 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Aspiration of the Hip Joint before Revision Total Hip Arthroplasty. Clinical and Laboratory Factors Influencing Attainment of a Positive Culture* , CHAPEL HILL, NORTH CAROLINAInvestigation performed at the Department of Orthopaedic Surgery, University of North Carolina at Chapel Hill, Chapel Hill
The value of routine aspiration of the hip joint before revision of a hip arthroplasty remains controversial. We reviewed the results of such aspirations in an attempt to determine clinical or laboratory factors that could help the surgeon to identify hips that are infected and that should be aspirated preoperatively. One hundred and fifty consecutive revision total hip arthroplasties were performed by one of us. Preoperative aspiration was not performed or data were excluded for eight hips; no fluid was obtained from one of these hips (0.7 per cent of the 150). Of the remaining 142 hips, 128 had preoperative aspiration once and fourteen, twice. Twenty-one (15 per cent) of the 142 hips were infected, as demonstrated by the intraoperative culture. The intraoperative culture for two of these hips, however, was considered to be false-positive. The initial aspiration was considered to be positive only if an organism grew on the solid medium or if grossly purulent fluid was obtained. The initial aspiration was positive for nineteen hips; on culture of specimens from one hip, Bacteroides thetaiotaomicron grew in the liquid medium only; and purulent fluid was obtained from one hip but no organisms grew on culture. Fourteen aspirations were repeated for various reasons, most commonly to confirm the presence of an unusual organism. The repeat aspiration did not change the diagnosis for these hips. When the two hips with a false-positive intraoperative culture were excluded, preoperative aspiration had a sensitivity of 92 per cent, a specificity of 97 per cent, and an accuracy of 96 per cent. Seventeen of the nineteen truly infected hips were associated with an abnormally elevated erythrocyte-sedimentation rate (mean, 80.8 millimeters per hour). However, fifty-eight (50 per cent) of the 116 hips that were not infected, and for which the results were available, also had an abnormally elevated erythrocyte-sedimentation rate (mean, 32.0 millimeters per hour). This difference was significant (p = 0.001, Fisher exact test). The peripheral leukocyte count was not helpful in predicting infection. Hips in which the implants had been in situ for more than five years were less likely to be infected (p = 0.008, Fisher exact test) than those in which the implants had been in situ for five years or less. None of the infected hips in which the implants had been in situ for more than five years were associated with a normal erythrocyte-sedimentation rate. In this study, preoperative aspiration of the hip joint had an excellent sensitivity and specificity with regard to the prediction of infection. On the basis of our findings, we now favor a selective approach to aspiration, as determined by the erythrocyte sedimentation rate and the amount of time that the implant has been in situ.
Routine aspiration of the hip joint before revision total hip arthroplasty is controversial. Although it had been accepted for many years for the evaluation of a painful hip after an arthroplasty2,6,7,9,10 its value has recently been questioned. In a report by Barrack and Harris of 270 hips that had routine preoperative aspiration, only six (2 per cent) were determined to be infected. Because of a 13 per cent prevalence of false-positive culture of specimens of the aspirated fluid, the authors recommended that aspiration be performed only for selected patients as determined by a clinical history or radiographic findings suggestive of infection. However, the authors did not evaluate the utility of certain clinical variables such as the amount of time that the implants had been in situ or routine laboratory variables such as the peripheral leukocyte count or erythrocyte sedimentation rate in the determination of infection. The purpose of the present study was to determine the sensitivity, specificity, and accuracy of routine aspiration of the hip joint before revision total hip arthroplasty at our institution. In addition, we sought to determine the clinical variables and laboratory studies that are helpful in the prediction of hips that will have a positive culture of specimens obtained with preoperative aspiration and to identify the hips for which preoperative aspiration is indicated.
One hundred and fifty consecutive revision total hip arthroplasties were performed by the senior one of us (P. F. L.) at our institution between 1983 and 1992. Preoperative aspiration was not performed or the data were excluded for eight hips for various reasons: two hips were revised for an acute fracture of the bone around a cemented femoral component, one hip was in a patient who was taking an antibiotic orally when the aspiration was performed, two hips had failure of a screw-plate device, two hips had had a previous resection arthroplasty, and no fluid was obtained from one hip (0.7 per cent) at the time of the aspiration. Therefore, this study includes 142 hips (140 patients) in which aspiration was performed before a revision total hip arthroplasty. Two patients who had had a bilateral arthroplasty were evaluated for infection in each hip; 138 patients had had a unilateral arthroplasty. Twenty-one hips had revision of a failed hemiarthroplasty, bipolar arthroplasty, or cup arthroplasty. The remaining 121 hips had revision of a total hip replacement. Seventy hips were in women, and seventy-two were in men. The mean age of the patients at the time of the revision arthroplasty was fifty-nine years (range, twenty-three to eighty-seven years). The indication for the initial procedure was osteoarthrosis in forty-four hips (31 per cent) (forty-four patients), osteonecrosis in thirty-eight hips (27 per cent) (thirty-seven patients), trauma or post-traumatic osteoarthrosis in thirty-five hips (25 per cent) (thirty-five patients), and rheumatoid arthritis in ten hips (7 per cent) (ten patients). The remaining fifteen hips (11 per cent) (fourteen patients) had another diagnosis. The implants had been in situ for a mean of 7.4 years (range, four months to 20.0 years). The implants had been in situ for less than two years in twelve hips (8 per cent), for two to five years in thirty-seven hips (26 per cent), and for more than five years in ninety-three hips (65 per cent). The patients had had a mean of 1.5 operations (range, one to four operations) on the hip before the aspiration. Throughout the period of this study, the routine evaluation usually included a peripheral leukocyte count (141 hips; 139 patients) and a Westergren erythrocyte-sedimentation rate (135 hips; 135 patients). The data were missing for the leukocyte count for one hip and for the erythrocyte sedimentation rate for seven hips. The routine aspiration of the hip joint was performed under fluoroscopic control on an outpatient basis by a skeletal radiologist in the radiology department of our hospital. A clean procedure room was used. The anterior region of the hip was prepared with an iodine solution and was draped with sterile towels and an iodoform-impregnated plastic sheet. An 18-gauge spinal needle with a ten-milliliter syringe was used to aspirate fluid. If no fluid was aspirated on the initial attempt, the needle was redirected once or twice into the joint to obtain fluid. The fluid was promptly sent to the laboratory for culture and sensitivity tests. A differential cell count was not performed. The hips were not washed with saline solution or water, and repeat aspiration was not performed because we do not believe that fluid can be obtained consistently with this technique or that the fluid is likely to be representative of fluid from the hip joint. The revision arthroplasties were performed in a standard operating room without laminar airflow. From 1985 to 1992, the operating-room personnel wore body-exhaust suits. The hip was draped with an iodoform-impregnated plastic sheet. Antibiotics were withheld until joint fluid and tissue had been obtained for culture and sensitivity tests. A mean of 2.3 (range, one to ten) intraoperative swabs were obtained for culture. Of the twenty-one hips with positive intraoperative cultures, three had had one culture; three, two cultures; three, three cultures; and twelve, four to ten cultures. Again, a differential cell count of the joint fluid was not performed. Tissue from multiple areas was sent for routine histological evaluation. Cefazolin (or vancomycin hydrochloride, if the patient was allergic to penicillin) was then routinely administered for five days, at which time the final result of the intraoperative cultures was generally available. A hip was considered to be infected if grossly purulent fluid was found at the time of the revision or if there was growth on the solid medium of the intraoperative cultures. When more than one intraoperative culture swab had been obtained, the organism had to grow on at least half of the cultures for the hip to be considered infected. Growth in only the liquid medium (thioglycollate broth) of the intraoperative cultures was considered to be due to contaminants and was disregarded unless there were special microbiological considerations, such as growth of anaerobic bacteria (for example, Bacteroides species). The organisms were not evaluated for glycocalyx, and polymerase chain reaction was not performed for any of the samples. The reliability of the intraoperative culture as the standard for the diagnosis of an infection about the hip at our institution was determined in a separate, unpublished retrospective study. In 1983 and 1984, the senior one of us routinely obtained one or two specimens of fluid during sixty primary total hip arthroplasties. No previous operations had been performed on these hips, and infection was not suspected. A total of eighty-eight intraoperative cultures were performed. Only one hip had a positive intraoperative culture (for Peptostreptococcus) on the solid medium, and the intraoperative cultures for two hips were positive in the liquid medium only (one for Staphylococcus epidermidis and one for Bacillus subtilis). The underlying diagnosis for these three hips was osteoarthrosis, the prostheses had been implanted with cement, and there was no evidence of infection after two to ten years of follow-up. Thus, only one (1 per cent) of the eighty-eight intraoperative cultures was false-positive on the solid medium, and only two (2 per cent) were false-positive in the liquid medium. The medical records of the 140 patients in the present study were reviewed to document the results of the culture of preoperatively aspirated fluid, the intraoperative culture, the histological evaluation of permanent sections, the white blood-cell count, and the measurement of the erythrocyte sedimentation rate. In addition, the condition of the previous incision; symptoms at the time of presentation; a history of systemic symptoms, such as fever, chills, or sweats; and the patient's temperature on admission to the hospital were recorded. All but three patients were followed and examined at a mean of forty months (range, twenty-four to 108 months) postoperatively by the senior one of us or another orthopaedic surgeon. Two patients had died, within two years after the revision, of causes unrelated to the revision arthroplasty, and one patient refused to return for follow-up after one year. The microbiological and clinical data for these three patients were included in the analysis.
Statistical Analysis
Intraoperative Cultures and Cultures of Aspirated Fluid Preoperative aspiration was performed once for 128 hips and was repeated for fourteen. Thus, the results of a total of 156 aspirations were evaluated. The initial aspiration was positive, with growth on the solid medium, for seventeen hips. The organisms grown on culture of the aspirated fluid were identical to those grown on culture of the intraoperative specimens. Staphylococcus epidermidis was isolated from seven hips; Staphylococcus aureus, from six; Enterobacter cloacae, from two; Klebsiella pneumoniae, from one; and Peptostreptococcus, from one. Streptococcus sanguis was isolated from one hip; it grew, in the liquid medium only, on culture of purulent fluid obtained from the second aspiration. The first aspiration had also produced purulent fluid, but no growth had occurred on culture. Multiple cultures of intraoperative material showed growth of this organism on the solid medium. Bacteroides thetaiotaomicron (a strictly anaerobic organism) grew, in the liquid medium only, on culture of purulent fluid from one hip. Propionibacterium acnes grew, in the liquid medium only, on culture of material obtained from the first aspiration of three hips. Culture of material obtained from the repeat aspirations showed no growth in either the liquid or the solid medium, and multiple cultures of intraoperative specimens from these three hips showed no growth in either the liquid or the solid medium. Fourteen hips were aspirated a second time, for specific indications. Culture of aspirated fluid from five hips showed growth in the liquid medium only. Cultures of specimens from the repeat aspirations as well as those of the intraoperative specimens, however, showed no growth. An unusual organism grew on culture of aspirated fluid from five hips, and culture of the specimens from the second aspiration showed the same organism. Grossly purulent fluid was obtained from one hip during the first aspiration, but nothing grew on culture. Streptococcus sanguis grew in the liquid medium only of the culture of fluid from the second aspiration. This organism was also isolated on the solid medium of the intraoperative cultures. Three hips had been referred for operative treatment because of a so-called history of infection or because there was a clinical suspicion of infection; however, nothing grew on the culture of specimens from either the first or the second aspiration or on culture of the intraoperative specimens. With all twenty-one hips with a positive intraoperative culture on the solid medium considered to have been infected, the 156 aspirations had 125 true-negative, twenty-three true-positive, four false-positive, and four false-negative results. Therefore, preoperative aspiration of the hip joint had a sensitivity of 85 per cent, a specificity of 97 per cent, and an accuracy of 95 per cent. If the false-positive intraoperative cultures for the two hips are reclassified as true-negative, then the aspirations had 127 true-negative, twenty-three true-positive, four false-positive, and two false-negative results, and preoperative aspiration had a sensitivity of 92 per cent, a specificity of 97 per cent, and an accuracy of 96 per cent.
Erythrocyte Sedimentation Rate
Peripheral Leukocyte Count
Clinical Variables Infected hips were significantly associated (p = 0.0002) with the systemic symptoms of fever, chills, and sweats, but only four of the nineteen infected hips were associated with these symptoms and none of the 123 uninfected hips were. Combining the variables of an abnormally elevated erythrocyte-sedimentation rate and the duration that the implants had been in situ revealed a significant difference between the infected and uninfected hips. Thirty-six (29 per cent) of the 123 uninfected hips had had the implants in situ for more than five years and were associated with a normal erythrocyte-sedimentation rate, and fourteen (11 per cent) had had the implants in situ for five years or less and were associated with an abnormally elevated erythrocyte-sedimentation rate. None of the infected hips that had had the implants in situ for more than five years were associated with a normal erythrocyte-sedimentation rate, and ten infected hips that had had implants in situ for five years or less were associated with an abnormally elevated erythrocyte-sedimentation rate (Mantel-Haenzel chi square = 21.7; p < 0.001). For the hips in which the implants had been in situ for five years or less, the aspiration had a sensitivity of 95 per cent, a specificity of 95 per cent, and an accuracy of 95 per cent, with growth on the solid medium or grossly purulent fluid. For the hips associated with an abnormally elevated erythrocyte-sedimentation rate, the aspiration had a sensitivity of 90 per cent, a specificity of 100 per cent, and an accuracy of 98 per cent, when growth on the solid medium only was considered positive. The sensitivity, specificity, and accuracy were 100 per cent with regard to growth on the solid medium or grossly purulent fluid from aspiration of the hip.
Complications
Latent or low-grade infection should always be suspected and considered in the differential diagnosis of a painful hip after an arthroplasty3. If an infection is detected preoperatively, both the surgeon and the patient are better prepared for the operative procedure and appropriate specific antibiotic therapy can be initiated soon after the intraoperative specimens are sent for culture. At present, most surgeons favor so-called two-stage procedures to treat infection after total hip arthroplasty, with reimplantation of the prostheses after completion of an appropriate course of specific antibiotic therapy. The usefulness of routine aspiration of the hip joint for the detection of infection is a controversial issue. O'Neill and Harris reported that aspiration proved reliable in excluding infection in fifty-nine of sixty hips. Phillips and Kattapuram reported that preoperative aspiration had a sensitivity of 91 per cent and a specificity of 82 per cent in 141 hips. When fluid was obtained, infection was correctly determined in ten of eleven hips. However, a recent report by Barrack and Harris from the same institution refuted these findings. In a retrospective study of 270 hips, only six (2 per cent) were determined to be infected. Thirty-two (13 per cent) of the 254 hips that did not have an infection and had been aspirated successfully had a false-positive culture of aspirated fluid. Only two (6 per cent) of the thirty-four hips that had a positive culture of fluid from the initial aspiration had a true-positive result. Barrack and Harris concluded that aspiration should not be performed routinely but rather should be done only if the clinical history suggests infection or if radiographs demonstrate focal lysis, diffuse lysis, or periostitis. However, those authors did not report the sensitivity and specificity of these radiographic findings for the diagnosis of infection and they did not investigate the use of the peripheral leukocyte count, the erythrocyte sedimentation rate, or clinical variables in the diagnosis of infection. The authors also stated that, because there were so few infections, aspiration of the hip was not cost-effective. However, several other studies support routine preoperative aspiration at the sites of prosthetic hip joints. In a study comparing the results of 147 cultures of preoperative aspirate with those of intraoperative cultures, Tigges et al. reported a 10 per cent prevalence of infection (fourteen of 147 hips) after arthroplasty. Preoperative aspiration had a sensitivity of 92.8 per cent, a specificity of 91.7 per cent, and an accuracy of 91.8 per cent. Because there were eleven false-positive results, the positive predictive value was only 54.2 per cent. However, the report did not make clear whether organisms that grew only in the liquid medium were considered contaminants. The cost of aspiration of the hip was only $293, which was approximately 20 per cent of the cost of an indium-labeled leukocyte scan. Tigges et al. concluded that aspiration of the hip is an accurate and cost-effective method of evaluating a potential periprosthetic infection. In a prospective study comparing the relative value of three-phase bone-scanning, the erythrocyte sedimentation rate, and aspiration, Levitsky et al. found that preoperative aspiration was the most useful single test in the workup for a painful joint after a total hip or knee arthroplasty. The prevalence of infection was 10 per cent (seven of seventy-two joints). Kraemer et al. compared the sensitivity and specificity of sequential bone and gallium scans and aspiration of the hip with those of intraoperative culture (the "gold standard". The prevalence of infection was 27.8 per cent (twenty of seventy-two hips studied). Hips in which the components had been inserted without cement were more likely to be infected than those in which the components had been cemented. Aspiration of the hip had a sensitivity of 57 per cent and a specificity of 95 per cent. It had an inexplicably high prevalence of false-negative results. Sequential gallium bone scans, as described by Kroop et al., were not found to be effective for determination of infection. The gallium scan used by Kraemer et al. had a sensitivity of 38 per cent and a specificity of 100 per cent. In the present study, preoperative aspiration of the hip joint had a sensitivity of 92 per cent, a specificity of 97 per cent, and an accuracy of 96 per cent. These findings are similar to those of Tigges et al. Nineteen (13 per cent) of 142 hips were associated with a true-positive intraoperative culture and were definitely infected. This prevalence was much closer to that reported in several studies3,5,11 than to the low prevalence (2 per cent) reported by Barrack and Harris. We did reclassify the false-positive intraoperative cultures for two hips as true-negative. For both hips, only one culture swab had been obtained intraoperatively and the gross and histological examination of the tissues showed fibrosis without acute or chronic inflammation. In our study, routine culture of specimens obtained at the time of primary total hip arthroplasty did not have a high prevalence of false-positive results, in contrast to the findings of Barrack and Harris. We attribute the excellent sensitivity and specificity of preoperative aspiration of the hip joint in our study to several factors, including the meticulous technique and diligence of the skeletal radiologists who obtained the fluid. We considered organisms that grew only in the liquid medium to be contaminants unless there were special microbiological considerations, such as growth of anaerobic bacteria. The aspiration was repeated for fourteen hips for various reasons, but this did not change the diagnosis that had been determined on the basis of the first aspiration. Specifically, when culture of the preoperative aspirate from five hips showed growth of organisms (considered contaminants) only in the liquid medium, there was no growth on culture of material obtained at the second aspiration or culture of the intraoperative specimens. The erythrocyte sedimentation rate was abnormally elevated in association with seventeen of the nineteen truly infected hips, with a mean rate of 80.8 millimeters per hour for the infected hips. Fifty-eight (50 per cent) of the 116 uninfected hips were associated with an elevated erythrocyte-sedimentation rate, with a mean rate of only 32.0 millimeters per hour. This difference was significant (p = 0.001). Only ten patients in this study had rheumatoid arthritis, and only one of them had an infection. One of the fourteen patients who had the initial procedure performed for a miscellaneous diagnosis had ankylosing spondylitis, and one hip in this patient was infected. If these eleven patients are excluded, the analysis with the Fisher exact test still holds true. However, the peripheral leukocyte count was not helpful, as only five of the nineteen infected hips were associated with an elevated leukocyte count. We were unable to find an association between infected hips and several clinical variables, such as age, gender, the initial diagnosis, the patient's temperature on admission, or the clinical symptoms. However, the amount of time that the implants had been in situ was inversely associated with infection. The hips in which the implants had been in situ for more than five years were significantly (p = 0.008) less likely to be infected. In addition, multiple regression analysis showed significant (p < 0.001) differences between the infected and uninfected hips when an abnormal erythrocyte-sedimentation rate was combined with the amount of time that the implants had been in situ. None of the infected hips in which the implants had been in situ for more than five years had a normal erythrocyte-sedimentation rate. Thus, these two variables may be used to identify hips for which aspiration is probably not necessary before a revision operation. It has been suggested that the diagnosis of infection can be inferred from the clinical history or from plain radiographs showing diffuse or focal osteolysis1. We did not specifically re-examine all of the radiographs of the hips that were aspirated, as this was beyond the scope of this study. It has been reported that diffuse and focal osteolysis occur frequently in both infected and uninfected hips8,10. Periosteal elevation (periostitis) is so rarely seen that it is not helpful in the diagnosis of the infection. As far as we know, the sensitivity and specificity of plain radiographs for the diagnosis of infection have not been determined in a prospective study. Although Barrack and Harris stated that all six infected hips in their study had radiographic findings suggestive of infection, it was not reported how often these findings were seen in the hips that were not infected. Technetium bone scans were not performed for any hips in the present study. On the basis of the findings in this study, we conclude that aspiration of the hip joint before a revision procedure is valuable and accurate. We no longer perform a routine aspiration for hips in which the implants have been in situ for more than five years if the patient's erythrocyte sedimentation rate is normal. We recommend preoperative aspiration for all hips that are painful five years or less after an arthroplasty or if the erythrocyte sedimentation rate is abnormally elevated. This selective approach to preoperative aspiration is currently being studied in a prospective manner.
NOTE: The authors thank Stuart Gansky, M.S., and Gary Koch, Ph.D., for help with the statistical analysis.
*No benefits in any form have been received or will be received from a commercial party related directly 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. |