Effects of Joint Blood Volume and Hemostasis Correction on Vascular Changes in the Joint after Induced Hemarthrosis in Hypocoagulable Mice

Introduction:Hemophilia causes recurrent bleeds into 'target joints' despite clotting factor replacement. Vascular remodeling associated with hemarthrosis is thought to contribute to vessel "leakiness" and perpetuated bleeding, but the underlying mechanisms remain unknown. Here, we studied the effects of joint blood volume and timing of hemostasis correction on vascular changes after induced hemarthrosis in FVIII-deficient mice, and in BALB/c mice treated temporarily with warfarin and anti-FVIII to create reversible hemostasis suppression.

Methods: Mice were injured by sub-patellar puncture of the right knee to induce hemarthrosis. FVIII-deficient mice were treated with two doses of recombinant human FVIII (rhFVIII) or saline with an 8-12 hour interval. The first dose of rhFVIII was given i) 2 hours before injury (prophylaxis), ii) 2 hours after injury (early), iii) 8 hours after injury (intermediate), or iv) 16 hours after injury (late). BALB/c mice were administered 10 µg/ml warfarin in drinking water for 1 week plus 0.25 mg/kg anti-FVIII 2 hours prior to injury. This dose of warfarin yielded a ~4-fold increase in prothrombin time that was reversed with 100 IU/kg 4-Factor prothrombin complex concentrate on day 2 post-injury in one group of mice, and maintained in another. Clearance of anti-FVIII within 4 days was indicated by normalization of blood loss in tail clip experiments. On day 2 after injury, the hematocrit was determined as a measure of joint bleeding and incapacitance was monitored by left:right weight bearing. Vascularity was assessed at baseline and 1-2 weeks post-injury by high resolution musculoskeletal ultrasound with Power Doppler (MSKUS/PD) and histology with Safranin-O-Fast Green staining. Statistical analyses were conducted by Mann-Whitney test.

Results: In FVIII-deficient mice, knee injury significantly reduced the mean hematocrit from 47 % to ~25 % for mice receiving late or no rhFVIII treatment (p<0.001). Hematocrit remained at 47 % with prophylaxis, and decreased to 40 % (p=0.01) and 37 % (p<0.0001) with early and intermediate rhFVIII treatment, respectively. In the saline-only group, hemarthrosis caused a 3.1-fold increase in left:right weight bearing of the hind legs (p=0.008). Earlier treatments with rhFVIII alleviated this incapacitance and mice receiving rhFVIII prophylaxis had no deficit. In contrast, vascular changes were similar across treatment groups, regardless of hematoma size. Mean vessel number increased ~3-fold compared to baseline for all groups (p<0.05). This was accompanied by moderate increases in mean vessel diameter and percentage of vessels with a diameter ≥ 20 µm. Vascular perfusion measured by MSKUS/PD increased significantly (~2.5-fold) with early (p=0.01), intermediate, late or no rhFVIII treatment (p<0.001), but not with prophylaxis.

Bleeding tendency and weight bearing deficits of the injured leg in BALB/c mice treated with warfarin and anti-FVIII were comparable to those observed in FVIII-deficient mice. Mean hematocrit decreased to 28 % (p<0.0001) and left:right weight bearing ratio increased to 2.6 (p=0.003). Mean vessel number in injured knees increased to a similar extent in the warfarin continuation and warfarin reversal groups (~1.5-fold; p=0.02). In contrast, mean vessel diameter and percentage of vessels with a diameter ≥ 20 µm were elevated only in mice continued on warfarin (p=0.008 and p=0.02, respectively).

Conclusions: FVIII prophylaxis or early treatment reduced blood loss into joints of FVIII-deficient mice and preserved joint function. However, neovascularization and vascular changes consistent with remodeling were not proportional to joint blood volume and were mostly unresponsive to short-term FVIII replacement, regardless of timing. Similarly, neovascularization in joints of hemostatically compromised BALB/c mice after induced hemarthrosis persisted despite normalization of hemostasis. Changes in vessel architecture, however, could be abrogated. This suggests that long-term hemostasis correction may protect against irreversible vascular remodeling in the joint, thought to underlie re-bleeding and development of hemophilic arthropathy. Further investigations are required to determine the effects of prolonged hemostasis correction on re-bleeding tendencies and joint health in hemophilia.