Hematologic risk assessment (HRA) involves evaluating and managing needs for red blood cell (RBC) transfusion as well as the integrity of a patient’s coagulation ability, including intrinsic bleeding disorders or coagulopathies and anticoagulant or antiplatelet medications. HRA is an essential component of preoperative assessment for virtually all patients undergoing surgery, but it is particularly critical for those refusing blood and blood products, in patients for whom an insufficient number of RBC units is available, in patients with a history of a bleeding disorder, in those with preoperative laboratory evidence of compromised blood coagulation, and in patients on anticoagulant or antiplatelet drugs ( Box 10.1 ). The goal of HRA is to minimize transfusion needs and prevent complications such as hemorrhage and thrombosis. Efficient collaboration and communication among surgeons, anesthesiologists, and at times other specialists, such as hematology consultants, are prerequisites to successful perioperative management of patients at elevated risk. This chapter covers the following topics: the preoperative assessment for RBC transfusion needs and mitigation of RBC transfusion risk, preoperative assessment of coagulation abnormalities, and perioperative management of antiplatelet and anticoagulant medications.
Patients refusing allogeneic red blood cell (RBC) transfusions
No or insufficient number of allogeneic RBCs available
Patients with a history of a bleeding disorder
Patients with a prolonged prothrombin time (PT)
Patients with a prolonged activated partial thromboplastin time (aPTT)
Patients with a low platelet count
Patients with a platelet function defect
Patients on antiplatelet drugs
Patients on anticoagulant medications
Red Blood Cell Transfusion
Hemoglobin (Hgb) contained within RBCs is the primary transport mechanism of oxygen throughout the circulatory system, responsible for maintaining systemic tissue oxygenation. Thus, decreases in Hgb content of the blood, termed “anemia,” can lead to decreased oxygen delivery (DO 2 ) and end-organ ischemia. DO 2 is dependent on the following equation:
DO2=cardiac outputCO×arterial oxygen contentCaO2
where SaO 2 is arterial oxygen saturation and PaO 2 is partial pressure of oxygen dissolved in arterial blood.
DO 2 and subsequent extraction must be equal to or greater than oxygen consumption; otherwise, a deficit exists, resulting in tissue ischemia. As can be seen in the previous equation, under physiologic conditions, arterial oxygen content is primarily dependent on Hgb concentration, with only a small fraction relying on dissolved oxygen. Therefore, the rationale for RBC transfusion is to maintain an adequate Hgb concentration for systemic DO 2 . In healthy, resting volunteers, acute isovolemic anemia was tolerated to an Hgb of 5 g/dL, and this is primarily compensated by increases in heart rate, stroke volume, and cardiac index, as well as increased oxygen extraction. However, in a study of 1958 patients who declined blood transfusion for religious reasons, odds of death rose inversely with declining preoperative Hgb, demonstrating the deleterious effects of anemia during the physiologic stress of the perioperative period, which was especially evident in patients with underlying cardiovascular disease. The appropriate Hgb target within the surgical population continues to be debated in randomized clinical trials. The TRICC (Transfusion Requirements in Critical Care) trial was one such randomized clinical trial in critically ill patients. That trial compared a restrictive (Hgb < 7 g/dL) versus liberal (Hgb < 10 g/dL) transfusion trigger. Although the TRICC trial did not find a statistically significant difference between the two groups, it found a trend toward lower 30-day mortality in the restrictive group, and subgroup analyses found a statistically significant decrease in 30-day mortality among the less critically ill and those older than age 55 years. However, the appropriate Hgb target likely differs among patients in varying comorbidities and in surgical subpopulations. In a recent meta-analysis comparing critically ill versus surgical patients, while a restrictive transfusion strategy resulted in reduced 30-day mortality among critically ill patients, the opposite effect was found among surgical patients. In the face of the available evidence, the American Society of Anesthesiologists Task Force on Blood Component Therapy concluded that no single Hgb trigger should dictate transfusion decisions, instead using individual patient risks for complications. The transfusion target for most patients likely falls somewhere in the range of an Hgb of 7–9 g/dL, with transfusion being rarely indicated for Hgb concentrations greater than 10 g/dL and almost always indicated when Hgb falls below 6 g/dL.
RBC transfusion is not without risks. These include transfusion-transmissible infection (viral, bacterial, and prion disease), allergic and immunologic reactions (e.g., febrile transfusion reaction, hemolytic reactions, alloimmunization, and autoimmunization), mistransfusion, transfusion-related acute lung injury, and transfusion-associated circulatory overload. According to 2016 US Food and Drug Administration data on fatalities reported following blood collection and transfusion, the highest number of fatalities related to transfusion was due to transfusion-associated circulatory overload.
In addition to the inherent risks of transfusion, for religious reasons or otherwise, not all patients are accepting of RBC transfusion. A common example is the patient who is a Jehovah’s Witness. While it is important to clarify with each individual patient which, if any, blood products they are willing to accept, Jehovah’s Witnesses generally refuse the transfusion of whole blood, blood cells, and plasma, stemming from strict obedience to the belief that the soul abides in the blood. The refusal or acceptance of “minor” components of the blood (such as albumin and coagulation factors) is left up to the individual patient, and substances produced by genetic engineering are generally accepted. Moreover, the integrity of the vascular tree is paramount because Jehovah’s Witnesses believe the soul cannot remain outside the body. Therefore, cardiopulmonary bypass or normovolemic hemodilution is accepted as long as the continuity of the blood circulation is preserved.
Patients who have received transfusions in the past may develop antibodies directed against relevant red cell surface antigens, leading to incompatibilities with future transfusions. Some patients raise a high level of a clinically important alloantibody that is directed against a very common antigen (e.g., present in > 90% of individuals). It might then become almost impossible to find blood units to which they do not react, or not enough for a proposed surgery.
In these situations in a patient refusing RBC transfusion or for whom there is insufficient availability of compatible RBC units, prior to proceeding with surgery, it becomes necessary to evaluate the patient’s maximum allowable blood loss (mABL), the periprocedural bleeding risk ( Table 10.1 ), and any alternatives to transfusion ( Fig. 10.1 ). For patients undergoing procedures with elevated bleeding risk, the mABL can be calculated using the following formula, assuming that normovolemia is maintained :
|High (2-day risk of major bleed 2%–4%)||Low (2-day risk of major bleed 0%–2%)|
|Heart valve replacement||Cholecystectomy|
|Coronary artery bypass||Abdominal hysterectomy|
|Abdominal aortic aneurysm repair||Gastrointestinal endoscopy ± biopsy, enteroscopy, biliary/pancreatic stent without sphincterotomy, endosonography without fine-needle aspiration|
|Neurosurgical/urologic/head and neck/abdominal/breast cancer surgery||Pacemaker and cardiac defibrillator insertion and electrophysiologic testing|
|Bilateral knee replacement||Simple dental extractions|
|Laminectomy||Carpal tunnel repair|
|Transurethral prostate resection||Knee/hip replacement and shoulder/foot/hand surgery and arthroscopy|
|Kidney biopsy||Dilation and curettage|
|Polypectomy, variceal treatment, biliary sphincterectomy, pneumatic dilation||Skin cancer excision|
|PEG placement||Abdominal hernia repair|
|Endoscopically guided fine-needle aspiration||Hemorrhoidal surgery|
|Multiple tooth extractions||Axillary node dissection|
|Vascular and general surgery||Hydrocele repair|
|Any major operation (procedure duration > 45 minutes)||Cataract and noncataract eye surgery|
|Bronchoscopy ± biopsy|
|Central venous catheter removal|
|Cutaneous and bladder/prostate/thyroid/breast/lymph node biopsies|
EBV=body weightkg×average blood volumemL/kg,
with EBV being estimated blood volume, Hgb i being initial hemoglobin, and Hgb f being final hemoglobin (transfusion trigger). Blood volume in nonobese patients is calculated as 70–75 mL/kg in adult males and 60–65 mL/kg in adult females. Blood volume is 80 mL/kg in infants, 85 mL/kg in full-term neonates, and 95 mL/kg in premature neonates.
The next step is to compare the mABL with the perioperative bleeding risk and the historical perioperative blood loss for the planned operation performed by the local surgical team. This comparison should be with the historical perioperative blood loss rather than the intraoperative blood loss, accounting for potentially significant postoperative blood loss. If the mABL is considerably greater than the historical perioperative blood loss, no further measures are necessary and the surgery can be performed safely, even in the absence of RBCs. However, if the mABL is similar to or smaller than the expected perioperative blood loss, additional measures are required.
As previously described, blood conservation strategies can be used for reasons including patient refusal of blood products or lack of blood availability, but the end goal is the same: to avoid allogeneic RBC transfusion. The preoperative assessment of these patients is particularly important and aims at optimizing Hgb concentration at the time of surgery and planning for blood conservation measures to be employed intraoperatively.
Ideally, a baseline Hgb should be measured 4 weeks in advance of all elective procedures aside from minor surgery with low risk of blood loss. Anemia has been found to be an independent risk factor for morbidity, mortality, longer hospital stay, and reduced quality of life and should be addressed for all patients. Shander et al. proposed an algorithm for evaluation and management of anemia (see Fig. 10.1 ). For anemic patients, Hgb should be increased by preoperative iron therapy, vitamin B 12 , folic acid, or erythropoietin if time allows ( Box 10.2 ), with the goal of therapy being to increase Hgb levels by 1 g/dL each week. This reduces the need for allogeneic blood transfusion, allows preoperative autologous blood donation (PABD) despite anemia, and increases the amount of blood collected by intraoperative isovolemic hemodilution. Erythropoietin and iron therapy have been used successfully in individual cases where blood transfusion was refused or impossible.
Preoperative pharmacologic preparation
Erythropoietin (EPO), 150–300 IU/kg, six doses in 3 weeks (alternative: 600 IU/kg, three doses in 7–10 days)
Iron, 100–300 mg/day, intravenous or oral
Folic acid, 5 mg/day
Vitamin B 12 , 15–30 μg/day
Preoperative autologous blood donation (PABD)
Intraoperative alternatives to blood transfusions
Acute normovolemic hemodilution
Cell salvage and retransfusion techniques (Cell-Saver)
Maintained normovolemia with crystalloids ± colloids
Hyperoxic ventilation (inspired oxygen 100%)
Deep anesthesia with muscle relaxation
Antifibrinolytic substances (aprotinin, ɛ-aminocaproic acid, tranexamic acid)
Coagulation factors (fresh frozen plasma; fibrinogen; factors II, VII, IX, and X)
Recombinant factor VIIa
Acceptance of minimal hemoglobin values
6 g/dL in healthy individuals
8 g/dL in aged or compromised patients
Adaptation of surgical procedure
PABD should be considered only if large blood losses are expected (> 2000 mL) and if the likelihood of transfusion exceeds 50%. In many cases, PABD is effectively hemodilution because only a fraction of the donated erythrocytes are regenerated preoperatively. To maximize the advantages of PABD, a longer interval before surgery or the use of erythropoietin and iron is required to allow regenerative erythropoiesis. If PABD is not an option, or if the patient does not respond to iron and/or erythropoietin, preoperative acute normovolemic hemodilution (ANH) and cell saving should be considered (see Fig. 10.1 ). ANH can be performed until the lowest acceptable Hgb is reached; most often, however, it is performed only to Hgb levels of 8 to 9 g/dL. Similar to calculating mABL, the Hgb that is mass harvested during ANH (HgM ANH ) can be calculated as follows :
where EBV is the patient’s estimated blood volume, Hgb i is the Hgb concentration prior to ANH, Hgb f is the Hgb concentration after ANH (target Hgb), and Hgb avg is the arithmetical average between the starting and final Hgb. After ANH, any further blood loss occurs from a lower starting Hgb, and thus mABL should be recalculated.
Accepting a lower minimal Hgb in otherwise healthy individuals will also help to avoid the need for transfusion, but this approach should be taken cautiously. Hyperoxic ventilation will increase DO 2 and has been shown to decrease myocardial ischemia and cognitive dysfunction due to ANH. However, the lower the Hgb, the more important it is to maintain normovolemia (see Box 10.2 ). Only at normovolemia are physiologic compensatory mechanisms maximally efficacious. Therefore, crystalloid replacement should be given at a 1:3 ratio and colloid given at 1:1. In addition, maintaining normothermia and avoiding hypothermia are of great importance in decreasing overall blood loss.
If preoperative treatment cannot augment RBC mass sufficiently, and despite use of ANH and cell-saving techniques the expected blood loss becomes greater than mABL, alternative treatment strategies, including less invasive and nonsurgical treatment modalities, must be considered (see Box 10.2 ).
Hemostasis is a complex mechanism involving the vessel wall, circulating cellular elements (particularly platelets), and circulating soluble factors such as coagulation factors. The characteristics of blood flow (laminar vs turbulent, flow velocity, pressure gradients, wall tension, and elasticity) define vascular bed specificity, which describes how different aspects of hemostasis can be relevant in different vascular beds. Hemostasis in general can be described as occurring in four phases:
Vasoconstriction: When the vessel wall is damaged, the vessel diameter reduces, thereby diminishing the size of the breach and bringing the circulating elements of coagulation into proximity of the endothelium.
Primary hemostasis: Circulating platelets adhere to subendothelial collagen via von llebrand factor (vWF) and the platelets’ receptor glycoprotein (GP) Ib. Activated platelets subsequently change from discoid to spherical; form pseudopods, exposing procoagulant receptors to form a primary hemostatic plug; and release the contents of their granules, secreting multiple factors that enhance further platelet activation and coagulation. Platelets’ GPIIb/IIIa receptors and negatively charged phospholipids are the anchors by which platelets adhere to one another.
Secondary hemostasis: The various coagulation factorsand cofactors interact on the platelets’ surfaces to form insoluble fibrin strands ( Fig. 10.2 ) that will mediate clot retraction and result in formation of a stable thrombus. The traditional division of this reaction into an extrinsic, an intrinsic, and a common pathway is a simplification of a complex series of interactions. The simplification is useful to illustrate the underlying mechanisms of common coagulation tests (e.g., prothrombin time [PT] and activated partial thromboplastin time [aPTT]). Several enzymes inhibit the coagulation cascade: activation of protein C, which is initiated by the coagulation cascade itself, tissue factor pathway inhibitor, and antithrombin.
Recanalization: After the endothelial continuity has been reestablished, the blood clot is broken down by the fibrinolytic system, which is catalyzed by plasmin, and blood flow is restored through the vessel. Plasminogen, the precursor of plasmin, circulates freely in plasma and is activated by tissue plasminogen activator and urokinase-type plasminogen activator, which are secreted by vascular cells.
Preoperative Risk Assessment
Strong emphasis is placed on preoperative evaluation to identify risk factors for requiring blood transfusion or adjuvant therapies. This evaluation should include review of medical records, patient or family interview, physical examination, review of existing laboratory results, and ordering additional laboratory tests when indicated. Multiple studies have demonstrated that routine preoperative hemostatic testing is of low clinical value, is not predictive of postoperative complications, and is not cost effective. Further, the surgical procedure itself does not constitute an indication for coagulation testing in both cardiac and noncardiac surgery.
Rather, laboratory testing should be ordered on an individual-patient basis (see “Medical History and Physical Examination” section). Patients identified as being at risk should then undergo further laboratory testing, which may include PT, aPTT, platelet count (PC), platelet function testing using a platelet function analyzer (PFA-100; Siemens Medical Solutions, Malvern, PA), and a functional vWF assay. This strategy of testing the at-risk subpopulation ( Fig. 10.3 ) was shown to reduce transfusion needs when applied in combination with a standardized therapeutic regimen and is consistent with the practice guidelines issued by the American Society of Anesthesiologists.
Medical History and Physical Examination
Clinical evaluation is one of the best tools to identify patients at risk for bleeding. The patient’s personal and family history can provide important clues to the presence and type of bleeding disorder. Several questionnaires have been published. The Koscielny questionnaire ( Box 10.3 ) was established retrospectively and later validated in a prospective study, each study including more than 5000 patients. On the basis of this questionnaire, 5021 (88.8%) of 5649 patients were identified as having a negative bleeding history, of which the only laboratory test abnormality found was a prolonged aPTT in 9 patients (0.2%) due to lupus anticoagulant. No other test identified a bleeding disorder in the patients with a negative bleeding history, highlighting the utility of clinical history.
Have you ever experienced strong nose bleeding without prior reason?
Did you ever have—without trauma—“blue spots” (hematoma) or “small bleeding” (at the torso or other unusual regions of the body)?
Did you ever have bleeding of the gums without apparent reason?
How often do you have bleeding or “blue spots” (hematoma): more than one or two times per week or just one or two times per week?
Do you have the impression that you have prolonged bleeding after minor wounds (e.g., razor cuts)?
Did you have prolonged or grave bleeding during or after operations (e.g., tonsillectomy, appendectomy, or during labor)?
Did you ever have prolonged or grave bleeding after a tooth extraction?
Did you ever receive blood packs or blood products during an operation? If so, please define the operation(s).
Is there a history of bleeding disorders in your family?
Do you take analgesic drugs or drugs against rheumatic disease? If so, please specify.
Do you take other drugs? If so, please specify.
Do you have the impression that you have prolonged menstruation (> 7 days) or a high frequency of tampon change? [to be answered only by women]
If a positive family history is present, the patient should be asked about the intensity and type of bleeding and the hereditary pattern. If a bleeding tendency is present, recent medication that might interfere with normal coagulation (e.g., non-steroidal anti-inflammatory drugs, aspirin) or measures that may have been taken to stop the bleeding (e.g., nasal tamponade, vitamin K), recent illnesses, and recent transfusion should all be asked about. Concomitant liver or renal disease, hypersplenism, hypothyroidism, excessive alcohol use, connective tissue disorders, malignancy, and hematologic diseases also influence hemostasis and should be investigated. Physical signs that can indicate pathologic states associated with increased bleeding include purpura, hematomas, jaundice, hepatomegaly, splenomegaly, and adenopathy.
The clinical presentation of bleeding diatheses varies, and different signs can be related to different disorders. The most frequent manifestations are cutaneous and mucosal bleeding. Oozing, hemarthrosis, and muscular hematomas are other types of bleeding that can be found in several bleeding disorders. The clinical presentations of the major bleeding disorders are listed in Table 10.2 .
|Von Willebrand disease|
|Hematomas||Single-factor congenital deficiency|
|Hemarthrosis||Hemophilia A and B|
|FII, FVII, FX deficiency|