Haemophilia (2010), 16 (Suppl.
3), 13–18
REVIEW ARTICLE
Wound healing in haemophilia – breaking the vicious cycle
M. HOFFMAN* and D. M. MONROE 
*Pathology and Laboratory Medicine Service, Durham VA Medical Center, Durham, NC, USA; and  Division of
Hematology/Oncology, University of North Carolina, Chapel Hill, NC, USA
Summary. Our group has been studying how haemo-
stasis interacts with repair processes and also how to
optimize treatment of bleeding disorders in a mouse
model of haemophilia B. We have found that cuta-
neous wounds heal more slowly in haemophilic mice
than in wild-type mice, and also exhibit histological
abnormalities, even after closure of the skin defect.
The haemophilic wounds showed reduced influx of
inflammatory cells and increased angiogenesis. Even
after surface closure, the haemophilic animals expe-
rienced repeated episodes of re-bleeding and progres-
sive accumulation of iron in the wound bed and
deeper tissues. A dose of replacement or bypassing
therapy sufficient to establish initial haemostasis did
Normal wound healing
Wound healing has been divided into four overlap-
ping phases: (i) haemostasis; (ii) inflammation; (iii)
proliferation; and (iv) remodelling or resolution
(Fig. 1). The haemostatic or coagulation phase com-
mences as soon as vessels are damaged during
wounding. Platelets adhere to the extracellular
matrix and provide primary haemostasis as well as
releasing a wide range of growth factors and cyto-
kines as they undergo degranulation. Activated
platelets provide the surface on which the coagula-
tion enzymes are activated, leading to a burst of
thrombin generation and formation of a stable fibrin
clot. Thrombin and the fibrin clot formed during the
process of haemostasis both play important roles in
healing. Thrombin has direct cytokine and growth
factor-like activities that promote chemotaxis and
Correspondence: Maureane Hoffman, MD, PhD, Pathology and
Laboratory Medicine Service, Durham VA Medical Center, Dur-
ham, NC 27705, USA.
Tel.: +1 919 286 6925; fax: +1 919 286 6818;
e-mail: maureane@med.unc.edu
Accepted after revision 1 March 2010
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd
not normalize wound healing. In fact, daily dosing for
7 days was required to normalize wound closure.
Thus, normal healing requires adequate haemostatic
function for an extended period of time. We have
hypothesized that this is because angiogenesis during
healing predisposes to bleeding, especially in the
setting where haemostasis is impaired. Thus, normal-
izing haemostasis, until the process of angiogenesis
has resolved, may be required to prevent re-bleeding
and additional tissue damage.
Keywords: angiogenesis, haemophilia, inflammation,
iron, tissue factor, wound healing
proliferation of monocyte/macrophages, keratino-
cytes, fibroblasts and endothelial cells [1–4]. While
other factors can influence fibrin clot stability, the
rate of thrombin generation during haemostasis
impacts the effectiveness of healing through its effect
on fibrin structure [5]. Thrombin not only clots
fibrinogen, but also activates factor XIII and the
thrombin-activatable fibrinolysis inhibitor (TAFI) to
enhance clot stability. The fibrin clot itself aids
healing by providing the scaffold on which the cell
influx and tissue repair take place, and its degrad-
ation products also have chemotactic activity for
cells involved in the repair process [6]. Thus, the
haemostatic process not only stops blood loss, but
also delivers biologically active molecules to the
wound site and sets the stage for the subsequent
phases of healing.
The inflammatory phase begins with the influx of
neutrophils that starts within minutes after injury.
This is followed, during the first day after wounding,
by an influx of monocytes from the blood, which
rapidly mature into macrophages in the tissues.
Neutrophils and macrophages play an important
role in defence against infection, and macrophages
secrete a wide range of cytokines. However, a lack of
13
14 M. HOFFMAN and D. M. MONROE
neutrophils [7] or macrophages [8] does not prevent
wound healing in the absence of infection.
The proliferative phase of healing is characterized
by epithelial proliferation, angiogenesis and fibro-
blast proliferation. The fibrin clot and debris filling
the wound site are replaced by granulation tissue, and
the surface defect is closed by migration and prolif-
eration of keratinocytes. The Ôgranulation tissueÕ is a
highly cellular reparative tissue that fills the wound
site. It is very delicate and bleeds easily because it
contains large numbers of new vessels, but little
stabilizing connective tissue. During development of
granulation tissue, capillary sprouts invade the
wound clot from adjacent vessels and organize into
a microvascular network. Initially, the angiogenic
vessels are very delicate and leaky, but are progres-
sively stabilized by the surrounding stroma and
pericytes. At the same time, fibroblasts proliferate
within the granulation tissue and begin to lay down
collagen and other connective tissue components to
reinforce the tensile strength of the repaired tissue.
The remodelling phase can continue for weeks or
months; as proliferation stops, the inflammatory
infiltrate resolves, many of the vascular sprouts
regress and the structure of the wound site collagen
reorganizes. The bulging contour of the wound site
returns to normal, or even retracts, as the cells are
replaced by hypocellular connective tissue (scar).
Impaired healing in haemostatic defects
There are many theoretical reasons to believe that
impaired haemostasis could lead to impaired wound
healing, but there are limited experimental data.
Studies in rabbits showed that healing after tooth
extraction is delayed in anticoagulated animals [9].
Cutaneous wound healing is not delayed in knockout
mice deficient in fibrinogen [10] or TAFI [11],
although the healed wounds in these mice display
histological abnormalities. Thus, it appears that
formation and persistence of a suitable fibrin clot
play some role in healing, but that thrombin gener-
ation is likely to make a more significant contribution.
Interestingly, thrombocytopaenic mice do not have a
defect in closure of a cutaneous wound, although they
do show altered inflammatory cell influx [12].
Our group found that cutaneous wound healing is
impaired in a mouse model of haemophilia B (HB)
[13]. The haemophilic mice exhibit delayed cutane-
ous wound healing with abnormal histology, includ-
ing (i) subcutaneous haematoma formation; (ii)
delayed macrophage influx; (iii) delayed re-epitheli-
alization; and (iv) an unexpected increase in wound
site angiogenesis [13].
Haemophilia (2010), 16 (Suppl. 3), 13–18
Studies in rabbits that were depleted of plasma
fibrinogen, either before or after the formation of a
haemostatic clot, suggested that impairing fibrin
formation after initial haemostasis did not impair
wound healing at all [14–16]. By analogy, we
hypothesized that temporarily restoring the initial
haemostatic burst of thrombin generation in HB
mice would allow formation of a normal haemostatic
clot, and thereby normalize the subsequent phases of
wound healing. Instead, we found that temporarily
restoring thrombin generation with a single dose of
factor IX (FIX) replacement or activated factor VII
(FVIIa) bypassing therapy at the time of wounding
did not correct the delayed epithelial closure time
[17].
Restoring thrombin generation at the time of
wounding did enhance macrophage influx compared
with untreated HB mice. It appears that two major
mechanisms operate to promote macrophage influx
in response to a cutaneous wound. Thrombin plays
an important role in promoting rapid macrophage
influx in wild-type (WT) mice. By contrast, in HB
mice, who are able to produce little thrombin at the
wound site, macrophage influx is largely a response
to haemorrhage [13]. The pattern of macrophage
influx in the treated HB mice is a composite of the
pattern seen in WT and HB mice; there was an early
influx of macrophages related to thrombin genera-
tion during haemostasis and a late influx in response
to recurrent haemorrhage.
The pattern of macrophage influx was linked to the
clearance of iron from the sites of wounding. Mac-
rophages ingest and degrade red blood cells and
release iron from haemoglobin. Ferrous iron in
haemoglobin is not detected by tissue iron stain
(Prussian Blue). However, iron is converted to the
ferric oxidation state and complexed to ferritin after
being released from haem. Tissue iron staining in WT
animals reached peak intensity in the wound bed
4 days after wounding as the macrophages degraded
the red blood cells deposited there at the time of
wounding. Iron is cleared from the wound bed
(dermis) after 12 days. Iron begins to appear in the
deeper tissues (below the dermis) in WT mice 6 days
after wounding, as iron-laden macrophages carry it to
draining lymph nodes. Iron clearance from the deeper
tissues is completed by 16 days after wounding.
In contrast, untreated HB mice had a slow rise in the
intensity of iron staining within the wound bed,
peaking 10 days after wounding. The peak corre-
sponded to the delayed macrophage influx. Acceler-
ating the initial macrophage influx – by treatment with
FIX or FVIIa at the time of wounding – correlated with
an earlier onset of haem degradation by macrophages.
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd
 WOUND HE ALING IN HAEMOPHILIA 15

Both treated groups showed the appearance of iron

staining within the wound bed by day 4 after wound-
ing. However, as both treated groups experience re-
bleeding into the tissues during healing, iron accumu-
lates and persists in the wound beds and deeper tissues,
as it does in untreated HB mice.
While the rate of angiogenesis was similar in HB
and WT mice, the HB wounds had greater numbers
of vessel profiles than did their WT counterparts – a
difference we have provisionally attributed to the
pro-inflammatory and pro-angiogenic effects of tis-
sue iron. Restoring initial thrombin generation did
not normalize the angiogenic response in HB mice. In
fact, FIX treatment led to a doubling in the number
of vessels in the wound bed compared with all other
groups. This increased angiogenic response might be
due to the combination of enhanced thrombin
generation at the time of initial haemostasis, and
the effects of bleeding that develops after FIX has
been cleared. These results suggest the possibility
that inadequate haemostatic therapy might actually
increase the subsequent risk of bleeding by enhancing
angiogenesis.
A perplexing finding of our study was that early
haematomas developed at the site of biopsy wound
placement, but later, haematomas were not directly
adjacent to the wound site. Those late haematomas
were in a tissue plane that we have affectionately
referred to as the Ômacrophage highwayÕ. This tissue
plane is deep to the hair follicles around the thin
layer of subcutaneous skeletal muscle, as illustrated
in the upper panel of Fig. 2. This tissue plane is the
location of small arterioles, venules and lymphatic
vessels, the sites at which inflammatory cells leave
Fig. 2. These views are from healed wounds in haemophilia B
the blood to enter the injured tissues, and (in the case
mice. The top panel is from a wound placed 15 days earlier and
the bottom panel from a wound placed 11 days earlier. The top
panel indicates the location of the Ômacrophage highwayÕ, which is
the tissue plane just beneath the skeletal muscle. On the right side
of the top panel, an area in the Ômacrophage highwayÕ is expanded
by inflammation, granulation tissue and fibrosis. In the lower pa-
nel, a haematoma in the Ômacrophage highwayÕ is indicated, which
is not adjacent to the wound site.

of macrophages) leave the tissues on their way to

Fig. 1. The phases of wound healing. This schematic shows the
relative duration and primary cellular participants in the process of
wound healing.
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd
draining lymph nodes. The pre-existing vessels are
also sites from which new angiogenic vessels sprout
during healing. It appears that angiogenic vessels
within the Ômacrophage highwayÕ may be the source
of late bleeding. As shown in the lower panel of
Fig. 2, late haematomas are found adjacent to the
Ômacrophage highwayÕ, but a small distance away
from the healed wound site. Thus, we hypothesize
that the process of angiogenesis is a predisposing
factor for late bleeding in haemophilic subjects.
Haemophilia (2010), 16 (Suppl. 3), 13–18
16 M. HOFFMAN and D. M. MONROE
Delayed bleeding during healing
In the initial stages of angiogenesis, the mature vessels
that provide sites for new vessel sprouting exhibit
vasodilatation, increased permeability to plasma
proteins, detachment of supporting cells and loosen-
ing of the adjacent extravascular matrix [18]. The
endothelial sprouts migrate through the developing
granulation tissue, acquire lumens and fuse with
other sprouts to form functional capillary loops [18].
Several factors may predispose to bleeding during this
process, including destabilization of existing vessels
as the sprouts develop, the delicate state of the
neovessels and the activity of proteolytic enzymes
within the granulation tissue [19]. Even normal mice
show extravasation of red blood cells in areas of
angiogenesis. This predisposition to bleeding during
angiogenesis could partially explain the development
of haematomas in the vicinity of apparently uninjured
subcutaneous vessels in the HB mice.
However, at least one additional mechanism may
contribute to bleeding during angiogenesis – the
reduced expression of perivascular tissue factor (TF)
after wounding [20]. Tissue factor is normally
expressed around the outside of blood vessels
[21,22]. In vitro data suggest that products of the
coagulation cascade [23] and inflammatory media-
tors released following injury promote TF expression
[24]. Therefore, we expected that TF would be highly
expressed in the tissue near a wound site. Instead, we
found that TF disappears from around vessels near
wounds in both WT and HB mice the first day after
wounding [20]. The TF coat is not re-expressed until
8–10 days after wounding. The reappearance of TF
is delayed in HB mice compared with WT mice. The
angiogenic vessels within the granulation tissue also
do not express TF in either WT or HB mice during
healing.
We propose that a lack of TF around angiogenic
vessels may reflect a mechanism to prevent thrombo-
sis of newly formed, delicate and leaky vessels. This
leads to an increased risk of bleeding from granulation
tissue. Such a bleeding tendency is not a significant
liability in normal individuals. However, in haemo-
philia, the absence of TF near wound sites could lead
to repeated haemorrhage in and around the wound
site, even after the surface defect has been healed.
Does angiogenesis or inflammation lead to
bleeding in haemophilia?
Our data suggest that angiogenesis is a predisposing
factor for ongoing/recurrent bleeding. However, it is
very difficult to separate the effects of inflammation
Haemophilia (2010), 16 (Suppl. 3), 13–18
per se from the effects of angiogenesis, as the two
processes usually go hand-in-hand. An integral part
of the inflammatory response is the influx of leuco-
cytes from the blood. This process is driven by the
release of a variety of mediators from injured cells
and resident tissue macrophages, as well as the
generation of biologically active products of com-
plement and coagulation proteins. Inflammatory cell
influx involves the margination of leucocytes via
adhesion to endothelial cells near the injury site,
passage through the endothelial layer and proteolysis
of the basement membrane. It is easy to imagine that
chemotaxis of leucocytes into an area of inflamma-
tion may predispose to bleeding in a host with
impaired haemostatic function. This has, in fact,
been shown to be the case in thrombocytopaenic
mice, where either inflammation or angiogenesis
alone leads to bleeding [25,26]. Interestingly, throm-
bocytopaenic mice do not have a defect in closure of
a cutaneous wound, although they do show altered
inflammatory cell influx [12].
We have adapted a cantharidin model of inflam-
mation [27] for use in mice. It has previously been
used to study leucocyte trafficking and induces no
visible angiogenesis. Preliminary data using this
model suggest that, in haemophilic mice, in contrast
to thrombocytopaenic mice, even massive inflamma-
tory cell influx alone does not provoke significant
bleeding.
Relevance to joint haemorrhage
Repeated bleeding into a joint cavity is the inciting
factor for synovial and cartilage changes in haemo-
philic arthropathy [28,29]. The progressive accumu-
lation of iron from red blood cells during successive
intra-articular haemorrhages triggers synovial
inflammation and may be a direct stimulus for the
proliferation of synovial cells [30,31]. Free iron
clearly promotes inflammation and cellular prolife-
ration, and is thought to play a role in the synovitis
and excess angiogenesis observed in patients with
haemophilia following repeated haemorrhage into a
Ôtarget jointÕ [32,33]. Such a joint that has suffered
repeated haemorrhage not only develops synovial
proliferation, but also florid angiogenesis that serves
as a source of repeated bleeding.
Conclusions
Our wound healing studies have led us to formulate
the following hypothesis: late haematomas in our HB
mice (and recurrent joint haemorrhage in humans)
are caused by a vicious cycle in which iron deposited
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd

by bleeding stimulates increased angiogenesis, which
then predisposes to more bleeding.
The phenomenon of late (re-)bleeding at sites of
angiogenesis has implications for the treatment of
haemophilic manifestations, such as haemophilic
arthropathy. The initial insult of bleeding into a
joint is similar to the punch biopsy, as it sets into
motion the cycle of inflammation and angiogenesis
that promotes further bleeding. Bleeding during the
process of angiogenesis could then lead to a vicious
cycle of inflammation, increased angiogenesis and
more bleeding.
The most efficient way of preventing inflammation,
angiogenesis and recurrent bleeding is to prevent
haemorrhage from occurring in the first place. While
extended factor therapy may prevent additional
bleeding related to angiogenesis, only up-front pro-
phylactic therapy can prevent the initial bleeding
episode that triggers the subsequent bleeding cycle. It
remains to be seen whether anti-inflammatory or
anti-angiogenic therapies might prevent recurrent
bleeding and subsequent arthropathy.
Disclosures
Our studies were supported, in part, by Novo
Nordisk and by the US Department of Veterans
Affairs.
References
1 Gospodarowicz D, Brown KD, Birdwell CR, Zetter BR. Control of
proliferation of human vascular endothelial cells. Characterization
of the response of human umbilical vein endothelial cells to
fibroblast growth factor, epidermal growth factor, and thrombin. J
Cell Biol 1978; 77: 774–88.
2 Maruyama N, Hirano F, Yoshikawa N, Migita K, Eguchi K, Ta-
naka H. Thrombin stimulates cell proliferation in human fibro-
blast-like synoviocytes in nuclear factor-kappaB activation and
protein kinase C mediated pathway. J Rheumatol 2000; 27: 2777–
85.
3 Bar-Shavit R, Kahn AJ, Mann KG, Wilner GD. Identification of a
thrombin sequence with growth factor activity on macrophages.
Proc Natl Acad Sci USA 1986; 83: 976–80.
4 Algermissen B, Sitzmann J, Nürnberg W, Laubscher JC, Henz BM,
Bauer F. Distribution and potential biologic function of the
thrombin receptor PAR-1 on human keratinocytes. Arch Dermatol
Res 2000; 292: 488–95.
5 Wolberg AS. Thrombin generation and fibrin clot structure. Blood
Rev 2007; 21: 131–42.
6 Richardson DL, Pepper DS, Kay AB. Chemotaxis for human
monocytes by fibrinogen-derived peptides. Br J Haematol 1976;
32: 507–13.
7 Simpson DM, Ross R. The neutrophilic leukocyte in wound repair
a study with antineutrophil serum. J Clin Invest 1972; 51: 2009–
23.
8 Martin P, DÕSouza D, Martin J et al. Wound healing in the PU.1
null mouse – tissue repair is not dependent on inflammatory cells.
Curr Biol 2003; 13: 1122–8.
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd
WOUND HE ALING IN HAEMOPHILIA 17
9 Vinckier F, Vermylen J. Wound healing following dental extrac-
tions in rabbits: effects of tranexamic acid, warfarin anti-coagula-
tion, and socket packing. J Dent Res 1984; 63: 646–9.
10 Drew AF, Liu H, Davidson JM, Daugherty CC, Degen JL. Wound-
healing defects in mice lacking fibrinogen. Blood 2001; 97: 3691–
8.
11 te Velde EA, Wagenaar GT, Reijerkerk A et al. Impaired healing of
cutaneous wounds and colonic anastomoses in mice lacking
thrombin-activatable fibrinolysis inhibitor. J Thromb Haemost
2003; 1: 2087–96.
12 Szpaderska AM, Egozi EI, Gamelli RL, DiPietro LA. The effect of
thrombocytopenia on dermal wound healing. J Invest Dermatol
2003; 120: 1130–7.
13 Hoffman M, Harger A, Lenkowski A, Hedner U, Roberts HR,
Monroe DM. Cutaneous wound healing is impaired in hemophilia
B. Blood 2006; 108: 3053–60.
14 Brandstedt S, Olson PS. Effect of defibrinogenation on wound
strength and collagen formation. A study in the rabbit. Acta Chir
Scand 1980; 146: 483–6.
15 Brandstedt S, Olson PS. Lack of influence on collagen accumula-
tion in granulation tissue with ÔdelayedÕ defibrinogenation. A study
in the rabbit. Acta Chir Scand 1981; 147: 89–91.
16 Brandstedt S, Rank F, Olson PS. Wound healing and formation of
granulation tissue in normal and defibrinogenated rabbits. An
experimental model and histological study. Eur Surg Res 1980; 12:
12–21.
17 McDonald A, Hoffman M, Hedner U, Roberts HR, Monroe DM.
Restoring hemostatic thrombin generation at the time of cutaneous
wounding does not normalize healing in hemophilia B. J Thromb
Haemost 2007; 5: 1577–83.
18 Conway EM, Collen D, Carmeliet P. Molecular mechanisms of
blood vessel growth. Cardiovasc Res 2001; 49: 507–21.
19 Lijnen HR, Van Hoef B, Lupu F, Moons L, Carmeliet P, Collen D.
Function of the plasminogen/plasmin and matrix metalloproteinase
systems after vascular injury in mice with targeted inactivation of
fibrinolytic system genes. Arterioscler Thromb Vasc Biol 1998; 18:
1035–45.
20 McDonald AG, Yang K, Roberts HR, Monroe DM, Hoffman M.
Perivascular tissue factor is down-regulated following cutaneous
wounding: implications for bleeding in hemophilia. Blood 2008;
111: 2046–8.
21 Drake TA, Morrissey JH, Edgington TS. Selective cellular
expression of tissue factor in human tissues. Implications for
disorders of hemostasis and thrombosis. Am J Pathol 1989; 134:
1087–97.
22 Fleck RA, Rao LV, Rapaport SI, Varki N. Localization of human
tissue factor antigen by immunostaining with monospecific, poly-
clonal anti-human tissue factor antibody. Thromb Res 1990; 59:
421–37.
23 Hoffman M, Cooper ST. Thrombin enhances monocyte secretion
of tumor necrosis factor and interleukin-1 beta by two distinct
mechanisms. Blood Cells Mol Dis 1995; 21: 156–67.
24 Ryan J, Brett J, Tijburg P, Bach RR, Kisiel W, Stern D. Tumor
necrosis factor-induced endothelial tissue factor is associated with
subendothelial matrix vesicles but is not expressed on the apical
surface. Blood 1992; 80: 966–74.
25 Kisucka J, Butterfield CE, Duda DG et al. Platelets and platelet
adhesion support angiogenesis while preventing excessive hemor-
rhage. Proc Natl Acad Sci USA 2006; 103: 855–60.
26 Goerge T, Ho-Tin-Noe B, Carbo C et al. Inflammation induces
hemorrhage in thrombocytopenia. Blood 2008; 111: 4958–64.
27 Day RM, Harbord M, Forbes A, Segal AW. Cantharidin blisters: a
technique for investigating leukocyte trafficking and cytokine
production at sites of inflammation in humans. J Immunol Meth-
ods 2001; 257: 213–20.
28 Roy S, Ghadially F. Pathology of experimental haemarthrosis. Ann
Rheum Dis 1966; 25: 402–15.
Haemophilia (2010), 16 (Suppl. 3), 13–18
18 M. HOFFMAN and D. M. MONROE
29 Mainardi CL, Levine PH, Werb Z, Harris ED Jr. Proliferative
synovitis in hemophilia: biochemical and morphologic observa-
tions. Arthritis Rheum 1978; 21: 137–44.
30 Hakobyan N, Kazarian T, Jabbar AA, Jabbar KJ, Valentino LA.
Pathobiology of hemophilic synovitis I: overexpression of mdm2
oncogene. Blood 2004; 104: 2060–4.
31 Wen FQ, Jabbar AA, Chen YX, Kazarian T, Patel DA, Valentino
LA. c-myc proto-oncogene expression in hemophilic synovitis: in
Haemophilia (2010), 16 (Suppl. 3), 13–18
vitro studies of the effects of iron and ceramide. Blood 2002; 100:
912–6.
32 Hakobyan N, Kazarian T, Valentino LA. Synovitis in a murine
model of human factor VIII deficiency. Haemophilia 2005; 11:
227–32.
33 Hoots WK, Rodriquez N, Boggio L, Valentino LA. Pathogenesis of
haemophilic synovitis: clinical aspects. Haemophilia 2007;
13(Suppl. 3): 4–9.
Ó 2010 The Authors
Journal Compilation Ó 2010 Blackwell Publishing Ltd