 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1999 The Journal of Bone and Joint Surgery, Inc. Prospective Analysis of Preoperative and Intraoperative Investigations for the Diagnosis of Infection at the Sites of Two Hundred and Two Revision Total Hip Arthroplasties* , VANCOUVER, BRITISH COLUMBIA, CANADAInvestigation performed at the Departments of Orthopaedics and Pathology, University of British Columbia, and Vancouver Hospital and Health Sciences Center, Vancouver
Background: Total hip arthroplasty is a commonly performed procedure in the United States and Canada that is associated with a definite risk of postoperative infection. Moreover, diagnosing an infection after total hip arthroplasty can present a challenge as there are no preoperative tests that are consistently sensitive and specific for infection in patients who need a revision arthroplasty. The present prospective study was performed to evaluate a variety of investigations for the diagnosis of infection at the site of a previous arthroplasty in order to determine if any combination of diagnostic studies could be used to determine which patients are at risk for a postoperative wound infection. Methods: We prospectively analyzed the preoperative and intraoperative investigations used for the diagnosis of infection in 178 patients who had a total of 202 revision hip replacements. Clinical data were collected preoperatively. Investigations to determine the presence or absence of infection included a white blood-cell count, measurement of the erythrocyte sedimentation rate, measurement of the level of C-reactive protein, preoperative aspiration of the joint, intraoperative gram-staining and culture of periprosthetic tissue, a white blood-cell count in synovial fluid, and examination of intraoperative frozen sections. Frozen sections were analyzed in a blinded fashion without knowledge of clinical or laboratory data. Patients receiving antibiotics at the time of aspiration or collection of specimens for intraoperative culture were excluded from the analysis of those investigations, regardless of the results of the cultures. A positive result (suggestive of infection) was clearly defined for each of the investigations. Results: Thirty-five hips (17 percent) were determined to be infected on the basis of clinical findings and positive results, according to the defined criteria, of investigations. With inflammatory conditions excluded, the sensitivity, specificity, positive predictive value, and negative predictive value were 0.82, 0.85, 0.58, and 0.95, respectively, for the erythrocyte sedimentation rate and 0.96, 0.92, 0.74, and 0.99, respectively, for the level of C-reactive protein. All patients who had a periprosthetic infection had an elevated erythrocyte sedimentation rate or level of C-reactive protein, but not always both. When patients who were receiving antibiotics were excluded, the results of aspiration of the joint were 0.86 for sensitivity, 0.94 for specificity, 0.67 for the positive predictive value, and 0.98 for the negative predictive value. Intraoperative studies revealed sensitivities, specificities, positive predictive values, and negative predictive values of 0.19, 0.98, 0.63, and 0.89, respectively, for gram-staining of specimens of the most inflamed-appearing tissue; 0.36, 0.99, 0.91, and 0.90, respectively, for the white blood-cell count in synovial fluid; and 0.89, 0.85, 0.52, and 0.98, respectively, for a neutrophil count in synovial fluid of more than 80 percent. The sensitivity, specificity, positive predictive value, and negative predictive value were 0.80, 0.94, 0.74, and 0.96, respectively, for the frozen sections and 0.94, 0.97, 0.77, and 0.99, respectively, for the intraoperative cultures. Conclusions: The combination of a normal erythrocyte sedimentation rate and C-reactive protein level is reliable for predicting the absence of infection. Aspiration should be used when the erythrocyte sedimentation rate or the C-reactive protein level is elevated or when a clinical suspicion of infection remains. We found the gram stain to be unreliable. Examination of intraoperative frozen sections is useful in equivocal cases or when hematological markers may be falsely elevated because of an inflammatory or other condition.
The rate of infection after total hip replacement has decreased considerably since 1964, when Charnley reported a rate of infection of more than 9 percent (nineteen of 199)5. A variety of measures are responsible for this decrease18,27. However, infection is still the cause of failure after 1 percent (seventy-one of 5081 and twenty-seven of 2084) to 2 percent (ninety-four of 5500 and thirty-six of 1798) of primary total hip arthroplasties, and the rate of failure due to infection is even higher after revision procedures6,10,15,17,24,25,29,38,40. Even though these rates of infection are low, the fact that a large number of hip arthroplasties are performed each year means that infection will develop in a number of patients after total hip arthroplasty. The cost of caring for patients who have a postoperative infection places a substantial burden on the health-care system41. Infection at the site of an arthroplasty is costly to treat because additional operations and a prolonged duration of hospitalization are often necessary. In the United States, the annual cost to treat the 3500 to 4000 infections that develop after arthroplasties each year is between 150 and 200 million dollars11. With an aging population that will need an increasing number of arthroplasties, rational methods to prevent, diagnose, and treat this complication must be found in order to reduce both the cost of total hip arthroplasty to society and the substantial impact of an unrecognized infection, which may lead to failure, on a patient's quality of life. Infection after total hip arthroplasty can present a diagnostic challenge. There are no preoperative tests that are consistently 100 percent sensitive and specific; thus, the diagnosis of infection depends on the surgeon's judgment with respect to the clinical presentation, findings on physical examination, and interpretation of the results of investigations. The consequences of misdiagnosis are substantial. Reimplantation of a prosthesis into an infected host bed, without appropriate débridement, is likely to result in persistent infection28. Numerous tests are available for the investigation and diagnosis of a failed total hip replacement. These tests can be divided into preoperative and intraoperative investigations. Preoperative investigations include hematological screening tests (white blood-cell count, measurement of the erythrocyte sedimentation rate, and measurement of the level of C-reactive protein), aspiration of the hip joint, plain radiography, and radionuclide imaging studies. Intraoperative investigations include analysis of synovial fluid, gram-staining of specimens of the most inflamed-appearing tissue, histological evaluation of frozen sections of the most inflamed-appearing tissue, and intraoperative cultures of periprosthetic tissue. The most commonly used preoperative investigations are laboratory screening tests, such as measurements of the erythrocyte sedimentation rate and the level of C-reactive protein, as well as aspiration of the hip joint. Measurement of the erythrocyte sedimentation rate is often criticized for its lack of specificity due to false elevations resulting from other conditions, such as rheumatoid arthritis or other connective-tissue diseases, and its lack of sensitivity due to normal values despite known infection11,37,39,42. In the past, aspiration of the hip joint has been recommended as a routine investigation to exclude the possibility of periprosthetic infection16,33,37. However, more recently, many investigators have favored a more limited role for aspiration; they have recommended that it be used either to confirm a clinical suspicion of infection or to support or negate the results of other preoperative investigations3,13,23,31,44. Although there have been some recent prospective studies on the efficacy of certain diagnostic tests19,26,37, most reports on the diagnostic value of tests that are available to exclude or diagnose infection generally are based on a review of retrospective studies of one or more diagnostic tests2,3,7,12,14,23,34,36,44. Important factors that are difficult to control in retrospective studies include the use of antibiotics before the preoperative aspiration and before the operation as well as associated conditions that may lead to an elevated erythrocyte sedimentation rate or level of C-reactive protein. These factors may be responsible for the poor specificity reported for the erythrocyte sedimentation rate and the poor sensitivity reported for preoperative aspiration in some series3,13. Moreover, the value of some reports has been reduced further because conclusions were based on a small number of periprosthetic infections3,14. Finally, confidence intervals often were not included in these studies. To our knowledge, no other authors have prospectively analyzed the results of a large number of investigations in the same group of patients. The present study was undertaken to evaluate, in a prospective manner, the value of a variety of investigations for the diagnosis of infection at the site of a total hip arthroplasty.
A prospective analysis was performed on 178 patients who had a total of 202 revision hip replacements between April 1, 1994, and March 31, 1996. The mean age of the patients at the time of the revision was sixty-five years (range, twenty-seven to eighty-eight years). There were 122 women and fifty-six men. Of the twenty-four patients who had more than one procedure, four had a revision of the contralateral hip and twenty had a two-stage revision of the ipsilateral hip because of infection. Additional patients had a two-stage exchange because of infection, but one of the two procedures was not included because it was performed outside the study period. Most patients had had multiple procedures performed on the hip before the index revision: 119 hips had had at least two previous operations and ninety-seven had had at least one previous revision procedure. The primary diagnosis was osteoarthritis in ninety-nine hips, posttraumatic arthritis in twenty-five, a connective-tissue disorder (rheumatoid arthritis, systemic lupus erythematosus, ankylosing spondylitis, or psoriatic arthritis) in twenty-three, osteonecrosis in nineteen, conversion to a total hip arthroplasty because of a failed hemiarthroplasty in seventeen, congenital dysplasia in fourteen, and other disorders (Legg-Perthes disease, slipped capital femoral epiphysis, or spontaneous fusion due to septic arthritis) in five. All patients had a variety of preoperative and intraoperative investigations, which will be described, to determine the presence or absence of infection. Each patient provided a thorough history and had a physical examination. Recorded data included any local or systemic signs or symptoms of infection, either immediately after the index procedure or at the time of presentation, as well as the use of antibiotics before either the aspiration of the hip or the revision procedure. Preoperative investigations to determine the presence of infection included a white blood-cell count, measurement of the erythrocyte sedimentation rate, measurement of the level of C-reactive protein, and aspiration of the hip joint. Intraoperative investigations included a white blood-cell count in synovial fluid, gram-staining of specimens of the most inflamed-appearing tissue, culture of periprosthetic tissue, and histological evaluation of frozen sections from intraoperative samples of periprosthetic tissue. A positive result (suggestive of infection) or a negative result (not suggestive of infection) was defined for each investigation.
Preoperative Investigations
Hematological Investigations White blood-cell count: A white blood-cell count was considered to be elevated when it was more than 11.0 x 109 per liter. The number of polymorphonuclear cells was considered to be increased (a so-called left shift) when more than 75 percent of the total white blood-cell count consisted of granulocytes. Erythrocyte sedimentation rate and level of C-reactive protein: For the purposes of this analysis, an erythrocyte sedimentation rate of more than thirty millimeters per hour and a C-reactive protein level of more than ten milligrams per liter were considered to be suggestive of infection and were deemed a positive result. The erythrocyte sedimentation rate and the level of C-reactive protein from all patients who had a connective-tissue disorder were excluded from the analysis in order to minimize the predictable false-positive elevations.
Aspiration of the Hip Joint A protocol for repeat aspiration was used when the findings based on the initial aspiration were not consistent with the clinical impression. If there were no clinical signs or symptoms of infection and the hematological investigations revealed negative findings but the aspiration revealed positive findings, then the aspiration was repeated. For the purposes of analysis, the results of an aspiration were considered positive (suggestive of infection) if there was growth in any of the cultures, including late growth of bacteria on subculture. This definition was chosen because in a clinical situation in which only preoperative investigations are available or in which aspiration yields only one or two samples, any positive result on culture of aspirate raises concern about a potential infection.
Intraoperative Investigations
White Blood-Cell Count in Synovial Fluid
Analysis of Frozen Sections
Culture and Gram-Staining Cultures of tissue specimens and those of material obtained by swabbing of the prosthesis were analyzed separately. In both analyses, a result was considered positive when more than one-third of the specimens showed growth. If only one specimen (of a minimum of three) showed growth, the result was considered negative (a contaminant) and not suggestive of infection. A gram stain was considered to have a true-positive result if the organism that was seen was consistent with the organism grown on culture. When a hip had a periprosthetic infection but the result was negative on culture (because of preoperative antibiotic therapy), a negative gram stain was considered a true-negative result and not a false-negative result.
Diagnosis of Infection
Sensitivity, specificity, positive and negative predictive values, and 95 percent confidence intervals were calculated for each of the tests as well as for combinations of tests.
Thirty-five (17 percent) of the 202 hips that had a revision arthroplasty in the present study were infected. The high prevalence of infection in this series reflects the fact that our institution is a tertiary referral center for total joint arthroplasties complicated by infection. Of the thirty-five hips (thirty-five patients) that had a periprosthetic infection, twenty-seven were suspected of having an infection on the basis of the history and the physical examination. Thirteen of the twenty-seven patients had wound drainage at the time of presentation, ten had acute pain and systemic signs and symptoms of infection at the time of presentation, three reported a history of postoperative wound infection after the most recent operation on the hip, and one had acute pain without signs or symptoms of infection but there was a clear history of infection elsewhere in the body preceding the onset of symptoms. In the eight patients who did not have a history or clinical findings suggestive of infection, the erythrocyte sedimentation rate or the level of C-reactive protein was elevated above the reference values of thirty millimeters per hour and ten milligrams per liter, respectively. Therefore, a diagnosis of infection was apparent on the basis of the history, physical examination, or hematological investigation for all thirty-five patients. Ten of the thirteen patients who had wound drainage (purulent drainage in which the sinus extended down to the periprosthetic space) at the time of presentation also had at least three positive investigations. Two of the remaining three patients who had wound drainage had positive cultures; however, these were excluded from the analysis as the patients were receiving concurrent antibiotics. The remaining patient who had wound drainage had negative cultures, but the patient had been receiving suppressive antibiotics for more than two months because of an acute infection following a revision hip replacement. Twenty-one patients (twenty-one hips) had a minimum of three positive investigations. Nine of them also had signs or symptoms of infection (fever, chills, acute onset of pain in the hip, or purulent fluid within the joint). One patient was first seen with a history of infection followed by an acute onset of pain in the right hip, which had been functioning well for eleven years. The aspiration of the hip, which had been performed elsewhere, revealed positive findings, and the patient had been managed with suppressive antibiotics for five months before being seen by us for additional management. Turbid fluid was present within the hip joint intraoperatively. In addition, examination of the frozen sections revealed acute inflammation. The intraoperative cultures were negative because of the prolonged antibiotic suppression. The results of each preoperative and intraoperative investigation were analyzed individually, and the sensitivity, specificity, and positive and negative predictive values are given with the 95 percent confidence interval in parentheses.
White Blood-Cell Count
Erythrocyte Sedimentation Rate
Level of C-Reactive Protein
Preoperative Aspiration
White Blood-Cell Count in Synovial Fluid
Gram Stain
Analysis of Frozen Sections
Intraoperative Cultures
Combined Results
The exclusion of the presence of infection at the site of a failed total hip replacement is imperative for operative planning. A variety of investigations can be used to assist in the diagnosis or exclusion of infection. Ideally, the number of investigations should be kept to a minimum while at the same time ensuring an accurate diagnosis. In addition, investigations should be cost-effective and of minimum inconvenience or risk to the patient. The white blood-cell count rarely aids in the diagnosis of infection. Systemic infection is rare. Only seven (20 percent) of the thirty-five patients (thirty-five hips) who had an infection had an elevated white blood-cell count. Of the seven patients (seven hips) who had an elevated white blood-cell count, six had only mild elevation and only one had a count of more than 15 x 109 white blood cells per liter. In a tertiary-care setting, patients who have a periprosthetic infection are frequently referred from other centers, often after the diagnosis is established. Patients are often receiving antibiotics and are first seen when the infection is subacute or chronic. The lack of systemic symptoms and the delay in presentation may account for the normal white blood-cell count in most patients who have an infection at the site of a hip replacement. Canner et al. also noted a low prevalence of leukocytosis in patients who had an infection after a joint arthroplasty4. They reported that only eight of fifty-two patients who were first seen with a periprosthetic infection had an elevated white blood-cell count. The erythrocyte sedimentation rate and the level of C-reactive protein are nonspecific markers of acute inflammation. The erythrocyte sedimentation rate is a measure of erythrocyte rouleaux formation, whereas C-reactive protein is an acute-phase reactant produced by the liver. Both may be elevated in association with inflammatory, infectious, or neoplastic processes. These investigations are often criticized for their lack of specificity. However, if a careful history is obtained, with possible causes of false elevations (such as an active connective-tissue disease or a recent operation9,19,42) taken into account, these investigations become useful as a safe and economical screening tool with which to exclude infection. When patients who had inflammatory arthritis were excluded from the present study, the sensitivity and specificity of the erythrocyte sedimentation rate were 0.82 (twenty-eight of thirty-four) and 0.85 (117 of 137), respectively. Four of the twenty false-positive results in this series may have been due, in part, to a recent operation (within three months); however, three patients who had had a recent operation also had a negative finding on measurement of the erythrocyte sedimentation rate. The level of C-reactive-protein may be more reliable in this regard, as it returns to normal faster than the erythrocyte sedimentation rate does1,19,39,42. In the present series, the level of C-reactive protein was a more accurate indication than the erythrocyte sedimentation rate, with a sensitivity of 0.96 (twenty-five of twenty-six) and a specificity of 0.92 (107 of 116). Of particular interest are the false-negative results for the patients who had an infection. Often these patients had been either managed with suppressive antibiotics for a prolonged period of time or had a chronic low-grade infection. Of the six false-negative results of the measurements of the erythrocyte sedimentation rate, four were for patients who had had long-term management with antibiotics. The one false-negative result for the level of C-reactive protein was also for a patient who had been managed with suppressive antibiotics; however, the measurement of the erythrocyte sedimentation rate revealed a positive finding (seventy-two millimeters per hour). There were no periprosthetic infections associated with both a normal erythrocyte sedimentation rate and a normal level of C-reactive protein. Although measurement of the level of C-reactive protein is more efficacious than measurement of the erythrocyte sedimentation rate for the diagnosis and exclusion of infection, the ease of measurement of the erythrocyte sedimentation rate and the minimum cost associated with the test, as well as our observation that none of the infected hips had negative findings on both tests, suggest that the combined use of these two investigations is necessary in the assessment of a failed hip arthroplasty. These results are similar to those of Sanzén and Carlsson, who reported that only one of twenty-three patients with a periprosthetic infection had both a normal erythrocyte sedimentation rate and a normal level of C-reactive protein, with use of values of more than thirty millimeters per hour and ten milligrams per liter, respectively, as criteria for an abnormal finding39. The remaining twenty-two patients had a positive finding on one test or the other. Recently, the role of aspiration has generated considerable debate. In the past, aspiration was advocated for all patients who had a failed hip replacement16,33,37. At present, many3,13,23,31,44 favor a more limited role, with the test used to confirm a clinical suspicion of infection or as an adjuvant investigation when hematological values, such as the erythrocyte sedimentation rate and the level of C-reactive protein, may be falsely elevated because of conditions such as a connective-tissue disease, concurrent infection (for example, a urinary tract infection), injury, or a recent operation. An additional benefit of preoperative aspiration when periprosthetic infection is suspected is the identification of the organism. This information allows better operative planning, as the surgeon can choose the appropriate antibiotic-impregnated cement if cement is to be used. Identification of the organism also is useful when a patient is not managed with a revision operation but instead is managed with suppressive antibiotics or when a patient has a systemic infection and needs antibiotics before being operated on. The reported sensitivity and specificity of preoperative aspiration of the hip have varied widely. The sensitivity of preoperative aspiration has ranged from 0.50 to 0.93 (two of four3, three of six13, thirty of thirty-three36, twenty-three of twenty-five23, and thirteen of fourteen44), whereas the specificity has ranged from 0.82 to 0.97 (eighty-nine of 10936, 141 of 16013, sixty of sixty-three37, and 127 of 13123). This wide variation is partly a function of the different methods that have been used to calculate these results. Some variations in methodology include the criteria used for interpretation of a positive result (growth of bacteria in fluid compared with that on a solid medium); the number of aspiration samples that determine a positive result; inclusion of repeat aspirations with the initial aspirations to determine the result; the lack of a so-called gold standard for comparison (many authors compare results with those of intraoperative cultures alone, which also have false results); and unrecognized or unreported use of antibiotics before aspiration, an inherent problem in many retrospective studies. In the present study, thirteen of 193 hips were in patients who were managed with antibiotics before the aspiration was performed. Twelve of the thirteen had an infection, but only six had a positive result on aspiration. According to our study protocol, all patients receiving antibiotics were excluded from the analysis. When use of antibiotics is controlled, preoperative aspiration has a reasonably high sensitivity (0.86) and specificity (0.94). Histological analysis of frozen sections for the diagnosis of infection was described, in 1976, by Mirra et al.30. Although some investigators have recommended against its use because of poor sensitivity12, authors of the most recent reports have concluded that it is a reliable intraoperative test for differentiation between aseptic failure and failure due to infection2,14,26. Reported sensitivities have ranged from 0.18 to 1.00 (two of eleven12, sixteen of nineteen26, twenty of twenty-two2, and nine of nine14) and specificities, from 0.90 to 0.99 (114 of 12612, eighty-one of eighty-four2, and 154 of 15626). In larger studies with more infections, including the present series, the sensitivities have ranged from 0.80 to 0.91 (twenty-eight of thirty-five, sixteen of nineteen26, and twenty of twenty-two2) and the specificities, from 0.94 to 0.99 (157 of 167, eighty-one of eighty-four2, and 154 of 15626). The variable results may be due in part to the methodology of the studies, including the criteria used to define infection. Many authors have compared the results of the evaluation of frozen sections with those of intraoperative cultures2,12,14,26. This comparison may not be an accurate reflection of periprosthetic infection, as suggested by the false results of the cultures in the present study and in those reported by others3,13,16,32,34,43. Furthermore, our pathologist was blinded to the diagnosis and the planned treatment, something that has not been reported in other studies. A discrepancy between the findings on evaluation of frozen sections and the final histological result is attributable to sampling error. This discrepancy occurred in the evaluations of nine hips in the present study. In the investigation of seven of the nine hips, the evaluation of the frozen sections yielded an incorrect result and the final histological result was correct. To minimize these sampling errors, tissue submitted for preparation of frozen sections should be taken from areas that appear most suggestive of infection. Evaluation of frozen sections is a useful investigation, particularly when the erythrocyte sedimentation rate or the level of C-reactive protein may be falsely elevated, provided that an experienced musculoskeletal pathologist is available to interpret the results. The value of analysis of synovial fluid for the diagnosis of septic arthritis is well established. Although some reports have suggested that there is considerable variability and overlap among various diagnoses, including septic arthritis19,21, there is general agreement that a white blood-cell count in synovial fluid of more than 50.0 x 109 per liter and a neutrophil count of more than 80 or 90 percent is suggestive of a bacterial infection of the joint20-22,27,35. In a thorough search of the literature, we could not find a source that established whether these values are similar to or different from those for infection involving total joint arthroplasties. Windsor and Insall stated that if the white blood-cell count in the synovial fluid has more than 25.0 x 109 polymorphonuclear leukocytes per liter and the differential count is more than 75 percent of these white cells, infection should be suspected at the site of a failed total knee arthroplasty45. However, the basis for these values was not given. Eftekhar recommended use of a total white blood-cell count of 80.0 x 109 per liter and a polymorphonuclear count of 80 percent as suggestive of infection11. Again, the basis for these values was not provided. In the present study, values that are suggestive of bacterial arthritis (a total count of more than 50.0 x 109 per liter and more than 80 percent neutrophils) were used for analysis. The total white blood-cell count in synovial fluid was found to be an unreliable indicator of infection because of the frequent occurrence of false-negative results (eighteen of twenty-eight periprosthetic infections), with a sensitivity of only 0.36. An elevated neutrophil count, although a more sensitive indicator (0.89), had a positive predictive value of only 0.52 because of the high number of false-positive results (twenty-three of forty-eight positive results). Although synovial fluid is easily obtainable on exposure of the pseudocapsule, the poor sensitivity and poor positive predictive value as well as the delay in obtaining the result intraoperatively make this investigation less reliable and impractical for routine use. We found that the gram stain was an unreliable investigation, with a sensitivity of only 0.19 (five of twenty-seven). The results of gram-staining were assessed by comparison with the results of intraoperative cultures of tissue alone. A negative finding on gram-staining of a specimen from a patient with an infection who has negative findings on culture because of antibiotic therapy should not be considered to be false-negative. Even so, there were twenty-two false-negative results on gram-staining of the specimens from the thirty-five patients who had an infection. The gram stain correctly identified the infecting organism in five of the infected hips (true-positive results) and was negative for eight infected hips for which the cultures were negative because of concurrent antibiotic therapy (true-negative results). Of the 167 hips that did not have an infection, 164 had a negative result on gram-staining. A surprising finding was the three positive gram stains in the group without infection. All preoperative and intraoperative cultures were negative for these three hips, and none of the other diagnostic parameters were suggestive of infection. The false-positive gram stains can only be explained by laboratory contamination or error on the part of the interpreter. There was no other evidence of infection in the patients for whom the gram stain was false-positive. These results suggest that the gram stain should not be relied on to help to establish a diagnosis of infection or be used intraoperatively to determine treatment. A similar conclusion was reached by Chimento et al., who recently reported that thirty-two of 169 hips with a negative finding on gram-staining had an infection, resulting in a sensitivity of zero7. Intraoperative cultures have traditionally been used as the so-called gold standard for the diagnosis of infection. Although intraoperative cultures were the most accurate of the investigations, six (3 percent) of the 180 tissue cultures yielded a false result (five false-positive results and one false-negative result). Others have reported rates of false-positive results of intraoperative cultures ranging from 2.4 to 31.5 percent (twenty-eight of 118743, 111 of 43715, and 233 of 74032) in association with primary hip replacements. In more recent studies, the rates of false-positive results of intraoperative cultures have been reported to be between 13 and 31 percent (twenty-two of 16613, fifty-four of 2693, and forty of 12934) in association with revision hip replacements. The low prevalence of false-negative results of the cultures (one tissue culture and four cultures of material obtained by swabbing of the prosthesis) in the present study was probably due to the exclusion of patients who were receiving antibiotics from the analysis of the results. Twenty-two hips were excluded for this reason. Of those excluded hips, eight had positive findings on culture of tissue specimens and six had positive findings on culture of material obtained by swabbing of the prosthesis. There was no substantial difference between the results on culture of tissue compared with those on culture of material obtained by swabbing of the prosthesis. We see no advantage in obtaining material by swabbing of the prosthesis because the intraoperative administration of antibiotics would be delayed if the surgeon chose to withhold the antibiotics until the prosthesis was removed, and this may place the patient at increased risk for infection8. Instead, tissue specimens should be obtained with clean instruments immediately after the pseudocapsule is opened, and they should be taken from the most inflamed-appearing tissue. In summary, the exclusion of the diagnosis of infection as a cause of failure is imperative to determine the management of patients who need a revision total hip replacement. The key to making the correct diagnosis with reasonable certainty is not the use of a single investigation but rather the use of a correct combination of investigations. In most instances, a diagnosis of infection can be made or excluded on the basis of a carefully obtained clinical history and the measurement of the erythrocyte sedimentation rate and the C-reactive protein level. If the history is not suggestive of infection and if both of these inflammatory markers are negative, then infection is unlikely. When one of these inflammatory markers is positive and the other is negative, we recommend repeating the serological investigations or proceeding with aspiration. The interpretation of an intraoperative frozen section by an experienced musculoskeletal pathologist adds reasonable assurance of a correct diagnosis. When infection is suspected or the inflammatory markers are elevated, aspiration of the hip joint before the operation is indicated. This recommendation also applies to patients with a connective-tissue disease who have an elevated erythrocyte sedimentation rate and C-reactive protein level because of active inflammatory disease. Aspiration may also be indicated if a reliable interpretation of a frozen section is not available intraoperatively. Intraoperative cultures, although still the most accurate, are also plagued by false-positive and false-negative results. The occurrence of false results makes intraoperative culture alone a poor gold standard for comparison with other investigations in the diagnosis of infection associated with a failed total hip arthroplasty. A diagnosis of infection should be based on clearly defined criteria, with consideration of both the clinical presentation and the well defined results of preoperative and intraoperative investigations.
*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. |