вход по аккаунту


Periosteal new bone formation in a canine neuropathic model of osteoarthritis.

код для вставкиСкачать
Vol 40, No 10, October 1997, pp 1756-1759
0 1997. American College ot Rheumatology
Objective. To characterize, for the first time,
periosteal new bone formation in a well-established
canine model of accelerated osteoarthritis (OA) with
features of neuropathic arthropathy.
Methods. Seven dogs underwent left L4-S1 dorsal
root ganglionectomy (DRG), followed 3 weeks later by
transection of the anterior cruciate ligament of the
ipsilateral knee (ACLT). Eight weeks thereafter, a postmortem examination was performed to assess the severity of cartilage changes of OA and the formation of new
bone on the distal femur and proximal tibia in the
cruciate-deficient limb.
Results. As described previously, extensive fullthickness ulceration of the articular cartilage was
present in the unstable knee of every dog. The femoral
shaft immediately proximal to the condyles in the
unstable limb was consistently wider (mean +. SD
diameter 22.4 f 2.2 mm) than that in the contralateral
limb (19.9 & 1.3 mm; P = 0.01). Xeroradiography and
histologic examination of the distal femur revealed
extensive formation of woven bone on the periosteal
surfaces of the medial, lateral, and anterior aspects of
the femoral shaft in the OA limb of every dog. These
bony changes were not seen in radiographs of dogs that
underwent DRG with the cruciate ligament left intact
(n = 8) or of neurologically intact dogs that underwent
ACLT (n = 7) and were examined 24 weeks after
Conclusion. Formation of new periosteal bone on
Supported in part by NIH grant AR-20582.
Stephen L. Myers. MD, Kenneth D. Brandt, MD, Marjorie
Albrecht. BS: Indiana University School of Medicine, and Indiana
University Multipurpose Arthritis aiid Musculoskeletal Diseases Center, Indianapolis; Brian CXonnor, PhD: Indiana University School of
Medicine, Indianapolis; William R. Widmer. DVM: School of Veterinary Medicine, Purduc University, Webt Lafayette, Indiana.
Address rcprint requests to Stephen Myers, MD, Division of
Rheumatology, 541 Clinical Drive, Room 492; Indianapolis. IN 46202.
Suhmittcd for publication November 8, 10Yh; accepted in
revised form May 28, 1997.
the distal femur and tibia is a feature of this model of
accelerated OA that is not seen in the conventional
ACLT model of OA in the neurologically intact dog. This
observation suggests that interruption of sensory input
from the limb may affect the regulation of osteogenesis
in the mechanically unstable joint.
Neuropathic arthropathy (NA) is a complication
of disorders associated with neurosensory loss, such as
diabetes mellitus, leprosy, tabes dorsalis, and syringomyelia (1). NA has been described as “osteoarthritis . . .
with a vengeance” (2) because the breakdown of articular cartilage often progresses rapidly while large osteophytes and juxtaarticular new bone develop (2,3). Serial
radiographs usually allow NA to be distinguished from
idiopathic O A (1,2).
Attempts to create NA by a variety of surgical
procedures that interrupt the sensory input from the
limb have met with only limited success (1,4). For
example, extensive deafferentation of the hind limb in
dogs by unilateral L4-S1 dorsal root ganglionectomy
(DRG) produced no changes in the ipsilateral knee over
a followup interval as long as 16 months (4). A strikingly
different result was obtained, however, when DRG was
followed by transection of the ipsilateral anterior cruciate ligament (ACLT) (5,6). Under these conditions, OA,
i.e., large, full-thickness ulcers of the articular cartilage,
developed within a few weeks of surgery (5,6). Other
features of this model of “accelerated” OA included
prominent osteophytes, and a detritic synovitis that was
characteristic of NA or end-stage OA (6). This result
stands in dramatic contrast to the morphologic changes
of early OA typically seen in the knees of neurologically
intact dogs 12-32 weeks aftcr ACLT ( 5 ) . The present
report provides the initial description of periosteal osteogenesis, i.e., the formation of new woven bone, on the
distal femur and tibia as an additional feature of the
DRG-ACLT model of OA, distinguishing it from the
standard ACLT model of OA in the neurologically intact
Surgical procedures. Seven adult male mongrel dogs
(weighing 25-30 kg) were used in this study. Each dog underwent left LA-S1 DRG, followed 3 weeks later by ACLT of the
ipsilateral knee in accordance with guidelines of the Indiana
University School of Medicine Animal Care and Use Committee. The surgical procedures were performed exactly as described previously (5). After recovery from surgery, each dog
was transferred to a spacious pen (1.8m X 1.8m X 1.Xm) in
which it could ambulate freely for the next 8 weeks, until it
was killed by injection of a barbituric acid derivative
(Beuthanasia-D; Schering-Plough Animal Health, Kenilworth NJ).
Radiography. Radiographs of both hind limbs of each
dog were obtained prior to surgery to assure that the epiphyses
were closed and to exclude preexisting joint pathology. Because fracture of the posterior margin of the tibial plateau with
posterior dislocation of the unstable, deafferented limb may
occur occasionally in this model, a second radiograph of the
left hind limb of 1 dog was obtained when this animal suddenly
developed limitation in the use of that limb 5 weeks after
ACLT. Dislocation of the knee was confirmed, and this dog
was killed at that juncture. A veterinary radiologist (WRW)
examined the radiographs of each dog in the present study and,
for comparison, reviewed hind limb radiographs of 15 dogs
from a previously reported study ( 5 ) for evidence of periosteal
reaction. The latter radiographs, which served as historical
controls, were obtained 24 weeks after DRG in 8 dogs whose
cruciate ligaments were intact or 24 weeks after ACLT in 7
dogs that were neurologically intact. All DRG and ACLT
surgery in the past and present series was performed by the
same investigator (BO’C) using a standardized surgical
method, and all animals were housed under similar conditions
and permitted exercise ad libitum in their pens.
Procurement of articular tissues. Immediately after
the dogs were killed, both knees were opened and the distal
femur and proximal tibia of each knee were removed with a
bone saw and photographed. The location and extent of
cartilage ulcers and fibrillation and the severity of osteophytosis in each knee were recorded (6). Radiographs of the
disarticulated femur and tibia were obtained in 2 randomly
selected dogs.
In each hind limb, the width of the femoral diaphysis 5
cm proximal to the distal end of the femur and slightly
proximal to the superior margin of the trochlea, and the
maximum width of the metaphysis, were measured to the
nearest 0.5 mm with a calipers. A ring of bone 4-6 mm thick
was removed with a saw to obtain a cross-section of the
diaphysis at the level at which its width was measured. Xeroradiographs of the specimens from the right and left femur of
each dog were prepared with an Elscint Mammography System
(TruFocus, Los Gatos, CA) (7). For comparison, the femoral
diaphysis of the OA limb of 7 neurologically intact dogs that
were killed 16 weeks after ACLT was sawed in the same
manner, and xeroradiographs on these cross-sections of bone
were prepared. The rings of femoral shaft from 2 DRG-ACLT
Figure 1. Radiograph of the left leg of a dog 5 weeks after transection
of the anterior cruciate ligament and 7 weeks after ipsilateral dorsal
root ganglionectomy, showing periosteal new bone formation on the
anterior aspect of the distal femur and posterior aspect of the proximal
tibia (arrows).
dogs were embedded in plastic, after which 50-F.m sections
were prepared with a diamond saw for histologic examination.
In all 6 DRG-ACLT dogs killed after 8 weeks of
knee instability, the neurosurgical procedure resulted in
loss of the ipsilateral patellar reflex, and marked loss of
the response to pinprick over the ipsilateral paw and
posterior leg. At the time of killing, as described previously (5,6), extensive full-thickness ulcers of the articular
cartilage were present in the unstable knee. In 5 of the 6
animals, large osteophytes were present on the medial or
lateral patellar ridge, or both, and on the tibial plateau
and intercondylar notch.
Extensive periosteal new bone formation was
seen on the anterior aspect of the distal femur and
posterior aspect of the proximal tibia in the knee radiograph of the single DRG-ACLT dog that underwent
radiographic examination 5 weeks after ACLT became
of an acute fracture-dislocation (Figure 1). No periosteal
reaction was seen on the radiographs of dogs that
underwent ACLT without DRG, or D R G without
The extent of periosteal new bone formation in
DRG-ACLT dogs was assessed postmortem by comparing the O A femur with that of the contralateral limb.
Radiographs of the disarticulated femur and tibia from 2
animals showed periosteal new bone extending 10-15
cm proximal to the joint line on the medial and lateral
Figure 2. Xeroradiographs of 5 mni-thick cross-sections of the distal
femurs from the deafferented, cruciate-deficient limbs (lower row) and
contralateral limbs (upper row) of 3 dogs ( I ; 2; 3), showing formation
of periosteal woven bone (arrow) on the anterior, medial, and lateral
aspects of the cruciate-deficient specimens.
aspects of the femoral shaft. Similar changes were
observed on the lateral and posterior aspects of the tibia
in the unstable limb, -2-3 cm distal to the plateau. N o
periorteal reaction was present in the contralateral limb.
The femoral diaphysis immediately proximal to
the femoral condyles in the OA limb of the 6 DRGACLT animals was consistently wider (mean 5 SD
diameter 22.4 2 2.2 mm) than that in the contralatera!
limb (19.9 ? 1.3 mm; P = 0.01). Some widening of the
metaphysis was also present in the OA knee, as compared with the contralateral knee (mean 5 SD diameter
37.3 2 2.0 mm and 36.4 t 1.6 mm, respectively; P =
0.03). Xeroradiography of cross-sections of the femur at
this site revealed new bone on the medial, lateral, and
anterior aspects of the shaft (Figure 2) in the OA limb of
every animal. No change in the diameter of the endosteal canal was apparent. Comparable cross-sections
of the femur obtained from neurologically intact dogs
with OA 16 weeks after ACLT showed no evidence of
periosteal new bone.
Histologic sections of the distal femur from 2
DRG-ACLT animals confirmed the formation of subperiosteal woven bone in the unstable limb (Figure 3).
No evidence of periosteal inflammation, fracture, or
subperiosteal hematoma was noted.
changes, since they were not present in radiographs of
dogs subjected to DRG but not ACLT. The proliferation
of new bone and the consequent substantial increase in
the diameter of the OA femur seen in this study differ
from the findings in the widely used ACLT model of
OA. In that model, when fluorescent compounds that
become incorporated into newly calcified tissue were
administered 2-6 weeks after ACLT to assess osteophyte formation, only a thin line of periosteal new
bone was seen on the distal femur 8 weeks after surgery
(8). The less sensitive xeroradiographic technique
used to evaluate the distal femur in the present study
showed a distinct cuff of periosteal new bone on the
ipsilateral femur after DRG-ACLT, but no such change
in specimens from neurologically intact cruciatedeficient dogs.
Bone formation is a normal response to strain or
loading of bone (9), but studies of tubular bones, e.g., rat
tibia, show that pcriosteal and endocortical bone formation can be induced selectively by varying the magnitude
and vector of a compressive or bending load applied to
the bone (10). Although skeletal loading has not been
measured directly in DRG-ACLT dogs, kinematic studies reveal abnormal extension of the knee in the deafferented limb at touchdown (ll), and suggest that
mechanical factors could contribute to the up-regulation
of periosteal bone formation.
The proliferation of periosteal new bone in
DRG-ACLT animals resembled that seen in hypertrophic osteoarthropathy (12), in that the production of
new bone was not accompanied by significant remodeling of the existing cortical bone. On the other hand, the
subchondral bony plate in the DRG-ACLT model
We describe herein the proliferation of periosteal
woven bone as the destruction of articular cartilage
proceeds in this canine model of accelerated OA. Extensive deafferentation of the canine limb alone without
knee instability is insufficient to induce these bony
Figure 3. Cross-section of distal femur from the cruciate-deficient
limb, showing apposition of woven bone (W) on the surface of the
cortex (C). No evidence of an inflammatory cell infiltrate is seen
(hematoxylin and eosin stained; original magnification X 10).
becomes thin and porous (13) and is readily distinguished from the thickened subchondral plate typically
seen in OA. Notably, in patients with NA, periosteal new
bone formation and progressive resorption of juxtaarticular bone may coexist (2,3). The mechanisms underlying
the abnormal bone production and resorption in this
conditions could involve circulating mediators of osteogenesis, although the success of vagotomy or adrenergic
blockade in reducing the severity of symptoms of some
patients has raised the possibility that neurogenic factors
could be involved (2,12).
Bone and periosteum are innervated by a complex network of proprioceptive nerve endings, sensory
afferent neurons, and sympathetic neurons that can
deliver neuropeptides to the joints and periarticular
tissues. Considerable evidence indicates that these mediators, including vasoactive intestinal peptide, neuropeptide Y, calcitonin gene-related peptide, substance
P, and norepinephrine, can affect bone metabolism
(14,15), and that neural lesions interfere with bone
physiology (16). Because DRG extensively disrupts sensory input from the knee to the ccntral nervous system,
this procedure can be expected to significantly alter the
spectrum of neuropeptides delivered to the periosteum.
Evidence that this is the case would support our speculation that the neurologic deficit produced in the DRGACLT model alters the mechanical threshold for bone
formation, and thereby enhances periosteal new bone
Slowman-Kovacs SD, Braunstein EM, Brandt KD: Rapidly progressive Charcot arthropathy following minor joint trauma i n
patients with diabetic neuropathy. Arthritis Rheum 33:412-417,
Resnick D: Neuropathic osteoarthopathy. In, Diagnosis of Bone
and Joint Disorders. Edited by D Resnick. Philadelphia, WB
Saunders, 1995
3. Brower AC, Allman RM: The neuropathic joint: a neurovascular
bone disorder. Radio1 Clin North Am 19:571-580, 1981
4. O’Connor BL, Palmoski MJ, Brandt KD: Ncurogenic acceleration
of dcgcncrative joint lesions. J Bone Joint Surg Am 67A:562-571,
5. O’Connor BL, Visco DM, Brandt KD, Albrecht M, O’Connor AB:
Sensory nerves only temporarily protect the unstable canine knee
joint from osteoarthritis: evidence that sensory nervcs reprogram
the central nervou? system after cruciatc ligament transection.
Arthritis Rheum 36: I 154-1 163, 1993
6. Yu LP Jr, Smith GN Jr, Brandt KD, Myers SL, O’Connor BL,
Brandt DA: Reduction of the severity of caninc osteoarthritis by
prophylactic treatment with oral doxycyclinc. Arthritis Rheum
35:1150-115O, 1992
7 Vogelwcid CM, Dreesen J: Radiographic imaging of rodents using
a portable mammographic system. Contemp Top 3436-87, 1095
8. Gilbertson EMM: Development of pcriarticular ostcophytcs in
experimentally induced osteoarthritis in the dog. Ann Rheum Dis
34: 12-25, 1975
9 Burr DB, Schaffler MB, Yang KH, Lukoschek M, Sivaneri N,
Blaha JD, Radin EL: Skeletal change in response to altered strain
environments: is woven bone a response to elevated strain? Bone
10:223-233, 1989
10 Turner CH, Forwood MR, Rho J-Y, Yoshikawa T: Mechanical
loading thresholds for lamellar and woven bone formation. J Hone
Miner Res 9:87-97, 1994
11 Vilensky JA, O’Connor BR, Brandt KD, Dunn EA, Rogers PI:
Serial kinematic analysis of the canine knee after L4-Sl dorsal root
ganglionectomy: implications for the cruciate deficiency model of
ostcoarthritis. J Rheumatol 21:2113-2117, 1994
12 Segal AM, Mackenzic AH: Hypertrophic osteoarthropathy: a
10-year retrospective analysis. Semin Arthritis Rheum 12:220-232,
13 Dedrick DK, Goulet RW, O’Connor BL, Brandt KD: Preliminary
report: increased porosity of the subchondral plate in an accelerated canine model of ostcoarthritis. Osteoarthritis Cartilage 5:7174, 1997
14 Hill EL, Elde R: Distribution of CGRP-, VIP-, D beta €I-, SP-, and
HPY-immunoreactive nerves in the periosteum of the rat. Cell
Tissue Res 264:469-480, 1991
15 Hukkanen M, Konttinen YT, Rees RG, Gibson SS, Santavirta S.
Pollak JM: Innervation of bone from healthy and arthritic rats by
substance P and calcitonin gene-related peptide containing sensory fibers. J Rheumatol 19:1252-12.50, 1992
16 Hukkancn M, Konttinen YT, Santavirta S, Paavolainen P, Gu XH,
Terenghi G, Polak JM: Rapid proliferation of calcitonin generelated peptidc-immunoreactivc nerves during healing o f rat tibia1
fracture suggests neural involvement in bone growth and healing.
Neuroscience 54969-979, 1993
Без категории
Размер файла
445 Кб
periosteum, mode, formation, neuropathy, osteoarthritis, new, canine, bones
Пожаловаться на содержимое документа