Transfusion Medicine: Time for a Change: Patient Blood Management and the Ontario ONTraC Program

Despite increasing recognition of the potential adverse effects of blood transfusion there is evidence that the procedure is over-utilized and often inappropriately and unnecessarily administered. Patient Blood Management (PBM) is a recent world-wide initiative to optimize transfusion therapy by employing preoperative and perioperative maneuvers and alternatives to avoid unnecessary blood transfusion. These approaches are, however, not restricted to the surgical setting, but may be applied whenever transfusion is contemplated. A key component of PBM is the appropriate management of anemia, a prime determinant of transfusion. Another is tolerance of anemia through the use of restrictive transfusion thresholds for transfusion and of a single unit at a time policy. Among a number of options at surgery, antifibrinolytics have proven effective at reducing blood loss and transfusion. Burgeoning literature demonstrates that PBM (also known as Blood Conservation or as ‘Bloodless Medicine/Surgery’) is costeffective while improving patient outcomes and conserving the blood supply. Data from the ONTraC program in Ontario are described herein to illustrate the effectiveness of PBM approaches.


Introduction
Countless lives have been saved over the many years that people have been receiving blood transfusions. Transfusions replenish blood lost through trauma, illness or surgery. They are a medical safety net. But the field of transfusion medicine is changing. Instead of being viewed as an inert recharging of fluid, we now appreciate that a blood transfusion is essentially a liquid organ transplant, and like any other organ transplant, it has its risks.
While we often hear that transfusion is safer than it has ever been, that doesn't mean blood transfusion is 100% safe. Blood occupies a somewhat privileged position in modern medicine, in that it has evolved as a treatment option without the equivalent research scrutiny -at least on patient outcomes -that other treatments are subjected to. Studies over the past 25 years have shown that simply receiving a transfusion is a risk factor and may be associated with poorer outcomes in terms of increased length of stay in hospital, postoperative infections, and risk of multi-organ failure leading to ICU admission and/or death [1][2][3][4][5][6]. Red blood cell (RBC) transfusion is associated with a risk-adjusted increased risk for every postoperative morbid event: mortality, renal failure, prolonged ventilator support, serious infection, cardiac complications and neurologic events [2] and even one unit of blood is enough to cause problems [5,6] and the effects are dose-dependent [5][6][7]. While there are still unanswered questions on how, when and why blood transfusions are administered, there are situations in which a transfusion is the only option and can make the difference between life and death. But known risks should outweigh perceived benefits every time. There is a need for a precautionary approach-nothing should be done to a patient if there is no evidence of benefit. Primum non nocere.
There is a growing body of evidence that RBC transfusion is overused and incurs avoidable costs to healthcare organizations and avoidable harm to patients. Blood transfusion has been said to be one of five most overused procedures in medicine and the most frequently performed therapeutic procedure [8][9][10]. While the overall use of red cells is decreasing in many countries, large audits show much inappropriate use (up to 50%) [11] and considerable inter-institutional variability (ranging from 7.5 to 98 percent transfusion rates in cardiac surgery) [12]. Recently, multiple societies have endorsed recommendations from the Choosing Wisely Campaigns [13][14][15][16] that include avoiding transfusion for arbitrary hemoglobin (Hb) thresholds in the absence of symptoms of active coronary disease, heart failure or stroke, and not performing repetitive CBC and chemistry testing in the face of clinical and lab stability.

What is Patient Blood Management (PBM)?
PBM (also known as Blood Conservation or Bloodless Medicine/ Surgery) is an evidence-based, multidisciplinary approach aimed at optimizing the care of patients who might need transfusion [17][18][19]. PBM represents an international initiative to improve patient outcomes by promoting the use of transfusion of blood and components only when clearly required. This minimizes unnecessary transfusion, reducing the risk of adverse events and conserving limited donor blood for patients with the greatest need. The principles of PBM have been described as the so called '3 pillars' namely: (i) Optimizing the patient's own blood (e.g. pre-operative optimization of hemoglobin (Hb) and hemostasis), (ii) Minimizing surgical blood loss, and (iii) Optimizing the patient's physiological reserve in relation to anemia (including use of restrictive transfusion triggers). The key principles of PBM apply not just to surgery, but to all patients who may need a blood transfusion in both elective and emergency clinical settings. Accordingly, PBM covers all aspects of decision-making in transfusion therapy, including patient evaluation and clinical management with use of appropriate indications and triggers, minimization of blood loss and optimization of the patient's own red cell mass. Anesthetists, transfusionists and intensivists are increasingly driving this focus aimed at using blood transfusion in the most optimum way. By reducing the need for allogeneic blood transfusions and minimizing risks and unnecessary work, PBM can also reduce health-care costs.

Effect of PBM on Costs
PBM can reduce costs both at the blood supply and the hospital levels. The business case for reducing costs by addressing overuse is compelling. Transfusion is expensive. In addition to production, testing and shipping costs, there are further costs associated with transfusion: hospital Transfusion Service type and screen, crossmatch, storage and disbursement, nurse administration and monitoring of the transfused patient, increased LOS and infections in transfused patients. Recent studies have estimated the cost of a transfusion at about USD $1100 [7,20]. There are, however, proven strategies (e.g. a restrictive approach to RBC transfusion, a one-unit dose in nonbleeding patients, and avoiding "routine" daily lab testing in hospitalized patients) and proven tools (e.g. order sets and decision support) that can be used to reduce unnecessary and inappropriate transfusions [21][22][23] and reduce transfusion costs [7,[24][25][26].

How is PBM Done?
There may be many ways to achieve PBM. One important change has been to address underlying anemia before any elective surgery. Also important is tolerance of anemia: studies of restrictive transfusion have essentially shown that a restrictive policy did not adversely affect patient outcomes. While it used to be thought that if a patient needed a unit of blood, you might as well give them two, that paradigm has changed. At surgery, a variety of techniques and approaches can be employed to minimize blood loss. In some hospitals, there is also a procedure known as 'cell salvage' in which the patient's own lost blood is collected during surgery, filtered and transfused back.
Pharmacologics (such as antifibrinolytics like tranexamic acid) and fibrin glues and sealants can play important roles in reducing blood loss at surgery. And good surgical technique remains critical.

Pre-operative/perioperative
• Pre-operative autologous donation (PAD) • Anemia management • Restrictive transfusion trigger PAD Advantages of PAD may include prevention of disease transmission, of red cell alloimmunization and of some adverse reactions. PAD may also supplement the blood supply, allow provision of blood to patients with alloantibodies, and provide reassurance to patients about blood risks. On the other hand, PAD does not affect the risk of bacterial contamination or the risk of ABO incompatibility due to clerical error, and may be associated with donor reactions e.g. vasovagal reactions, while being more costly than allogeneic blood. Furthermore, PAD may contribute to perioperative anemia and increased likelihood of transfusion. Consequently, the use of PAD has diminished and it is rarely employed today as the drawn blood is usually not administered and it is an unnecessary expense [27]. Rather let the patient serve as their own 'blood bank' . But, when judiciously used, PAD may avoid allogeneic transfusion.

Preoperative anemia evaluation and readiness for surgery
Anemia should be considered a serious and treatable medical condition rather than simply an abnormal laboratory value. Preoperative anemia, even when mild, is an independent risk factor that can increase perioperative mortality and morbidity, including increased risk of hospitalization or readmission, prolonged hospital LOS), loss of function, diminished quality of life [28,29]. Despite clear evidence of the harmful effects of unwarranted allogeneic blood, anemia is a major, but modifiable, risk factor for allogeneic blood transfusion.
Anemia is defined by the World Health Organization as a Hb below 120 g/L in women and 130 g/L in men. Specific thresholds apply to pregnant women and children. Anemia prevalence increases with age to 10% of the general population above 65 years of age and >20% of above 85 years of age [30,31]. In the literature, anemia rates in presurgical patients range from 5 to 75 percent, depending on the procedure. For example, a large review of 31,857 patients undergoing elective vascular surgery identified 47% as having preoperative anemia [32]. The prevalence of anemia in hospitalized patients ranges from that in the general population to much higher, depending on the reason for admission, co-morbidities and patient factors such as age and gender; the reported prevalence is at least 25-50% and may be higher, with the elderly and those with chronic conditions at increased risk [33,34]. Also, 74% of hospitalized patients will develop a hospitalacquired anemia [35] with 95% of patients admitted to the intensive care unit (ICU) developing anemia by the third ICU day [36]. A surgical procedure with a moderate or high blood loss will further aggravate the anemia and deplete iron stores.
Early screening for preoperative anemia will allow for diagnostic workup and treatment; this may be done by the surgeon or family physician as soon as prospective surgery is contemplated. In the absence of overt anemia, laboratory tests to evaluate iron stores may be done in patients with co-morbidities associated with depletion of iron stores, such as chronic heart failure, chronic renal failure, rheumatoid arthritis, inflammatory bowel disease or chronic genitourinary and gastrointestinal blood loss.
A standardized approach for the detection, evaluation and management of anemia in the preoperative setting is an important aspect of an effective PBM program. A complete PBM plan should always include an individual risk assessment but algorithms for anemia management may be useful. Figure 1 is an example algorithm. A time frame of at least three to four weeks prior to surgery is recommended to allow sufficient time to diagnose and optimize treatment of anemia based on the identified cause. A comprehensive history and physical examination should be performed with attention to important risk factors for anemia such as advanced age, small body size, female gender, chronic renal, hepatic or connective tissue diseases, and sources of blood loss. Particular attention should be paid to anticoagulant and antiplatelet medications with a clear plan of how to modify these therapies in the few days before surgery, as well as to laboratory testing to determine the cause of the anemia, in order to institute appropriate treatment. Treatment may involve administration of oral or intravenous iron, erythropoiesisstimulating agents (ESAs), folate or vitamin B12. It is important to note that while oral iron is effective in treating iron deficiency, this takes time and often the time is insufficient in the pre-surgical setting and intravenous iron is preferred. Referral to a specialist such as a hematologist or gastroenterologist may be necessary and for some patients it may be prudent to consider deferral of a truly elective surgical procedure to when the patient is appropriately optimized.
Patients with preoperative anemia should be followed in the postoperative period to ensure appropriate continued management of the anemia.

Restrictive transfusion approach
As transfusion practice has come under more scrutiny, there has been increased interest in the "tolerability" or risk of anemia. For example, Carson et al. [37] showed that even mild anemia was associated with increased mortality risk and this was higher for individuals with cardiovascular disease. On the other hand, there is considerable evidence that low levels of Hb can be tolerated in healthy subjects. Hematocrits (Hct) of 10-20% have been achieved in animals and humans using normovolemic hemodilution without untoward effects [38,39]. It has been observed that it is not until the Hb concentration falls below 50 to 60 g/L that morbidity and mortality increases substantially [40].
With recognition that less, or no, transfusion can be associated with improved patient outcomes (e.g. shorter length-of-stay and fewer infections than in transfused patients) and with an increasing acceptance of tolerating lower Hb levels in patients, restrictive transfusion practices are being increasingly applied. These consist primarily of (1) a single unit at a time policy and (2) accepting a lower Hb trigger for transfusion. Use of a restrictive transfusion protocol can be regarded as an important blood-saving and cost-saving strategy which does not lead to adverse events in the non-high risk patient and can readily be achieved at no additional cost.

Single unit transfusions
Single unit RBC transfusion is the practice of prescribing only one unit at a time, with clinical reassessment of the patient prior to prescribing a subsequent unit, in patients who are not critically bleeding. Traditionally, single-unit RBC transfusions were believed to be insufficient to treat anemia, but recent data suggest that they lead to a safe reduction of transfusion requirements. Each transfusion should be an independent clinical decision based on the risk, benefits and alternatives and transfusion should be to alleviate patient signs and symptoms of anemia. A single unit of RBC will often relieve acute symptoms of anemia and/or raise hemoglobin level in a stable patient to a level above the guideline trigger, in which case transfusion may be unnecessary or can be postponed or replaced by other anemia treatment. A single unit policy may have the added advantages of conserving the blood supply and of reducing the risk of Transfusion Associated Circulatory Overload (TACO) which is among the highest risks associated with RBC transfusion, of particular concern in the elderly, females and patients with heart or kidney failure, or a positive fluid balance [41,42]. Hence, where transfusion is indicated, a single unit of RBC, followed by clinical reassessment to determine the need for further transfusion, is appropriate. Transfusion therapy in severely bleeding patients may require separate guidelines.

Restrictive transfusion threshold/trigger
Restrictive transfusion thresholds (triggers) are an effective method of reducing and conserving RBC use and improving patient outcomes [43]. RBC transfusion should not, however, be dictated by a Hb 'trigger' alone, but rather based on assessment of the patient's clinical status. While PBM Guidelines provide guidance on Hb 'triggers' this should always be in the context of the patient's clinical condition. The threshold for RBC transfusion in both medical settings and in the postoperative surgical period has evolved over the years. A number of large, randomized clinical trials and prospective observational studies have assessed the effectiveness of allogeneic RBC transfusion and demonstrated that restrictive RBC transfusion practices result in at least equivalent patient outcomes as liberal approaches, and may actually reduce morbidity and mortality rates in some patients. Figure  2 shows some of the randomized prospective trials demonstrating the reduction in transfusion rates when a restrictive trigger is employed [44][45][46][47][48][49].  [44][45][46][47][48][49] showing reduction in transfusion rate when a restrictive (in red) versus more liberal (in blue) Hb trigger is employed.
Meta-analyses of 19 trials showed that restrictive transfusion strategies were associated with a significant reduction in hospital mortality, 30-day mortality, pulmonary edema, bacterial infections and re-bleeding [50]. In the absence of acute myocardial or cerebrovascular ischemia, postoperative transfusion may be inappropriate for patients with a Hb level of >80 g/L; other Guidelines suggest a Hb level of 70 g/L [51]. Some use a trigger of 70 g/L for most patients and a trigger of 80 g/L for patients with evidence of potential added oxygen need e.g. those with cardiovascular or cerebrovascular disease.
It is important to recognize that the Hb level selected should be viewed as a threshold rather than a trigger; if the Hb is below the threshold level, a transfusion may be considered but is not mandatory. Nevertheless, many clinicians mainly use Hb values to guide transfusion decisions. Physician compliance with clinical practice guidelines is, however, often incomplete, as multiple barriers limit guideline adherence. However, recently programs utilizing clinical decision support (CDS) directed toward more appropriate RBC transfusion practice have been successfully implemented and have been effective in reducing RBC utilization with equivalent or improved patient outcomes [22].

PBM at the hospital level
Establishing a PBM strategy needs leadership and support at all levels, from policymakers and managers, to executive management and health professionals from various clinical disciplines within hospitals. Patient input on achieving patient acceptance and participation is often useful. At an operational level, the cornerstone of a PBM program is the multidisciplinary team. General practitioners, surgeons, anesthetists, nurses, hematology/transfusion medicine, pharmacy and laboratory staff all have important roles to play in surgical PBM and should be engaged in the development and implementation of the PBM strategy. PBM strategies also apply in the non-surgical setting. Table 1 shows key elements of a PBM program at the hospital level.  Institutional support is critical to project success, as it provides access to the resources required to change current hospital culture and practices. Examples of areas that people may rally around include the financial costs of RBC overuse to the organization, the potential impact on LOS and hospital-acquired infections associated with transfusion, the costs and mortality associated with untreated anemic patients, or the potential harm associated with iatrogenic anemia. Although useful, these arguments should not necessarily be the primary selling point -the major impact is on managing patient risks for patient safety.

Role of Transfusion Practitioner
The Transfusion Practitioner (TP) [69] or equivalent (e.g. PBM Practitioner/Nurse/Coordinator) has a critical role to play in developing a PBM culture within healthcare establishments. Although only one part of the team developing and implementing the required PBM strategies, they have a multifaceted role in engaging with scientific, laboratory and clinical colleagues, as well as, importantly, the patients. Very often the TP is the conduit for information, pulling together available resources, reviewing activities undertaken by transfusion colleagues in other centers, collecting audit data and evaluating the benefit of activities.
TPs have made a significant contribution in helping to improve transfusion practice at a local, regional and national level by promoting safe transfusion practice. To do this, information must be provided to clinical colleagues to engage with them and plan strategies. The many different elements that constitute PBM can be tailored for the different clinical specialties that use blood transfusion services and, through good clinical leadership, the TP can coordinate a number of initiatives and work streams such as pre-surgery optimization, medical management in iron deficient anemia, or electronic solutions for blood ordering and prescribing. The TP may not lead these projects as they are owned and managed at a local level by the users of blood, but the TP can have the global view of PBM and the benefits it brings. Currently there is a lack of definitive evidence on the role of the TP within PBM and much of the information is anecdotal. It is acknowledged that a greater focus of the role of the TP in PBM is needed within the printed material and resources available. In the UK the TP role was recommended as part of the Department of Health Better Blood Transfusion strategy. Over the years many UK Trusts have appointed to the TP role, employing experienced staff with a Nursing, Midwifery or Biomedical Science background. TPs also play a major role within Hospital Transfusion Committees (HTCs) and hospitals may also have a separate PBM Committee. HTCs should play a key role in blood conservation and evidence does exist for improved use of blood components due to interventions such as transfusion policies, clinical audits, education of clinicians.
TPs add value with regards to training and education, transfusion safety and clinical practice. However, the role and responsibility of the TP varies widely and has changed significantly for some since it was introduced. There is significant variation in how TPs spend their time. A book by the American Association of Blood Banks (AABB), 'Transfusion Medicine's Emerging Positions: Transfusion Safety Officers and Patient Blood Management Coordinators' , provides some guidance on such positions (whether hospital-based or bloodcenter-based), the professionals who make good candidates, the scope of their responsibilities, their key role in improving patient outcomes and the influence they have on various hospital departments.

The Ontario experience
In Ontario (population approx. 13,000,000), the ONTraC (Ontario Transfusion Coordinators) network program funded by the Ontario Ministry of Health and Long-term Care has placed PBM coordinators in 25 community and teaching hospital throughout the province [7,70,71]. These hospitals account for about 75% of the blood used in the province and the intent is to educate physicians, nurses and patients about best transfusion practices and alternatives to transfusion. The ONTraC TPs are expected to spend 50% of their time on implementing PBM practices, 25% on staff and patient education, and 25% on data collection [7]. Figure 3 shows the reduction in annual provincial transfusion rates in Ontario in knee arthroplasty and coronary artery bypass graft (CABG) surgery since initiation of the program in 2002. Not only has there been a significant reduction in proportion of patients transfused, but those who are transfused are receiving less blood (for example, CABG patients who were transfused in 2002 received a mean of 3.3 units RBC; in 2015 transfused patients received a mean of 2.5 units). Overall, in the initial four ONTraC targeted surgical procedures (knee, hip, CABG and radical prostatectomy) transfusions have decreased between 60 and 90% from baseline, and many additional diagnoses and surgeries (e.g. gynecologic and obstetric procedures, heart valves, spine and colon surgery) have been added. In 2015, RBC usage in Ontario was 27.4 units per thousand population, among the lowest internationally [72]; by contrast, data collected in an AABB survey showed that RBC usage in 2013 in the USA was 40.3 units per thousand population [73]. In the ONTraC program there is a focus on correction of preoperative anemia. Data from ONTraC shows that the likelihood of receiving a transfusion increases exponentially when the preoperative Hb level is below 130 g/L (Figure 4). Similar results have been reported by others. Hence we try to achieve this level if possible in the preoperative management of anemia. Whilst not always possible, the higher we are able to get the Hb level, the less likely the patient is to receive a transfusion. Lead time is necessary for the anemia treatment. A longer lead time results in fewer transfusions: for example, in 2014 knee arthroplasty patients who had a lead time of <7 days had a transfusion rate of 5.5%, versus those with a lead time of 7-15 days of 4.7%, 15-21 days of 3.7% and >21 days of 1.8%. We try to get to see the patient at least 3-4 weeks prior to surgery. Family physicians can play an important role in the management of the anemia. The ONTraC TPs also emphasize the need for tolerance of anemia and 'restrictive' transfusion thresholds when appropriate and, in Ontario, for knee arthroplasty for example, whereas in 2002 the mean nadir Hb (transfusion trigger surrogate) was 84 g/L, it has decreased to 72 g/L.
Another focus has been on the use of the antifibrinolytic drug, tranexamic acid. Whilst this was extensively used in cardiac surgery, despite increasing literature advocating its use in orthopedic surgery, it was relatively infrequently employed in the latter. Table 2 Table 2: Effect of tranexamic acid (TXA) use on provincial transfusion rates in knee arthroplasty.
The annual cost savings potentially attributed to the ONTraC program for the four initial targeted procedures only compared to baseline are over CAD 15 million in purchase of RBCs and over CAD 45 million in savings to the healthcare system overall [70]. The cost of the program is approximately CAD 3 million. Hence, in addition to catering to patients' preferences, the program reduces risks, improves outcome and is cost-effective and cost-efficient.
Further information on the ONTraC program can be found (along with much information on PBM, including patient pamphlets) at www.ontracprogram.com.

Conclusion
Pathways supporting PBM are simple in design, labor costs relatively low, capital investments small, quality and outcomes gains high, savings for the organization large, and the greater public good is served. So why is it not more extensively practiced? This is in large part because of physician training and 'traditional' practices, i.e., transfusion was the default position for anemia, and, secondly, the belief that blood is safe -risks have decreased with improved HIV and hepatitis testing. We need to replace the 'Traditional Concept' that blood products are an effective therapeutic intervention with the 'New Concept' that transfusion of blood products is rather an undesirable outcome. We have learned that transfusion rates are not patientdependent, but are institution (geography) dependent and while 30% of the variation may be surgeon dependent, 70% is hospital specific and that there are 'high-transfusion' hospitals and 'low transfusion' hospitals [74,75] -so we need to change the 'culture' in hospitals, although this is often not easy.
We have recognized that although hazardous, anemia is frequently ignored, exacerbated by wasteful practices such as excess phlebotomy, or inappropriately treated with inappropriate transfusion. However, anemia is easily treated, and effective protocols have been developed to guide physicians in their care of anemic patients. Improving the management of anemia therefore represents a great opportunity to improve patient outcomes and reduce costs and adverse events. In the operative period, the benefits of appropriate use of antifibrinolytics, of cell salvage and of topical hemostats, among other approaches, have been shown. The principles of PBM extend beyond surgery and are multidisciplinary, best applied by a team approach involving a large variety of health care personnel and patients. There are challenges in embarking on a PBM project, given the scope and complexity of the issue, but a systematic approach to the process (including forming a multidisciplinary team, obtaining institutional support, assessing baseline performance, thoughtful design and implementation of interventions, and careful monitoring) increases the chance of success. PBM is cost-effective and presents an opportunity to improve outcomes, reduce risks and save costs all at the same time.
It is interesting to note that recognition of the risks of transfusion is not new. A letter to the Lancet in 1949 [76] stated: "blood transfusion … a mass produced remedy which daily presents fresh problems. In the hands of experts it is virtually safe, and very valuable; but there is little doubt that … many deaths supposed to have occurred 'in spite of transfusion' have really been caused by it. In fact there are few risks when the doctor fails to insert a needle … into the vein; they begin to mount once he succeeds. " And in the 'immortal words of Dr. Beal in 1976 [77]: "Blood transfusion is a lot like marriage. It should not be entered into lightly, unadvisedly or wantonly, or more often than is absolutely necessary. "