03-06-1076.

qxd 6/24/2004 11:42 AM Page 233

Use of the caudolateral curvilinear osteophyte

SMALL ANIMALS
as an early marker for future development
of osteoarthritis associated with hip dysplasia
in dogs
Michelle Y. Powers, DVM; Daryl N. Biery, DVM, DACVR; Dennis F. Lawler, DVM;
Richard H. Evans, DVM, MS; Frances S. Shofer, PhD; Philipp Mayhew, BVM&S;
Thomas P. Gregor, BS; Richard D. Kealy, PhD; Gail K. Smith, VMD, PhD

joint becomes functionally lax (subluxation during

weight bearing), a cascade of events begins and leads to
Objective—To determine the relationship between cartilage degeneration and osteoarthritis.
the caudolateral curvilinear osteophyte (CCO) and The first radiographic signs of hip dysplasia have
osteoarthritis associated with hip dysplasia in dogs. been described as femoral head subluxation (laxity) and
Design—Longitudinal cohort study. a lag in development of the acetabular rim that can be
Animals—48 Labrador Retrievers from 7 litters. detected as early as 30 to 60 days of age.1 In dogs as
Procedure—In each of 24 sex- and size-matched pairs young as 16 weeks of age, the degree of hip joint laxity
fed the same diet, a restricted-fed dog was fed 25% is a risk factor for osteoarthritis.2 It has been suggested
less than a control dog for life. The dogs’ hips were that subluxation of the coxofemoral joint during weight
evaluated in the standard ventrodorsal hip-extended bearing places tension on the joint capsule, leading to
radiographic projection at 16, 30, and 52 weeks of age trauma and thickening of the joint capsule, and new
and then yearly for life. Histologic examination of hip bone (osteophytosis) forms in the capsule along the cap-
joint tissues was performed on 45 dogs. sular border.3 As described by others,1,4-7 the caudolater-
Results—Median age at death was 11.2 years. al curvilinear osteophyte (CCO) on the femoral neck is
Adjusting for feeding group, dogs with a CCO were 1 site of this osteophytosis. It has also been referred to as
3.7 times as likely to develop radiographic signs of Morgan’s line.6 The CCO is a curvilinear radiopaque line
osteoarthritis than those without a CCO. Stratified by
diet, 100% of the control dogs with a CCO developed
that varies in thickness and length. It is superimposed
radiographic signs of osteoarthritis and 55% of on the caudolateral surface of the femoral neck on the
restricted-fed dogs with a CCO developed radio- ventrodorsal hip-extended radiograph. Presently, this
graphic signs of osteoarthritis. The CCO was the first radiographic sign is not considered indicative of
radiographic change seen in 22 of 29 (76%) dogs with osteoarthritis or CHD in the absence of subluxation or
osteoarthritis. Overall, 35 of 37 (95%) dogs with a other radiographic signs of osteoarthritis.
CCO had histopathologic lesions of osteoarthritis. Slow progress in decreasing the incidence of CHD
Conclusions and Clinical Relevance—Results indi- during the last 30 years has been attributed to several
cate a relationship between a CCO on the femoral neck factors; among them are the insensitivity for detection
and subsequent development of radiographic signs of of hip joint laxity by use of the standard ventrodorsal
osteoarthritis in Labrador Retrievers evaluated over hip-extended projection,8,9 variation among evalua-
their life span. A CCO is an important early radiograph- tions by radiologists,10 and the late appearance of radio-
ic indication of osteoarthritis associated with canine hip graphic signs pathognomonic for osteoarthritis of CHD
dysplasia. (J Am Vet Med Assoc 2004;225:233–237)
that are not seen until middle age or older.5,11,12
Common radiographic criteria used for diagnosing
osteoarthritis of CHD are periarticular osteophytosis,
C anine hip dysplasia (CHD) is a complex, poly-
genic, multifactorial disease that results in
osteoarthritis of the hip joints. The degree of involve-
subchondral bone sclerosis, and joint remodeling.
In addition, environmental factors such as body
ment ranges from minute changes in cartilage and syn- weight have been reported by Kealy et al11,13,14 to delay
ovium to marked remodeling of the hip joint. If the the onset and severity of radiographic signs of
osteoarthritis. Programs that promote the eradication
From the Department of Clinical Studies, School of Veterinary of CHD use phenotypic radiographic markers to judge
Medicine, University of Pennsylvania, Philadelphia, PA 19104-
6010 (Powers, Smith, Shofer, Mayhew, Gregor); Nestlé Purina Pet
whether a dog is clinically normal or has CHD; breed-
Care & Statistical Service, Checkerboard Square, St Louis, MO ers then make the decision whether that dog is to be
63164 (Lawler, Kealy); and Veterinary Pathology Service, 30802 bred or not. Therefore, subtle changes in radiographic
S Coast Hwy, H-1, Laguna Beach, CA 92651 (Evans). Dr. Powers’ appearance are being explored to find more reliable
present address is the Department of Veterinary Clinical Sciences, indicators to diagnose CHD in young dogs.
College of Veterinary Medicine, Washington State University, Recognition of early markers of osteoarthritis would
Pullman, WA 99164-6610.
Supported by and conducted at Nestlé Purina Company.
help to reduce the pool of dogs with false-negative
Presented in abstract form at the Annual Conference of the results that are used for breeding or working purposes.
Veterinary Orthopedic Society, The Canyons, Utah, March 2002. The importance of the CCO as an early indicator
Address correspondence to Dr. Powers. of CHD has been a topic of controversy since its first

JAVMA, Vol 225, No. 2, July 15, 2004 Scientific Reports: Original Study 233

Unauthenticated | Downloaded 06/12/24 11:54 PM UTC

03-06-1076.qxd 6/24/2004 11:42 AM Page 234

description in 1961.4 Some radiologists consider the weights after 8 years were 33.7 ± 6.02 kg and 24.2 ±
SMALL ANIMALS

CCO as the earliest radiographic change, whereas oth- 3.79 kg (74.1 ± 13.2 lb and 53.2 ± 8.3 lb; P < 0.01) for
ers dismiss it as an incidental finding, particularly in CF and RF dogs, respectively.14 Throughout the study,
the absence of subluxation. Recent evidence indicates mean body weight of dogs in the 2 groups closely par-
that dogs with a CCO are 7.9 times more likely to have alleled differences in caloric intake (ie, weight of RF
definitive radiographic signs of osteoarthritis contem- dogs was approx 75% of weight of CF dogs). Mean age
poraneously.7 That study evaluated 25,986 dogs of dif- at appearance of the CCO was 2.3 and 4.0 years in the
ferent ages and breeds at 1 point in time, and no longi- CF and RF groups, respectively.
tudinal relationship was established. The authors also Of the 24 dogs in the RF group, 20 had a CCO and
noted a less discrete radiopaque line and termed it a 4 did not (Figure 1). Eleven of 20 (55%) dogs with a
puppy line. The puppy line is seen in the same location CCO developed radiographic signs of osteoarthritis,
as the CCO (in dogs < 18 months of age) but has a whereas only 1 of 4 dogs that did not have a CCO
slightly different appearance than the CCO. developed radiographic signs of osteoarthritis.
The purpose of the study reported here was to eval- Of the 24 dogs in the CF group, 18 had a CCO and
uate in a fixed cohort of dogs whether the CCO and 6 did not. All of the dogs with a CCO developed radio-
puppy line were risk factors for the development of graphic signs of osteoarthritis. Two of the 6 dogs with-
osteoarthritis of the hip joint later in life. out a CCO developed radiographic signs of osteoarthri-
tis. Therefore, 38 of all dogs in the study had a CCO, of
Materials and Methods which 29 (76%) developed radiographic signs of
Forty-eight Labrador Retrievers, the progeny from 7 lit- osteoarthritis by the end of study (Figure 2).
ters, were used in a life-long, paired-feeding study.14 Their fam- Among dogs with a CCO, 29 of 38 developed
ily lines had high prevalences of CHD. Dogs were paired by sex other radiographic changes of osteoarthritis. The CCO
and body weight within each litter prior to random assignment was the first radiographic change seen in 22 of 29
to dietary treatment. Each pair of dogs was housed in 2 X 19-m (76%) dogs that developed radiographic signs of
indoor-outdoor kennel runs with concrete floors. The amount
of exercise was not controlled. All dogs were fed the same dry osteoarthritis. In 27 of 29 (93%) dogs with a CCO, the
diet. Each restricted-fed (RF) pair mate received 75% of the CCO either was the first sign or appeared simultane-
amount of the food consumed by the corresponding control- ously with other radiographic changes.
fed (CF) pair mate. Diet analysis and feeding schedules were as Adjusted for diet, all dogs with a CCO were 3.7
described elsewhere.13 At 3.25 years of age, 2 adjustments were times (95% CI, 1.7 to 8.0 times) more likely to devel-
incorporated into the feeding protocol. The diet was switched
from a 27% protein puppy growth formula to a 21% protein
adult formula, and the amount of food was reduced and fed on
the basis of estimated energy requirements. This change was
made to prevent insidious development of obesity in the CF
dogs. The ideal body weight for each CF dog was estimated on
the basis of skeletal size in reference to other dogs of the same
breed. The CF dogs were fed 62.1 kcal of metabolizable ener-
gy/kg of estimated ideal body weight. The RF dogs continued
to receive 75% of the amount fed to the corresponding CF pair
mate.
The dogs were administered general anesthesia and evalu-
ated radiographically by use of the standard ventrodorsal hip-
extended projection at 16, 30, and 52 weeks of age and then
annually for life. A board-certified radiologist evaluated all radi-
ographs by use of a subjective scoring system for the extent and
severity of osteoarthritis (consistent with criteria of the
Orthopedic Foundation for Animals) from the ventrodorsal hip-
extended projections. Left and right hip joints of each dog were
scored independently on the basis of sclerosis of the craniodor-
sal portion of acetabular subchondral bone, osteophytes on the
cranial aspect of acetabular margin, osteophytes on the caudal
aspect of the acetabular margin, and femoral periarticular osteo-
phytes. The CCO was not considered a sign of osteoarthritis of
the hip joint. The radiologist who scored the radiographs and
the histopathologist who performed joint studies on 45 of the 48
dogs were unaware of group assignments.11,13
Statistical analyses—A Cochran-Mantel-Haenszel test was
used to evaluate the CCO as a risk factor for osteoarthritis. Data
were evaluated for relative risk (RR) with 95% confidence inter-
vals (CIs). All analyses were performed by use of a statistical
software program.a

Results Figure 1—Ventrodorsal radiographic view of a caudolateral curvi-

linear osteophyte (CCO; arrow) on the femoral neck of a dog
Median life span was 11.2 years for CF dogs and with otherwise good hip joint conformation. Notice that the
13.0 years for RF dogs (P < 0.01). Mean ± SD body CCO appears as a thick radiopaque curvilinear line.

234 Scientific Reports: Original Study JAVMA, Vol 225, No. 2, July 15, 2004

Unauthenticated | Downloaded 06/12/24 11:54 PM UTC

03-06-1076.qxd 6/24/2004 11:42 AM Page 235

SMALL ANIMALS
Figure 2—Ventrodorsal radiographic view of a CCO (arrow) on
the femoral neck of a dog with otherwise good hip joint confor- Figure 3—Ventrodorsal radiographic view of a radiopaque line
mation. Notice that the CCO appears as a well-defined curvilin- (ie, a puppy line [arrow]) in a young dog. Notice that the puppy
ear line that is less radiopaque than the CCO in Figure 1. line is more subtle, diffuse, and shorter, compared with the
CCOs in Figures 1 and 2.

op radiographic signs of osteoarthritis than those with-

out a CCO (P < 0.01). Stratified by diet, 100% of dogs femoral neck as the CCO. However, it was more sub-
in the CF group with a CCO developed radiographic tle, more diffuse, and shorter, compared with the CCO
signs of osteoarthritis (P = 0.001), whereas only 55% of (Figure 3). Mean age at which it was first visible was
RF dogs with a CCO developed radiographic signs of 4.3 and 5.5 months in the CF and RF dogs, respec-
osteoarthritis (P = 0.59). tively. The puppy line was not seen after 17 months of
Necropsies and histologic examinations of the age. The proportion of dogs with a puppy line that
coxofemoral joint and surrounding soft tissues were were ≤ 5 months of age when the line was first seen
performed on all 48 dogs. Coxofemoral joints of 45 of was 16 of 24 and 11 of 24 for the CF and RF groups,
these dogs were available for histopathologic evalua- respectively. Ignoring diet, dogs with a puppy line
tion. Three dogs died prior to 8 years of age. Tissues were not more likely to develop radiographic signs of
were obtained and fixed in chilled neutral-buffered osteoarthritis than those without a puppy line (RR,
10% formalin at 1:20 volumes. The bony tissues were 0.78; 95% CI, 0.5 to 1.1) and were not more likely to
decalcified, embedded, sectioned, and stained. develop a CCO (RR, 0.88; 95% CI, 0.7 to 1.2). When
Of the 19 dogs in the RF group with a CCO and for stratified by diet, dogs with a puppy line (both
which results of histopathologic examinations of hip groups) were not more likely to develop osteoarthritis
joints were available, 18 had histopathologic evidence than dogs without a puppy line (RR, 0.93 and 0.59 for
of osteoarthritis. In the CF group, 17 of 18 dogs with a CF and RF groups, respectively). Similarly, dogs with
CCO had histopathologic changes consistent with a puppy line (both groups) were not more likely to
osteoarthritis. Among all dogs in the study, 35 of the 37 develop a CCO than dogs without a puppy line (RR =
(95%) dogs that had a CCO and for which histologic 0.78 and 0.97 for CF and RF groups, respectively).
findings were available developed histologic evidence
of osteoarthritis (Figure 2). Of the 2 dogs for which no Discussion
histologic evidence of osteoarthritis was reported, 1 This study attempted to determine the association
dog did have obvious radiographic changes of between a CCO and radiographic and histologic signs
osteoarthritis. The affected sites likely were missed of osteoarthritis of the hip joint. Because all but 2 of
during the histologic sectioning process. Among 10 the dogs that we evaluated eventually developed
dogs that did not have a CCO, histologic findings were histopathologic lesions of osteoarthritis, the inherent
available for only 8 dogs. Five of 8 dogs did not have a selection bias associated with lack of sufficient
CCO and developed radiographic evidence of histopathologically normal joints (albeit inadvertent)
osteoarthritis, although all 8 dogs had histologic evi- precluded a full assessment of the clinical importance
dence of osteoarthritis. of a CCO. The initial study design was intended to
A so-called puppy line was seen radiographically examine the effects of limited food intake on subluxa-
on the ventrodorsal hip-extended projection in 39 of tion of the femoral head and the radiographic appear-
48 (81%) dogs when they were < 1 year of age. This ance of secondary changes up to 2 years of age. The
radiographic sign was visible in the same region of the study dogs derived from mating pairs included 5 clini-

JAVMA, Vol 225, No. 2, July 15, 2004 Scientific Reports: Original Study 235

Unauthenticated | Downloaded 06/12/24 11:54 PM UTC

03-06-1076.qxd 6/24/2004 11:42 AM Page 236

cally normal dogs, 3 dogs with CHD, and 1 dog with CCO, variability in positioning for the hip-extended
SMALL ANIMALS

unknown phenotype. The mating of a clinically normal view (such as external rotation) that can obscure the
dog and a dog with CHD would produce approximate- CCO,6 or severe radiographic changes of osteoarthritis
ly 26% to 51% offspring with CHD.13 In addition, at the that obscure the CCO. Although the hip-extended
time of this study, the distraction index (DI) was not view is the most sensitive radiographic projection for
used to select parents. In the cohort of 48 dogs, the detecting the CCO, the sensitivity is undeniably
lowest DI was 0.36.15 Consequently, there were no dogs < 100%. Therefore, if the CCO is present, it likely rep-
that had a DI < 0.3; therefore, all dogs were at risk to resents an early sign of osteoarthritis of the hip joint.
develop osteoarthritis.8 Our study revealed that the However, lack of the CCO does not confirm that a dog
CCO had high sensitivity, but in the absence of dogs is not susceptible to osteoarthritis of the hip joint.
with low DI in our sample, it was not possible to deter- Given the high frequency of CHD, we believe it is pru-
mine the specificity of CCO for prediction of dent to consider even a faint CCO as cause for suspi-
osteoarthritis. Another study7 revealed an association cion of CHD and degenerative joint disease. Multiple
between the CCO and DI, with < 20% incidence of follow-up radiographic evaluations throughout the life
CCO in dogs with low DI, thus corroborating the find- of the dog would be indicated in these instances.
ings in the longitudinal investigation reported here. A radiographic finding that closely resembles the
Results of our study support a relationship between CCO has been termed the puppy line.7 This line was
a CCO on the femoral neck and subsequent develop- observed in our study in dogs ≤ 18 months of age and
ment of osteoarthritis of the hip joints in Labrador then either disappeared or transformed to a CCO. This
Retrievers evaluated over their life span. Another study7 line is thought to be nonpathologic and should not be
revealed that dogs with a CCO had 7.9 times the risk of confused with a CCO. In our study, there was no sta-
having radiographic signs of osteoarthritis, compared tistical relationship between dogs having a puppy line
with dogs without a CCO. In our study, dogs with a and later development of osteoarthritis. Results indi-
CCO had 3.7 times the risk of developing radiographic cated that regardless of feeding group, dogs with a
signs of osteoarthritis; 95% of these dogs developed his- puppy line were not more likely to develop osteoarthri-
tologic evidence of osteoarthritis. tis or a CCO, compared with dogs without the line. A
It has been hypothesized that the CCO is a result clear distinction between CCO (an indicator of
of excessive laxity of the hip joint,3,6 which causes osteoarthritis) and a puppy line (presently believed to
osteophytes to form within the insertion lines of the be a nonpathologic finding) is needed so that dogs are
joint capsule. On a ventrodorsal hip-extended radio- not unnecessarily disqualified as potential breeding or
graphic view, the CCO is seen extending as a curvilin- working animals. A puppy line should not be consid-
ear line superimposed on the femoral neck. The CCO ered an indication for medical or surgical treatment in
can be uni- or bilateral and is variable in radiographic the absence of appropriate clinical signs. Dogs ≤ 18
prominence, ranging from a faint radiographic opacity months of age should be reevaluated at an older age to
to a thicker radiopaque line. confirm the presence of a CCO or other radiographic
Results of necropsies reveal that the bony prolifer- signs of osteoarthritis. Environmental factors have the
ation that occurs at the site of the CCO is much more potential to influence positively or negatively the phe-
prominent than that visible radiographically.5 Its loca- notype of dogs that are genotypically predisposed to
tion and appearance can also be affected by changes in hip dysplasia. Interestingly, restricting the food intake
the positioning of the dog in the conventional hip- of Labrador Retrievers delayed or prevented the radi-
extended radiograph, especially if external femoral ographic expression of osteoarthritis,12,14,17 but the diet
rotation occurs.6 did not influence the frequency of CCOs; only time of
The CCO as a single finding noted on a radiograph onset of the CCO was affected. This may suggest that
should be considered of importance. We believe that the CCO is a more sensitive radiographic marker for
the CCO is early evidence of secondary degenerative susceptibility to osteoarthritis that is not confounded
joint disease caused by joint laxity. by environmental factors such as restricted feeding.
There is no gold standard for diagnosing CHD. Studies examining the heritability of the CCO will help
Recently, arthroscopy has been used to evaluate hip determine the variability of the CCO as a function of
joints for signs of osteoarthritis, but arthroscopy is a environmental factors and indicate whether the CCO is
more invasive technique than radiography and far indeed a radiographic phenotype of CHD that is close-
more expensive. Studies using arthroscopy may help to ly linked to the dog’s genotype.
further evaluate the development of the CCO, but until
such data are available, histopathologic examination is a
SAS statistical software, version 8.0, SAS Institute Inc, Cary, NC.
the best technique for evaluation of osteoarthritis. Of
the 45 dogs for which histopathology results were References
available, 43 (96%) dogs had histologic evidence of 1. Riser WH. The dog as a model for the study of hip dyspla-
osteoarthritis. Two dogs without histologic evidence of sia. Philadelphia: S. Karger AG, Arnold-Bocklin-Strasse, 1975;
osteoarthritis also had a CCO; 1 of these dogs had 229–334.
obvious radiographic changes of osteoarthritis, and 2. Smith GK, Popovitch CA, Gregor TP, et al. Evaluation of
risk factors for degenerative joint disease associated with hip dyspla-
this discrepancy was thought to be attributable to a sia in dogs. J Am Vet Med Assoc 1995;206:642–647.
sectioning artifact. All 8 dogs that did not have a CCO 3. Riser WH, Rhodes WH, Newton CD. Hip dysplasia: theories
also developed histologic evidence of osteoarthritis. of pathogenesis In: Newton C, ed. Textbook of small animal orthope-
This could result from variability in radiopacity of the dics. Philadelphia: JB Lippincott Co, 1985;953–980.

236 Scientific Reports: Original Study JAVMA, Vol 225, No. 2, July 15, 2004

Unauthenticated | Downloaded 06/12/24 11:54 PM UTC

03-06-1076.qxd 6/24/2004 11:42 AM Page 237

4. Whittington K, Banks WC, Carlson WD, et al. Report of 11. Kealy RD, Lawler DF, Ballam JM, et al. Five-year longitudi-

SMALL ANIMALS
panel on canine hip dysplasia. J Am Vet Med Assoc 1961;139:791–806. nal study on limited food consumption and development of
5. Ackerman N. Hip dysplasia in the Afghan Hound. Vet Radiol osteoarthritis in coxofemoral joints of dogs. J Am Vet Med Assoc 1997;
1982;23:88–97. 210:222–225.
6. Morgan JP. Canine hip dysplasia: significance of early bony 12. Smith GK. Influence of body condition on canine
spurring. Vet Radiol 1987;28:2–5. osteoarthritis (degenerative joint disease), in Proceedings. Purina
7. Mayhew PD, McKelvie PJ, Biery DN, et al. Evaluation of a Nutr Forum 2001;9–14.
radiographic caudolateral curvilinear osteophyte on the femoral 13. Kealy RD, Olsson SE, Monti KL, et al. Effects of limited
neck and its relationship to degenerative joint disease and distraction food consumption on the incidence of hip dysplasia in growing dogs.
index in dogs. J Am Vet Med Assoc 2002;220:472–476. J Am Vet Med Assoc 1992;201:857–863.
8. Smith GK, Gregor TP, Rhodes WH, et al. Coxofemoral joint 14. Kealy RD, Lawler DF, Ballam JM, et al. Evaluation of the
laxity from distraction radiography and its contemporaneous and effect of limited food consumption on radiographic evidence of
prospective correlation with laxity, subjective score, and evidence of osteoarthritis in dogs. J Am Vet Med Assoc 2000;217:1678–1680.
degenerative joint disease from conventional hip-extended radiogra- 15. Smith GK, Kealy RD, Biery DN, et al. Effects of restricted
phy in dogs. Am J Vet Res 1993;54:1021–1042. feeding on onset, incidence and severity of hip dysplasia and
9. Adams WM, Dueland TR, Meinen J, et al. Early detection of osteoarthritis in dogs: diagnostic, therapeutic, and genetic ramifica-
canine hip dysplasia: comparison of two palpation and five radio- tions, in Proceedings. Nestlé Purina Symp 2002;21–26.
graphic methods. J Am Anim Hosp Assoc 1998;34:339–347. 16. Ackerman N, Nyland T. Radiographic diagnosis of canine
10. Smith GK, Biery DN, Rhodes WH, et al. Between- and with- hip dysplasia. Calif Vet 1977;9–15.
in-radiologist accuracy of subjective hip scoring of the ventrodorsal 17. Kealy RD, Lawler DF, Ballam JM, et al. Effects of diet restric-
hip-extended radiograph, in Proceedings. Int Symp Hip Dysplasia tion on life span and age-related changes in dogs. J Am Vet Med Assoc
Osteoarthritis Dogs 1996. 2002;220:1315–1320.

Selected abstract for JAVMA readers from the

American Journal of Veterinary Research
Estimation of heritability of atopic dermatitis in Labrador and Golden Retrievers
Stephen C. Shaw et al

Objective—To estimate the heritability of atopic dermatitis in Golden and Labrador Retrievers.
Animals—429 dogs related to 13 dogs with atopic dermatitis. July 2004
Procedure—Atopic dermatitis was defined on the basis of the type and frequency of clinical signs
recorded in the clinical records, and each dog was classified with atopic dermatitis or probable atopic
dermatitis or as nonatopic. By use of data from atopic and nonatopic dogs, regression analyses of
See the midmonth
parental status on offspring status were performed to estimate heritability. issues of JAVMA
Results—There was no difference in the frequency of atopic dermatitis between sexes or between for the expanded table
breeds. There was a marked association between the atopic status of the parent and that of the off- of contents
spring, particularly for sires. By use of data from 32 litters in which the status of both parents was
known and considering only those dogs classified with atopic dermatitis or as nonatopic, the heri- for the AJVR
tability (± SE) of atopic dermatitis was estimated to be 0.47 (± 0.17). or log onto
Conclusions and Clinical Relevance—Atopic dermatitis has a strong genetic component, and www.avma.org
breeding of dogs with clinical signs of atopic dermatitis should be discouraged. (Am J Vet Res 2004;
65:1014–1020) for access
to all the abstracts.

JAVMA, Vol 225, No. 2, July 15, 2004 Scientific Reports: Original Study 237

Unauthenticated | Downloaded 06/12/24 11:54 PM UTC