Total knee replacement with and without emicizumab: a unique comparison of perioperative management

Emicizumab is a bispecific monoclonal antibody that mimics the activity of factor VIII (FVIII) and is approved for the prevention of bleeding in persons with severe FVIII deficiency with or without inhibitors.^1,2  However, the reported experience with surgeries in persons on emicizumab is relatively limited to the published data from the Haven 1-4 trials, which include a total of 233 surgeries with 18 major surgeries in 18 patients and 215 minor surgeries among 115 patients.³  All 5 arthroplasties performed during these trials received a prophylactic factor with either FVIII concentrate or recombinant FVIIa (rFVIIa). More detailed and “real-world” experiences of arthroplasties in persons with FVIII deficiency and inhibitors while on emicizumab include an abstract by Santagostino et al⁴  who described a total hip arthroplasty in a 56-year-old man, a case report by Kizilocak et al⁵  of a 25-year-old man who had a total knee replacement, and a recent case report by Seaman and Ragni⁶  of the successful management of a man who underwent a total hip replacement.

We describe a 48-year-old man with severe FVIII deficiency, hemophilic arthropathy, and an inhibitor to FVIII with a titer of 33 Bethesda units who, following extended treatment of a prior joint infection, underwent reimplantation of an artificial right knee after having achieved a steady state on emicizumab. He had the same surgery on his left knee several years before emicizumab therapy was initiated. This coincidence provided us with a unique opportunity to compare differences in the perioperative management of an identical procedure on the same individual by the same surgeon.

Our patient had previously undergone removal of both his left and right knee components in April 2012 because of a methicillin-resistant Staphylococcus aureus infection acquired following a life-threatening episode of septic shock requiring hospitalization for 2 months in 2009. After a prolonged course of antibiotic therapy, he underwent left total knee reimplantation in September 2012. Based on his significant bleeding history and his requirements for rFVIIa for previous procedures, he received a 120 μg/kg bolus of rFVIIa before the first incision followed by a 90 μg/kg per hour continuous infusion (CI) and 3 additional boluses of 90 μg/kg intraoperatively. Estimated blood loss was 1480 mL. Postoperatively, he received rFVIIa 90 μg/kg per hour by CI on days 1 through 5, 60 μg/kg per hour on day 6, 50 μg/kg per hour on day 7, 35 μg/kg per hour on day 8, and 25 μg/kg per hour on day 9. He then received rFVIIa 90 μg/kg boluses every 3 hours on day 10, 180 μg/kg every 6 hours on days 11 and 12, and 180 μg/kg every 8 hours on days 13 and 14. During this time, he received an additional 3 rFVIIa 90 μg/kg bolus doses for ongoing oozing from his incision. His total rFVIIa requirement was ∼18 795 μg/kg. On postoperative day 2, his hemoglobin (Hb) had dropped by 6.7 g/dL from his preoperative Hb and he received 2 U of packed red blood cells (RBCs; pRBCs). On postoperative day 4, his Hb fell another 3 g/dL and he received 1 additional unit of pRBCs (Table 1).

In September 2016, he was started on emicizumab 1.5 mg/kg subcutaneously weekly following a loading dose of 3 mg/kg weekly for 4 weeks on the Haven-1 trial.

His first surgical procedure while on emicizumab was in January 2018 and involved the removal of a right femoral cephalomedullary antibiotic spacer that had been placed in March 2015 following the development of an infected cephalomedullary nail in his femur. The day of surgery, he received his usual weekly scheduled emicizumab prophylaxis plus 90 μg/kg rFVIIa prior to the first incision and 1 bolus of rFVIIa 90 μg/kg 2 hours later intraoperatively. Estimated blood loss during the procedure was <50 mL. He was then maintained on rFVIIa 50 μg/kg per hour by CI postoperatively days 1 to 4, rFVIIa 25 μg/kg per hour on day 5, and rFVIIa 90μg/kg every 3 hours on day 6. He was then discharged to continue with rFVIIa 90 μg/kg every 6 hours through post day 7. He had no bleeding complications following this procedure.

In December 2018, he underwent reimplantation of a right total knee. He received his usual weekly scheduled dose of emicizumab and continued on his weekly maintenance schedule perioperatively. The day of surgery, he received a 90 μg/kg bolus of rFVIIa prior to the first incision with 3 boluses of 90 μg/kg every 2 hours intraoperatively. Despite extensive manipulation in the operating room and a prolonged surgery time, hemostasis was rated as good to excellent and estimated blood loss was only 500 mL. Postoperatively, he was started on rFVIIa 50 μg/kg per hour by CI. On postoperative day 1, immediately after physical therapy, he developed bleeding into his right knee with rapid resolution of bleeding following 8 bolus doses of rFVIIa 90 μg/kg every 2 to 3 hours. He was maintained on rFVIIa 50 μg/kg per hour on postoperative days 2 to 5 and 25 μg/kg per hour on days 6 to 11. He then received rFVIIa 75 μg/kg boluses every 4 hours on day 12 and 75 μg/kg boluses every 6 hours on day 13, followed by 75 μg/kg rFVIIa prior to physical therapy (PT) sessions for 2 additional weeks with good recovery. His total rFVIIa requirement of this time frame was ∼10 170 μg/kg. Notably, on postoperative day 3, his Hb had dropped by 8.3 g/dL from his preoperative Hb. He required only 1 U of pRBCs with stabilization and improvement in his anemia thereafter while on emicizumab (Table 1). There were no thrombotic complications during his course based on clinical examination.

This case describes similar orthopedic surgeries by the same surgeon on the same patient before and after initiating emicizumab. Controlling for these variables provides a unique opportunity for a direct comparison of his rFVIIa usage and his perioperative bleeding with and without emicizumab. As anticipated, our patient required significantly less rFVIIa and RBC replacement for his knee surgery while on prophylactic treatment with emicizumab, consistent with overall better control of bleeding perioperatively. Focusing exclusively on the cost of treatment of his hemophilia during these 2 procedures, this was accompanied by a 45.8% reduction in rFVIIa usage for a total savings of $1.8 million based on the current average wholesale price of rFVIIa and emicizumab.

Of the arthroplasties reported to date, only 1 other knee replacement has been described while on emicizumab prophylaxis. That report describes requiring a grand total of ∼7421 μg/kg rFVIIa given as bolus dosing for 14 days postoperatively and then prior to PT sessions up through postoperative day 35.⁵  In our case, we describe safe use of CI with boluses of rFVIIa postoperatively and prior to PT through postoperative day 30 while on emicizumab. Our case differed in regards to a higher rFVIIa need of ∼10 170 μg/kg over a similar time frame when taking into consideration that our patient was older, had prior scar tissue related to prior knee replacements complicated by joint infections, and developed a postoperative bleed secondary to PT necessitating additional unplanned bolus dosing of rFVIIa.

Our report complements previous literature and provides further descriptive experience for hemophilia providers caring for similar patients receiving total knee replacements, consistent with the need for dissemination of such information per December 2018 recommendations from the Medical and Scientific Advisory Council (MASAC) of the National Hemophilia Foundation.⁷  As this case highlights, rFVIIa will still be needed perioperatively for those with inhibitors; however, frequency and dose are likely to be significantly less than without emicizumab. Although the ideal dosing regimens of rFVIIa have not yet been identified, the use of emicizumab in these patients is likely to result in significant cost savings with major surgical procedures such as joint replacements.

This work was supported by the Livingston Trout and Mellinger Medical Research Fund.

Contribution: M.S.E., C.D., and M.E.E. analyzed the data/experience, formulated the conclusions, and wrote and edited the paper.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Matthew S. Evans, Penn State Health Milton S. Hershey Medical Center, 500 University Dr, Hershey, PA 17033; e-mail: mevans1@pennstatehealth.psu.edu.