Effect of Estrus and Other Factors on Apparent Hip Joint Conformation

INTRODUCTION

Before we get into the meat of the matter, let’s review the OFA-type definitions of canine hip dysplasia, referred to here as HD.

Mild HD: Significant subluxation, only partial coverage by the acetabular rim, and usually no arthritic changes (yet). A fairly recent development in OFA’s operating rules is that if a dog rated “Mild HD” is under 30 months old at the time, the owner may submit a film at a later age in hopes of getting a “normal” reading and that prized certificate, even though most dogs will not show improvement in this position. However, due to the inaccuracies and changeability of the hip-extended position, some dogs may appear to show “improved hip conformation with increasing age” (Reed & Keller, approx. 1999). This is based on the as-yet unproven idea that inactivity or estrus may cause joints to be looser than they would be at other times, but as you will see, this concept, if substantiated, would be highly dependent on the hip-extended view.

Moderate HD is diagnosed if there is considerably more joint space, or there are secondary arthritic or remodeling changes and/or osteophytes and sclerosis in addition to the noticeable subluxation.

Severe HD is either gross subluxation, or much DJD (degenerative joint disease), or both. The OFA, as well as most hip registries around the world, still rely on the old leg-extended ventro-dorsal view, which unnaturally positions the dog and artificially tightens the hip joint capsule. The PennHIP position and technique, described in some of my other articles, uses a position more alike the dog’s natural stance and gait, and demonstrates latent or covert laxity that is usually hidden in the old view.

In evaluating the hips of dogs genetically predisposed to dysplasia, certain factors could make the hips in the leg-extended radiographs appear worse, compared to another dog with presumably the same genes for hip joint construction, such as a littermate or a dog with otherwise near-identical genotype yet reared differently. These include tight extensor and adductor muscles and ligaments, poor muscle tone or development, atrophy or dystrophy, transitional vertebrae, high-calorie diet and rapid growth rate, excessive activity before skeletal maturity, anteversion or valgus deformity, poor positioning on the radiography table, narrow pelvis, and possibly (though not likely) estrus. A few of these could even give a slightly poorer radiographic impression of a dog with a relatively lower genetic makeup (fewer bad genes) for HD.

When using the old ventro-dorsal view, if you don’t rotate the femurs adequately to make the stifles move internally (medially) enough, the amount of true laxity can be hidden and a false negative reading can be given. Another problem in getting a good screening from hip-extended pelvic radiographs may be not twisting and aligning the femurs and patellas quite right. Conversely, if a dog with tight sockets is radiographed without rotating the femurs sufficiently, the femoral neck may appear shortened and at a valgus angle, both of which may cause some less-experienced vets to give a dysplastic diagnosis to a “normal” set of hips. Thus, in this v-d position, you can find both false negative and (to a minor extent) false positive readings if the position is not ideal and other signs are absent.

Conversely, factors that make for a better-looking hip-extended radiograph than might be expected, given the probable genetic makeup, include keeping a dog lean and well muscled (especially during the fast-growth period and leading up to the date of radiography) and taking the picture for certification at a young age (as is permitted in other parts of the world). Some factors that influence the hip-extended appearance of the pelvic joints are obviously environmental, such as “conditioning” by some roadwork, swimming, and relatively low body weight. Other reasons for a dog’s hip-extended radiographs looking “better” at a later date might include a thickening of the joint capsule, the first film having been taken when a bitch was in estrus or close to whelping date (a controversial and unverified assumption), and anesthesia used the second time and not the first time. When many of these differences are found at the same time, there is a possibility of a small difference between the particular film and a relatively accurate diagnosis in the AVMA’s hip-extended view. Not all of these “reasons” are accepted by everyone, especially any effect of estrus. In any case, these and other influences on diagnosis of hip joint laxity are avoided by using the newer and superior distraction positioning/radiographic technique as recommended at the University of Pennsylvania Veterinary College.

EFFECT OF ANESTHESIA AND CONDITIONING

It has been stated by Corley (letter in “Health News and Views” feature, The Rottweiler Quarterly, Fall 1991) that anesthetized dogs might show slightly looser hip joints, but that the difference may be insignificant. However, the older hypothesis goes, if you combine a number of factors such as radiographing while the bitch is in estrus or close to whelping (either before or after), doing the evaluation early in the morning, a lack of muscle tone in the dog because of a sedentary life, and deep anesthesia, it’s possible that a number of dogs would show more laxity than otherwise (in the hip-extended view). Is this because they have what I call covert or latent laxity, which may show up only at a later age, without those factors? Perhaps we can’t answer this with total assurance, as not enough data may have been compiled, but my observations make me suspicious. Depending on the individual dogs, with hip-extended radiography in the late afternoon after exercise, on a dog in good athletic condition, you might find these are more likely to have tighter muscles and ligaments when seen in the supine (OFA-style) position. Some have theorized that lean German Shepherd imports from Germany, where they have been trained and well-exercised several days a week, may have benefited from this regimen when they were radiographed and certified at about 13 months age there. Certainly, an uncomfortable number have “passed” the SV requirements then, only to become (or be discovered as) dysplastic a year or more later. As to sedation, the OFA has found that it makes little or no difference in most dogs. Because of the force applied to demonstrate laxity, PennHIP requires anesthesia for their distraction technique.

EFFECT OF ESTRUS

A magazine editor and others asked me to comment on the effect of estrus (heat cycle). For many years we breeders have been assuming that there regularly were differences in observable hip joint laxity between films made in or near estrus and those made in complete anestrus (not close to a heat cycle). We also told each other that to get the best picture and the best chances of “passing OFA”, we should take the pictures in the afternoon, after the dog had been exercised and “toned up”, on a dog with good muscle development from regular exercise over a longer period of time, not within a couple weeks of estrus or several weeks of pregnancy, and without sedation. This was based on intuitive supposition, and the OFA even made statements to the effect that they were true concepts (Reed & Keller, undated brochure, approx 1999-2000; and Canine Practice 1993, volume 18). There are two holes in this otherwise neat scenario.

One is that there had never been any fully accepted scientific study to prove the idea that the actual joint capsule ligaments are truly tighter or looser than at other times. It simply “made sense” to many of us, because conventional wisdom shows us that we are slightly longer when we wake up than after a day of gravity compressing our spinal discs, and that hormones in late pregnancy cause a woman’s pelvic symphysis to relax or bend so that there is more room for the child exiting the birth canal. But does that necessarily mean that the other joints are likewise affected, and that ligaments or associated tissues somehow relax more than they would in other circumstances? No. It is possibly a faulty assumption, with hardly anything to support it. In fact, a more definitive 1997 prospective study (Veterinary Comparative Orthopaedics and Traumatology, F.K.Schattauer Publ. Co.) of fluctuating hormone levels throughout the estrus cycle and the effect on hip joint laxity indicated that hip laxity is not changed by estrus cycle using either the OFA or the PennHIP method. The main conclusion: “Neither statistically nor clinically significant changes in hip laxity were observed.” And, intra-class correlations suggested that “…any hormone-related variation in hip laxity, if present, was minimal (<7% for distraction index, 18% for Norberg angle).” Further, “Although it is statistically impossible to prove the null hypothesis [i.e., no effect], if an effect existed within this sampling of dogs, it was small and clinically insignificant.” There is no significant change in DI (distraction index is the most accurate diagnostic technique) attributable to a bitch being in heat, or to the other reasons, when this method is used. The OFA’s contentions, as well as our breeders’ assumptions that estrus loosens joints, are unfounded, according to this 1997 report.

Still, there may again be that other problem with our assumptions, which is that we were basing them on the old and imprecise leg-extended position. As you can imagine, this natural position for humans but unnatural position for quadrupeds tends to wind up and artificially tighten the muscles, tendons, and ligaments in the area of the hip. We do not always get repeatable results with that position. With the knees flexed and the distractor unit employed in the PennHIP technique, however, a more objective measurement of laxity is seen time after time, with few and minor exceptions and extremely minor and infrequent changes from one visit to the radiologist and another. See the comments under “Lack of Reliability.” The dog’s full laxity is seen each time the dog is radiographed in this P-H position.

Hormones

It would seem to make sense that estrogens have an effect on joints, but the only scientific study I am aware of on the subject (that has what is accepted as statistical significance) indicates that it does not. The published reports in the 1960s indicated “exogenously administered estrogen” caused some increase in hip joint laxity, but injection with steroids or hormones is not the same as changes resulting from the bitch’s own endocrine system. There was one study by Keller et al in 1993 that indicated a possible connection, but the correlation was not considered “statistically significant” by colleagues. In order to make a strong conclusion that would be accepted by scientists, two things must happen:

  1. There must be statistical significance, and
  2. Other researchers must get similar results.

Neither of these conditions for scientific conclusion has yet been attained. Corley in 1987 said, in a non-peer-reviewed press release, that some “bitches demonstrate subluxation while in season, but appear normal when not in season.” But, other than some support from the Keller article, this has yet to be verified or quantified in repeated or scientific studies. Besides, I have seen the same variation when dogs were simply re-radiographed in the same timeframe, such as during the same visit or a few weeks apart. If there ever is sufficient evidence to come to a scientific conclusion that the two are related in a cause-and-effect way, my guess is that there would be more apparent correlation in this artificially tightened leg-extended view, and nothing statistically important in the distraction view.

Can a bitch’s heat cycle alter the outcome of the radiographic diagnosis? Not likely. There is not enough strong scientific evidence that estrus affects laxity of the joints, despite what we used to think. If it affects anything, it would be in the OFA-type legs-extended position, but even that has not been shown to have statistical significance in any controlled study. So, the conclusion of the estrus-laxity question is (again) that true laxity does not hide from the distraction procedure, that it very well might be hidden in the leg-extended view, and that the role of hormones in canine hip joint laxity (whatever it might be in other species and anatomical locations) is not present, or is immeasurable. We need verification of either viewpoint, as one or two small studies on each side of the question are not enough for us to make nearly absolute statements.

LACK OF RELIABILITY (REPEATABILITY)

Occasionally, there are reports of some disturbing and puzzling aberrations in diagnosing from radiographs taken in the hip-extended (OFA-type) position, and it may take a little detective work or repeated radiography, perhaps at a later age or different position, to resolve the mystery. Rendano and Lust at Cornell reported on one example: a dog radiographed in “the standard view” twice the same day had a diagnosis of dysplasia at one examination and normal at the other, the difference (and erroneously? assumed to be the reason) being the use of anesthesia. In most dogs, anesthesia would probably not make any noticeable difference. In fact, at the University of PA, many dogs were repeatedly radiographed in the ventro-dorsal position, often with different results, even though only board-certified radiologists were used in the tests.

Some dogs diagnosed dysplastic at 12 months have appeared normal in the hip-extended view at 2 years, so it had been customary to explain away these few cases by differences in estrus, exercise, position, etc. Much more often are the cases wherein dogs will have subluxation but no osteophytic or other degenerative changes, many of these being examples of an age- or breed-related phenomenon. There are even differences between palpation showing laxity yet radiographs indicating normal hips, and vice-versa. These inconsistencies have led some orthopedic specialists to diagnose one way and a radiologist another, but these are quite rare. And remember, this is the hip-extended view we are talking about, with its lower reliability.

More than one breeder has tried to test the “heat theory” and they report that it appears to be true in their dogs. A couple of correspondents told me shortly before I finished this article that they had their bitches radiographed as they were “coming in”; the evaluation by the vets was that the hips were “close but probably wouldn’t pass OFA.” Then they were tested again 60-90 days after they were finished with the “season” (per OFA encouragement) and both were rated “good”. However, knowing how males and non-estrus females can be read differently by different people or by the same evaluator just hours or days apart, we cannot ascribe different readings to actual differences in the hips. There are too many “buts” and questions about such testimonials to allow us to count them as reliable, although they may be good reason to investigate scientifically. You know how people often tend to trust in testimonials rather than controlled scientific studies. If vets would compile good records, make sure all control protocols were followed, and have the results published, it would be the start of a definitive answer to this question. So far, nobody has done this, although there has been a start. Do not rule out the possibility of the same difference being seen simply because it was a different date that the second radiographs were made. Unless everything is exactly the same, it is not a controlled experiment. The amount of sedative, level of sedation, methodology of vet and technicians, etc. must be the same, and then repeated trials with the same results should follow. Further, other researchers must get the same results in order for it to be accepted that there is any connection between estrus and laxity. As of this writing, that corroboration has not been forthcoming.

Data compiled on 2,603 dogs by the GSD Council (Australia) and analyzed by geneticist Malcolm Willis (author of The German Shepherd Dog: A Genetic History, 1991) appeared in their March 1990 GSDCA National Review. About two thirds of the GSDs in the study were females, which hints that most breeding kennels keep more females and more non-breeding (and non-scoring) homes have males. This is a common pattern seen in most breeds. Data from the U.K. show males as having lower BVA scores (better hips), which at one time was thought to be at least partly due to some females being in or near estrus or postpartum at the time of radiography, at which time joints were supposedly slightly looser. However, as I mentioned earlier, although there was that one report by Keller and Bouchard that supported this effect of estrus on laxity or congruity, it has never been confirmed by others, and at least one study by Hassinger and others, performed with peer review and scientific controls, indicates that there is no basis for that idea. But there may be other reasons why the data are different than those reported earlier (JAVMA, vol. 160, 1972), which indicate no direct cause/effect relationship between severity of HD and sex of the dog.

In another assessment of nearly 6,000 German Shepherd Dog bitches and 3,000 males in the United Kingdom, Willis found the same percentages being submitted for scoring. Slightly higher scores (worse hips) for females were seen in Swedish data (Henricson and others at the 1972 OFA Canine Hip Dysplasia Symposium and Workshop in St. Louis MO, and Swenson et al in 1997). There are also breed differences that may complicate the picture. The prevalence of HD in female GSDs, Goldens, and Saints was significantly higher than for males of those breeds in one Swedish study by Swenson, Audell, & Hedhammar, and reported in JAVMA, vol. 210, 1997. However, similar differences were not found for Labrador Retrievers, Rottweilers, Newfoundlands, and Bernese Mountain Dogs.

A caveat: Internet forum opinions should usually be taken with a grain of salt, as most are dashed off without deep thought, preparation, or study. But they do influence some people’s opinions. On a 2005 Q&A session, a forum member expressed “serious doubts” because to her mind the number of dogs in a Hassinger/Smith report was not as many as she thought it should be. That’s the same weak argument that was first (ineffectively) used when PennHIP’s initial statistics showed it was a far more accurate method of prediction than OFA-type v-d positioning. My point in both instances was that if a procedure is far more accurate, a lower number of case histories can be much more informative than a large number of cases with inaccurate methods.

Because of long-held suppositions without data, an early (1997) study on hormone influence (estrus) on hip laxity was conducted by the University of Pennsylvania. (Hassinger KA, Smith GK, et al. Effect of the oestrus cycle on coxofemoral joint laxity. Veterinary Comparative Orthopaedics and Traumatology. (F.K.Schattauer Publ.) 1997;10:69-74. Also in Proceedings. An Conf Vet Orth Soc 1997;31). This was only the first systematic look at the relationship, so, naturally, it was questioned and suggested that others might want to repeat the experiment. The conclusions arrived at by Hassinger, Smith, and team should be accepted with the realization that the PennHIP method of determining HD was used, not currently “standard” ventro-dorsal view alone. It should be remembered that in the v-d view, a great deal of dysplasia (as defined by unacceptable laxity, the major risk factor in later DJD) is “missed.” But with the distraction technique in PennHIP, much more of this hidden laxity is discovered. As I have stated in my lectures many times, “Laxity does not hide from the PennHIP distraction view.” If anyone finds the funding to do repeated study on the effect of estrus on true laxity (seen in distraction) rather than that which is “hit or miss” (the old OFA v-d only view), it will only back up the conclusions of Hassinger & Smith. My advice to owners of dogs that might be bred is to take the attitude that hiding possible “bad news” (by using a less-accurate method) is not good for the sport or the individuals. Rather, that it is preferable to use PennHIP, which is a far more accurate prediction in young dogs, and which certainly seems to be unaffected by (or far less affected by) estrus.

The few and minor aberrations cited must not be taken as typical. Indeed, while there are always exceptions and variations in anything biological, the generalities are still valid. Estrus is highly unlikely to affect the view of hip joint conformation or relative laxity, especially but not limited to “distraction” (stress) radiography such as is used by PennHIP. Summary to this story of studies: a combination of improved diagnostic techniques and higher standards held by breeders will advance orthopedic soundness in the purebred dog.

Additional readings:

  • Corley EA. letter in “Health News and Views” feature, The Rottweiler Quarterly, Fall 1991.
  • Hassinger KA, Smith GK, et al. Effect of the oestrus cycle on coxofemoral joint laxity. Veterinary
    Comparative Orthopaedics and Traumatology. (F.K.Schattauer Publ.) 1997;10:69-74.
  • Keller GG, Bouchard G, et al. Influence of the estrus cycle on coxofemoral joint subluxation. Canine Practice 1993; 18:19-22.
  • Reed AL, Keller GG. What happens to a radiograph at the OFA: The submission and evaluation process. Undated brochure, approx 1999-2000
  • Any publications by PennHIP on the subject.
  • My book on orthopedic disorders (see the following advertisement).
Fred Lanting

Fred Lanting is an internationally respected show judge, approved by many registries as an all-breed judge, has judged numerous countries’ Sieger Shows and Landesgruppen events, and has many years experience as one of only two SV breed judges in the US. He presents seminars and consults worldwide on such topics as Gait-&-Structure, HD and Other Orthopedic Disorders, and The GSD. He conducts annual non-profit sightseeing tours of Europe, centered on the Sieger Show (biggest breed show in the world) and BSP.

Books by Fred Lanting