FABB Annual Meeting
How Do I Implement A Hospital- Based
Blood Management Program
Richard R. Gammon, MD Medical Director
Learning Objectives
• Discuss steps needed to implement a
patient blood management (PBM) program
• Give an overview of alternates to
allogeneic donations
• Cover case studies of hospitals that have
effectively implemented PBM programs
Patient Blood Management
Definitions
• AABB
– Evidence-based multidisciplinary approach to optimizing the care of patients who may need transfusion
• SABM
– Timely application of evidenced-based
medical and surgical concepts designed to maintain hemoglobin concentration, optimize hemostasis, and minimize blood loss in an effort to improve patient outcomes
SECTION ONE
Background
• Transfusion of blood products is lifesaving therapeutic intervention in many clinical settings • Certain risks associated
with these products
ranging from hemolysis caused by the
misadministration of a unit, to
• lung injury, to bacterial contamination
Background
• To avoid all of these risks of transfusion,
one must simply avoid all transfusions.
• Wide variety in transfusion practice that
exists suggests that suboptimal patient
care is occurring
– Over and underuse of blood products can be problematic
Background
• Recognizing the overuse, the University of
Pittsburgh Medical Center (UPMC)
• Implemented a six-point plan to reduce the
use of allogeneic blood products
• Total blood management (TBM)
– Strategy aimed at improving and streamlining the diagnosis and management of anemia,
thrombocytopenia, and coagulopathy
– Focus on reducing or eliminating allogeneic transfusions
TBM - Team
• Transfusion service,
• Anesthesiology
• Perfusion
• Change specialist
– UPMC’s Quality and Innovations Institute. – Liaison between the TBM task force and
UPMC executives and others who have the power to implement the task force’s
Point 1- Implement Evidenced-Based
Transfusion Triggers
• Topic of evidence-based transfusion triggers is vast • Many reports have described optimal use of blood
products in various medical and surgical patients. • TRICC trial
– Demonstrated that even sickest patients in the intensive care unit who were not undergoing acute cardiac events did not benefit from liberal red blood cell (RBC) transfusion strategy,
• Other studies
– Demonstrated that prophylactic plasma transfusion for patients with relatively low international normalized ratios (INRs) does not result in significant clinical benefit,
Implement Evidenced-Based Transfusion
Triggers
• As optimal use of our blood product
resources is becoming clearer.
• Not sufficient to simply generate this data—
the outcomes from these studies must be put
into clinical practice.
• Question becomes:
• How can we as transfusion medicine
specialists disseminate what we know about
how to use our products to those who
Point 2 – Prescriber Education and
Audits
• Made extensive use of the information
technology resources
– Available to us at UPMC to assist with both
physician education and auditing of transfusion practice.
• Using the computerized physician order entry
(CPOE) interface
• Implemented a warning screen
– Appears when prescriber attempts to order either RBCs or plasma on a patient whose most recent laboratory values are in excess of the UPMC
Avoid Warning Fatigue
• Selected a hemoglobin (Hb) level of more
than 8.5 g/dL and INR of less than 1.6 as
values above and below which, respectively,
twarning will appear to inform prescriber that
based on laboratory values transfusion does
not appear to be indicated.
• Prescriber
– Can override the warning with a simple mouse click for emergency situations
Audit Effect
• Plasma orders - June 2011
– Warning screen appeared 205 times and in 25% of these cases, order was canceled
• RBC warning screen
– Recent 6-month period ending May 31, 2011, fully 12% of the 5948 non–evidence-based orders were canceled
– Provided prescriber information on the best
practice via the warning screen at the time the decision to transfuse was occurring.
Audit Effect
• Translates to more than $600,000 of savings
when annualized over a year.
• Knowing on which patients the warning
screen appeared also facilitates the chart
reviews performed by the transfusion
committees as required by some
regulatory/professional agencies
• Warning screen appeared but order was not
canceled
– Clinical emergency was occurring
Audit Effect
• RBC and plasma alerts by themselves are probably not sufficient to eliminate non–
guideline-based transfusions. • Total number of alerts and
number of canceled orders
– August 2011 (1076 and 115, respectively) were nearly
identical to the corresponding mean monthly values
– December 2010 through May 2011 (991 and 115,
respectively).
• Continued provider education and other guideline
enforcement modalities appear to be required to optimize
Charting High-Users
• More targeted approaches to the enforcement of
transfusion guidelines have been initiated
• RBC transfusion practice among orthopedic surgeons at our facilities who perform total hip arthroplasties
(THAs) is highly variable • Thresholds among THA
patients using a functional endpoint, we have
produced what we call a “bubble” graph
Charting High-Users
• Reports are generated on a monthly basis
• Forwarded to the vice president of
operations for each individual hospital.
• Strategic considerations about the
surgeon’s economic value to the hospital
can be weighed with the cost of providing
excess blood products
Education
• Provide live teaching sessions
– Departmental grand rounds, resident and fellow lectures
– Consultation with prescribers on demand and when the blood bank identifies unusual or egregious
transfusion practice (e.g., ordering a double dose of platelets or ordering a single unit of cryoprecipitate).
• Our teaching sessions
– Associate a face with a voice on the end of the phone and provide a means to update our colleagues on
other blood management initiatives that are under way at UPMC.
Education – Committee Meetings
• Dedicated agenda item for blood
management
– Effective way of communicating progress in implementing and enhancing six-point plan, – Allows to anticipate roadblocks that might
Point 3 - Minimize Preoperative Donation
of Autologous RBCs
• Historically preoperative autologous donation (PAD) was considered a means to avoid allogeneic transfusion
– One or more of the patient’s autologous RBC units available on day of surgery
– Thought that if patients donated RBCs in advance of their surgery, then RBCs that would be produced by their marrow
between time of donation and the reinfusion of PAD units during or after surgery would represent a net gain of RBC mass, which might obviate the need for allogeneic RBC transfusion.
• Theory does not match the practice
– Suggestions that patients who reduce their RBC mass before their surgery through PAD may require more allogeneic RBCs than those who do not donate RBCs
PAD- Disadvantages
• Does not eliminate other more common adverse events
– Bacterial contamination of the PAD unit
– Accidental transfusion of an allogeneic RBC unit instead of PAD unit
– Febrile reactions
• Evidence that patients who undergo surgeries with some degree of preoperative anemia tend to have worse outcomes
– Removing a significant amount of Hb PAD might predispose patients to worse outcomes
– Goal of the procedure was to try to improve their outcome by avoiding allogeneic RBCs
PAD
• A PAD unit utilization rate of 50% is often
cited as the benchmark
• 2008, the PAD utilization rates at two
hospitals in system that had historically
collected large numbers of these units
– Combined 148 of 239 (62%)
Prescriber Audits
• Identifying surgeons who repeatedly
transfuse large quantities of RBCs
• Provide specific feedback and education
to these “outliers”
• Inspire desire to reflect and change their
own practice when they see how they
PAD - Education
• Completely eliminate the collection of PAD RBC units at one hospital by 2010
• Cutting number of units collected at the other hospital by nearly 40% by 2011
• Offering surgeons alternatives to PAD was essential to reducing their utilization of this potentially wasteful
practice.
• PAD persists at some of our hospitals suggests that perhaps
– System wide policy against the use of PAD, with the exception of those patients with rare blood groups and multiple alloantibodies, might be required to completely eliminate its use
Point 4 – Cell Salvage
• Blood that is shed from the surgical site
can be recovered using a suction device
and then filtered and washed before it is
returned
• Occur after
many different types of surgery
• Main value
– Ability to reduce the need for allogeneic RBCs by collecting and returning autologous shed blood
Cell Salvage
• Encouraged use among the surgeons who
routinely recommended PAD to patients
• Emphasizing benefits of not reducing their
patient’s Hb before surgery
• Ensuring that experienced perfusionists were
readily available during their cases
• Initial obstacles
– Erroneous perception that there were numerous contraindications to its use
– Lack of perceived effectiveness in avoiding allogeneic transfusion
Cell Salvage – Addressing
Contraindications
• Results of a locally conducted clinical
study that demonstrated better outcomes
were presented to the surgeons for their
review
• Perception lack of effectiveness of these
technologies
– More robust technician education and implementation of AABB perioperative
Cell Salvage – Cost Effectiveness
• Washing and returning of the cells optimally
takes place when a full bowl has been
collected
• Cost of processing a partially filled bowl might
not make the process worthwhile, nor is it
expected to return significant quantity of
RBCs to patient
• Close working relationship between TBM
subject matter experts transfusion medicine,
anesthesiology, and perfusion services was
forged with our surgeons
Point 5 - Preoperative Anemia
Optimization
• Anemic patients who undergo surgery
tend to have worse outcomes
• Major risk factor for requiring perioperative
allogeneic RBCs
• Elective surgery cases there is frequently
a sufficient length of time between the
scheduling of surgery and its actual date
to permit the diagnosis and treatment
POA- Audit
• 200 patients undergoing elective hip or
knee surgery at our facilities
– Approximately 27% of these patients were anemic before their surgery
– Nearly 7.5 times increased risk of requiring an allogeneic RBC transfusion compared to
patients who were not anemic before their surgery (p = 0.0001)
Anemia - Prevention
• Computer-based, system wide surgical scheduling program (Surginet) are electronically matched with their preoperative bloodwork that is typically drawn around the time of the scheduling.
• Electronic rule matches each surgical patient to laboratory values;
– Anemic patient is discovered
– Computer program automatically generates e-mail or fax to surgeon and primary care physician
– Indicating that their patient is anemic and suggesting that its etiology be investigated before the surgery
Anemia - Prevention
• The surgeon
– Weigh the extent of the anemia with
anticipated blood loss for procedure and
manage patients themselves
• Refer them to
specialist or primary care physician
Benefits – Advance Notice and Treatment
of Anemia
• Patient
– Alerts their physicians to the fact that they are anemic and also provides a mechanism for patients to be assessed and treated before their surgery
– Reducing the likelihood that they will require a perioperative transfusion.
Point 6- Point of Care Testing
• “laboratory on wheels” • Measure many different
laboratory measures at a site that is often remote from the main laboratory and closer to where the
patient is physically located. • Main advantages
– Turnaround times for
reporting of these measures is greatly reduced
– “Real-time” decision making on the need for blood
products
– Only microliter quantities of blood
POC Testing – Workflow Analysis
• Largest hospital revealed that for the
operating room (OR), mean turnaround time
for a Hb level -45 minutes
• Not conducive to real-time transfusion
decision
• Studies have shown that when POC devices
were used to inform transfusion decisions
– Cardiac surgery patients received fewer RBC units and had less morbidity through reduced
chest tube drainage and fewer returns to the OR to control excessive bleeding
POC – Decisions To Implement
• Initial capital expenditure of buying devices
themselves
• Ensure that these devices are placed under
the aegis of hospital department that has the
expertise to design and execute protocols
that validate and keep them in control
• Some cases - extension of the main
laboratory
POC Standardization
• Program to standardize devices used in the
ORs throughout hospitals
• Devices that must
remain stationary (like TEG)
– Machine located in the main laboratory or in dedicated space near the OR with its output viewable in real time in the actual ORs and
SECTION TWO
Phamacologic Therapies – Assist
Coagulation Function
Drug Brand Name Mechanism of
Action Suggested Dose Aminocaproic Acid Amicar (Xanodyne, Newport, KY) Inhibits the proteolytic activity of plasmin 5 g bolus followed by 1g/hr infusion
Desmopressin DDAVP
(Sanofi-Aventis, Bridgewater, NJ) Increases Factor VIII release 0.3 ug/kg Recombinant Factor VIIa (rVIIa)* Novoseven (Novo Nordisk, Princeton, NJ) Activates coagulation cascade Approximately 50-100 ug/kg**
Tranexamic acid Cyklokapron
(Pfizer, New York, NY) Inhibits the proteolytic activity of plasmin 10 mL/kg every 6-8 hours
Tranexamic Acid
• A synthetic derivative of the amino acid lysine • Binds to plasminogen
and plasmin, occupying the binding sites for
fibrinogen and fibrin
• Inhibits the activation of plasminogen to plasmin and interferes with the lysis of fibrinogen and fibrin
Tranexamic Acid
• In a total of 53 studies that included 3836
participants, tranexamic acid reduced the need for blood transfusion by a third (relative risk [RR] = 0.61)
• In a study of over 20,000 trauma patients in 40 countries, the patients received either
tranexamic acid (loading dose 1 g over 10
minutes, then an infusion of 1 g over 8 hours) or matching placebo
• All cause mortality was significantly reduced with tranexamic acid 14.5%, compared to placebo
Tranexamic Acid
• Early treatment (<= 1 hour from injury)
– Significantly reduced the risk of death due to bleeding by 32% (5.3% versus 7.7% in the placebo group [RR = 0.68])
• Treatment 1-3 hours
– Reduced the risk of death due to bleeding (4.8% versus 6.1% in the placebo group [RR = 0.79])
• Treatment >3 hours
– Increase the risk of death due to bleeding (4.4% versus 3.1% in the placebo group [RR = 1.44])
Tranexamic Acid
• No difference in vascular occlusion (pulmonary embolism, myocardial infarction) was observed • High dose tranexamic acid has a risk of seizures
in patients undergoing cardiac surgery
• Some data show a lower dose results in a reduction in the incidence of seizures
Total Hip Replacement
Postoperative Autologous Transfusion
rFVIIa
• A phase 3 randomized clinical trial in trauma patients was terminated early, at 573 of 1502 planned patients because of unexpectedly low mortality prompted by futility analysis (10.8% versus 27.5% planned/predicted)
• Mortality was 11.0% (rFVIIa) versus 10.7% (placebo)(p=0.93) blunt trauma: and 18.2% (rFVIIa) versus 13.2% (placebo)(p=0.40) penetrating trauma
• Blunt trauma rFVIIa patients received 7.8 ± 10.6 RBC units and placebo received 9.1 ± 11.3 units (p=0.04)
rFVIIa
• For counting units of any allogeneic products, blunt trauma rFVIIa patients received 19.0 ±
27.1 units through 48 hours and placebo 23.5 ± 28.0 units (p=0.04)
• No difference in thrombotic events was observed • Modern evidence-based trauma care lowers
mortality, paradoxically making outcome studies increasingly difficult
rFVIIa
• A study for the correction of coagulopathy before emergent craniotomy in 28 blunt trauma patients was performed
• The rFVIIa group consumed fewer costs of
RBCs ($756 per patient vs. $2916 per patient, p<0.001) and plasma ($369 vs. $927, p=0.001) • The rFVIIa group still consumed fewer total
costs of transfused blood products when costs of rFVIIa were included ($2557 versus $4110,
p=0.04)
rFVIIa - Conclusions
• Pharmacologic doses of rVIIa enhance the
thrombin generation on activated platelets
• May be of benefit in providing hemostasis
with profuse bleeding and impaired
thrombin generation
– Thrombocytopenia and functional platelet defects
– Used successfully with trauma or surgical bleeding.
rFVIIa - Conclusions
• Dose and timing remain to be defined
• Several prospective, controlled clinical trials have shown marginal or no benefit when compared to placebo in patients with trauma, hemorrhagic
stroke, cirrhosis and gastrointestinal bleeding • The decision on when and where to use
recombinant Factor VIIa continues to be one that must be made by individual physicians, assisted by their hospital pharmacotherapeutic or
Topical Hemostatic Agents
• Conventional methods include manual pressure, clamps, ligatures, tourniquets and cautery
• A military study showed several hundred
applications of tourniquets for 15 minutes – 2 hours had only rare complications
• No amputations resulted from their use
• Conventional methods are sometimes less effective in controlling bleeding from complex injuries and where access to the area of
Topical Hemostatic Agents
• Collagen-based hemostats • Gelatin-based hemostats • Cellulose-based hemostats • Polysaccharide-based hemostats • Albumin-derived hemostats • Inorganic hemostats • Fibrin-based hemostats • Polymeric hemostatsErythropoietin
• A randomized, double-blind, placebo-controlled, multicenter trial (EPO-2, N=1302) in anemic,
critically ill patients, demonstrated a 29-day
survival benefit in the trauma subgroup receiving Epoetin-alpha (mortality 8.9% versus 4.1% for the placebo group)
• A second trial (EPO-3, N=1460) confirmed this survival benefit
• A subsequent article evaluated the impact of baseline factors, including trauma-specific
Erythropoietin
• Patients received 40,000 U Epoetin-alpha or placebo weekly for a total of 4 (EPO-2) or 3 (EPO-3) doses, starting on ICU day 3
• Demographic and trauma variables at baseline were comparable
• Mortality was consistently reduced by ~50% in both studies
• EPO-2 day-29 adjusted hazard ratio (HR)=0.50 • EPO-3 day-29 adjusted HR=0.38
Erythropoietin
• In EPO-3, the effect of Epoetin-alpha on
mortality was independent of the number of units of blood transfused
• In addition to increasing RBC production,
erythropoietin has many other functions as well • Multiple sources produce erythropoietin,
including endothelium, vascular smooth muscle cells, kidney, liver and brain
Erythropoietin
• A main effect of erythropoietin is to reduce
apoptotic cell death and decrease inflammation • It also increases the production of neural stem
cells
• It preserves endothelium and contributes to angiogenesis
Erythropoietin
• EPO remains a valuable tool for patients
with special requirements, such as
Jehovah’s Witness patients for whom
blood transfusion is not an option
Erythropoietin
• Main disadvantages
– Expense compared with red blood cell units (may not be covered by insurance)
– Link with cardiovascular and thromboembolic events
• Until additional safety data are
forthcoming, however, off-label use cannot
be recommended
Blood Recovery Sponges
• Rinsing sponges used to wipe blood from
surgical field • Fully soaked laparotomy sponge – 100 mL RBC • Harvest in basin of saline solution- 75% recovery
SECTION THREE
