Choosing Wisely

Transfusion of the smallest effective dose of red blood cells (RBCs) is recommended because liberal transfusion strategies do not improve outcomes when compared to restrictive strategies. Unnecessary transfusion generates costs and exposes patients to potential adverse effects without any likelihood of benefit. Clinicians are urged to avoid the routine administration of 2 units of RBCs if 1 unit is sufficient and to use appropriate weight-based dosing of RBCs in children.

Thrombophilia testing is costly and can result in harm to patients if the duration of anticoagulation is inappropriately prolonged or if patients are incorrectly labeled as thrombophilic. Thrombophilia testing does not change the management of venous thromboembolisms (VTEs) occurring in the setting of major transient VTE risk factors. When VTE occurs in the setting of pregnancy or hormonal therapy, or when there is a strong family history plus a major transient risk factor, the role of thrombophilia testing is complex and patients and clinicians are advised to seek guidance from an expert in VTE.

Inferior vena cava (IVC) filters are costly, can cause harm, and do not have a strong evidentiary basis. The main indication for IVC filters is patients with acute venous thromboembolism (VTE) and a contraindication to anticoagulation, such as active bleeding or a high risk of anticoagulant-associated bleeding. Lesser indications that may be reasonable in some cases include patients experiencing pulmonary embolism (PE) despite appropriate, therapeutic anticoagulation, or patients with massive PE and poor cardiopulmonary reserve. Retrievable filters are recommended over permanent filters with removal of the filter when the risk for PE has resolved and/or when anticoagulation can be safely resumed.

Blood products can cause serious harm to patients, are costly, and are rarely indicated in the reversal of vitamin K antagonists. In non-emergent situations, elevations in the international normalized ratio are best addressed by holding the vitamin K antagonist and/or by administering vitamin K.

Computed tomography (CT) surveillance in asymptomatic patients in remission from aggressive non-Hodgkin lymphoma may be harmful through a small but cumulative risk of radiation-induced malignancy. It is also costly and has not been demonstrated to improve survival. Physicians are encouraged to carefully weigh the anticipated benefits of post-treatment CT scans against the potential harm of radiation exposure.

Due to a decreasing probability of relapse with the passage of time and a lack of proven benefit, CT scans in asymptomatic patients more than two years beyond the completion of treatment are rarely advisable.

Anticoagulation is potentially harmful and costly. Patients with a first venous thromboembolism (VTE) triggered by a major, transient risk factor such as surgery, trauma, or an intravascular catheter are at low risk for recurrence once the risk factor has resolved and an adequate treatment regimen with anticoagulation has been completed. Evidence-based and consensus guidelines recommend three months of anticoagulation over shorter or longer periods of anticoagulation in patients with VTE in the setting of a reversible provoking factor. By ensuring a patient receives an appropriate regimen of anticoagulation, clinicians may avoid unnecessary harm, reduce health care expenses, and improve quality of life.

This Choosing Wisely® recommendation is not intended to apply to VTE associated with non-major risk factors (e.g., hormonal therapy, pregnancy, travel-associated immobility, etc.), as the risk of recurrent VTE in these groups is either intermediate or poorly defined.

Patients with sickle cell disease (SCD) are especially vulnerable to potential harms from unnecessary red blood cell transfusion. In particular, they experience an increased risk of alloimmunization to minor blood group antigens and a high risk of iron overload from repeated transfusions.

Patients with the most severe genotypes of SCD with baseline hemoglobin (Hb) values in the 7-10 g/dl range can usually tolerate further temporary reductions in Hb without developing symptoms of anemia. Many patients with SCD receive intravenous fluids to improve hydration when hospitalized for management of pain crisis, which may contribute to a decrease in Hb by 1-2 g/dL. Routine administration of red cells in this setting should be avoided. Moreover, there is no evidence that transfusion reduces pain due to vaso-occlusive crises.

In patients with asymptomatic, early-stage chronic lymphocytic leukemia (CLL), baseline and routine surveillance computed tomography (CT) scans do not improve survival and are not necessary to stage or prognosticate patients. CT scans expose patients to small doses of radiation, can detect incidental findings that are not clinically relevant but lead to further investigations, and are costly. For asymptomatic patients with early-stage CLL, clinical staging and blood monitoring is recommended over CT scans.

In patients with suspected heparin-induced thrombocytopenia (HIT), use the "4T’s" score to calculate the pre-test probability of HIT. This scoring system uses the timing and degree of thrombocytopenia, the presence or absence of thrombosis, and the existence of other causes of thrombocytopenia to assess the pre-test probability of HIT. HIT can be excluded by a low pre-test probability score (4T’s score of 0-3) without the need for laboratory investigation.

Do not discontinue heparin or start a non-heparin anticoagulant in these low-risk patients because presumptive treatment often involves an increased risk of bleeding, and because alternative anticoagulants are costly.

Treatment for immune thrombocytopenic purpura (ITP) should be aimed at treating and preventing bleeding episodes and improving quality of life. Unnecessary treatment exposes patients to potentially serious treatment side effects and can be costly, with little expectation of clinical benefit. The decision to treat ITP should be based on an individual patient’s symptoms, bleeding risk (as determined by prior bleeding episodes, and risk factors for bleeding, such as use of anticoagulants, advanced age, high-risk activities, etc.), social factors (distance from the hospital/travel concerns), side effects of possible treatments, upcoming procedures, and patient preferences.

In the pediatric setting, treatment is usually not indicated in the absence of mucosal bleeding regardless of platelet count. In the adult setting, treatment may be indicated in the absence of bleeding if the platelet count is very low. However, ITP treatment is rarely indicated in adult patients with platelet counts greater than 30,000/microL unless they are preparing for surgery or an invasive procedure, or have a significant additional risk factor for bleeding.

In patients preparing for surgery or other invasive procedures, short-term treatment may be indicated to increase the platelet count prior to the planned intervention and during the immediate post-operative period.

Preoperative hemostatic screening in healthy pediatric patients with no personal or family history of excessive bleeding does not effectively identify those who will have unexpected surgical bleeding. Artifacts or disorders that do not affect bleeding risk may be identified, such as factor XII deficiency or an infection-associated, transient lupus anticoagulant. Hemostatic testing adds cost and may introduce additional stress, either due to blood sampling or if a child has “abnormal” results.   

In asymptomatic (i.e, non-bleeding) pediatric patients with hypoproductive thrombocytopenia conditions such as bone marrow failure or malignancy with a platelet count > 10,000/mcL, transfusion is not clinically indicated unless signs, symptoms, or increased risk factors of bleeding are present. This practice is consistent with recommendations from the clinical guidelines of multiple associations (National Institute for Health and Care Excellence, British Society for Hematology, American Society of Clinical Oncology, and American Society of Hematology). The risk of spontaneous bleeding is low at platelet counts > 10,000/mcL. Unnecessary transfusions put patients at risk for transfusion reactions, alloimmunization, blood borne infections, and refractoriness to future platelet transfusions. This recommendation does not apply in anticipation of an invasive procedure.

Testing for inherited forms of thrombophilia does not influence the initial management of a first episode of provoked venous thrombosis and should not be performed routinely. The results of such testing have not been shown to either predict recurrence of venous thrombosis or inform the intensity or duration of anticoagulant therapy. Thrombophilia testing has substantial financial cost, and a positive result has the potential for misinterpretation of risk assessment leading to undue psychological distress or impact on childbearing plans, as well as possible life insurance discrimination for affected patients.

In pediatric patients with asymptomatic, iron deficiency anemia, do not transfuse packed red blood cells (pRBC) in the absence of hemodynamic instability or active bleeding. Unnecessary pRBC transfusions put patients at risk for complications, such as transfusion reactions, blood borne infections and volume overload. The judicious use of pRBCs transfusions would also be associated with cost savings for healthcare systems.

In pediatric patients with asymptomatic autoimmune neutropenia, there is insufficient evidence to support the routine use of granulocyte colony stimulating factor (G-CSF) as a prophylaxis strategy to improve health outcomes. The use of G-CSF in this population should be guided by clinical evaluation. The unnecessary routine use of G-CSF could lead to intolerable side effects, such as bone pain, as well as avoidable healthcare cost.

While deep vein thrombosis (DVT) and pulmonary embolism (PE) are relatively common clinically, they are rare in the absence of elevated blood D-Dimer levels and certain specific risk factors. Imaging, particularly computed tomography (CT) pulmonary angiography, is a rapid, accurate, and widely available test, but has limited value in patients who are very unlikely, based on serum and clinical criteria, to have significant value. Imaging is helpful to confirm or exclude PE only for such patients, not for patients with low pre-test probability of PE.

Source: American College of Radiology (ACR). Wording reflects that of the ACR recommendation; other societies have similar recommendations and some explicitly recommended D-Dimer testing prior to imaging.

There is no indication to order these tests, and there is no benefit to be derived in obtaining them in someone that does not have any history of bleeding or abnormal clotting and in the absence of any family history. This testing is not a part of the infertility workup. Furthermore, the testing is costly, and there are risks associated with the proposed treatments, which would also not be indicated in this routine population.

Source: American Society for Reproductive Medicine.

Hospitalized patients frequently have considerable volumes of blood drawn (phlebotomy) for diagnostic testing during short periods of time. Phlebotomy is highly associated with changes in hemoglobin and hematocrit levels for patients and can contribute to anemia. This anemia, in turn, may have significant consequences, especially for patients with cardiorespiratory diseases. Additionally, reducing the frequency of daily unnecessary phlebotomy can result in significant cost savings for hospitals.

Source: Society for Hospital Medicine – Adult Hospital Medicine. Wording reflects that of the Adult Hospital Medicine recommendation; other societies have similar recommendations.

Blood transfusion has become a routine medical response despite cheaper and safer alternatives in some settings. Pre-operative patients with iron deficiency and patients with chronic iron deficiency without hemodynamic instability (even with low hemoglobin levels) should be given oral and/or intravenous iron.

Source: American Association of Blood Banks (AABB).

Positron emission tomography (PET) and PET-CT are used to diagnose, stage, and monitor how well treatment is working. Available evidence from clinical studies suggests that using these tests to monitor for recurrence does not improve outcomes and therefore generally is not recommended for this purpose. False positive tests can lead to unnecessary and invasive procedures, overtreatment, unnecessary radiation exposure, and incorrect diagnoses.

Until high-level evidence demonstrates that routine surveillance with PET or PET-CT scans helps prolong life or promote well-being after treatment for a specific type of cancer, this practice should not be done.

Source: American Society of Clinical Oncology