Patient assessment of Bleeding Disorders
Patients undergoing elective hip or knee arthroplasty are at risk for bleeding and bleeding-associated complications. In the absence of reliable evidence, it is the opinion of this work group that patients be assessed for known bleeding disorders like hemophilia and for the presence of active liver disease which further increase the risk for bleeding and bleeding-associated complications.

Rationale
Bleeding complications related to the soft tissue envelope around the surgical site and the effects of bleeding on functional outcomes are an important concern. A hematoma can lead to joint stiffness and a compromised functional outcome or to a periprosthetic joint infection (with its associated morbidity). Although these potential risks have historically not been addressed in other guidelines on this topic, given the seriousness of these concerns, this work group believed it necessary to address them.

We found very little data that addressed risk factors for bleeding in patients undergoing elective hip or knee replacement surgery (see Table 24 for the list of risk factors for which we sought evidence and for a summary of these results. Two studies of very low quality (see Table 50 in Appendix XIII) addressed patients with hemophilia, with the only comparative study finding it to be a significant predictor of hemarthrosis. One comparative study of very low quality addressed cirrhosis of the liver and found it to be a significant predictor of perioperative blood loss (Table 25). 

Therefore, patients with a known bleeding disorder or active liver disease may have an increased risk for bleeding. Evaluating patients for these factors has minimal cost and low risk to the patient; we believe that these actions are consistent with the current practice of most orthopaedic surgeons. Recommendation 7 discusses the recommended thromboprophylaxis strategy for these patients.

Evidence about whether factors other than the presence of a known bleeding disorder or active liver disease affect the risk for bleeding in patients undergoing primary hip and knee arthroplasty is unclear. Six low quality studies among non-arthroplasty surgical patients did not find convincing evidence that preoperative coagulation screening predicts postoperative bleeding (Table 50 in Appendix XIII summarizes our evaluation of the quality and applicability of these studies).
  • bleeding time predicted blood loss in one of three studies
  • fibrinogen predicted blood loss in one of three studies
  • platelet count predicted blood loss in one of six studies
  • prothrombin time predicted blood loss in one of six studies (Table 27).
In other very low quality (and, therefore, unreliable) studies of non-arthroplasty surgical patients (Table 26):
  • thrombocytopenia was a significant predictor of postoperative intracranial hematoma among intracranial surgery patients,
  • a history of gastrointestinal (GI) bleed was not a significant predictor of postoperative upper GI bleeding among non-ulcer surgery patients,
  • a history of bleeding with previous surgery did predict excessive bleeding among cardiac bypass patients, while
  • epistaxis and a history of bleeding with dental extraction each did not predict major bleeding among Type 1 von Willebrand disease patients undergoing surgery.
No data were found addressing the other risk factors (see Table 24 for the list of risk factors for which we sought evidence).

The data on hemorrhage-related complications are also sparse. Three low quality and fourteen very low quality studies addressed whether patients with one or more potential risk factors have higher rates of hemorrhage-associated complications. (The results of these studies are summarized in Table 28, Table 29 and Table 30, which provide a detailed description of these studies’ results. Our evaluation of their quality and applicability is shown in Appendix XIII, Table 51)  Low hemoglobin levels and more complex revision procedures did predict a higher risk of transfusion, but none of the factors studied could be directly tied to hemorrhage-associated complications such as deep periprosthetic joint infection.

Due to the inconclusive evidence regarding other risk factors for bleeding or hemorrhage-associated complications among elective hip and knee arthroplasty patients, we are unable to recommend for or against further risk stratification.

The clinician should be aware of established contraindications against the use of individual anticoagulant agents.

We excluded some of the studies we retrieved to address this recommendation. These studies, and the reasons for their exclusion are listed in Appendix XIV, Table 59- Table 60.