perhaps through regular clinical audit, to see if there was a sustained change.
While there was a significant improvement in compliance with the restrictive arm of the TRICS III study, this only reached 56.7% at best. There may be several factors for this, perhaps the most important one being that in the TRICS III study protocol a higher transfusion trigger was allowed in patients who were bleeding. In our study approximately 50% of noncompliant transfusion episodes were attributed to bleeding and, in a further 25%, the reason was unknown. This is consistent with findings of the clinician survey where ongoing active bleeding ranked highest across all three clinical areas as additional consideration for RBC transfusion.
There was some variability in transfusion practice between centres however this is less than previously reported and did not reach significance at T2, which perhaps adds weight to any plan for national transfusion guidelines.
We also found that reasons considered before prescribing an RBC transfusion did not differ across the surveys and relate to previously published work. When asked what additional factors were considered important, clinicians cited age and low body mass index. TRICS IV—a trial of restrictive versus liberal transfusion in younger patients undergoing cardiac surgery—is a large new study attempting to clarify the relationship between patient age and transfusion thresholds (ClinicalTrials.gov Identifier: NCT04754022).[[20]]
We found that the proportion of clinicians who reported that they would aim for a higher Hb reduced significantly from S1 to S3 which may reflect a change in self-reported practice based on the results of the TRICS III study.
These results also concur with the recommended haemoglobin threshold of 75g/L for cardiac surgery patients reported in the 2018 Patient Blood Management International Consensus.[[21]]
We undertook both the observational study and the clinician-reported survey to demonstrate both what had actually occurred in practice and what clinicians thought had changed. This adds strength to the study findings and impetus for the development of future practice guidelines.
As with all observational work this study does have several limitations. We did not link individuals across time points for the clinician surveys, so could not assess individual changes in practice but only those at an overall level. We also cannot say whether the findings of this study relate specifically to the trial results or whether other factors may have also influenced over the study periods including background temporal changes in practice. It should also be recognised that this study was undertaken in one country only and therefore may not be generalisable to other healthcare systems and practices. In addition, because all five cardiac surgical hospitals participated in this study, we did not have available control sites that did not participate to compare.
Our study suggests a rapid translation of the results of the TRICS III study into clinical practice in New Zealand. Participation of all five cardiac surgical centres in the study, rapid and targeted presentation of the findings in multi-disciplinary fora, and the use of local guidelines at each hospital may have contributed to this. It is anticipated that there may be savings to the public hospital system through a more consistent approach to blood transfusion.
In conclusion, following the publication of the findings of the TRICS III trial, clinicians involved in the care of cardiac surgery patients in New Zealand were more restrictive in their administration of red blood cell transfusions than before the trial findings were published. There was strong support for the development of clinical guidelines for transfusion of RBCs.
Summary
Abstract
Aim
Cardiac surgery is the largest perioperative user of donated blood products. There is significant uncertainty as to the optimal threshold for RBC transfusion in patients undergoing cardiac surgery with little evidence to guide practice. We wished to determine whether the results of a large randomised controlled trial had changed practice.
Method
A prospective observational study of red blood cell (RBC) transfusions of patients undergoing cardiac surgery utilising cardiopulmonary bypass was undertaken as well as a cross-sectional self-administered online practice survey of clinicians ordering red blood cell transfusions in all publicly funded cardiac centres in New Zealand.
Results
Significantly more transfusions were administered to a pre-transfusion haemoglobin <75g/L and thus considered in agreement with the restrictive arm of the TRICS III study after completion of TRICS III study enrolment and before results were known (T1)=44% when compared to after results were known (T2=56.7%, p=0.01). Most respondents in the clinician survey had participated in the TRICS III study.
Conclusion
After the publication of the findings of a large multi-national clinical trial, clinicians involved in the care of cardiac surgery patients were more restrictive in their administration of red blood cell transfusions than before the trial findings were published.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Josephson CD, Glynn SA, Kleinman SH, Blajchman MA. State-of-the-Science Symposium Transfusion Medicine Committee. A multidisciplinary “think tank”: the top 10 clinical trial opportunities in transfusion medicine from the National Heart, Lung, and Blood Institute–sponsored 2009 state-of-the-science symposium. Transfusion. 2011 Apr;51(4):828-41.
2) Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion. 1996 Aug;36(8):699-706.
3) Covin R, O'Brien M, Grunwald G, et al. Factors affecting transfusion of fresh frozen plasma, platelets, and red blood cells during elective coronary artery bypass graft surgery. Arch Pathol Lab Med. 2003 Apr;127(4):415-23.
4) Karkouti K, Wijeysundera DN, Beattie WS. Reducing Bleeding in Cardiac Surgery Investigators. Risk associated with preoperative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008 Jan 29;117(4):478-84.
5) Corwin H, Gettinger A, Pearl R, et al. The CRIT Study: Anemia and blood transfusion in the critically ill--current clinical practice in the United States. Crit Care Med. 2004 Jan;32(1):39-52.
6) Karkouti K, Wijeysundera DN, Yau TM, et al. The influence of baseline hemoglobin concentration on tolerance of anemia in cardiac surgery. Transfusion. 2008 Apr;48(4):666-72.
7) Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006 Jun;34(6):1608-16.
8) Scott B, Seifert F, Grimson R. Blood transfusion is associated with increased resource utilisation, morbidity and mortality in cardiac surgery. Ann Card Anaesth. Jan-Jun 2008;11(1):15-9.
9) Surgenor SD, Kramer RS, Olmstead EM, et al. The association of perioperative red blood cell transfusions and decreased long-term survival after cardiac surgery. Anesth Analg. 2009 Jun;108(6):1741-6.
10) Vincent J, Baron J, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA. 2002 Sep 25;288(12):1499-507.
11) Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999 Feb 11;340(6):409-17.
12) Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011 Dec 29;365(26):2453-62.
13) Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013 Jan 3;368(1):11-21.
14) Goodnough L, Soegiarso R, Birkmeyer J, Welch H. Economic impact of inappropriate blood transfusions in coronary artery bypass graft surgery. Am J Med. 1993 May;94(5):509-514.
15) Yap C-H, Lau L, Krishnaswamy M, et al. Age of Transfused Red Cells and Early Outcomes After Cardiac Surgery. Ann Thorac Surg. 2008 Aug;86(2):554-9.
16) Mazer CD, Whitlock RP, Fergusson DA, et al. Restrictive or liberal red-cell transfusion for cardiac surgery. N Engl J Med. 2017 Nov 30;377(22):2133-2144.
17) Mazer CD, Whitlock RP, Fergusson DA, et al. Six-Month Outcomes after Restrictive or Liberal Transfusion for Cardiac Surgery. N Engl J Med. 2018 Sep 27;379(13):1224-1233.
18) Greenhalgh T, Papoutsi C. Studying complexity in health services research: desperately seeking an overdue paradigm shift. BMC Med. 2018 Jun 20;16(1):95.
19) Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011 Dec;104(12):510-20.
20) U.S. National Library of Medicine. Transfusion Requirements in Younger Patients Undergoing Cardiac Surgery (TRICS-IV). 2022. Last updated January 24, 2022. Cited February 13, 2022. Available from https://clinicaltrials.gov/ct2/show/NCT04754022.
21) Mueller MM, Van Remoortel H, Meybohm P, et al. Patient Blood Management: Recommendations From the 2018 Frankfurt Consensus Conference. 2019 Mar 12;321(10):983-997.
For the PDF of this article,
contact nzmj@nzma.org.nz
Transfusion practice in patients undergoing cardiac surgery in New Zealand—impact of the TRICS III study (the TRICS TRIPS study)
View Article PDF
Cardiac surgery is the largest perioperative user of donated blood products accounting for 14–20% of total blood product consumption and consuming 50% of all red blood cell (RBC) transfusions given to surgical patients.[[1–3]] Perioperative anaemia and administration of RBC transfusions have been shown to be independent predictors of poor outcomes.[[4–10]] While restrictive strategies have reduced transfusion in other populations,[[11–13]] it may be inappropriate and unsafe to extrapolate these findings to cardiac surgery. Previously, inter-physician and inter-hospital variability in transfusion practice has been shown, often unrelated to patient risk factors.[[14,15]]
The Transfusion Requirements in Cardiac Surgery III (TRICS III) study was an international, open label, randomised-controlled, noninferiority study in moderate- to high-risk patients undergoing cardiac surgery conducted 2014–2017.[[16]] The study enrolled 5,243 participants (633 in New Zealand) and showed that a restrictive transfusion strategy (transfuse if haemoglobin level was <75g/L starting from induction of anaesthesia) reduced transfusion of allogeneic red blood cells and was non-inferior to a liberal strategy (transfuse if haemoglobin level was <95g/L in the operating room or Intensive Care Unit (ICU) or was <85g/L in the non-ICU ward) for a composite outcome of mortality and major morbidity.[[16,17]] Participants were enrolled between 14 November 2014 and 17 March 2017 in New Zealand and the study published 12 November 2017.
Of equal importance to the generation of evidence is the translation of findings into clinical practice. We wished to understand how the dissemination of the results of the TRICS III study impacted real-world clinical practice.
Methods
This study of the impact of knowledge translation involved two components:
1. Multi-time period, prospective observational study.
2. Cross-sectional, self-administered online questionnaire of clinicians ordering blood transfusions.
Prospective observational study of RBC administration
Setting
A prospective, observational study at all five publicly funded cardiac surgical centres in New Zealand.
Aim
To benchmark current RBC transfusion practice and determine the proportion of transfusions compliant with findings of the TRICS III study.
Conducted for four weeks at the following time points:
• October 2017 after completion of TRICS III study enrolment and before results were known (T1).
• May 2019 after the publication of TRICS III study results (T2).
Participants
Consecutive patients admitted to the ICU following cardiac surgery using cardiopulmonary bypass during the study periods. Each site collected data on eligible patients over a four-week period capped at 50 patients at each time point.
Ethical approval
Ethics (13/CEN/189/AM07) and local site governance approval were obtained before study commencement. The requirement for written informed consent was waived.
Data collection
Data was collected by trained research coordinators at each site and managed using REDCap (Research Electronic Data Capture) hosted at the Medical Research Institute of New Zealand.
Data included:
• Demographic data (ethnicity, age, gender), type of surgery.
• At the time of each transfusion episode, the following data was recorded by the clinical staff administering the transfusion: pre- and post-transfusion haemoglobin and haematocrit values (routinely collected values measured before transfusion and first value measured after transfusion), place of transfusion (theatre, ICU or ward) and the primary reason for transfusion from a pre-specified list.
Clinician survey
Aim:
To describe self-reported practice regarding the administration of RBCs.
The survey was conducted:
3. September 2018 after completion of study enrolment and before results were known.
4. May 2018 two months after the results of the TRICS III study were published.
5. May 2019 12 months after the results of the TRICS III study were published.
Participants
All cardiac surgeons, anaesthetists, perfusionists and ICU consultants in New Zealand that cared for patients undergoing cardiac surgery were invited to participate. An invitation to participate containing a link to the questionnaire was distributed by email to all participants. Completion of the survey implied consent to participate. Participants received a reminder, weekly for four weeks, to complete the survey. As clinicians could choose to participate in all surveys the sample at each time point was not identical, however, there was likely significant overlap, but information to link them together was not collected.
Data collection
An online questionnaire was administered using REDCAP. The survey consisted of demographic questions regarding participants including specialty, length of practice, number of cases performed per year and location. The respondent was asked to provide an answer regarding the administration of RBCs to hypothetical patients. Finally, respondents were asked if they had participated in the TRICS III study or not; if they knew the results of the study, and whether this knowledge had impacted their practice.
Data analysis
Data were summarised separately by time point and transfusion status for the prospective observational study of RBC administration, and by survey (time point) for the clinician surveys. All data were described using means (± standard deviation [SD]) for continuous variables or number (n) and percentage (%) for categorical variables. For categorical variables, comparisons between time points/groups/surveys were performed using chi-squared tests or Fisher’s exact test. To compare time points/groups/surveys by continuous variables, a t-Test or Kruskal–Wallis test (for measures with non-normal distribution) was applied. For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were documented as occurring for a major bleeding episode. Major bleeding was defined as clinical evidence of ongoing blood loss and a decrease in the haemoglobin concentration of 3 g/L in the preceding 12 hours or a requirement for at least three units of RBCs during the same period. For the question that asked clinicians to rank additional considerations when ordering RBCs, responses were analysed as follows: each factor ranked first was given three points; second-ranked received two points; third-ranked considerations received one point. Points were then summed for each factor.
All tests were independent and two-sided, with P-values <0.05 considered statistically significant. All statistical analyses were performed using Stata 16 (StataCorp LP, College Station, Texas).
Results
View Tables 1–4.
Prospective observational study of RBC administration
At baseline, (T1) 209 patients were admitted following cardiac surgery over a four-week period in October 2017. At T2, 181 patients were admitted following cardiac surgery over a four-week period in May 2019.
Participant characteristics.
Characteristics of participants were comparable between timepoints and typical of the patient population that undergoes cardiac surgery in New Zealand (Table 1).
Transfusion characteristics
There was no significant difference in the number of units transfused at each transfusion episode between time points (Table 2). A similar proportion of patients received a RBC transfusion in T1 (45.9%) and T2 (43.6%; Table 1). Participants who were transfused received a median (IQR, range) of 3 units (1–4; 1–17) in T1 and 4 units (2–4; 1–25) in T2.
This study showed a significant increase from T1 to T2 in the proportion of RBC transfusions that were compliant with the restrictive arm of the TRICS III study (T1=44% T2=56.7%, p=0.01). The increase in compliance in ICU was more marked (41.5% of transfusions compliant at T1 increasing to 60% at T2, p=0.004).
RBC transfusions were mostly ordered and administered in the ICU with most patients requiring only one unit (Table 2).
The mean pre-transfusion haemoglobin and haematocrit did not differ significantly between T1 and T2, respectively. However, the proportion of transfusions given when the pre-transfusion haemoglobin was <75g/L was significantly higher at T2 (Table 2, Figure 1).
Clinical indication for transfusion differed between T1 and T2 (p<0.001). The most common reason recorded for RBC transfusion was low haemoglobin followed by major bleeding. There was a higher proportion of major bleeding and shock/altered tissue perfusion in T1 vs T2 and higher reporting of minor bleeding and disseminated intravascular coagulation in T2 (Table 2).
For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were clinician-reported as a major bleeding episode (n=65). Findings remained unchanged.
Clinician survey
Survey 1 (S1) received 97 responses of which 90 were complete and included in the analysis (response rate=97/15; 62%). Survey 2 (S2) received 67 responses of which 63 were complete and included in the analysis (response rate=67/161; 41.6%). Survey 3 (S3) received 72 responses of which 64 were complete and included in the analysis (response rate=72/15;, 45.9%).
Respondent characteristics and existing guidelines
Most responses were received from anaesthetists followed by ICU Consultants (Table 3). Close to half of the respondents reported having no formal written guidelines for the administration of RBCs but where they did exist, a significant increase in perceived adherence was seen at later time points (p=0.001).
Responses to scenarios and transfusion decision making
When asked what modifying factors would allow acceptance of a higher transfusion threshold there was broad agreement between intra-operative and post-operative clinicians. Both groups rated emergency surgery, valve surgery and age over 70 years as the most important factors (Figure 2).
Participation in and knowledge of TRICS III study
Most respondents had participated in the TRICS III study, with the majority reporting that participation had not resulted in them changing their practice. After the publication of TRICS III, 75.4% of respondents felt that there should be New Zealand-specific guidelines developed for the administration of RBCs, and an additional 9.8% thought we should wait for international guidelines to be developed (Table 4).
View Figures 1–2.
Discussion
We conducted a national, multi-timepoint prospective observational study and clinician survey of transfusion practice after cardiac surgery to assess the impact of the dissemination of the findings of the TRICS III study.[[16]]
Our key findings were of a significant change in transfusion practice, with clinicians being more restrictive in their use of RBCs and strong support for the development of guidelines for transfusion practice.
There was a 29% relative increase in the proportion of RBC transfusions compliant with the findings of the TRICS III study. This occurred in theatres, ICUs and the postoperative wards, driven by a significant change in practice in the ICUs, which recorded an increase in compliance of 45%. One reason for this may be that the TRICS III study was run out of the ICUs in four of the five centres: therefore, ICU clinicians may have been more aware and engaged in disseminating and translating the results of the study. Another reason might be that ICUs had the highest reported rates of guidelines for RBC transfusion.
The findings of the TRICS III study appear to have been translated into practice more rapidly than may have been expected. Translating evidence and knowledge into practice is a challenging process due to the complexity of healthcare systems.[[18]] It has been demonstrated previously that change in healthcare can be slow with reports of it taking up to 17 years.[[19]] Our study was undertaken 12 months following the dissemination of the TRICS III study results. It would be interesting to undertake the observational study again, perhaps through regular clinical audit, to see if there was a sustained change.
While there was a significant improvement in compliance with the restrictive arm of the TRICS III study, this only reached 56.7% at best. There may be several factors for this, perhaps the most important one being that in the TRICS III study protocol a higher transfusion trigger was allowed in patients who were bleeding. In our study approximately 50% of noncompliant transfusion episodes were attributed to bleeding and, in a further 25%, the reason was unknown. This is consistent with findings of the clinician survey where ongoing active bleeding ranked highest across all three clinical areas as additional consideration for RBC transfusion.
There was some variability in transfusion practice between centres however this is less than previously reported and did not reach significance at T2, which perhaps adds weight to any plan for national transfusion guidelines.
We also found that reasons considered before prescribing an RBC transfusion did not differ across the surveys and relate to previously published work. When asked what additional factors were considered important, clinicians cited age and low body mass index. TRICS IV—a trial of restrictive versus liberal transfusion in younger patients undergoing cardiac surgery—is a large new study attempting to clarify the relationship between patient age and transfusion thresholds (ClinicalTrials.gov Identifier: NCT04754022).[[20]]
We found that the proportion of clinicians who reported that they would aim for a higher Hb reduced significantly from S1 to S3 which may reflect a change in self-reported practice based on the results of the TRICS III study.
These results also concur with the recommended haemoglobin threshold of 75g/L for cardiac surgery patients reported in the 2018 Patient Blood Management International Consensus.[[21]]
We undertook both the observational study and the clinician-reported survey to demonstrate both what had actually occurred in practice and what clinicians thought had changed. This adds strength to the study findings and impetus for the development of future practice guidelines.
As with all observational work this study does have several limitations. We did not link individuals across time points for the clinician surveys, so could not assess individual changes in practice but only those at an overall level. We also cannot say whether the findings of this study relate specifically to the trial results or whether other factors may have also influenced over the study periods including background temporal changes in practice. It should also be recognised that this study was undertaken in one country only and therefore may not be generalisable to other healthcare systems and practices. In addition, because all five cardiac surgical hospitals participated in this study, we did not have available control sites that did not participate to compare.
Our study suggests a rapid translation of the results of the TRICS III study into clinical practice in New Zealand. Participation of all five cardiac surgical centres in the study, rapid and targeted presentation of the findings in multi-disciplinary fora, and the use of local guidelines at each hospital may have contributed to this. It is anticipated that there may be savings to the public hospital system through a more consistent approach to blood transfusion.
In conclusion, following the publication of the findings of the TRICS III trial, clinicians involved in the care of cardiac surgery patients in New Zealand were more restrictive in their administration of red blood cell transfusions than before the trial findings were published. There was strong support for the development of clinical guidelines for transfusion of RBCs.
Summary
Abstract
Aim
Cardiac surgery is the largest perioperative user of donated blood products. There is significant uncertainty as to the optimal threshold for RBC transfusion in patients undergoing cardiac surgery with little evidence to guide practice. We wished to determine whether the results of a large randomised controlled trial had changed practice.
Method
A prospective observational study of red blood cell (RBC) transfusions of patients undergoing cardiac surgery utilising cardiopulmonary bypass was undertaken as well as a cross-sectional self-administered online practice survey of clinicians ordering red blood cell transfusions in all publicly funded cardiac centres in New Zealand.
Results
Significantly more transfusions were administered to a pre-transfusion haemoglobin <75g/L and thus considered in agreement with the restrictive arm of the TRICS III study after completion of TRICS III study enrolment and before results were known (T1)=44% when compared to after results were known (T2=56.7%, p=0.01). Most respondents in the clinician survey had participated in the TRICS III study.
Conclusion
After the publication of the findings of a large multi-national clinical trial, clinicians involved in the care of cardiac surgery patients were more restrictive in their administration of red blood cell transfusions than before the trial findings were published.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Josephson CD, Glynn SA, Kleinman SH, Blajchman MA. State-of-the-Science Symposium Transfusion Medicine Committee. A multidisciplinary “think tank”: the top 10 clinical trial opportunities in transfusion medicine from the National Heart, Lung, and Blood Institute–sponsored 2009 state-of-the-science symposium. Transfusion. 2011 Apr;51(4):828-41.
2) Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion. 1996 Aug;36(8):699-706.
3) Covin R, O'Brien M, Grunwald G, et al. Factors affecting transfusion of fresh frozen plasma, platelets, and red blood cells during elective coronary artery bypass graft surgery. Arch Pathol Lab Med. 2003 Apr;127(4):415-23.
4) Karkouti K, Wijeysundera DN, Beattie WS. Reducing Bleeding in Cardiac Surgery Investigators. Risk associated with preoperative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008 Jan 29;117(4):478-84.
5) Corwin H, Gettinger A, Pearl R, et al. The CRIT Study: Anemia and blood transfusion in the critically ill--current clinical practice in the United States. Crit Care Med. 2004 Jan;32(1):39-52.
6) Karkouti K, Wijeysundera DN, Yau TM, et al. The influence of baseline hemoglobin concentration on tolerance of anemia in cardiac surgery. Transfusion. 2008 Apr;48(4):666-72.
7) Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006 Jun;34(6):1608-16.
8) Scott B, Seifert F, Grimson R. Blood transfusion is associated with increased resource utilisation, morbidity and mortality in cardiac surgery. Ann Card Anaesth. Jan-Jun 2008;11(1):15-9.
9) Surgenor SD, Kramer RS, Olmstead EM, et al. The association of perioperative red blood cell transfusions and decreased long-term survival after cardiac surgery. Anesth Analg. 2009 Jun;108(6):1741-6.
10) Vincent J, Baron J, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA. 2002 Sep 25;288(12):1499-507.
11) Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999 Feb 11;340(6):409-17.
12) Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011 Dec 29;365(26):2453-62.
13) Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013 Jan 3;368(1):11-21.
14) Goodnough L, Soegiarso R, Birkmeyer J, Welch H. Economic impact of inappropriate blood transfusions in coronary artery bypass graft surgery. Am J Med. 1993 May;94(5):509-514.
15) Yap C-H, Lau L, Krishnaswamy M, et al. Age of Transfused Red Cells and Early Outcomes After Cardiac Surgery. Ann Thorac Surg. 2008 Aug;86(2):554-9.
16) Mazer CD, Whitlock RP, Fergusson DA, et al. Restrictive or liberal red-cell transfusion for cardiac surgery. N Engl J Med. 2017 Nov 30;377(22):2133-2144.
17) Mazer CD, Whitlock RP, Fergusson DA, et al. Six-Month Outcomes after Restrictive or Liberal Transfusion for Cardiac Surgery. N Engl J Med. 2018 Sep 27;379(13):1224-1233.
18) Greenhalgh T, Papoutsi C. Studying complexity in health services research: desperately seeking an overdue paradigm shift. BMC Med. 2018 Jun 20;16(1):95.
19) Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011 Dec;104(12):510-20.
20) U.S. National Library of Medicine. Transfusion Requirements in Younger Patients Undergoing Cardiac Surgery (TRICS-IV). 2022. Last updated January 24, 2022. Cited February 13, 2022. Available from https://clinicaltrials.gov/ct2/show/NCT04754022.
21) Mueller MM, Van Remoortel H, Meybohm P, et al. Patient Blood Management: Recommendations From the 2018 Frankfurt Consensus Conference. 2019 Mar 12;321(10):983-997.
Contact diana@nzma.org.nz
for the PDF of this article
Transfusion practice in patients undergoing cardiac surgery in New Zealand—impact of the TRICS III study (the TRICS TRIPS study)
View Article PDF
Cardiac surgery is the largest perioperative user of donated blood products accounting for 14–20% of total blood product consumption and consuming 50% of all red blood cell (RBC) transfusions given to surgical patients.[[1–3]] Perioperative anaemia and administration of RBC transfusions have been shown to be independent predictors of poor outcomes.[[4–10]] While restrictive strategies have reduced transfusion in other populations,[[11–13]] it may be inappropriate and unsafe to extrapolate these findings to cardiac surgery. Previously, inter-physician and inter-hospital variability in transfusion practice has been shown, often unrelated to patient risk factors.[[14,15]]
The Transfusion Requirements in Cardiac Surgery III (TRICS III) study was an international, open label, randomised-controlled, noninferiority study in moderate- to high-risk patients undergoing cardiac surgery conducted 2014–2017.[[16]] The study enrolled 5,243 participants (633 in New Zealand) and showed that a restrictive transfusion strategy (transfuse if haemoglobin level was <75g/L starting from induction of anaesthesia) reduced transfusion of allogeneic red blood cells and was non-inferior to a liberal strategy (transfuse if haemoglobin level was <95g/L in the operating room or Intensive Care Unit (ICU) or was <85g/L in the non-ICU ward) for a composite outcome of mortality and major morbidity.[[16,17]] Participants were enrolled between 14 November 2014 and 17 March 2017 in New Zealand and the study published 12 November 2017.
Of equal importance to the generation of evidence is the translation of findings into clinical practice. We wished to understand how the dissemination of the results of the TRICS III study impacted real-world clinical practice.
Methods
This study of the impact of knowledge translation involved two components:
1. Multi-time period, prospective observational study.
2. Cross-sectional, self-administered online questionnaire of clinicians ordering blood transfusions.
Prospective observational study of RBC administration
Setting
A prospective, observational study at all five publicly funded cardiac surgical centres in New Zealand.
Aim
To benchmark current RBC transfusion practice and determine the proportion of transfusions compliant with findings of the TRICS III study.
Conducted for four weeks at the following time points:
• October 2017 after completion of TRICS III study enrolment and before results were known (T1).
• May 2019 after the publication of TRICS III study results (T2).
Participants
Consecutive patients admitted to the ICU following cardiac surgery using cardiopulmonary bypass during the study periods. Each site collected data on eligible patients over a four-week period capped at 50 patients at each time point.
Ethical approval
Ethics (13/CEN/189/AM07) and local site governance approval were obtained before study commencement. The requirement for written informed consent was waived.
Data collection
Data was collected by trained research coordinators at each site and managed using REDCap (Research Electronic Data Capture) hosted at the Medical Research Institute of New Zealand.
Data included:
• Demographic data (ethnicity, age, gender), type of surgery.
• At the time of each transfusion episode, the following data was recorded by the clinical staff administering the transfusion: pre- and post-transfusion haemoglobin and haematocrit values (routinely collected values measured before transfusion and first value measured after transfusion), place of transfusion (theatre, ICU or ward) and the primary reason for transfusion from a pre-specified list.
Clinician survey
Aim:
To describe self-reported practice regarding the administration of RBCs.
The survey was conducted:
3. September 2018 after completion of study enrolment and before results were known.
4. May 2018 two months after the results of the TRICS III study were published.
5. May 2019 12 months after the results of the TRICS III study were published.
Participants
All cardiac surgeons, anaesthetists, perfusionists and ICU consultants in New Zealand that cared for patients undergoing cardiac surgery were invited to participate. An invitation to participate containing a link to the questionnaire was distributed by email to all participants. Completion of the survey implied consent to participate. Participants received a reminder, weekly for four weeks, to complete the survey. As clinicians could choose to participate in all surveys the sample at each time point was not identical, however, there was likely significant overlap, but information to link them together was not collected.
Data collection
An online questionnaire was administered using REDCAP. The survey consisted of demographic questions regarding participants including specialty, length of practice, number of cases performed per year and location. The respondent was asked to provide an answer regarding the administration of RBCs to hypothetical patients. Finally, respondents were asked if they had participated in the TRICS III study or not; if they knew the results of the study, and whether this knowledge had impacted their practice.
Data analysis
Data were summarised separately by time point and transfusion status for the prospective observational study of RBC administration, and by survey (time point) for the clinician surveys. All data were described using means (± standard deviation [SD]) for continuous variables or number (n) and percentage (%) for categorical variables. For categorical variables, comparisons between time points/groups/surveys were performed using chi-squared tests or Fisher’s exact test. To compare time points/groups/surveys by continuous variables, a t-Test or Kruskal–Wallis test (for measures with non-normal distribution) was applied. For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were documented as occurring for a major bleeding episode. Major bleeding was defined as clinical evidence of ongoing blood loss and a decrease in the haemoglobin concentration of 3 g/L in the preceding 12 hours or a requirement for at least three units of RBCs during the same period. For the question that asked clinicians to rank additional considerations when ordering RBCs, responses were analysed as follows: each factor ranked first was given three points; second-ranked received two points; third-ranked considerations received one point. Points were then summed for each factor.
All tests were independent and two-sided, with P-values <0.05 considered statistically significant. All statistical analyses were performed using Stata 16 (StataCorp LP, College Station, Texas).
Results
View Tables 1–4.
Prospective observational study of RBC administration
At baseline, (T1) 209 patients were admitted following cardiac surgery over a four-week period in October 2017. At T2, 181 patients were admitted following cardiac surgery over a four-week period in May 2019.
Participant characteristics.
Characteristics of participants were comparable between timepoints and typical of the patient population that undergoes cardiac surgery in New Zealand (Table 1).
Transfusion characteristics
There was no significant difference in the number of units transfused at each transfusion episode between time points (Table 2). A similar proportion of patients received a RBC transfusion in T1 (45.9%) and T2 (43.6%; Table 1). Participants who were transfused received a median (IQR, range) of 3 units (1–4; 1–17) in T1 and 4 units (2–4; 1–25) in T2.
This study showed a significant increase from T1 to T2 in the proportion of RBC transfusions that were compliant with the restrictive arm of the TRICS III study (T1=44% T2=56.7%, p=0.01). The increase in compliance in ICU was more marked (41.5% of transfusions compliant at T1 increasing to 60% at T2, p=0.004).
RBC transfusions were mostly ordered and administered in the ICU with most patients requiring only one unit (Table 2).
The mean pre-transfusion haemoglobin and haematocrit did not differ significantly between T1 and T2, respectively. However, the proportion of transfusions given when the pre-transfusion haemoglobin was <75g/L was significantly higher at T2 (Table 2, Figure 1).
Clinical indication for transfusion differed between T1 and T2 (p<0.001). The most common reason recorded for RBC transfusion was low haemoglobin followed by major bleeding. There was a higher proportion of major bleeding and shock/altered tissue perfusion in T1 vs T2 and higher reporting of minor bleeding and disseminated intravascular coagulation in T2 (Table 2).
For consistency with the rules used for suspension of the transfusion protocol in the TRICS III study, we also compared study time points after removing transfusion episodes that were clinician-reported as a major bleeding episode (n=65). Findings remained unchanged.
Clinician survey
Survey 1 (S1) received 97 responses of which 90 were complete and included in the analysis (response rate=97/15; 62%). Survey 2 (S2) received 67 responses of which 63 were complete and included in the analysis (response rate=67/161; 41.6%). Survey 3 (S3) received 72 responses of which 64 were complete and included in the analysis (response rate=72/15;, 45.9%).
Respondent characteristics and existing guidelines
Most responses were received from anaesthetists followed by ICU Consultants (Table 3). Close to half of the respondents reported having no formal written guidelines for the administration of RBCs but where they did exist, a significant increase in perceived adherence was seen at later time points (p=0.001).
Responses to scenarios and transfusion decision making
When asked what modifying factors would allow acceptance of a higher transfusion threshold there was broad agreement between intra-operative and post-operative clinicians. Both groups rated emergency surgery, valve surgery and age over 70 years as the most important factors (Figure 2).
Participation in and knowledge of TRICS III study
Most respondents had participated in the TRICS III study, with the majority reporting that participation had not resulted in them changing their practice. After the publication of TRICS III, 75.4% of respondents felt that there should be New Zealand-specific guidelines developed for the administration of RBCs, and an additional 9.8% thought we should wait for international guidelines to be developed (Table 4).
View Figures 1–2.
Discussion
We conducted a national, multi-timepoint prospective observational study and clinician survey of transfusion practice after cardiac surgery to assess the impact of the dissemination of the findings of the TRICS III study.[[16]]
Our key findings were of a significant change in transfusion practice, with clinicians being more restrictive in their use of RBCs and strong support for the development of guidelines for transfusion practice.
There was a 29% relative increase in the proportion of RBC transfusions compliant with the findings of the TRICS III study. This occurred in theatres, ICUs and the postoperative wards, driven by a significant change in practice in the ICUs, which recorded an increase in compliance of 45%. One reason for this may be that the TRICS III study was run out of the ICUs in four of the five centres: therefore, ICU clinicians may have been more aware and engaged in disseminating and translating the results of the study. Another reason might be that ICUs had the highest reported rates of guidelines for RBC transfusion.
The findings of the TRICS III study appear to have been translated into practice more rapidly than may have been expected. Translating evidence and knowledge into practice is a challenging process due to the complexity of healthcare systems.[[18]] It has been demonstrated previously that change in healthcare can be slow with reports of it taking up to 17 years.[[19]] Our study was undertaken 12 months following the dissemination of the TRICS III study results. It would be interesting to undertake the observational study again, perhaps through regular clinical audit, to see if there was a sustained change.
While there was a significant improvement in compliance with the restrictive arm of the TRICS III study, this only reached 56.7% at best. There may be several factors for this, perhaps the most important one being that in the TRICS III study protocol a higher transfusion trigger was allowed in patients who were bleeding. In our study approximately 50% of noncompliant transfusion episodes were attributed to bleeding and, in a further 25%, the reason was unknown. This is consistent with findings of the clinician survey where ongoing active bleeding ranked highest across all three clinical areas as additional consideration for RBC transfusion.
There was some variability in transfusion practice between centres however this is less than previously reported and did not reach significance at T2, which perhaps adds weight to any plan for national transfusion guidelines.
We also found that reasons considered before prescribing an RBC transfusion did not differ across the surveys and relate to previously published work. When asked what additional factors were considered important, clinicians cited age and low body mass index. TRICS IV—a trial of restrictive versus liberal transfusion in younger patients undergoing cardiac surgery—is a large new study attempting to clarify the relationship between patient age and transfusion thresholds (ClinicalTrials.gov Identifier: NCT04754022).[[20]]
We found that the proportion of clinicians who reported that they would aim for a higher Hb reduced significantly from S1 to S3 which may reflect a change in self-reported practice based on the results of the TRICS III study.
These results also concur with the recommended haemoglobin threshold of 75g/L for cardiac surgery patients reported in the 2018 Patient Blood Management International Consensus.[[21]]
We undertook both the observational study and the clinician-reported survey to demonstrate both what had actually occurred in practice and what clinicians thought had changed. This adds strength to the study findings and impetus for the development of future practice guidelines.
As with all observational work this study does have several limitations. We did not link individuals across time points for the clinician surveys, so could not assess individual changes in practice but only those at an overall level. We also cannot say whether the findings of this study relate specifically to the trial results or whether other factors may have also influenced over the study periods including background temporal changes in practice. It should also be recognised that this study was undertaken in one country only and therefore may not be generalisable to other healthcare systems and practices. In addition, because all five cardiac surgical hospitals participated in this study, we did not have available control sites that did not participate to compare.
Our study suggests a rapid translation of the results of the TRICS III study into clinical practice in New Zealand. Participation of all five cardiac surgical centres in the study, rapid and targeted presentation of the findings in multi-disciplinary fora, and the use of local guidelines at each hospital may have contributed to this. It is anticipated that there may be savings to the public hospital system through a more consistent approach to blood transfusion.
In conclusion, following the publication of the findings of the TRICS III trial, clinicians involved in the care of cardiac surgery patients in New Zealand were more restrictive in their administration of red blood cell transfusions than before the trial findings were published. There was strong support for the development of clinical guidelines for transfusion of RBCs.
Summary
Abstract
Aim
Cardiac surgery is the largest perioperative user of donated blood products. There is significant uncertainty as to the optimal threshold for RBC transfusion in patients undergoing cardiac surgery with little evidence to guide practice. We wished to determine whether the results of a large randomised controlled trial had changed practice.
Method
A prospective observational study of red blood cell (RBC) transfusions of patients undergoing cardiac surgery utilising cardiopulmonary bypass was undertaken as well as a cross-sectional self-administered online practice survey of clinicians ordering red blood cell transfusions in all publicly funded cardiac centres in New Zealand.
Results
Significantly more transfusions were administered to a pre-transfusion haemoglobin <75g/L and thus considered in agreement with the restrictive arm of the TRICS III study after completion of TRICS III study enrolment and before results were known (T1)=44% when compared to after results were known (T2=56.7%, p=0.01). Most respondents in the clinician survey had participated in the TRICS III study.
Conclusion
After the publication of the findings of a large multi-national clinical trial, clinicians involved in the care of cardiac surgery patients were more restrictive in their administration of red blood cell transfusions than before the trial findings were published.
Author Information
Acknowledgements
Correspondence
Correspondence Email
Competing Interests
1) Josephson CD, Glynn SA, Kleinman SH, Blajchman MA. State-of-the-Science Symposium Transfusion Medicine Committee. A multidisciplinary “think tank”: the top 10 clinical trial opportunities in transfusion medicine from the National Heart, Lung, and Blood Institute–sponsored 2009 state-of-the-science symposium. Transfusion. 2011 Apr;51(4):828-41.
2) Chiavetta JA, Herst R, Freedman J, et al. A survey of red cell use in 45 hospitals in central Ontario, Canada. Transfusion. 1996 Aug;36(8):699-706.
3) Covin R, O'Brien M, Grunwald G, et al. Factors affecting transfusion of fresh frozen plasma, platelets, and red blood cells during elective coronary artery bypass graft surgery. Arch Pathol Lab Med. 2003 Apr;127(4):415-23.
4) Karkouti K, Wijeysundera DN, Beattie WS. Reducing Bleeding in Cardiac Surgery Investigators. Risk associated with preoperative anemia in cardiac surgery: a multicenter cohort study. Circulation. 2008 Jan 29;117(4):478-84.
5) Corwin H, Gettinger A, Pearl R, et al. The CRIT Study: Anemia and blood transfusion in the critically ill--current clinical practice in the United States. Crit Care Med. 2004 Jan;32(1):39-52.
6) Karkouti K, Wijeysundera DN, Yau TM, et al. The influence of baseline hemoglobin concentration on tolerance of anemia in cardiac surgery. Transfusion. 2008 Apr;48(4):666-72.
7) Koch CG, Li L, Duncan AI, et al. Morbidity and mortality risk associated with red blood cell and blood-component transfusion in isolated coronary artery bypass grafting. Crit Care Med. 2006 Jun;34(6):1608-16.
8) Scott B, Seifert F, Grimson R. Blood transfusion is associated with increased resource utilisation, morbidity and mortality in cardiac surgery. Ann Card Anaesth. Jan-Jun 2008;11(1):15-9.
9) Surgenor SD, Kramer RS, Olmstead EM, et al. The association of perioperative red blood cell transfusions and decreased long-term survival after cardiac surgery. Anesth Analg. 2009 Jun;108(6):1741-6.
10) Vincent J, Baron J, Reinhart K, et al. Anemia and blood transfusion in critically ill patients. JAMA. 2002 Sep 25;288(12):1499-507.
11) Hébert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. N Engl J Med. 1999 Feb 11;340(6):409-17.
12) Carson JL, Terrin ML, Noveck H, et al. Liberal or restrictive transfusion in high-risk patients after hip surgery. N Engl J Med. 2011 Dec 29;365(26):2453-62.
13) Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. N Engl J Med. 2013 Jan 3;368(1):11-21.
14) Goodnough L, Soegiarso R, Birkmeyer J, Welch H. Economic impact of inappropriate blood transfusions in coronary artery bypass graft surgery. Am J Med. 1993 May;94(5):509-514.
15) Yap C-H, Lau L, Krishnaswamy M, et al. Age of Transfused Red Cells and Early Outcomes After Cardiac Surgery. Ann Thorac Surg. 2008 Aug;86(2):554-9.
16) Mazer CD, Whitlock RP, Fergusson DA, et al. Restrictive or liberal red-cell transfusion for cardiac surgery. N Engl J Med. 2017 Nov 30;377(22):2133-2144.
17) Mazer CD, Whitlock RP, Fergusson DA, et al. Six-Month Outcomes after Restrictive or Liberal Transfusion for Cardiac Surgery. N Engl J Med. 2018 Sep 27;379(13):1224-1233.
18) Greenhalgh T, Papoutsi C. Studying complexity in health services research: desperately seeking an overdue paradigm shift. BMC Med. 2018 Jun 20;16(1):95.
19) Morris ZS, Wooding S, Grant J. The answer is 17 years, what is the question: understanding time lags in translational research. J R Soc Med. 2011 Dec;104(12):510-20.
20) U.S. National Library of Medicine. Transfusion Requirements in Younger Patients Undergoing Cardiac Surgery (TRICS-IV). 2022. Last updated January 24, 2022. Cited February 13, 2022. Available from https://clinicaltrials.gov/ct2/show/NCT04754022.
21) Mueller MM, Van Remoortel H, Meybohm P, et al. Patient Blood Management: Recommendations From the 2018 Frankfurt Consensus Conference. 2019 Mar 12;321(10):983-997.
Contact diana@nzma.org.nz
for the PDF of this article
Transfusion practice in patients undergoing cardiac surgery in New Zealand—impact of the TRICS III study (the TRICS TRIPS study)
Cardiac surgery is the largest perioperative user of donated blood products accounting for 14–20% of total blood product consumption and consuming 50% of all red blood cell (RBC) transfusions given to surgical patients.
Subscriber Content
The full contents of this pages only available to subscribers.
Login, subscribe or email nzmj@nzma.org.nz to purchase this article.
