 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1998 The Journal of Bone and Joint Surgery, Inc. The Utility of Histological Examination of Tissue Removed during Elective Joint Replacement. A Preliminary Assessment*   , IOWA CITY, IOWAInvestigation performed at the Department of Pathology, University of Iowa Hospitals and Clinics, Iowa City
The utility of histological examination of tissue removed during elective joint replacement has not been determined. During a one-year period, tissue removed during 168 total joint replacements was submitted for histological examination. The clinical and histological diagnoses, the cost of the histological study, and the clinical course were determined for all joints. Degenerative joint disease, rheumatoid arthritis, and avascular necrosis accounted for 98 per cent of the histological diagnoses. There were sixteen discrepancies between the clinical and histological diagnoses. The histological diagnosis did not affect the treatment of fifteen of these joints. However, the treatment was altered for one joint that had a clinical diagnosis of degenerative joint disease and a histological diagnosis of osteomyelitis; on review, the initial histological diagnosis was determined to be incorrect. In 1996 dollars, the cost of histological examination for all 168 joints was $10,698.24. Although there would be considerable cost-savings on a population basis if histological examination were not performed, this savings must be weighed against the effect of a misdiagnosis on the management of a particular patient.
More than 120,000 elective prosthetic hip replacements are performed annually in the United States22. Although the tissue excised during the procedure is sent to pathology laboratories, there is no uniform practice with regard to the examination of this tissue. At some laboratories, all tissue specimens are examined histologically; at some, only a gross examination is performed; and at some, neither evaluation is done. This spectrum of practice patterns is a reflection of the contrasting viewpoints regarding histopathological tissue-processing. One opinion is that histological examination is always beneficial, for all types of tissue, because the information provided may affect the management of the patient (either immediately or in the long term) and may improve standards for clinical service through quality-control protocols10,13. Another opinion is that the information gained by histological examination of some tissue specimens, such as hernia sacs, gallbladders, tonsils, and tissue excised during joint replacement, is so rarely beneficial that examination is not warranted2,5,6,11,14,23. The debate over whether to histologically examine tissue that has been excised during a joint replacement has become increasingly important in this era of cost containment. The cost of histological processing and examination generally ranges from $50 to $200 dollars per patient. Even if the lowest cost is assumed, the total cost of examination of tissue excised during joint replacements in the United States is more than $6 million annually ($50 x 120,000). As far as we know, no one has performed a cost-benefit analysis of such histological examination16,19. The objectives of the present study were to compare the preoperative clinical diagnosis with the diagnosis based on histological examination of tissue removed during elective total joint replacement, and to investigate the cost-effectiveness of such histological examination, at our institution.
Before the initiation of the present study, the policy at our institution was that tissue removed during joint replacement was sent to the pathology laboratory for examination. The tissue was handled according to the guidelines of the College of American Pathologists, which inspects and accredits its pathology laboratories. There is no state legislation that stipulates how a specific specimen should be handled. The College of American Pathologists recommends that specimens be saved for at least three weeks before they are discarded. At our institution, tissues removed during joint replacement were saved for six to eight weeks. Institutional policy determines how specific specimens are examined. Some specimens may be examined grossly without histological examination, whereas others are evaluated histologically. At the time of this study, all tissue removed during joint replacement was examined both grossly and histologically. The pathology reports on all specimens coded as joint specimens from January 1, 1993, to December 31, 1993, were retrospectively reviewed. Of these joint specimens, seventy-nine were removed during an elective primary hip replacement and eighty-nine, during an elective primary knee replacement. Thus, there was a total of 168 procedures, which were performed in 156 patients. (Twelve patients had a bilateral procedure.) The patients included sixty-eight men and eighty-eight women who ranged in age from twenty-five to eighty-seven years (mean, sixty-three years). The preoperative clinical information, gross description of the specimen or specimens, and histological diagnosis were recorded for each patient. Clinical information was provided on the requisition form for 101 joints (60 per cent) in ninety-eight patients. The diagnosis on the requisition form was degenerative joint disease for seventy-six patients; rheumatoid arthritis for three; degenerative joint disease and rheumatoid arthritis for six; avascular necrosis for six; pain for two; and fracture, hemarthrosis associated with hemophilia, and cancer of the prostate for one each. The specimen of one patient was to be examined to rule out infection, and that of another was a bone graft. The preoperative diagnostic investigation, including other diagnostic costs, was not specifically addressed in this study. All tissue specimens were evaluated systematically by residents under faculty supervision. The cartilaginous joint surface, bone, and soft tissue were examined grossly, and any remarkable external features were described. Four-to-five-millimeter sagittal slices were cut from the joint tissue with a band saw for gross examination of the subchondral bone, cortex, medullary component, and margin of the resection. In general, one or two sections of bone, which included the joint surface, subchondral bone, and medullary tissue, were submitted for decalcification and histological examination. As with any tissue specimen, sampling error may have occurred. Soft-tissue fragments were submitted in a separate cassette and were not decalcified. The histological examination was performed by one of eight board-certified pathologists, an average-size pathology department. Nearly every pathology department has the capability to cut and examine bone specimens. All of the pathology reports provided descriptions of the gross and histological findings as well as a histological diagnosis. The initial histological diagnosis was degenerative joint disease for 149 joints; rheumatoid arthritis for one; degenerative joint disease and rheumatoid arthritis for four; avascular necrosis for one; degenerative joint disease and avascular necrosis for seven; degenerative joint disease, avascular necrosis, and rheumatoid arthritis for two; and chronic osteomyelitis, metastatic cancer of the prostate, small lymphocytic lymphoma with osseous involvement, and hemochromatosis with degenerative joint disease for one each. All of the histological slides were reviewed for this study by an experienced bone pathologist (R. A. R.) who was blinded with regard to the history, and the accuracy of the initial histological diagnoses was calculated by comparison with his definitive diagnoses. Clinical information about all patients was obtained by review of the medical chart. The preoperative clinical diagnosis, the initial histological diagnosis, and the diagnosis on review of the histological sections were recorded. The interval from the initial diagnosis to the diagnosis on review ranged from three months to five years (mean, 3.6 years). The crude agreement between the preoperative clinical diagnosis and the initial histological diagnosis was determined15,16,21. The initial histological diagnosis was considered discrepant when it differed from the preoperative clinical diagnosis. It was considered discordant when there was a discrepancy and the medical record showed that it had resulted in a change in treatment. Thus, a discordant histological diagnosis differed from the preoperative clinical diagnosis and affected the care of the patient. The 1996 technical (hospital) and professional (physician) costs were calculated. The technical costs were calculated on a per-block basis; the cost for decalcification and tissue-processing of a block was $17.78, and the cost for tissue-processing of a block with no decalcification was $15.38. The professional costs were calculated on a per-slide basis; the cost for the examination of a slide was $1.43 per minute, and it was assumed that a pathologist spent ten minutes with each slide. To determine total costs, the average number of blocks submitted per joint was calculated. The average number of bone (decalcified) and soft-tissue (non-decalcified) blocks was 1.8 and 0.2, respectively. The cost per discrepant diagnosis and the cost per discordant diagnosis were calculated.
There was exact concordance between the clinical and histological diagnoses for 152 (90 per cent) of the 168 joints, and the histological diagnosis was discrepant for sixteen (Table I). In four joints (2 per cent) with a discrepant diagnosis, additional findings (avascular necrosis in two joints, rheumatoid arthritis in one, and chronic osteomyelitis in one) that had not been suspected clinically were noted histologically. In the remaining twelve joints (7 per cent) with a discrepant diagnosis, additional findings (rheumatoid arthritis in seven joints and avascular necrosis in five) that had been suspected clinically were not documented histologically.
The histological diagnosis on review was different from the initial histological diagnosis for fourteen joints (8 per cent) (Table II). Four joints for which the initial diagnosis had included avascular necrosis were not considered to have histological evidence of avascular necrosis on review. Five joints that were found to have histological evidence of avascular necrosis on review had not been so diagnosed initially. On a second review of the specimens from these fourteen joints, the histological diagnosis made on the first review was thought to be more accurate than the initial histological diagnosis.
There was only one discordant diagnosis, as determined by review of the medical charts. The initial histological diagnosis for this joint was chronic osteomyelitis. The chart showed that there had been no perioperative signs of infection and intraoperative cultures had been negative. However, because of the histological diagnosis, the patient was started on a six-week course of antibiotics administered orally. Review of the histological sections showed intramedullary fibrous tissue, focal necrotic bone trabeculae, and focal plasma-cell and lymphocytic infiltrates. Neutrophils were not observed. These findings were thought to be more consistent with degeneration than with chronic osteomyelitis, and the initial histological diagnosis was thought to be incorrect1,24. There was no evidence of infection at any time before, during, or after the operation. Thus, it would appear that the initial histological diagnosis was an error rather than a true discordant diagnosis. For two patients, the diagnosis of a malignant lesion had been made clinically and was confirmed histologically. No occult malignant lesions were discovered. The total cost of the histological examination for all patients was $10,698.24. The cost per discrepant diagnosis (a total of sixteen in the series) was $668.64, and the cost per discordant diagnosis (one in the series) was $10,698.24, although on review this discordant diagnosis was thought to be an error.
In order to determine the value of histological examination of tissue removed during joint replacement, we evaluated the effect of the histological diagnosis on the management of the patients. Ninety-eight per cent (165) of the 168 joints had a histological diagnosis of degenerative joint disease, rheumatoid arthritis, or avascular necrosis, or a combination of these disorders. Of the remaining three joints, two had a clinical diagnosis of a malignant lesion that was confirmed histologically and one had a histological diagnosis of chronic osteomyelitis that was determined retrospectively to be incorrect. The crude agreement between the preoperative clinical diagnosis and the initial histological diagnosis was only 90 per cent (152 joints), reflecting the fact that certain diseases either are missed or cannot be identified with histological examination or are missed on clinical examination. The histological characteristics of degenerative joint disease, rheumatoid arthritis, and avascular necrosis may be similar and cannot always be distinguished from one another3,9,12,20. The histological term degenerative joint disease does not describe a unique lesion; rather, it is a description of the end point of a number of separate disease processes20. The histological examination is not the standard by which all other examinations should be measured; instead, it should be used in conjunction with the clinical impression to predict the likelihood of a specific disease7,17,18. The histological evaluations that demonstrated an increased likelihood that other disease processes (specifically, avascular necrosis) were present reflect the quality-assurance value of histological examination13. Avascular necrosis cannot be diagnosed with 100 per cent sensitivity on radiographs. Although two additional cases of avascular necrosis were diagnosed histologically, these diagnoses were never mentioned in the chart and there was no indication that these data were used in any quality-assurance program at our institution. The discrepancies between the histological diagnoses determined initially and on review suggest that there is substantial variability (8 per cent) between expert and general pathologists. On review, the only discordant diagnosis (chronic osteomyelitis) was corrected, and the other discrepant diagnoses did not appear to compromise care. Thus, the effect of the histological diagnosis on the treatment was minimum. However, because of the small number of joints in the present study, this finding is anecdotal. A larger series might show histological diagnosis to have a totally different effect on the management of patients. Over a one-year period, the management of only one patient was altered (incorrectly) on the basis of the histological examination; thus, the overall cost per discordant diagnosis was $10,698.24. In comparison to the overall cost of a joint replacement, the cost of a histological examination is small (1 per cent)4,8,25. However, with the costs at our institution used as a baseline, the annual cost for the histological examination of tissue for all joint replacements performed in the United States is more than $7.6 million ($64 x 120,000)22. The main reason to continue the practice of histological examination is that the diagnostic information may occasionally affect the management of the patient. The finding of a previously undiagnosed primary malignant lesion clearly affects management, but no such lesions were found in this study and, to the best of our knowledge, this finding has not been reported in the literature either. Malignant lesions that have been detected in tissues removed during joint replacement have been in patients who were known to have cancer10. The cost-benefit ratio of histological examination depends on the incidence of detection of previously undiagnosed disease and the improvement in an outcome measure (such as life expectancy) if this disease is detected. The national incidence of discordant diagnoses has not been determined, and the incidence may vary by geographic region and institution. More than one year of case data from one institution is needed to determine this incidence. The lack of case reports on so-called incidental cases suggests that the risk of a clinically important disease being missed if a histological examination is not performed is low. The present study is a preliminary cost-benefit analysis of histological examination of tissue removed during joint replacement. Additional studies, including retrospective analyses involving more patients from more institutions, are needed before sweeping conclusions can be made. Additional studies also are needed to determine the cost of missing a rare disease (such as tuberculosis); the cost of the preoperative diagnostic workup; the legal cost of omitting a histological examination when such an examination may have been beneficial; and the effect, on the cost of long-term care, of an initial histological diagnosis that reveals important clinical information. For example, an unexpected diagnosis of avascular necrosis may not affect immediate care, but it may affect the long-term care of the contralateral hip. Cost-benefit studies must take into consideration multiple outcome measures, such as the life expectancy of the patient and of the implant.
*No benefits in any form have been received or will be received from a commercial party related or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |