TO THE EDITOR:
We read with interest "Subclinical Slipped Capital Femoral Epiphysis. Relationship to Osteoarthrosis of the Hip" (79-A: 1489—1497, Oct. 1997), by Goodman et al., which was based on extensive measurement of intact femora from a cohort of skeletons of individuals who were born around the turn of the century. The authors clearly showed that there were more osteophytes in the 8 percent of the specimens that had a so-called subclinical slip deformity than there were in the specimens that did not have this deformity.
To call these specimens osteoarthritic is not consistent with the classic clinical study by Danielsson, who found no arthritic deterioration over ten years in patients who had osteophytes about the hip and a joint space of normal width2. Loss of cartilage, which is necessary to relate osteophytes to osteoarthritis, could not be observed in the study by Goodman et al. because the cartilage was destroyed during the preparation process.
Furthermore, the rates of osteoarthritis reported by Goodman et al. do not match those of clinical studies.The authors reported that 26 percent (seventy-nine) of the 306 age-matched control hips (which did not have a deformity) had grade-2 or 3 osteophytes, which is five times the 5 percent rate of osteoarthritis of the hip that they cited for the current general population. If the ninety-five control hips that had grade-1 osteophytes are also considered to be osteoarthritic, more than half of the control hips had osteoarthritis. Is it possible that the indigent population from that era had such a prevalence of osteoarthritis of the hip?
The anatomical deformity that was described in the gross specimens resembles the deformity of slipped capital femoral epiphysis, but whether this is, in fact, a forme fruste of slipped capital femoral epiphysis is not clear. There are a number of discrepancies between the described deformity and slipped capital femoral epiphysis. The authors noted that the annual incidence of slipped capital femoral epiphysis per 100,000 population has been reported to be as high as 7.8 percent, whereas slipped capital femoral epiphysis occurs in only 1 in 800 males or 1 in 2000 females. The reported male-to-female ratio of 1.3:1 does not match the ratio seen in association with clinical slipped capital femoral epiphysis. The authors did not mention anteversion in their specimens, whereas slipped capital femoral epiphysis is associated with decreased anteversion and sometimes with frank retroversion4.
In the entire series (the specimens with a post-slip morphology and the controls), about 6 percent of the specimens had grade-3 osteophytes. If these specimens were, in fact, osteoarthritic, only 17 percent of all osteoarthritic specimens would have had a post-slip morphology. In the clinical studies reported by Murray, Stulberg et al., and Solomon, 30 percent (fifty-nine of 1967) to 40 percent (eighty of 2005 and thirty of seventy-five8,9) of osteoarthritic patients had a tilt or pistol-grip deformity.
Goodman et al. correctly cited Resnick6, who demonstrated that severe osteoarthritis and remodeling of an osteoarthritic hip is not a tilt deformity. Therefore, one would not expect to see the osteoarthritis that Resnick described in the 8 percent of the specimens in the present study that had a post-slip deformity but rather in the 92 percent of the specimens that had no deformity. To disprove Resnick's work, the authors would have had to analyze the osteoarthritis in the 92 percent of the specimens that had no slip.
The visual scale of subclinical slips presented in the study, including the flattening of the acetabulum and the increased concavity of the posterior region of the head-neck junction, could be useful for clinical studies if these features can be recognized by three-dimensional computed tomography scans. The demonstration of hips that had the developmental deformity and no osteophytes emphasizes the need to look for other factors that cause this disease.
Franklin T. Hoaglund, M.D.; Lynne S. Steinbach, M.D.: Department of Orthopaedic Surgery (F. T. H.) and Division of Musculoskeletal Radiology (L. S. S.), University of California at San Francisco Medical Center, 500 Parnassus Avenue (MU-320W), San Francisco, California 94143-0728
Dr. Goodman, Dr. Feighan, Dr. Smith, Dr. Latimer,Dr. Buly, and Dr. Cooperman reply:
We appreciate the interest, comments, and questions of Dr. Hoaglund and Dr. Steinbach regarding our study. Their many studies of slipped capital femoral epiphysis were motivating factors in our pursuit of this study. Their question about the relationship of osteophytes to osteoarthritis is well taken. We agree that the most accurate way to grade arthritis revolves around multiple factors, including clinical symptoms, loss of cartilage, the formation of osteophytes and subchondral cysts, erosion, and flattening of the femoral head and acetabulum. Much of this information was not available to us for our study, as we used an osteological collection that was devoid of cartilage and no medical history or premorbid radiographic data were available. We believe that our arthritis scale is a valid rating system and that the visible osteophytes were a manifestation of osteoarthritis, given that there also was erosion and that subchondral cysts were evident on the radiographs. We also point out that this is certainly a weak point of our study and that the prevalence of osteoarthritis was meant to be used as a comparison between the post-slip and control groups and not as an absolute value.
Although we have no way of knowing, we certainly believe that this indigent population could have a higher percentage of osteoarthritis than the general population. Thus, it may have been an ideal population to use for a study of osteoarthritis that focused on differentiating between post-slip and control groups.
With regard to the question about the occurrence of slipped capital femoral epiphysis in the population, there was a serious typographical error in this section of the paper. The section should read: "The known prevalence of slipped capital femoral epiphysis in the population is about two per 100,000. Kelsey et al. found the annual incidence in Connecticut to be 3.41 per 100,000 and the annual incidence in New Mexico to be 0.7 per 100,000. Those authors noted that, in Connecticut, among patients twenty-five years old or less, the prevalence of slipped capital femoral epiphysis is 7.8 per 100,000 black male patients, 6.7 per 100,000 black female patients, 4.7 per 100,000 white male patients, and 1.6 per 100,000 white female patients."3
The male-to-female ratio in our study was 1.3:1 compared with the reported ranges1 of 1.7:1 to 5.8:1. One could theorize that the closer ratio in our study reflects the fact that we have identified a forme fruste of slipped capital femoral epiphysis. Other factors besides gender, such as weight, activity level, or hormonal differences, could change the actual number of hips that go on to true slipped capital femoral epiphysis.
All of the post-slip specimens had true retroversion—that is, posterior angulation of the head on the neck. Antetorsion-retrotorsion was not measured directly. Anecdotally, many of the post-slip specimens had retrotorsion as well.
The comparison of the rate of osteoarthritis in our entire series with those in other studies is not valid. The average age of our specimens at the time of death was forty-five years. Twenty-three (25 percent) of the ninety-three post-slip specimens from individuals who were fifty-six years old or more at the time of death had grade-3 osteoarthritis compared with eleven (12 percent) of the ninety-three controls, for an overall rate of 18 percent. The reported rates of 30 to 40 percent5,7-9 for slipped capital femoral epiphysis caused by post-slip deformity occurred in arthritic populations. In order to compare our series with those, it would have been necessary to identify an arthritic population in the osteological collection first and then to determine the percentage that had post-slip deformity in order to find the percentage of specimens in which osteoarthritis was related to slipped capital femoral epiphysis. In our study, thirty-four specimens (twenty-three that had a post-slip deformity and eleven controls) had grade-3 osteoarthritis; thus, 68 percent of the specimens with grade-3 osteoarthritis had a post-slip deformity. It must be kept in mind that specimens that had dysplasia were excluded from our study, whereas they were included in the studies by Murray5, Stulberg et al.8,9, and Solomon7.
Resnick stated that the slip deformity was due to the formation of osteophytes in the femoral head, not to a pre-slip lesion or nonclinical slip. Thus, the prevalence of deformity in Resnick's example would increase with age as the formation of osteophytes increases. We found that the prevalence of the slip deformity was the same in specimens from individuals who were forty years old or less, forty-one to fifty-five years old, and fifty-six years old or more. Thus, the deformity is not due to the formation of osteophytes or to aging. It is due to a deformity that is present before skeletal maturity. Whether this deformity is a congenital bone shape or a developmental deformity that occurs in childhood or adolescence cannot be determined from our study of specimens from skeletally mature individuals.
Certainly, the point made by Dr. Hoaglund and Dr. Steinbach regarding the femoral and acetabular findings and the use of three-dimensional computed tomography scans is an excellent observation. We agree, and we point out that many factors must contribute to osteoarthritis and that the pre-slip deformity is a risk factor but is not an absolute determinant for osteoarthritis with aging.
David A. Goodman, M.D.; John E. Feighan, M.D.; Angela D. Smith, M.D.; Bruce Latimer, Ph.D.; Robert L. Buly, M.D.; Daniel R. Cooperman, M.D.: Department of Orthopedics, University Hospitals of Cleveland, 11100 Euclid Avenue, Cleveland, Ohio 44106