In the general patient population, 10% receive a transfusion during their inpatient stay.^1,2 Among obstetric patients, fewer than 2% of patients require transfusion from birth through 6 weeks’ postpartum.³ Postpartum hemorrhage (PPH), defined as blood loss of ≥750 mL after cesarean delivery or ≥500 mL after vaginal birth,⁴ may complicate 3% to 7% of deliveries and accounts for nearly 20% of all in-hospital deaths following delivery.^5–7 The rate of severe PPH (defined as PPH accompanied by transfusion, hysterectomy, or surgical repair of the uterus) per 1000 deliveries has increased between 2001 and 2008, from 1.9 to 4.2.⁷ In addition to red blood cell (RBC) transfusion, obstetric transfusion may involve transfusion of platelets, plasma, and cryoprecipitate (CRYO).⁸ Massive transfusion protocols (MTPs) are increasingly used in the obstetric hemorrhage setting.^9–13 This chapter will therefore familiarize the OB/GYN hospitalist with pretransfusion testing as well as principles of blood component therapy in the obstetrics patient.

TRANSFUSION SAFETY PRACTICES

Specimen Labeling and Type Checks

Improper identification of the patient or mislabeling of the tube results in wrong blood in tube (WBIT) errors, which are detected at rates of 1.2 to 17 per 1000 specimens.^14,15 WBIT errors, the rates of which rise during urgent transfusion scenarios, place patients at risk for potentially fatal ABO-incompatible transfusion events.^16,17 Type checks enhance patient safety by providing a second opportunity to verify the patient’s ABO type prior to transfusion of non–Group O red cells.¹⁸ Type checks are triggered when non–Group O patients have never previously had their blood group confirmed. Type checks are also performed to enable electronic crossmatch.

RISKS OF TRANSFUSION

Infectious and Noninfectious Transfusion Risks

Blood component therapy has witnessed an era of increasing safety.¹⁹ Infectious risks are quite low,^20–23 and far surpassed by noninfectious risks^24–31 (see Table 21-1). Zika virus, an arbovirus in the flaviviridae family³² causes infections with mild or asymptomatic disease in affected individuals but bears serious implications to the fetus.³³ Centers are currently complying with a recent US Food and Drug Administration (FDA) guidance³⁴ on Zika by testing of donor samples using one of two nucleic acid amplification-based ZIKV testing platforms: Procleix Zika Virus Assay—codeveloped by Hologic (Hologic, Inc, Marlborough, MA) and Grifols (Grifols Biologics Inc., Barcelona, Spain), and the Cobas Zika Test (Roche Diagnostics, Basel, Switzerland). Units testing positive are discarded and donors deferred for 120 days following a positive test or resolution of symptoms, whichever is longer. Pathogen reduction technology (PRT), more widely implemented in other regions, reduces the infectivity of contaminating pathogens, including ZIKV, within blood products and is currently only available for platelets and plasma. PRT is now gaining momentum in the United States.³⁵

PRACTICE POINTS

• 
  

  WBIT errors may result in fatal mistransfusion events.

  
  
• 
  

  Transfusion complications are more likely to be noninfectious than infectious.

  
  
• 
  

  Knowledge of these risks is useful when obtaining informed consent for transfusion.

PATIENT-LEVEL TESTING

Laboratory testing is prompted by family history or signs and symptoms of bleeding disorders or anemia, ongoing anticoagulant or antiplatelet therapy, or signs of active bleeding. Initial testing may include complete blood count (CBC), standard tests of coagulation (prothrombin time—PT, international normalized ratio—INR, partial thromboplastin time—PTT), and fibrinogen level. Other more specialized testing also may be ordered as indicated.

STANDARD COAGULATION TESTS

The PT, later incorporating the INR, was originally intended for monitoring of Vitamin K antagonist therapy,³⁶ while the PTT was originally intended for diagnosis of patients with signs of hemophilia.³⁷ These tests have become widely used as unintended surrogates for other outcomes, including severity of liver disease (Child-Pugh Turcot score),³⁸ survival in liver disease (Model for End-stage Liver Disease score),³⁹ and “prediction” of perioperative bleeding risk.

The tempting conclusion to make, that normal results predict zero risk of excess bleeding and that prolonged results absolutely predict bleeding (and therefore must be corrected), is not scientifically supported.^40,41 Laboratory orders therefore should be driven by clinical risk assessment. Testing based upon viscoelastic methodologies (TEG-analyzer, Haemonetics Corp, Braintree, MA; and ROTEM, TEM International, Munich, Germany) is gaining interest, but additional work needs to be done prior to widespread application in the obstetrics arena.^42,43 The inclusion of platelet count and fibrinogen in the assessment of bleeding patients is particularly crucial, as derangements are predictive for blood utilization.⁴⁴ See Table 21-2^45–60 for additional considerations.

PRACTICE POINTS

• 
  

  Coagulation abnormalities, including thrombocytopenia, may result from pregnancy-specific conditions such as HELLP syndrome, acute fatty liver of pregnancy, and gestational thrombocytopenia as well as nonpregnancy specific conditions such as sepsis-mediated disseminated intravascular coagulation (DIC) or idiopathic thrombocytopenic purpura (ITP).

  
  
• 
  

  Plasma would only rarely be indicated in the setting of isolated prolongation of the PTT.

BLOOD GROUP ANTIGENS AND ANTIBODIES

As a general concept, patients are at risk of forming antibodies following exposure (via pregnancy, transfusion, or transplantation) to antigens that they do not themselves express. The ABO blood group antigen system is the exception to this, as it is characterized by naturally occurring antibodies directed against nonexpressed A or B antigens. Non-ABO antigens are arranged into blood group systems. Commonly recognized antigens within selected blood group systems include, for Rh—D, E/e, and C/c; Kell—K/k; Duffy—Fy^a and Fy^b; Kidd—Jk^a and Jk^b; and MNS—M, N, S/s, and U. Clinically relevant alloantibodies are those known to cause hemolytic transfusion reactions or hemolytic disease of the fetus and newborn (HDFN).

The antibody screen assesses for the presence of alloantibodies, and the identification panel determines their specificities.⁶¹ Antibodies nonreactive at 37°C and those not traditionally considered clinically significant (i.e. anti-M, anti-N, anti-P1, anti-Le^a, or anti-Le^b) do not necessitate the selection of antigen negative units for crossmatching.⁶¹ In addition to clinically relevant alloantibodies, ABO antibodies may cause HDFN, particularly when the mother is Group O and the fetus is non-O.⁶²

ABORh, Type-and-Screen, Type-and-Crossmatch

Transfusion-level testing permits the selection of compatible blood components for transfusion. For red cell transfusions, the standard tests include (1) ABO-Rh, or type; (2) ABO-Rh and antibody screen, or type-and-screen; (3) ABO Rh, antibody screen and crossmatch, or type-and-cross(match); and (4) ABO Rh, antibody screen and hold (units), or type and hold.

Antibody screen and identification tests represent the rate-limiting step to red cell transfusion. Even if not currently detectable, historically detected antibodies are honored (meaning that red cell units selected for crossmatch are negative for the corresponding antigen). If antigen-negative units are not immediately available, procurement from area vendors becomes necessary.

For circumstances where standard pretransfusion testing would introduce life-threatening delays to transfusion, institutions have in place emergency release orders. A waiver attesting to emergency need, signed by the ordering physician, allows a blood bank to forego standard pretransfusion testing. If a previous type has been confirmed and there is knowledge of preexisting antibodies, then uncrossmatched type-specific units known to be negative for the corresponding antigens may be selected if time permits. If not, then uncrossmatched Group O, Rh-negative units will be chosen.

To assess for compatibility, the crossmatch tests patient serum against red cells from units intended for transfusion. The ordering clinician specifies how many units to crossmatch—for example: “T&C 4 units.” If required, additional units may be crossmatched against any remaining specimen, provided it was properly labeled and has not yet expired. Crossmatched units are labeled for the patient and removed from general inventory until transfusion or expiration of the crossmatch specimen (72 hours). The Type-and-Hold order is a variation in which—following type and screen—a physician-specified number of units are allocated (but not crossmatched) for the patient in anticipation for a procedure or transfusion.

Red Blood Cell Transfusion

In otherwise stable patients, red cells are administered in 1-unit increments for treatment of symptomatic anemia. For such patients, restrictive hemoglobin triggers in the range of 7 to 8 g/dL and incorporation of symptom assessment into transfusion decisions are supported by the literature⁶³ and have been found to significantly reduce transfusion exposure, with no impact on 30-day mortality or morbidity.⁶⁴

In obstetrics, iron supplementation (enteral or parenteral) may reduce the need for transfusion. There is growing experience with the use of intravenous (IV) iron formulations during pregnancy, including Iron Sucrose—Pregnancy Category B (Venofer, American Regent, Shirley, NY) and Low-Molecular-Weight Iron Dextran—Pregnancy Category C (various manufacturers).^65–67 A recent Cochrane analysis concluded that parenteral iron produces better hematological responses than enteral supplementation, but further study is needed to better understand safety and outcomes.⁶⁸ In other perioperative settings, use of IV iron, and in some cases augmentation with erythropoietin, has been found to effectively reduce transfusion risk.⁶⁹

PRACTICE POINTS

• 
  

  Review prior transfusion history, anemia symptoms, and results of CBC evaluation, and, if indicated, type and screen.

  
  
• 
  

  Coordinate with the blood bank to ensure timely availability of blood, particularly for patients with red cell antibodies.

  
  
• 
  

  Discuss alternatives to red cell transfusion with patients if clinically appropriate.

Hospital transfusion committees set forth evidence-based guidelines for transfusion that incorporate local and regional practices. Such guidelines typically include thresholds for transfusions of plasma (i.e. INR > 1.7 with bleeding), platelets (i.e. platelet count < 20 K/mcL in hospitalized cancer patients or < 50 K/mcL prior to most surgical procedures or < 100 K/mcL for neurosurgery), and CRYO (i.e. bleeding with fibrinogen level < 110 mg/dL). While most centers currently issue plateletpheresis (aka apheresis platelet units), a small proportion of centers continue to supply whole blood–derived platelet concentrates.

PLATELETS

Platelets are transfused for patients with quantitative (i.e. thrombocytopenia) or qualitative abnormalities in platelet function. A single whole blood–derived platelet concentrate (PC) must contain a minimum of 5.5 × 10¹⁰ platelets.⁷⁰ Typically, four to six individual PCs are pooled to arrive at a standard adult platelet dose, hence engendering the common platelet ordering parlance (e.g. “a six-pack of platelets”). A single apheresis platelet unit, on the other hand, must contain a minimum of 3.0 × 10¹¹ platelets,⁷⁰ hence reflecting an essentially equivalent number of platelets as a pool of six individual PCs. The actual numbers of platelets contained within platelet products—whether whole blood–derived or apheresis-collected—are higher than the quality-control-defined minimum values. At centers dealing in apheresis platelets, therefore, the standard adult dose is 1 unit per transfusion. There is no clinical outcome difference between the use of either product. Platelet transfusion in the setting of ITP is reserved only for critical site bleeding (i.e. intracranial); primary therapies for ITP instead include corticosteroids and IV immune globulin. Patients refractory to these options benefit from hematology consultation.

PLASMA

Plasma is dosed at 15 to 20 mL/kg, with most whole blood–derived plasma components having a volume of around 300 mL.⁷¹ A 20-mL/kg dose in a 70-kg patient would represent 1400 mL of plasma, or about 5 units. Many patients with stable or mild prolongations in PT/INR may still retain adequate hemostatic integrity.^72,73 A randomized trial comparing a single preprocedural dose of 12 mL/kg fresh frozen plasma (FFP) to no FFP in critically ill patients with INR 1.5 to 3.0 undergoing placement of central venous catheters, percutaneous tracheotomy, chest tube, or abscess drainage demonstrated no difference in bleeding between groups.⁷⁴ In a post-hoc analysis of specialized coagulation data from this study, it was concluded that although factor levels were increased, the effects of plasma transfusion on thrombin generation and viscoelastometry were very limited and failed to augment the procoagulant state.⁷⁵ Refer to Table 21-2 for further considerations regarding prolongation of standard coagulation values. Recall that severe reductions of common-pathway factors—especially fibrinogen—may also contribute to prolongation of the PT (INR) and PTT.

In the setting of obstetric hemorrhage, massive transfusion protocols incorporate fixed-ratio transfusion red cells, plasma, and platelets.^76,77 Adequate use of transfusion therapy in the setting of postpartum hemorrhage is crucial for the provision of high-quality care.^78,79 The key point with plasma, therefore, is to simultaneously avoid both overtransfusion of nonbleeding patients and undertransfusion of massively bleeding patients.

Alternatives to Plasma

Prothrombin complex concentrates (PCCs) are pathogen-inactivated factor concentrates in lyophilized form that are resuspended in sterile water at the time of administration. PCCs are categorized as 3-factor (e.g. Profilnine, Grifols Biologicals, Los Angeles, CA) or 4-factor concentrates (e.g. KCENTRA, CSL Behring, Marburg, Germany) based upon relative factor VII content. PCCs are sometimes used in an off-label manner for perioperative bleeding⁸⁰ and may be especially valuable in extremely volume-sensitive patients. Doses of 4-factor concentrates in the range of 25 to 50 FIX U/kg provide high concentrations of vitamin K–dependent factors (II, VII, IX, X) and result in more rapid correction of the INR than plasma, with reduced risk of volume overload.^81–83 Use of these agents require further study in the setting of obstetric hemorrhage.⁸⁴ While concentrates are available outside the United States for Factor XI deficiency, FXI deficiency is treated with plasma within its borders.⁵³

Cryoprecipitate and Its Alternatives

The key procoagulant factors contained in CRYO have gradually been made available either as pathogen-reduced factor concentrates (derived from pooled plasma) or recombinant, single-factor preparations. These preparations are used for prophylaxis or bleeding due to deficiencies of fibrinogen (RiaSTAP, CSL Behring, Marburg, Germany), von Willebrand factor (various products), factor VIII (various products), or FXIII (Corifact, CSL Behring, Marburg, Germany).⁸⁵ The use of CRYO is now mainly confined to the restoration of fibrinogen levels in bleeding patients.⁸⁶ Historically, CRYO was supplied as single units (each derived from a single unit of FFP) to be pooled at the time of transfusion. Currently, many centers are supplying prepooled CRYO in 5-unit increments.

Dosing is not standardized; rather, it ranges from 10 to 20 units per adult dose (or, roughly 1 unit of CRYO for every 5 kg of body weight).⁸⁷ Alternatively, CRYO doses can be calculated.⁸⁸ This is accomplished by first estimating the total plasma volume (in dL), then estimating the patient’s mg fibrinogen deficit, and finally calculating the number of CRYO units to administer as follows: (1) TPV in dL = (Weight in kg × 70 mL/kg) × (1-Hct) × (1dL/100 mL); (2) Fibrinogen Deficit (in mg) = (Desired Fibrinogen in mg/dL—Current Fibrinogen in mg/dL) × TPV in dL; (3) CRYO Units to Administer = Fibrinogen Deficit in mg/250 mg per CRYO Unit. If the local center deals in prepooled CRYO, then the calculated dose is rounded up or down to the nearest whole increment.

Practice Points

• 
  

  Hospital transfusion guidelines set forth the criteria for blood product transfusion.

  
  
• 
  

  Plasma is used to address multiple factor deficiencies, whereas specific factor deficiencies may best be addressed with plasma derived from recombinant factor concentrates.