DNM1 mutation status, sex, and sterilization status of a cohort of Labrador retrievers with and without cranial cruciate ligament rupture
© The Author(s). 2017
Received: 25 October 2016
Accepted: 26 January 2017
Published: 2 February 2017
Exercise-induced collapse (EIC) due to DNM1 mutation and rupture of the cranial cruciate ligament are both common syndromes in the Labrador retriever breed. A cohort of 313 Labradors was recruited based on their CCLR status and were subsequently genetically tested for EIC. Epidemiological aspects of the cohort were also described, including sex, sterilization status, and age at sterilization.
No sex difference was observed in dogs susceptible to EIC (homozygous for the mutant genotype) compared to dogs not susceptible to EIC (heterozygotes and dogs homozygous for the normal genotype). No evidence for association was detected between CCLR status and EIC status (p =0.357), although the sample cohort was not of sufficient size to entirely rule out an association. A significant difference (p = 0.031) was observed in the sex distribution of dogs affected with CCLR compared to those without CCLR. An increased number of female CCLR cases were observed compared to the number of female controls; male CCLR cases and controls were approximately the same number. When CCLR status was examined in each sex, no significant differences were observed between those that were sterilized and those that weren’t. However, for female dogs that were sterilized, CCLR cases were significantly higher in dogs sterilized at one year of age or younger compared to those sterilized when over the age of one year (p = 0.0021, OR 4.30, 95% CI 1.55–12.72); for males, this finding was suggestive, but not statistically significant (p = 0.0913, OR 3.57, 95% CI 0.809–14.476).
CCLR is not associated with a large increase in EIC occurrence. Statistically, these two syndromes cannot be proven to be unrelated; however, concomitant occurrence of CCLR and EIC in Labrador retrievers is rare, despite the high prevalence of both syndromes in this breed. Epidemiological findings suggest that females may be over-represented in CCLR cases and that early sterilization (≤1 year) may increase the risk of Labradors developing CCLR later in life (particularly in females). These results should be considered preliminary and require confirmation in larger populations of Labrador retrievers.
KeywordsExercise-induced collapse Cranial cruciate ligament Dog Sex Sterilization Neuter Spay
Plain english summary
Tear or rupture of the cranial cruciate ligament (CCL) in dogs is equivalent to humans tearing their anterior cruciate ligament (ACL), a knee ligament. CCL rupture (CCLR) in dogs can occur from trauma, but may also have genetic risk factors, because some breeds seem to be predisposed, including the Labrador Retriever. The definitive cause of CCLR has not been determined, but probably has many factors.
Labrador retrievers affected with exercise-induced collapse (EIC) have two defective copies of the DNM1 gene. A genetic test can determine if a collapse episode is due to the DNM1 mutation and predicts which dogs are susceptible to this form of collapse. During an EIC episode, a dog will usually first lose control of its hindlimbs, and may stumble while trying to continue running, all of which could stress the knee joint. Further, the DNM1 mutation interferes with nerve signal transmission, which might impact the body’s ability to protect the integrity of joints, such as the knee. We wondered if EIC and CCLR, which are both common in the Labrador retriever breed, might be related. Therefore, we assembled a group of Labradors that were carefully screened to either be cases (have torn their CCL) or controls (normal knees) and subsequently genetically tested them for EIC. In addition, this was a good opportunity to describe the sex, sterilization status, and age-at-sterilization for this group of dogs.
Our results show that dogs with CCLR do not have EIC in a markedly higher frequency compared to dogs without CCLR. Therefore, if both syndromes are observed in the same dog, it’s probably just a coincidence due to how common both conditions are in the breed. From the epidemiological data, we observed a higher proportion of females in the CCLR cases compared to males. When we looked only at sterilized dogs, and divided them by what age they were sterilized, we saw a significantly higher proportion of CCLR cases in those dogs that were spayed or neutered early in life (before 1 year of age) in females. Ideally, these epidemiological findings should be confirmed in a larger population of dogs.
Exercise-induced collapse (EIC) and cranial cruciate ligament rupture (CCLR) are often observed in active Labrador retrievers. Dogs with EIC are normal when resting, but can manifest a ‘wobbly’ gait, which typically progresses to loss of control (flaccid paraparesis) of the hindlimbs, after a short duration of strenuous exercise . The collapse episode typically resolves within 5–30 min . This characteristic manifestation of collapse has an autosomal recessive mode of inheritance and has been strongly associated with a mutation in the dynamin 1 gene (DNM1) . Depending on the sub-population of Labradors examined, the percentage of EIC homozygous mutant dogs varies from 1.8% (service Labradors - bred to assist humans with specific needs) to 13.6% (conformation Labradors - bred for showing), and the percentage of heterozygous (carrier) dogs varies from 17.9% (service Labradors) to 38.0% (field trial/hunt test Labradors) , indicating that this mutation and syndrome are common within this breed.
Cranial cruciate ligament rupture (CCLR) is a significant cause of pelvic limb lameness , with affected dogs displaying pain, instability of the joint, and altered kinematics . A breed predisposition for CCLR has been reported in Labrador retrievers, with a 5.5-fold increased risk of developing CCLR . Labrador retrievers and Labrador crosses are reported to account for 23% of all large breed CCLR cases (out of 228) in one study , and the Labrador retriever is reported as the most common breed (16% of cases, 70 out of 426 dogs) observed with CCLR in another study . The definitive etiology of non-traumatic (i.e. idiopathic) CCLR in dogs is unknown, but it is likely multifactorial, and it may include genetic influences .
Dogs experiencing EIC-specific collapse lose coordination, often use a crouched posture (knee flexion) with their pelvic limbs , and may drag themselves by their forelimbs. According to Taylor et al., dogs with EIC often continue to run, dragging their crouched rear legs despite being in a collapse episode . Taken together, these scenarios may subject the cranial cruciate ligament (CCL) to unknown additional mechanical stressors or damaging forces that could influence CCL integrity. Further, since the DNM1 protein, dynamin 1, functions in synaptic transmission, and genetically susceptible EIC dogs experiencing a collapse episode lose proper control of neurotransmission in the hindlimbs, this prevention of neurological input to the muscles may compromise the integrity of the CCL in affected dogs. The objective of this study was to determine if there is an association between these two common problems in the Labrador retriever breed by assembling a cohort of well-phenotyped CCLR cases and controls, which we would then genetically test for EIC, and from which we could also examine and report epidemiological aspects, such as sex and sterilization status.
Animals and phenotyping
North American dogs were recruited for this study as either CCLR cases or controls, with no consideration to sex or sterilization status, and were subsequently tested to determine their EIC genetic status. Therefore, no phenotype information was obtained regarding any dogs’ collapse status. After informed client consent was obtained, and each dog’s purebred Labrador retriever status was established (either by registration in or eligibility for registration in the American Kennel Club or the Canadian Kennel Club), case or control status was determined by stifle joint palpation by a board certified veterinary surgeon. The vast majority of CCLR cases (>90%) were defined by surgical confirmation of CCL tear, either unilateral or bilateral, with the remaining cases defined via physical exam, stifle palpation, and radiographs. Controls were defined as greater than 7 years of age  with no history of pelvic limb lameness or stifle surgery of any kind, and having no abnormalities in either knee via orthopedic examination (palpation only). Ages of case dogs ranged from 11 months to 15 years at the time of enrollment, although variable amounts of time may have elapsed post-injury. Information about each dog’s sex, sterilization status, and the age at which the dog was sterilized (divided into two categories: ≤ 1 year and > 1 year) was collected, however, all of these data were not available for every dog. When available, each dog’s age at sterilization, typically provided by the owner, was recorded. Dogs were recruited between 2010–2011 by the authors at the University of Minnesota’s referral hospital; a sub-set of samples were recruited during the same time frame by veterinary surgeons at the following specialty referral hospitals: Upstate Veterinary Specialties Clinic, in Latham, NY; Western College of Veterinary Medicine, in Saskatchewan, Canada, and Blue Pearl Referral Medicine, in Eden Prairie, MN. Controls were recruited contemporaneously with cases at all locations. The cohort consisted primarily of pet line Labradors.
DNA was obtained either from whole blood or cheek swabs using standard extraction methods. Each DNA sample was calibrated to a standardized concentration and subjected to PCR amplification of DNM1 exon 6, which contains the EIC mutant allele. PCR products were digested with restriction enzyme SmlI and visualized on 2% agarose gels as previously described . EIC genotypes were categorized as either two copies of the mutant allele (homozygous E or EE), two copies of the normal allele (homozygous N or NN), or one copy of each allele (heterozygous/carrier, EN).
The prevalence of EIC susceptibility in the entire cohort was calculated. Prevalence of CCLR in the Labrador breed could not be accurately calculated with this study due to ascertainment bias in recruiting for affected dogs. Descriptive statistics were performed for the distributions of sex for CCLR cases versus controls and for EIC genetic susceptibility; descriptive statistics were also performed for sterilization status and age at sterilization for CCLR cases versus controls. A Fisher’s exact test was used to test for association between sex and CCLR, sex and EIC, EIC and CCLR, sterilization status and CCLR for both sexes, and age at sterilization and CCLR for both sexes; odds ratios (OR) with a 95% confidence interval (CI) were calculated for each test. In addition, the test for the differences in proportions was used to compute the 95% confidence intervals on the EIC distribution in CCLR cases versus controls, and the Agresti-Coull method  was used to compute the confidence intervals on the proportion of sex distribution in CCLR cases versus controls. R (version 3.2.3)  was used to perform all statistical tests, and a p-value less than or equal to 0.05 was considered significant.
CCLR case and control Labradors grouped by EIC genotype and sex
EIC Genotype and Sex
Based on their genotyping data, 11 dogs (3.5%) were susceptible to EIC (EE), 91 dogs (29.0%) were EIC carriers (EN), and 211 dogs (67.5%) were homozygous normal (NN). The sex distribution of dogs susceptible to EIC (7 females and 4 males) (Table 1) was not statistically different (p = 0.554, OR 1.53, 95% CI 0.38–7.28) from the sex distribution of dogs without EIC (EN + NN, 161 females and 141 males). This is consistent with an autosomal mutation; DNM1 is located on canine chromosome 9 . The allele frequency for the mutant (E) allele in this population was 18.05%.
CCLR case and control Labrdors grouped by EIC phenotype
EN + NN (Non-susceptible)
CCLR case and control Labradors grouped by sex and sterilization status
Sex and Sterilization Status
Sterilized CCLR case and control Labradors grouped by sex and age at sterilization
Age at Sterilization
> 1 year
> 1 year
Dogs homozygous for the EIC mutation were identified in 8 of 174 CCLR-affected Labradors (4.6%, 95% CI 2.21–8.95%) and 3 of 139 CCLR-normal Labradors (2.2%, 95% CI 0.46–6.44%), which is not a significant difference. Because the number of dogs affected with EIC in the overall pet population of Labradors is low, it is not surprising that we only detected a small number of dogs concomitantly affected with EIC susceptibility and CCLR. It was entirely unknown prior to this study how frequently dogs would be concomitantly affected with EIC and CCLR, therefore several hundred were recruited. Ultimately, though, this study was under-powered to draw strong conclusions about association of EIC and CCLR, due to the relative rarity of dogs with both conditions. This sample cohort demonstrates that CCLR and EIC-susceptibility might be unrelated (p = 0.357, Fisher’s exact test), and, if they are associated, that CCLR is not associated with large increases in EIC occurrence. The test for differences in the proportion of EIC-susceptibility in CCLR cases versus CCLR controls (a 2.4% difference was observed) tells us that this difference could go as much as 2% in one direction and 7% in the other direction, and this is not likely to be a clinically significant difference.
The overall EIC genotype distribution was 67.5% homozygous for the wild-type allele, 29.0% heterozygous, and 3.5% homozygous for the mutant allele (EIC-susceptible dogs). Because our entire cohort was recruited based on CCLR status, it is inappropriate to use this group to calculate prevalence of CCLR in the Labrador breed. However, since these dogs were only subsequently tested for EIC, and they were not recruited based on collapse phenotype, this represents an opportunity to examine EIC frequency in the breed. Our finding that 3.5% of 313 dogs were EE is consistent with the overall-breed frequency of 3% as previously reported , and it is also consistent with the reported percentage of 2.9% EE in “pet” Labrador lines . The 29.0% carrier rate observed in the present study is also consistent with previously reported carrier frequencies: 37%  and 27.9% . These results confirm that the mutant EIC allele is very common in the Labrador breed.
Actual collapse status of the 11 EIC-susceptible dogs is unknown due to the lack of follow-up investigation, which was not undertaken due to the post hoc nature of the EIC testing and the fact that penetrance of EIC-susceptible dogs actually experiencing collapse has been examined in other work. The EIC-associated mutation is not fully penetrant; previous work has shown that dogs homozygous for the EE allele and therefore collapse-susceptible can be phenotypically normal and not experience collapse. One study reported 9% of EE dogs had no history of collapse , while another reported that an average of 83.6% of EE Labradors had a collapse event by 4 years of age , indicating that an average of 16.4% of homozygous mutant dogs had not collapsed by that age. The authors suggested that this could be due to the dogs never having been exposed to sufficient exercise or excitement to trigger the collapse episode, or due to modifying genetic and/or environmental factors. Therefore, it is possible to speculate that not all 11 EE dogs in the present study experienced a collapse episode.
In our cohort, which was recruited without respect to sex or sterilization status, a sex distribution difference was observed between dogs with CCLR versus dogs without CCLR. However, while this was statistically different (p = 0.031, OR 1.65, 95% CI 1.03–2.66), it may not be clinically significant. With hundreds of dogs in the study, a more precise calculation of statistical significance is possible; this can allow detection of smaller (and clinically irrelevant) differences. The group of CCLR cases was 59% female while only 47% of the CCLR control dogs were female; given the confidence interval for these proportions (38.7% to 66.2%, using the Agresti-Coull method), there could be almost twice as many female CCLR cases as female CCLR controls in the global population of Labradors retrievers. This difference, if observed, likely would have clinical significance. However, one could argue that the confidence interval also includes the possibility of 50% case:50% control for both sexes. In effect, these findings can only suggest that female dogs may be predisposed to CCLR.
Some previous studies comprising multiple breeds of dog have not detected a sex effect on rupture of the cranial cruciate ligament. For example, Duval et al. did not detect a difference in prevalence of CCLR between females and males in a sample cohort of 201 dogs . Guthrie et al. similarly reported that sex did not significantly affect the incidence of developing or presenting with bilateral CCLR in a report comprising 426 dogs representing 44 different breeds . While both of these studies included sterilization status as a variable, and Duval et al. did detect an increased risk for CCLR in neutered males and females compared to intact males and females, respectively, neither of these studies recorded the dogs’ age at sterilization, which may have resulted in different findings. Conversely, in a 2011 study by Adams et al., females were found to be twice as likely to suffer CCLR compared to males (189 cases, multiple breeds) . Very few studies have focused exclusively on Labradors with CCLR. One study looking specifically at Labrador retrievers determined that sex did not affect likelihood or rate of rupturing the contralateral CCL after tearing the first CCL, or presenting initially with bilateral CCL ruptures, however this study only represented 94 Labradors and had no CCLR-normal dogs . A Labrador-specific study by Hart et al. that included CCLR cases and controls reported 2.4% of males (19/801) were CCLR-affected and 2.5% of females (17/681) were CCLR-affected .
Taken together, it is unclear whether the statistically significant difference in sex distribution of CCLR-cases versus CCLR-controls in the present study is clinically significant or not, particularly since previous work offers mixed results and many previous studies did not consider age at sterilization as a variable. The question of sex differences for CCLR deserves further investigation. As a final perspective, it is interesting to note that the rate of anterior cruciate ligament (ACL) rupture in humans is three times higher in females compared to males . There are likely multiple reasons for this sex difference in humans.
Human orthopedic disease studies typically do not need to consider the effects of lost gonadal hormones (at least in premenopausal women), whereas the majority of US dogs are sterilized, often at a young age, and there has been concern that the early loss of gonadal hormones may influence orthopedic disease development later in life. The removal of gonadal hormones can delay long-bone growth plate closure [18, 19], which may result in stifle joint angle or other changes that predispose the dog to CCLR. One study examining CCLR across several breeds found that sterilized dogs of either gender were significantly more likely than sexually intact dogs to have CCLR . The present study had sterilization status information for all but two dogs and counters these findings, as no statistically significant differences were observed between dogs that were intact versus those that were sterilized when examining CCLR frequency separately amongst females and males (Females: p = 0.485, OR 2.25, 95% CI 0.41—22.89; Males: p = 0.141, OR 0.503, 95% CI 0.19—1.27). A more balanced population of intact dogs to sterilized dogs would allow drawing of stronger conclusions; as it stands, the confidence intervals, particularly with the females, are wide.
Since there were so few intact dogs in the present study, the subset of all sterilized dogs were examined for the age at which they sterilized, and whether this was associated with CCLR status. Differences were observed between those sterilized at a young age (≤ 1 year) versus an older age (>1 year). For both sexes, more CCLR cases were observed in dogs sterilized at a young age compared to those sterilized at an older age, and these differences were statistically significant for the females, though they were not quite statistically significant in the males.
Other studies have also examined the age at sterilization of CCLR cases. In one study, which looked exclusively at Golden Retrievers, a significantly higher percentage of both male and female early-neutered (sterilized at < 12 months) dogs had CCLR compared to intact dogs and to late-neutered (≥ 12 months) dogs . Another study, which examined exclusively German Shepherds, found a significantly higher occurrence of CCLR in both males and females that were neutered at less than one year of age compared to intact dogs of the same sex . A Labrador retriever-specific study found that males neutered at ages < 6 months had significantly more CCLR than intact males and that CCLR was increased in females with early sterilization, but not at a significant level . These studies, with the exception of the Golden Retriever study , make statistical comparisons between a neutered-young group and the intact group, while the present study is comparing the frequency of CCLR cases exclusively in sterilized dogs (less than one year versus over one year at sterilization); therefore not all results are entirely comparable.
While it is tempting to simply conclude that sterilization at a younger age predisposes a dog to CCLR, and the present cohort supports such a conclusion, it should not be ignored that there were many CCLR controls, both male and female, that were similarly neutered at a young age. Also, it is important to note that in this study typically the owner, and not a medical record, provided the age at the time of their pet’s sterilization, introducing a potential inaccuracy. In addition, if there is a relationship between age of sterilization and CCLR occurrence, there is no current biologic explanation for a dichotomous (e.g., < or > 12 months of age) relationship. It is difficult to draw firm conclusions regarding what impact sterilization, at any age, has on development of CCLR with only this cohort, but, taken together with results from previous studies, these data suggest that early sterilization might be a risk factor for eventual development of CCLR. Prospective evaluation of this association would address potential errors or bias in data collection and allow for a more precise evaluation of the relationship.
The lack of follow-up information on these dogs is a limitation of this study. Without lifetime follow-up, it is possible that a subset of any of the controls (sterilized or intact) may have developed CCLR at a time after their enrollment in the study. Additionally, a subjective body condition score (BCS) was not consistently or uniformly assessed in this cohort of dogs. Excess weight has been suggested to contribute to CCLR [6, 14], but does not appear to be predictive for whether or not a Labrador will rupture the contralateral CCL after tearing the first CCL . Certainly carrying excess weight is assumed to add to the biomechanical stresses on a joint.
Finally, given the size of the present cohort (~300 dogs, and for the age at sterilization data <200 dogs), these results should be considered preliminary, and in need of further confirmation in larger populations. Such limitations deserve to be addressed in future work attempting to better understand the risks and causes of CCLR in Labrador retrievers, which hopefully will shed further light on what risk, if any, is truly contributed by EIC status, sex, sterilization status, and age at sterilization.
We did not identify a significant association between a diagnosis of CCLR and having the EIC-susceptible genotype. There are reasons other than DNM1 mutation homozygosity which might cause a dog to collapse during exercise  any of which, along with an EIC collapse episode, could still contribute to a random individual dog tearing its CCL. But, our study was ultimately underpowered to definitively exclude association between EIC and CCLR; instead, we demonstrate that, even if the two syndromes are related, CCLR is not associated with a clinically significant increase in EIC. The high prevalence of both EIC and CCLR in the Labrador retriever population has likely resulted in the coincidental occurrence of both diseases in some individuals. Our cohort of Labradors also provided interesting epidemiological results in terms of sex distribution, sterilization status, and age at sterilization. Results suggest that females may be over-represented in the CCLR case category compared to males, and that sterilization per se is not a risk factor for development of CCLR, while early age at sterilization may increase a dog’s risk of developing CCLR, particularly for females. The finer points of these epidemiological aspects are beyond the scope of this study and deserve further investigation, with the addition of BCS data and even genetic data, in light of the multifactorial nature of CCLR.
Anterior cruciate ligament
Body condition score
Cranial cruciate ligament
Cranial cruciate ligament rupture
The authors gratefully acknowledge James Mickelson and Cathy Carlson for assistance with manuscript preparation.
This study was supported by funding from the Veterinary Orthopedic Society and the Tata Group Endowment at the University of Minnesota. The funding agencies had no part in the design of the study, the collection, analysis, and interpretation of data, or in the writing of the manuscript.
Availability of data and materials
All data for the study is included within the text and tables of this manuscript.
KJE and MGC conceived and designed the study. KJE and KMM designed and conducted the experiments. KJE conducted the analysis and wrote the manuscript. AKR provided statistical analysis. All authors have contributed to manuscript revisions and approved the final manuscript.
Katie Minor is one of the co-inventors of the EIC genetic test and receives a portion of EIC testing proceeds. Drs. Ekenstedt, Rendahl, and Conzemius declare that they have no competing interests.
Consent for publication
This study was approved by the University of Minnesota’s Institutional Animal Care and Use Committee. All dogs were enrolled and samples obtained with informed dog owner consent.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Taylor SM, Shmon CL, Adams VJ, Mickelson JR, Patterson EN, Shelton GD. Evaluations of labrador retrievers with exercise-induced collapse, including response to a standardized strenuous exercise protocol. J Am Anim Hosp Assoc. 2009;45(1):3–13. doi:10.5326/0450003.View ArticlePubMedGoogle Scholar
- Patterson EE, Minor KM, Tchernatynskaia AV, Taylor SM, Shelton GD, Ekenstedt KJ, et al. A canine DNM1 mutation is highly associated with the syndrome of exercise-induced collapse. Nat Genet. 2008;40(10):1235–9. doi:10.1038/ng.224.View ArticlePubMedGoogle Scholar
- Minor KM, Patterson EE, Keating MK, Gross SD, Ekenstedt KJ, Taylor SM, et al. Presence and impact of the exercise-induced collapse associated DNM1 mutation in Labrador retrievers and other breeds. Vet J. 2011;189(2):214–9. doi:10.1016/j.tvjl.2011.06.022.View ArticlePubMedGoogle Scholar
- Whitehair JG, Vasseur PB, Willits NH. Epidemiology of cranial cruciate ligament rupture in dogs. J Am Vet Med Assoc. 1993;203(7):1016–9.PubMedGoogle Scholar
- Cook JL. Cranial cruciate ligament disease in dogs: biology versus biomechanics. Vet Surg. 2010;39(3):270–7. doi:10.1111/j.1532-950X.2010.00653.x.View ArticlePubMedGoogle Scholar
- Duval JM, Budsberg SC, Flo GL, Sammarco JL. Breed, sex, and body weight as risk factors for rupture of the cranial cruciate ligament in young dogs. J Am Vet Med Assoc. 1999;215(6):811–4.PubMedGoogle Scholar
- Harasen G. Canine cranial cruciate ligament rupture in profile: 2002-2007. Can Vet J. 2008;49(2):193–4.PubMedPubMed CentralGoogle Scholar
- Guthrie JW, Keeley BJ, Maddock E, Bright SR, May C. Effect of signalment on the presentation of canine patients suffering from cranial cruciate ligament disease. J Small Anim Pract. 2012;53(5):273–7. doi:10.1111/j.1748-5827.2011.01202.x.View ArticlePubMedGoogle Scholar
- Wilke VL, Zaldivar-Lopez S, Ekenstedt KJ, Evans RB, Conzemius MG. Genetic tests to identify cranial cruciate ligament disease susceptibility in the Newfoundland and Labrador retriever breeds. JSciMed (Vet Med Res). 2015;2(2):1018.Google Scholar
- Taylor SM, Shmon CL, Shelton GD, Patterson EN, Minor K, Mickelson JR. Exercise-induced collapse of Labrador retrievers: survey results and preliminary investigation of heritability. J Am Anim Hosp Assoc. 2008;44(6):295–301. doi:10.5326/0440295.View ArticlePubMedGoogle Scholar
- Reif U, Probst CW. Comparison of tibial plateau angles in normal and cranial cruciate deficient stifles of Labrador retrievers. Vet Surg. 2003;32(4):385–9. doi:10.1053/jvet.2003.50047.View ArticlePubMedGoogle Scholar
- Agresti A, Coull B. Approximate is better than “exact” for interval estimation of binomial proportions. Amer Statistician. 1998;52(2):119–26. http://dx.doi.org/10.1080/00031305.1998.10480550.Google Scholar
- R Core Team. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/. Accessed 30 Jan 2017.
- Adams P, Bolus R, Middleton S, Moores AP, Grierson J. Influence of signalment on developing cranial cruciate rupture in dogs in the UK. J Small Anim Pract. 2011;52(7):347–52. doi:10.1111/j.1748-5827.2011.01073.x.View ArticlePubMedGoogle Scholar
- Buote N, Fusco J, Radasch R. Age, tibial plateau angle, sex, and weight as risk factors for contralateral rupture of the cranial cruciate ligament in Labradors. Vet Surg. 2009;38(4):481–9. doi:10.1111/j.1532-950X.2009.00532.x.View ArticlePubMedGoogle Scholar
- Hart BL, Hart LA, Thigpen AP, Willits NH. Long-term health effects of neutering dogs: comparison of Labrador Retrievers with Golden Retrievers. PLoS One. 2014;9(7):e102241. doi:10.1371/journal.pone.0102241.View ArticlePubMedPubMed CentralGoogle Scholar
- Sutton KM, Bullock JM. Anterior cruciate ligament rupture: differences between males and females. J Am Acad Orthop Surg. 2013;21(1):41–50. doi:10.5435/JAAOS-21-01-41.View ArticlePubMedGoogle Scholar
- Salmeri KR, Bloomberg MS, Scruggs SL, Shille V. Gonadectomy in immature dogs: effects on skeletal, physical, and behavioral development. J Am Vet Med Assoc. 1991;198(7):1193–203.PubMedGoogle Scholar
- Grumbach MM. Estrogen, bone, growth and sex: a sea change in conventional wisdom. J Pediatr Endocrinol Metab. 2000;13 Suppl 6:1439–55.PubMedGoogle Scholar
- Witsberger TH, Villamil JA, Schultz LG, Hahn AW, Cook JL. Prevalence of and risk factors for hip dysplasia and cranial cruciate ligament deficiency in dogs. J Am Vet Med Assoc. 2008;232(12):1818–24. doi:10.2460/javma.232.12.1818.View ArticlePubMedGoogle Scholar
- Torres de la Riva G, Hart BL, Farver TB, Oberbauer AM, Messam LL, Willits N, et al. Neutering dogs: effects on joint disorders and cancers in golden retrievers. PLoS One. 2013;8(2):e55937. doi:10.1371/journal.pone.0055937.View ArticlePubMedPubMed CentralGoogle Scholar
- Hart B, Hart L, Thigpen A, Willits N. Neutering of German Shepherd Dogs: associated joint disorders, cancers and urinary incontinence. Vet Med Sci. 2016;2:191–9. doi:10.1002/vms3.34.View ArticleGoogle Scholar
- Furrow E, Minor KM, Taylor SM, Mickelson JR, Patterson EE. Relationship between dynamin 1 mutation status and characteristics of recurrent episodes of exercise-induced collapse in Labrador Retrievers. J Am Vet Med Assoc. 2013;242(6):786–91. doi:10.2460/javma.242.6.786.View ArticlePubMedGoogle Scholar