Bone 37 (2005) 669 – 677

www.elsevier.com/locate/bone

Do serological tissue turnover markers represent callus formation

during fracture healing?
P. Seebeck a , H.J. Bail a , C. Exner a , H. Schell a , R. Michel b , H. Amthauer b ,
H. Bragulla c , G.N. Duda a,⁎
a
Center for Musculoskeletal Surgery, Charité-University Medicine Berlin, Free and Humboldt-University of BerlinAugustenburger Platz 1,
D-13353 Berlin, Germany
b
Radiology, Charité-University Medicine Berlin, Germany
c
Institute for Veterinary Anatomy, Free University of Berlin, Germany

Received 2 February 2005; revised 6 June 2005; accepted 13 June 2005

Available online 25 August 2005

Abstract

Serological parameters of bone and fibrous tissue turnover were demonstrated to monitor the course of fracture healing. The aim of this
study was to evaluate the correlation between the serological parameter levels during fracture healing and callus development in a
standardised ovine model of fracture healing.
Two years old female sheep received a standardised 3 mm tibial bone defect stabilised by an external fixator. The serological levels of the
C-terminal propeptide of procollagen type I (PICP), bone specific alkaline phosphatase (bALP), total alkaline phosphatase (tALP),
osteocalcin, tartrate-resistant acid phosphatase (TRAP), calcium, phosphate and the N-terminal peptide of procollagen type III (PIIINP) were
observed over a 9-week healing period. The course of fracture healing was monitored radiographically, and the callus composition was
evaluated histologically at 2, 3, 6 and 9 weeks post-surgery. The serological results were compared with an untreated control group.
Additionally, the maximum values during healing were compared with juvenile values to gauge the level of the serological response.
The histological and radiographical results demonstrated callus formation without complications. All serological parameters showed broad
inter-individual variations, and the response to the standardised fracture scenario was strongly individual. Maximum values during fracture healing
did not reach the juvenile levels. The fractured as well as the control animals showed significant changes in the parameter levels. No correlations
were observed between the histological course of healing and the course of bone formation markers whilst the TRAP level was reduced during
bony callus formation. The PIIINP level increased when the amount of soft callus tissue decreased during healing. The observed bone formation
markers were not suitable as general markers to detect the course of fracture healing, whilst PIIINP was able to reflect soft callus degradation.
© 2005 Elsevier Inc. All rights reserved.

Keywords: Fracture healing; Callus formation; Tissue turnover; Serological markers; Sheep model

Introduction reflect the healing progress and to differentiate between

Ever since the analysis of bone turnover markers was
introduced into the clinical investigation of systemical bone
disorders, the possibility to assess the course of fracture
healing using these markers has been suggested [10,18,21].
In clinical studies, several parameters were demonstrated to
⁎ Corresponding author. Fax: +49 30 450 559969.
E-mail address: georg.duda@charite.de (G.N. Duda).
8756-3282/$ - see front matter © 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.bone.2005.06.008
normal and delayed fracture healing [10,17,20]. Addition-
ally, some parameters seem to be suitable for diagnosing
infections at the fracture site [34].
Callus formation and maturation is an interlocking chro-
nology of the formation and degradation of different tissue
types [3,23,29,31]. The early soft callus containing mostly
fibrous tissue undergoes successive degradation, whilst the
hard callus is growing constantly. Direct bone formation
starts at the callus periphery by intramembranous
670 P. Seebeck et al. / Bone 37 (2005) 669–677
ossification, whilst bone formation via endochondral ossifi-
cation can be observed at the interface between fibrous tissue
and the newly formed woven bone. Finally, the bony callus is
remodelled reconstituting the original bone shape [32,35].
Tissue formation can be characterised by serological
markers since specific peptides are released into serum
during collagen formation. The carboxy-terminal propeptide
of procollagen type I (PICP) reflects bone formation, whilst
the amino-terminal propeptide of type III procollagen
(PIIINP) is released during the formation of fibrous tissue
as well as during its degradation [15,16,28]. The activity of
bone forming (osteoblasts) and bone resorbing cells
(osteoclasts) can be monitored by their specific enzymes.
The serological level of bone specific alkaline phosphatase
rises during matrix mineralisation, whilst the osteocalcin
level is thought to be sensitive to osteoblast maturation
[6,7,12,33]. Osteoclastic activity can be detected by tartrate-
resistant acid phosphatase type 5 b (TRAP) [13,14].
In a previous experimental study, the observed serological
parameters showed no correlation with the mechanical
course of fracture healing [19]. In addition, there is strong
evidence that local skeletal injury exerts an overall
osteogenic response so that it still remains unclear whether
bone turnover markers reflect the local fracture healing
scenario or potentially a systemic response to the fracture
[9,11]. Furthermore, the level of bone turnover parameters is
strongly influenced by the permanent skeletal bone remodel-
ling which in turn is influenced by various external factors,
e.g. patient activity and age or diurnal variation of
parameters [2,24,26,27,36].
The aim of this study was to evaluate the correlation
between the serological course of bone turnover markers and
callus development during bone healing in a standardised
ovine fracture healing model to reduce external factors for
serological variations to their minimum. Furthermore, the
serological results during fracture healing were compared
with an untreated control group to allow to differentiate the
influence of the fracture scenario from general factors which
itself may shroud or even hide serological changes. To gauge
the level of the serological response to a fracture, the fracture
healing group was compared with samples of juvenile
individuals representing an active bone metabolism.
Methods
Animals and surgical procedure
All animal experiments were carried out according to the
policies and principles established by the Animal Welfare
Act, the NIH Guide for Care and Use of Laboratory Animals
and the National Animal Welfare Guidelines. The study was
approved and monitored by the local legal representative1.
1
Landesamt für Arbeitsschutz, Gesundheitsschutz und technische
Sicherheit, Berlin: G 0224/01.
Female merino-mix sheep (2 years of age) were
acclimatised for 6 weeks. All animals were housed indoor
in loose groups of 6–8 sheep and received identical
husbandry and feeding. The stable was exposed to the
external light/dark cycle. Prior to study inclusion, all sheep
were screened by radiography and blood analysis for
evidence of skeletal diseases.
Eight animals served as an untreated control group.
Thirty-two animals underwent a standardised midshaft
osteotomy of the right tibia which was distracted to a 3-
mm gap and stabilised with a monolateral external fixator.
The monolateral external fixator consisted of 6 Schanz'
screws (∅ 5 mm, 3 inserted proximally, 3 inserted distally
of the osteotomy) and 2 steel tubes (∅ 10 mm) and was
mounted medially to stabilise the defect. In all animals, the
soft tissue was preserved to exclude additional influences
of tissue damage on the process of bone healing. The
osteotomy wound was sutured, and the leg, together with
the external fixator, was covered with a tube bandage. The
animals received an analgetic (Finadyne®, Essex, Ger-
many) for 3 days postoperatively. Daily pin care involved
cleaning of the insertions of the Schanz' screws with
Ethacridinlactate (Rivanol®, 5%, Chinosol, Germany). The
operated animals were randomly divided into four groups
(n = 8) which were sacrificed at 2, 3, 6 and 9 weeks
postoperatively.
Eight 5-month-old female lambs served as juvenile
control group representing an active bone metabolism. All
animals belonged to the identical husbandry.
Serological analyses
In the 9-week fracture healing group (n = 8) and the
untreated control group (n = 8), blood was collected by
puncture of the Vena cephalica antebracchii. Samples were
taken directly prior to sedation for surgery, 2 days post-
surgery and at weekly intervals thereafter in the operated
group. In the control group, the samples were collected in
identical intervals. In both groups, the first sample served as
the individual reference value. From the eight lambs,
singular samples were collected to serve as juvenile controls.
Each of the blood samples was collected between 10:00 and
12:00 a.m. to avoid the influence of diurnal variations.
Serum samples were immediately separated by centrifuga-
tion at 3000 rpm for 10 min, the supernatant was pipetted
into 1 ml Eppendorf tubes and stored at −80°C. All samples
were collected and stored in an identical manner until
parameter analysis.
The serum concentrations of total alkaline phosphatase
(tALP), calcium and phosphate were determined by
standardised procedures in the local clinical laboratory.
A haemogram served as infection control. The concentra-
tion of the C-terminal propeptide of procollagen type
I (PICP), bone specific alkaline phosphatase (bALP),
osteocalcin and the N-terminal peptide of procollagen type
III (PIIINP) were determined by commercially available
 P. Seebeck et al. / Bone 37 (2005) 669–677 671

Table 1
Preoperative normal values, maximum values during fracture healing and juvenile reference values of all analysed parameters
Parameter Adult normal values (n = 16) Maximum values (n = 8) Juvenile values (n = 8)
Median Range CV [%] Median Range Median Range CV [%]
a,c b
PICP [μg/l] 163.5 137.7–253.6 17.9 229.2 178.9–308.3 291.9 275.2–443.7 17.8
bALP [μg/l] 16.9a,c 6.4–42.9 61.7 31.8b 12.1–47.6 87.6 56.1–143.1 28.8
bALP [U/l] 44a,c 17–112 61.7 82.9b 31.6–124.1 228 146–373 28.8
tALP [U/l] 137a 63–424 53.0 191b 143–273 384 214–475 25.6
Ratio bALP/tALP 31.4a,c 12.1–59 46.5 42.2 27.2–81.8 70.8 49.4–78.5 17.0
Osteocalcin [μg/l] 8.0c 2.8–15.7 44.1 14.1 7.0–16.4 12.1 10.5–14.7 11.6
TRAP [U/l] 0.28a,c 0.14–0.52 38.9 0.39b 0.28–0.52 1.95 1.1–5.9 62.4
Calcium [mmol/l] 2.6c 2.3–2.8 5.3 2.8 2.6–2.9 2.7 2.5–2.7 2.7
Phosphate [mmol/l] 2.1a,c 1.6–2.8 17.9 2.3 2.0–2.8 2.6 2.0–2.8 10.0
PIIINP [μg/l] 4.6a,c 3.5–10.5 39.4 8.8b 5.4–11.4 22.7 10.6–39.5 44.6
a = normal vs. juvenile values (P ≤ 0.03), b = preoperative vs. maximum values (P ≤ 0.01), c = maximum vs. juvenile (P ≤ 0.043).

human radioimmunoassays (RIA)2. All used assays were
described to cross-react with ovine samples [1,5,19,27]. The
validation for cross-reactivity with sheep as well as the
measurement of tartrate-resistant acid phosphatase (TRAP)
was performed by sba-sciences, Finland using a competitive
ELISA. All samples were measured in duplicate on the same
day, and the mean values per day were calculated. The inter-
individual variation was calculated from the results of all
animals per group at one time point.
Radiographical monitoring of fracture healing
Radiographic images were taken directly after surgery and
at weekly intervals thereafter, at the same days as blood
samples were collected. The radiographs were taken in the
postero-anterior direction at the standing animal keeping the
operated limb in an unloaded position. All radiographs were
taken in a standardised manner under identical exposition
conditions (70 kV, 2.5 mAs) and with a standardised film to
focus distance of 1 m. All radiographs were analysed for signs of
callus mineralisation and bony bridging of the osteotomy gap.
Histomorphometrical analysis of callus development
The callus regions of all sheep were sawed in the frontal
plane and embedded in methylmethacrylate (Technovit®
9100, Heraeus Kulzer, Germany). Serial histological sections
were cut. One section per animal was stained with Safranin
Orange/von Kossa to visualise soft callus (fibrous tissue plus
cartilage) and hard callus tissue (newly formed mineralised
cancellous bone). Computerised histomorphometric analysis
was performed using an image analysis system (KS400,
Zeiss, Germany) as described previously [25]. A region of
interest (ROI) was defined to cover a maximum callus area in
all specimen. Thus, the ROI included the 3 mm diaphyseal
bone gap plus 4.5 mm into the proximal and distal direction
in its vertical dimension which added up to a standardised
ROI height of 12 mm in total. The horizontal ROI width was
2
PICP & PIIINP: Orion Diagnostica, Espoo, Finland. bALP: Beckman
Coulter Inc., Brea CA. Osteocalcin: DiaSorin Inc, Stillwater MN.
adjusted individually at the level of the largest callus
diameter. Within the ROI, the areas of bone and soft tissue
were determined quantitatively. All histological measure-
ments were performed by two experienced independent
observers whose results were averaged.
Statistical analyses
All absolute data were expressed as medians and the
corresponding minimum and maximum values respectively.
To level inter-individual variations, all turnover parameters
were expressed as a percentage of their preoperative values.
The results within graphical presentations were given as
median courses with corresponding inter-quartile ranges. To
determine statistical differences between the operated and
the control group as well as statistical relevant changes of the
turnover parameters over time, a two-factorial analysis of
variance for repeated measurements was performed using
SAS (SAS 8.2, SAS Institute Inc., Cary, NC). The pre-
operative values and the maximum values during fracture
healing were compared using the Wilcoxon test. The adult
and the juvenile values as well as the four histological groups
were compared using the Mann–Whitney U test. Those tests
were performed using SPSS (SPSS 10.0, SPSS Inc.,
Chicago, IL). For all tests, a P-value of less than 0.05 was
taken as a significant difference, multiple testing was
adjusted according to Bonferoni–Holm.
Results
All animals showed regular callus formation free of
complications. After surgery, the animals initially unloaded
the operated limb. Over the course of healing, a steady return
to full weight bearing was determined. No infections were
detected at the fracture or at the pin sites.
Except calcium, all parameters showed broad inter-
individual variations already preoperatively, ranging from
10 to 62% (Table 1). A broad inter-individual variation could
be also observed for the serological courses of all parameters
during the entire 9-week period of observation.
672 P. Seebeck et al. / Bone 37 (2005) 669–677

The juvenile values of all parameters except osteocalcin

and calcium were significantly higher compared to the adult
control values (P ≤ 0.03). Osteocalcin trended (P = 0.052) to
be different between the juvenile and the adult group. The
maximum values of PICP, bALP, tALP, TRAP and PIIINP
observed during the course of fracture healing did not reach
levels seen in juveniles (P ≤ 0.01). In contrast, the maximum
values of osteocalcin, calcium and phosphate reached the
juvenile level (Table 1). Except tALP, all parameters'
maximum values observed during fracture healing were
significantly larger compared to their relevant preoperative
normal values (P ≤ 0.043).

PIIINP

The control group showed significant time-dependent

changes of its PIIINP level during the observed 9-week
period (P b 0.001): an initial increase was observed for the
first two postoperative weeks (Fig. 1b). Thereafter, the
PIIINP level decreased beyond the first values until the eighth
week followed by a second increase. The animals of the
fracture group showed a decrease of their PIIINP levels for
until the fourth postoperative week followed by an increase
around the fifth postoperative week. Thereafter, the PIIINP
values returned to the preoperative level. No significant
differences could be observed for the serological courses of
PIIINP between the fracture and the control group.

PICP

The animals of the control group showed significant time-

dependent changes of the PICP level during the observed 9-
week period (P b 0.001): an increase of their PICP level was
observed in the first 3 weeks (Fig. 2a). Thereafter, the PICP
values declined until they reached the level of the first
measurement day in the fifth week followed by a fluctuation
around the initial level until the end of the observed period of
9 weeks. In the fracture group, the course of PICP showed an
initial drop within the first postoperative week (Fig. 2a).
Thereafter, the PICP values reached the preoperative level
followed by a steady from the fifth postoperative week
onwards. No significant differences could be observed for
the serological courses of PICP between the fracture and the
control group.
Fig. 1. (a) Histological course of fracture healing: hard and soft callus tissue
expressed as a percentage of the total callus area. The hard callus was
growing constantly, whilst the soft callus tissue underwent successive
degradation. (b) Systemic level of PIIINP during fracture healing: black:
fracture healing group; dashed line: control group. The bars are indicating
the soft callus area at the analysed selected time points. In the fracture
healing group, PIIINP showed an initial drop after surgery. The course of
PIIINP fits well with the histological development of soft callus tissue
during fracture healing.

Osteocalcin observation period. The animals of the fracture group

The control group showed a significant decrease in its
osteocalcin level around the fourth week of observation
(P = b0.001; Fig. 2b). Thereafter, the osteocalcin level
stayed slightly below the initial level until the end of
showed a decreased osteocalcin level compared to the
preoperative values throughout the entire healing period
(Fig. 2b). No significant differences could be observed for
the serological courses of osteocalcin between the fracture
and the control group.

Fig. 2. Serological courses of PICP (a), osteocalcin (b), calcium (c), phosphate (d) bALP (e), tALP (f), bALP/tALP ratio (g) and TRAP (h) during fracture
healing: black: fracture healing group; dashed line: control group. The bars are indicating the hard callus area at the analysed selected time points. In the untreated
control group, the bone turnover rate decreased during the observed period. In the fracture group, all bone turnover parameters except calcium showed an initial
drop after surgery. A relative increase could be observed for bALP as well as for the bALP/tALP ratio. Both osteocalcin and phosphate showed an initial drop
after surgery, and their courses were closely related to each other.
P. Seebeck et al. / Bone 37 (2005) 669–677 673
674 P. Seebeck et al. / Bone 37 (2005) 669–677
Calcium
The control group showed significant time-dependent
changes of its calcium level during the observed 9-week
period (P b 0.001): an increase was observed in the fourth
week followed by a slight decrease (Fig. 2c). The operated
group showed a slight increase in their calcium values
between the fourth and fifth postoperative week (Fig. 2c). No
significant differences could be observed for the serological
courses of calcium between the fracture and the control
group.
Phosphate
The animals of the control group showed a significant
decrease in phosphate levels from the fourth week of
observation until the end of the observed 9-week period
(P b 0.001; Fig. 2d). For the fracture group, an initial
decrease could be observed until the fourth postoperative
week followed by a fluctuation around the preoperative level
thereafter (Fig. 2d). No significant differences could be
observed for the serological courses of phosphate between
the fracture and the control group.
Alkaline phosphatase
A significant overall decrease was observed for the
bALP as well as the tALP level during the 9-week period
in the control group (P b 0.001; Figs. 2e and f). In
Fig. 3. Histological course of fracture healing: 2 weeks post-surgery (a), 3 weeks post-surgery (b), 6 weeks post-surgery (c) and 9 weeks post-surgery (d). The
amount of hard callus tissue increased, whilst the amount of soft callus tissue decreased during fracture healing. Signs of callus remodelling were observed from
the sixth postoperative week onwards.
addition, the share of bALP from tALP decreased to half
of its initial amount from the fifth observed week on
throughout the end of the observed 9-week period (Fig.
2g). The fracture group showed also a slight decrease
during the entire healing period but not as pronounced as
the control group. Particularly between the fifth and ninth
postoperative week, the bALP values of the fracture group
were clearly larger than those of the control group (Fig.
2e). The bALP/tALP ratio kept constant for the entire
healing period (Fig. 2g). The serological courses of bALP
were significantly different between the two groups
(P = 0.03), whilst no significant differences could be
observed for the serological courses of tALP as well as
the bALP/tALP ratio between the fracture and the control
group.
TRAP
In the control group, a constant decrease of the TRAP
level could be observed throughout the entire observation
period of 9 weeks (Fig. 2h). The fracture group showed
significant time-dependent changes of its TRAP level
during the observed 9-week period (P = 0.03): an initial
decrease was observed within the first week followed by
a recovery to the preoperative values in the second
postoperative week (Fig. 2h). Thereafter, the TRAP levels
decreased again until the fourth postoperative week
followed by an increase until the ninth postoperative
week, where the TRAP level was slightly larger than the
 P. Seebeck et al. / Bone 37 (2005) 669–677 675

Table 2
Histomorphometrical course of fracture healing
Callus composition Weeks post-surgery
2 (n = 8) 3 (n = 8) 6 (n = 8) 9 (n = 8)
2 d d
Total callus [mm ] Median 124.9 173.8 139.5 110.9
Range 73.2–158.8 128.6–207.6 97.2–187.1 86.4
CV [%] 24.4 15.6 20.8 19.2
Hard callus [mm2] Median 44.9 55.0e 80.5e 78.8
Range 36.6–58.9 45.6–70.9 71.2–105.8 67.6–117.2
CV [%] 18.2 16.3 13.8 20.6
Soft callus [mm2] Median 73.8d 119.3d,e 62.0e 34.4
Range 30.1–112.2 79.8–151.5 25.9–88.2 15.4–50.9
CV [%] 35.4 18.6e 40.3e 41.5
Hard callus/soft callus Median 40.3/59.7 33.9/66.1 e 58.3/41.8 e 69.0/31.0
[% of total callus area] Range 29.3–58.9/41.1–70.7 27.0–38.0/62.0–73.0 45.7–73.3/26.7–54.3 65.1–87.9/12.1–34.9
CV [%] 24.6/17.4 12.4/6.1 17.5/25.3 11.8/31.9
The soft callus tissue increased significantly between the second and third postoperative week thus increasing the total callus area. The hard callus area increased
significantly between the third and sixth postoperative week, whilst the soft callus area increased during this period (dP = 0.003; eP ≤ 0.001).

preoperative values (Fig. 2h). The serological TRAP
courses were significantly different between the two
groups (P = 0.01).
Callus development during bone healing
Moderate inter-individual variations of callus composi-
tion could be seen for all parameters (Table 2). The total
callus area increased significantly between the second and
third postoperative week (P = 0.003) followed by a constant
decrease towards the ninth postoperative week (Fig. 3). For
the hard callus area, a small increase was observed
between the second and third postoperative week (Fig. 3)
followed by a strong increase between the third and sixth
postoperative week (P b 0.001; Fig. 3). Between the sixth
and ninth postoperative week, the hard callus area stayed
nearly identical (Table 2). The soft callus area increased
significantly between the second and third postoperative
week (P = 0.003) followed by a marked decrease of its
area between the third and sixth postoperative week
(P = 0.001) and a further decrease towards the ninth
postoperative week (Fig. 3). During the second and third
postoperative week, the total callus area was mainly
enlarged by the formation of fibrous tissue. Therefore, the
percentual composition of the callus showed no relevant
changes between the two time points. In contrast, the
reduction of the total callus area between the third and
sixth postoperative week went along with a massive
degradation of soft tissue (as expressed by a significant
decrease of the percentual soft callus amount, P b 0.001;
Fig. 1a) in combination with a significant formation of
mineralised callus tissue (expressed by a significant
increase of its percentual amount, P b 0.001; Fig. 1a).
Thereafter, the callus composition did not change sig-
nificantly any more, and remodelling of the bony callus
was observed (Fig. 3).
The radiographic analysis detected first signs of
mineralised callus formation within the third postoperative
week starting at the callus periphery. At six weeks post-
surgery, the mineralised callus area was increased, showing
a pronounced bulbous cuff of mineralised tissue at both
sides of the tibia. Between six and nine weeks of post-
surgery, the mineralised callus area decreased, indicating
signs of a beginning restoration of the original tibial shape.
The osteotomy gap was clearly visible during the entire 9-
week healing period.
Discussion
The aim of this study was to evaluate whether
serological parameters are suitable to describe the course
of fracture healing. An experimental animal model was
chosen to compare the serological results to the histological
course of healing as well as to exclude external factors
having a strong influence on bone turnover markers.
A bone gap was created and stabilised with a rather rigid
external fixator to ensure secondary fracture healing via
callus formation.
Corresponding to previous results reported, the serologi-
cal reaction to the fracture situation was highly individual
[19,30]. An initial drop in the level of all parameters was
observed indicating a depression of the body's entire
turnover capacity which was restored soon thereafter. In
previous clinical studies of fracture healing, the collection of
samples started when the fracture has already occurred.
Therefore, it could be assumed that the first sample was
always taken just within the phase of initial depression of the
entire metabolism. Thus, the following restoration of the
normal turnover capacity alone could have been interpreted
as increasing parameter levels.
In contrast to previous studies in human fracture patients,
no increase of the PICP level was observed in this study
[18,21,28]. In addition, PICP showed large week to week
variations in the control as well as in the fracture group.
Therefore, PICP appears to be unsuitable for the detection of
676 P. Seebeck et al. / Bone 37 (2005) 669–677
fracture healing, especially in the non-standardisable clinical
every day setting.
The bALP level as well as the bALP/tALP ratio were
indicating an increase in bone formation activity from the
fourth postoperative week onwards. This correlates well with
previous studies in human patients describing a systemic bALP
level peaking some weeks after fracture [10,18,20,22]. The
increasing bALP level at the fourth week correlated well with
the histological findings. But bALP also showed the largest
inter-individual variations amongst all evaluated parameters
which might reduce its general suitability. Nevertheless, bALP
should be preferred compared to tALP for the detection of
postoperative upregulation of the bone metabolism.
Osteocalcin as well as calcium and phosphate seemed to
be the least affected by day to day variations also showing
minor inter-individual differences. Therefore, they appear to
be more promising for the detection of even small serological
changes such as those caused by the healing of a fracture. In
addition, a strong correlation seemed to exist between these
three parameters.
In parallel, a decrease in the TRAP activity was observed
correlating with bony callus formation, suggesting that bone
resorption was lowered for the benefit of bone formation.
The TRAP activity slightly increased again thereafter when
remodelling of the newly formed bone started.
The results of this study were unable to support the
previously described findings of an increased level of PIIINP
during the early phase of fracture healing [17,20]. However,
an increase could be detected at a later time point correlating
with the histological detection of a marked decrease of soft
callus tissue. Whilst previous studies suggested that PIIINP
represents the formation as well as the degradation of fibrous
tissue, in this study, it seemed to be more sensitive for the
degradation of fibrous tissue.
Although it was evident that all parameters were
influenced by the fracture scenario, this would have been
either undetectable or misinterpreted for most parameters
without the additional information of the untreated control
group. Not only did the fracture scenario have influence on
the course of serological parameters, but obviously also
a reduced activity level caused by the study-specific housing
conditions (the animals underwent an activity restriction
after running free) as observed in the control group. For the
clinical situation, it can be assumed that changes in the
patients' behaviour or environment occurring after a fracture
might be more influential than the relatively small and
circumscribed fracture scenario.
The observed bone formation markers were not suitable
to detect the increase of hard callus tissue which is mainly
responsible for an increasing fracture stability during healing
[4,8], a finding which is supported by the results of
a previous study showing no correlation between the course
of bone formation markers and the mechanical course of
healing as documented by decreasing interfragmentary
movements indicating increasing callus stiffness caused by
the progress in hard callus formation [19].
It remains questionable whether the observed bone
formation markers are able to detect the course of hard
callus formation or rather a systemical increase in bone
turnover after a fracture since previous studies suggest
that a fracture scenario causes a systemical osteogenic
response within the entire bony skeleton [9,11,17,20].
Therefore, it can be supposed that also in this study the
process of bone modelling during healing might be
overshadowed by the general rise in bone metabolism.
The question arises whether previous clinical studies
using bone formation to monitor fracture healing were
able to detect the course of fracture healing or only
a variability of osteogenic responses to the fracture
situation. In addition, the serological courses for delayed
fracture healing observed in those studies might have
represented a poorer systemical response which in turn
might have led to delayed fracture healing.
The observed bone formation markers seemed to be
unsuitable as general clinical markers to monitor the
course of fracture healing. Their strong dependency on
external influencing factors which seemed to be too large
to deduce any correlations possessing universal validity
and their broad inter-individual variations should make it
difficult to establish normal values or at least thresholds
which are essential for routinely clinical application.
However, those markers might be suitable to estimate the
extent of the osteogenic response after a fracture has
occurred to indicate possible healing disturbances due to
systemical deficiencies at a very early stage and facilitate
early specific intervention. Additionally, they could be
suitable for experimental studies of bone healing since they
provide additional non-invasive information concerning the
individual healing status. Further analyses will have to
verify if PIIINP and TRAP are suitable to discriminate
between normal and delayed fracture healing and therefore
to detect fracture healing disturbances at an early time
point.
Acknowledgments
This study was supported by a grant of the German
Research Foundation (DFG KFO 102/1 and 102/2). The
authors would like to thank Dr. R. Ranta from sba-sciences
for the determination of TRAP and Mr. D.R. Epari for
editing the manuscript.
References
[1] Allen MJ. Biochemical markers of bone metabolism in animals: uses
and limitations. Vet Clin Pathol 2003;32:101–13.
[2] Allen MJ, Hoffmann WE, Richardson DC, Breur GJ. Serum markers of
bone metabolism in dogs. Am J Vet Res 1998;59:250–4.
[3] Brighton CT. The biology of fracture repair. Instr Course Lect
1984;33:60–82.
[4] Chakkalakal DA, Lippiello L, Wilson RF, Shindell R, Connolly
 P. Seebeck et al. / Bone 37 (2005) 669–677
JF. Mineral and matrix contributions to rigidity in fracture healing.
J Biomech 1990;23:425–34.
[5] Collignon H, Davicco MJ, Barlet JP. Metacarpal growth and systemic
markers of bone metabolism in the ovine fetus. Reprod Fertil Dev
1996;8:287–95.
[6] Delmas PD. Biochemical markers of bone turnover for the clinical
assessment of metabolic bone disease. Endocrinol Metab Clin North
Am 1990;19:1-18.
[7] Delmas PD, Beaudreuil J. Biochemical markers of bone turnover in
osteoporosis. Rev Rhum Engl Ed 1997;64:31S–6S.
[8] den Boer FC, Bramer JA, Patka P, Bakker FC, Barentsen RH, Feilzer
AJ, et al. Quantification of fracture healing with three-dimensional
computed tomography. Arch Orthop Trauma Surg 1998;117:345–50.
[9] Einhorn TA, Simon G, Devlin VJ, Warman J, Sidhu SP, Vigorita VJ.
The osteogenic response to distant skeletal injury. J Bone Joint Surg
Am 1990;72:1374–8.
[10] Emami A, Larsson A, Petren-Mallmin M, Larsson S. Serum bone
markers after intramedullary fixed tibial fractures. Clin Orthop
1999:220–9.
[11] Frost HM. The regional acceleratory phenomenon: a review. Henry
Ford Hosp Med J 1983;31:3–9.
[12] Garnero P, Delmas PD. Bone markers. Bailliere's Clin Rheumatol
1997;11:517–37.
[13] Halleen JM, Alatalo SL, Janckila AJ, Woitge HW, Seibel MJ, Vaananen
HK. Serum tartrate-resistant acid phosphatase 5b is a specific and
sensitive marker of bone resorption. Clin Chem 2001;47:597–600.
[14] Halleen JM, Ylipahkala H, Alatalo SL, Janckila AJ, Heikkinen JE,
Suominen H, et al. Serum tartrate-resistant acid phosphatase 5b, but not
5a, correlates with other markers of bone turnover and bone mineral
density. Calcif Tissue Int 2002;71:20–5.
[15] Haukipuro K, Risteli L, Kairaluoma MI, Risteli J. Aminoterminal
propeptide of type III procollagen in serum during wound healing in
human beings. Surgery 1990;107:381–8.
[16] Jensen LT, Garbarsch C, Horslev-Petersen K, Schuppan D, Kim K,
Lorenzen I. Collagen metabolism during wound healing in rats. The
aminoterminal propeptide of type III procollagen in serum and wound
fluid in relation to formation of granulation tissue. APMIS 1993;101:
557–64.
[17] Joerring S, Jensen LT, Andersen GR, Johansen JS. Types I and III
procollagen extension peptides in serum respond to fracture in humans.
Arch Orthop Trauma Surg 1992;111:265–7.
[18] Joerring S, Krogsgaard M, Wilbek H, Jensen LT. Collagen turnover
after tibial fractures. Arch Orthop Trauma Surg 1994;113:334–6.
[19] Klein P, Bail HJ, Schell H, Michel R, Amthauer H, Bragulla H, et al.
677
Are bone turnover markers capable of predicting callus consolidation
during bone healing? Calcif Tissue Int 2004;75:40–9.
[20] Kurdy NM. Serology of abnormal fracture healing: the role of PIIINP,
PICP, and BsALP. J Orthop Trauma 2000;14:48–53.
[21] Kurdy NM, Bowles S, Marsh DR, Davies A, France M. Serology
of collagen types I and III in normal healing of tibial shaft
fractures. J Orthop Trauma 1998;12:122–6.
[22] Lammens J, Liu Z, Aerssens J, Dequeker J, Fabry G. Distraction bone
healing versus osteotomy healing: a comparative biochemical analysis.
J Bone Miner Res 1998;13:279–86.
[23] Lane JM. Breakout session. 2: Fracture repair process. Clin Orthop
1998:S354–5.
[24] Lepage OM, Hartmann DJ, Eicher R, Uebelhart B, Tschudi P,
Uebelhart D. Biochemical markers of bone metabolism in draught
and warmblood horses. Vet J 1998;156:169–75.
[25] Lienau J, Schell H, Duda GN, Seebeck P, Muchow S, Bail HJ. Initial
vascularization and tissue differentiation are influenced by fixation
stability. J Orthop Res 2005;23:639–45.
[26] Liesegang A, Reutter R, Sassi ML, Risteli J, Kraenzlin M, Riond JL, et
al. Diurnal variation in concentrations of various markers of bone
metabolism in dogs. Am J Vet Res 1999;60:949–53.
[27] Liesegang A, Sassi ML, Risteli J. Diurnal variation in concentration of
various markers of bone metabolism in growing female goats and
sheep. Anim Sci 2003;77:197–203.
[28] Lotz J, Gaertner T, Hahn M, Prellwitz W. Collagen type I metabolism
after bone surgery. Arch Orthop Trauma Surg 1999;119:212–6.
[29] McKibbin B. The biology of fracture healing in long bones. J Bone
Joint Surg Br 1978;60-B:150–62.
[30] Ohishi T, Takahashi M, Kushida K, Hoshino H, Tsuchikawa T, Naitoh
K, et al. Changes of biochemical markers during fracture healing. Arch
Orthop Trauma Surg 1998;118:126–30.
[31] Oni OA. The bony callus. Injury 1997;28:629–31.
[32] Pennig D. The biology of bones and of bone fracture healing.
Unfallchirurg 1990;93:488–91.
[33] Risteli J, Melkko J, Niemi S, Risteli L. Use of a marker of collagen
formation in osteoporosis studies. Calcif Tissue Int 1991;49: S24–5.
[34] Southwood LL, Frisbie DD, Kawcak CE, McIlwraith CW. Evaluation
of serum biochemical markers of bone metabolism for early diagnosis
of nonunion and infected nonunion fractures in rabbits. Am J Vet Res
2003;64:727–35.
[35] Webb JCJ, Tricker J. Bone biology—A review of fracture healing.
Curr Orthop 2000;14:457–63.
[36] Bone Markers. Biochemical and clinical perspectives. London: Martin
Dunitz Ltd; 2001.