Imaging-Guided Bone Biopsy for Osteomyelitis: Are There Factors Associated with Positive or Negative Cultures?

Imaging-Guided Bone Biopsy for Osteomyelitis Musculoskeletal Imaging Original Research Jim S. Wu 1 Tetyana Gorbachova 2 William B. Morrison 3 Andrew H. Haims 4 Wu JS, Gorbachova T, Morrison WB, Haims AH Keywords: biopsy, bone, CT, fluoroscopy, infectious diseases, osteomyelitis DOI:10.2214/AJR.06.1286 Received September 27, 2006; accepted after revision January 26, 2007. 1 Department of Radiology, Beth Israel Deaconess Medical Center, 330 Brookline Ave., Shapiro 4th Fl., Boston, MA 02215. Address correspondence to J. S. Wu (jswu@bidmc.harvard.edu). 2 Department of Radiology, Hahnemann University Hospital, Philadelphia, PA. 3 Department of Radiology, Thomas Jefferson Medical College, Philadelphia, PA. 4 Department of Diagnostic Radiology, Yale University, New Haven, CT. AJR 2007; 188:1529 1534 0361 803X/07/1886 1529 American Roentgen Ray Society Imaging-Guided Bone Biopsy for Osteomyelitis: Are There Factors Associated with Positive or Negative Cultures? OBJECTIVE. The objective of our study was to identify the clinical and technical factors associated with positive or negative culture results in histologically positive cases of osteomyelitis obtained from imaging-guided bone biopsies. MATERIALS AND METHODS. A retrospective review was performed of 800 consecutive patients undergoing imaging-guided core bone biopsies at two institutions. Seventy-five biopsies were performed for suspected osteomyelitis and 41 patients had histologically proven osteomyelitis. A chart review was performed to determine whether the following factors affected the culture result: histologic type of osteomyelitis, antibiotic therapy before biopsy, fever (temperature 38.0 C), elevated WBC count ( 10 10 3 μl), elevated erythrocyte sedimentation rate (ESR) ( 10 mm/h), elevated C-reactive protein value (CRP) ( 6 mg/l), the size of the biopsy needle, and the amount of purulent fluid obtained at biopsy. RESULTS. Of the 41 cases of osteomyelitis, 14 (34%) had positive cultures. Eighteen (44%) of 41 cases were chronic osteomyelitis. Seventeen (41%) of 41 patients received antibiotics before biopsy, seven (17%) were febrile, five (12%) had an elevated WBC count, 16 (39%) had an elevated ESR, and six (15%) had an elevated CRP value. The biopsy needle size ranged from 11- to 18-gauge. These factors did not have any significant association with positive or negative culture results. Purulent fluid was aspirated in 10 (24%) of the 41 cases. In six (15%) of the cases, 2 ml of purulent fluid was aspirated and five (83%) of the six cases were associated with positive culture (p =0.02). CONCLUSION. The rate of positive culture results in histologically proven cases of osteomyelitis obtained from imaging-guided bone biopsies is low. Aspirating 2 ml of purulent fluid is associated with a significantly higher rate of positive cultures. steomyelitis is characterized by inflammation of the bone marrow and O adjacent bone and is often associated with cortical and trabecular destruction. It can be caused by bacteria, fungi, and a variety of other organisms [1]. Early diagnosis and treatment of osteomyelitis are essential because undiagnosed cases can lead to chronic pain, amputation, and death [2, 3]. Unfortunately, diagnosing osteomyelitis is difficult. Although clinical symptoms; inflammatory markers in the blood; and findings on MRI, leukocyte scintigraphy, and PET can suggest osteomyelitis, the definitive diagnosis of osteomyelitis is made by culturing an organism directly from the site of infection [3 6]. The identification of a causative organism by culture both confirms osteomyelitis and allows tailoring of antimicrobial therapy; however, cultures from samples obtained during surgery or by imaging guidance are often negative. Several studies suggest that 40 60% of histologically proven cases of osteomyelitis at surgery or biopsy are negative at culture, but many of these studies involve small patient populations [5 8]. The factors that predict positive or negative culture results are unknown. Treatment with antimicrobial therapy around the time of tissue sampling, small biopsy tissue volume, and sampling error are factors that may affect culture results [2, 9 12]. In several studies, investigators have stressed the importance of sending both histologic and microbiologic samples at the time of biopsy given the low rate of positive culture [5 7, 11]. Clinical and laboratory factors associated with osteomyelitis include fever, elevated WBC count, elevated erythrocyte sedimentation rate (ESR), and elevated C-reactive protein value (CRP). Unfortunately, many of these factors can be negative at clinical presentation, and some studies indicate that these factors are best re- AJR:188, June 2007 1529

served for monitoring treatment rather than determining diagnosis [3, 5, 11]. Imaging-guided bone biopsy with CT or fluoroscopy is a useful technique in diagnosing osteomyelitis and is the preferred initial technique for obtaining both histologic and microbiologic samples. This procedure can be performed on an outpatient basis, and complication rates are low [13 15]. For this study, we determined the rate of positive culture in histologically proven cases of osteomyelitis obtained from imagingguided biopsies and identified clinical or technical factors associated with positive or negative culture results. We hypothesized that the culture positivity rate would be low and that certain clinical and technical factors would affect the culture positivity rate. Materials and Methods We performed a retrospective review of 800 consecutive imaging-guided core bone biopsies at two large tertiary care medical centers. Between January 1998 and December 2005, 225 imaging-guided bone biopsies were performed at the first institution and 575 biopsies were performed at the second institution from January 1999 to March 2005. Of the 800 total bone biopsies performed under CT or fluoroscopy guidance, 75 patients had samples sent for both histologic and microbiologic analyses. In all 75 patients, osteomyelitis was suspected; however, the level of suspicion varied. In some patients, the diagnosis of osteomyelitis was nearly certain and the biopsy was performed to tailor antibiotic therapy. In other patients, osteomyelitis was very low on the differential and the culture samples were obtained only for diagnostic completeness. Patients referred for imaging-guided biopsies are typically greater diagnostic challenges than those seen at surgery. We did not include spine biopsies in this series because culture positivity rates may be higher in the spine than in the remainder of the skeleton [16, 17]. This study was approved by the institutional review boards of both institutions. Musculoskeletal radiologists at the two institutions obtained samples using standard coaxial bone biopsy techniques [18 20]. Most of the biopsies were performed with either a 15-gauge needle (1.7- mm-diameter bore) (Bonopty, Radi Medical Systems) or a 14-gauge needle (2.03-mm-diameter bore) (Elson/Ackerman, Cook). At the beginning of the procedure, the target lesion was localized under CT or fluoroscopy guidance. Once the patient had received appropriate local anesthetic and was under conscious sedation, a penetration cannula was placed adjacent to the outer cortex of the lesion. Subsequently, a biopsy cannula was used to obtain core bone samples. An average of three samples were obtained at both institutions. Aspiration of the lesion for purulent fluid is attempted during each biopsy. At least one core sample was sent for microbiologic analysis. We defined a positive case of osteomyelitis for this study as a histologic report indicating acute osteomyelitis, chronic osteomyelitis, or acute and chronic osteomyelitis. Cases of acute osteomyelitis showed the presence of acute inflammatory cells, congestion or thrombosis of medullary or periosteal small vessels, and necrotic bone. Cases of chronic osteomyelitis exhibited areas of woven bone and fibrosis with large numbers of lymphocytes, histiocytes, and plasma cells in the absence of neutrophils. Features of both were seen in cases of acute and chronic osteomyelitis. Forty-one of the 75 patients (55%) had a histologic diagnosis of osteomyelitis. For microbiologic analysis, we considered a culture positive if any organism grew. We performed a chart review of the 41 histologically positive cases of osteomyelitis to determine whether the following factors contributed to positive or negative culture results: histologic type of osteomyelitis (acute or chronic), antibiotic therapy before biopsy, fever (temperature 38.0ºC), elevated WBC count ( 10 10 3 μl), elevated ESR ( 10 mm/h), elevated CRP value ( 6 mg/l), biopsy needle size, and the amount of purulent fluid obtained at biopsy. For the purposes of this study, histology samples interpreted as acute osteomyelitis or as acute and chronic osteomyelitis were considered to be cases of acute osteomyelitis. A patient was considered to have received antibiotic therapy if any dose was given within 24 hours before the biopsy. Patients at both institutions are asked to discontinue antibiotic therapy for at least 24 hours before biopsy, if feasible. The definition of fever was a temperature of 38.0 C within 7 days of the biopsy. A subject had an elevated WBC count, ESR, or CRP value if the values were elevated at any time from initial presentation to the time of biopsy. We also documented the biopsy needle gauge and the amount of purulent fluid aspirated at biopsy. Analyses were conducted using SAS software (version 8.02, SAS Institute). We calculated p values from the chi-square test or Fisher s exact test for categoric variables and from the Wilcoxon s rank sum test for ordinal or continuous variables. Finally, we compared the histologic and microbiologic data from the surgical biopsy or débridement with the preceding imaging-guided biopsy results. Results Of the 75 cases of suspected osteomyelitis, there were 41 cases of osteomyelitis based on histology, and these 41 cases were retrospectively reviewed. In the remaining 34 cases that were negative for osteomyelitis based on histology, there were four cases with positive culture results. The histologic diagnoses for the 30 remaining cases with negative culture results were metastasis, n=3; lymphoma, n=2; myeloma, n=1; telangiectatic osteosarcoma, n=1; Ewing s sarcoma, n=1; nonossifying fibroma, n =1; Paget s disease, n=1; healing fracture, n=1; and no malignancy or osteomyelitis, n =19. Of the 41 histologically positive cases, 24 patients were male, and the mean age of all 75 patients was 40.8 years with a range of 3 82 years. The results are summarized in Table 1. Of the 41 histologically positive cases of osteomyelitis, 14 (34%) cases were positive at culture. Staphylococcus aureus was the most common organism cultured, detected in eight (57%) of the 14. The culture positivity rate did not vary with patient sex, by type of imaging guidance, or between the two institutions (Table 2). Moreover, no significant difference in the culture positivity rate was seen with regard to acute versus chronic osteomyelitis, antibiotic therapy before biopsy, fever, elevated WBC count, elevated ESR, elevated CRP value, or biopsy needle size (Table 2). Purulent fluid was aspirated in 10 (24%) of the 41 cases and five (50%) of 10 grew an organism at culture (p = 0.22). Of the 31 cases for which no purulent fluid was aspirated at biopsy, only nine (29%) of the 31 cultures were positive. Of the six patients from whom 2 ml of purulent fluid was aspirated, five (83%) had positive culture (p = 0.02) (Table 2). Surgical biopsy or débridement was performed in 10 (24%) of the 41 histologically positive cases of osteomyelitis: four surgeries in patients with positive cultures and six surgeries in patients with negative cultures (Table 3). In nine of the 10 surgical cases, acute or chronic osteomyelitis was the histologic diagnosis, compatible with the imaging-guided biopsy result. In the 10th case, the surgical specimen was nondiagnostic and no histologic result was given; however, the culture specimen was positive. Of the four cases with positive culture results from the imaging-guided biopsy, two grew S. aureus, which is identical to the imaging-guided biopsy culture results. No growth occurred in the other two cases; however, both of these patients received antibiotics during the time interval between the imaging-guided biopsy and the surgical procedure. Of the six culture-negative cases at imaging-guided biopsy that underwent surgery, four cases had no growth at culture similar to the imaging-guided biopsy results. The other two cases grew organisms from the surgical specimen. 1530 AJR:188, June 2007

Imaging-Guided Bone Biopsy for Osteomyelitis TABLE 1: Summary of 41 Patients Undergoing Imaging-Guided Biopsy with Histologically Proven Osteomyelitis Patient Imaging Clinical Finding Age (y) Guidance Used for Biopsy Biopsy Site Type of Osteomyelitis per Histology Results Microbiology Results ABX a Fever b WBC Count c ESR d CRP e Size of Biopsy Needle (gauge) No. Sex 1 M 10 CT Femur Acute 15 2 F 12 CT Pelvis sacrum Acute and chronic 11 3 M 47 CT Rib Acute 15 4 F 13 CT Tibia Chronic 15 5 M 3 CT Pelvis iliac Acute and chronic + + + + 14 6 F 78 CT Pelvis pubic symphysis Acute 15 7 F 82 CT Humerus Acute + + 15 8 F 72 CT Pelvis pubic symphysis Acute and chronic + 11 9 F 13 CT Clavicle Chronic + + + 15 10 F 21 CT Tibia Acute and chronic + + 15 12 11 F 6 CT Tibia Acute and chronic 15 12 F 73 FL Foot phalanx Chronic + + 14 13 M 52 FL Pelvis ischium Chronic 14 14 M 48 CT Foot calcaneus Acute + + + 18 1 15 M 45 FL Foot metatarsal Chronic NA 16 M 30 CT Pelvis iliac Acute + + + + 11 17 F 77 FL Foot phalanx Acute 14 18 M 76 FL Pelvis coccyx Chronic + + + 14 19 M 30 CT Pelvis ischium Chronic NA 20 F 41 CT Pelvis sacrum Chronic + NA 1 21 M 20 FL Femur Chronic + 14 22 M 53 CT Sternum Chronic + 14 23 M 41 FL Foot metatarsal Acute 14 24 M 21 CT Sternum Chronic + + + 15 25 F 27 FL Pelvis iliac Acute and chronic + 15 26 M 35 CT Fibula Acute + + + 14 1 27 M 66 FL Foot metatarsal Chronic + + + 14 1 28 M 32 CT Humerus Chronic Staphylococcus aureus + 15 50 29 M 70 CT Femur Acute S. aureus + + 15 30 F 23 CT Tibia Acute S. aureus 15 31 M 33 FL Pelvis ischium Chronic S. aureus + 14 32 M 30 CT Pelvis ischium Chronic S. aureus 14 33 M 23 CT Tibia Acute and chronic S. aureus + 14 5 34 M 66 CT Foot calcaneus Acute S. aureus 14 2 35 F 24 CT Pelvis ischium Acute S. aureus 15 36 M 59 CT Pelvis ischium Chronic Streptococcus viridans, 14 Bacteroides fragilis 37 F 45 CT Radius Acute Salmonella 15 4 38 M 17 FL Pelvis ischium Acute and chronic Corynebacterium + + + + 14 organisms 39 F 62 FL Foot metatarsal Chronic Corynebacterium + + 14 organisms 40 M 41 FL Foot metatarsal Chronic Corynebacterium organisms + + 15 41 F 54 CT Femur Acute Escherichia coli + 15 4 Note Plus sign (+) indicates present. FL = fluoroscopy, NA = not available. a A patient was considered to have received antibiotics (ABX) if any dose was given within 24 hours before the biopsy. b Fever = temperature 38.0 C within 7 days of the biopsy. c WBC count 10 10 3 μl. d Erythrocyte sedimentation rate (ESR) 10 mm/h. e C-reactive protein (CRP) 6 mg/l. Purulent Fluid (ml) AJR:188, June 2007 1531

TABLE 2: Culture Positivity Rates (%) for Various Factors in 41 Histologically Positive Cases of Osteomyelitiss Factor Present Absent p Histologic result of osteomyelitis 34 Patient sex Male 38 Female 29 Imaging guidance used for biopsy CT 36 Fluoroscopy 31 Institution 1 31 Institution 2 36 Type of osteomyelitis Acute 35 Chronic 33 Antibiotics administered before biopsy 24 42 0.23 Fever a 29 35 0.73 Elevated WBC count b 20 36 0.48 Elevated ESR c 38 32 0.72 Elevated CRP value d 17 37 0.33 Size of biopsy needle 14-gauge 41 15-gauge 41 Any purulent fluid aspirated 50 29 0.22 2 ml of purulent fluid aspirated 83 25 0.02 e a Fever = temperature 38.0 C within 7 days of the biopsy. b WBC count 10 10 3 μl. c Erythrocyte sedimentation rate (ESR) 10 mm/h. d C-reactive protein (CRP) 6 mg/l. e Statistically significant difference. Discussion Osteomyelitis is a challenging disease to diagnose and treat given its variable imaging and clinical characteristics. Although the definitive diagnosis of osteomyelitis is made by culturing an organism directly from the site of infection, cultures attained from biopsy and surgery can be negative in 40 60% of cases [5 8]. Only 34% of the histologically positive cases of osteomyelitis were positive at culture in this study. This result is similar to those of other studies [5 7, 11]. White et al. [6] evaluated the utility of sending both histologic and microbiologic biopsy samples and found eight positive cultures in 19 histologically or surgically proven cases of osteomyelitis, for a culture positivity rate of 42%. Similarly Schweitzer et al. [7] found an overall culture positivity rate of 50% for imaging-guided biopsies when evaluating the effect of lidocaine on culture yield in 28 cases of osteomyelitis. The culture positivity rate in the spine may be higher; thus, spinal Culture Positivity Rate (%) in Cases with Factor infection cases were excluded from our study. Chew and Kline [16] found 39 positive cultures of 43 histologically or surgically proven cases of osteomyelitis in the spine. Sampling of adjacent purulent disk material during the vertebral body biopsy was a proposed factor accounting for the high culture-positive rate. Although imaging-guided bone biopsies are relatively safe [13 15], the referring clinician and radiologist should be aware of the low culture positivity rate. Thus, in cases in which the diagnosis of osteomyelitis is obvious, performing a biopsy for the sole purpose of attaining an organism to guide antimicrobial treatment must be performed with caution. This is most important in cases in which the risk of adjacent tissue injury or seeding of noninfected tissue is high. The most common organism found at culture was S. aureus, seen in 57% of the cultures, which is consistent with the results of other studies [1, 11, 16, 21]. Dich et al. [21] evaluated 163 cases of osteomyelitis and found S. aureus to be the most common etiologic agent, seen in 61% of cases. With regard to clinical, laboratory, and biopsy-related factors associated with culture positivity or negativity, we found only one significant association: the aspiration of 2 ml of purulent fluid at biopsy. Histologic type of osteomyelitis (acute or chronic), antibiotic therapy before biopsy, fever, elevated WBC count, elevated ESR, elevated CRP value, and biopsy needle size did not have a significant association with positive or negative cultures. These results have several implications. First, the aspiration of 2 ml of purulent fluid at biopsy was the only factor associated with a higher culture positivity rate. When no purulent fluid was obtained, only 29% of cases were positive at culture. The culture positivity rate increased to 50% if any purulent fluid was aspirated and rose to 83% if 2 ml of purulent fluid was obtained. Intuitively, this finding that culture yield would increase as the amount of organisms obtained at biopsy increases seems reasonable. This finding suggests that performing biopsies in suspected cases of intraosseous or periosseous abscess is useful, especially if the abscess is large. Thus, if fluid collections can be seen on imaging, the culture yield from these biopsies may be higher and may warrant the procedure, especially in more hazardous situations. Second, radiologists commonly request that patients discontinue antimicrobial therapy for at least 24 hours before biopsy to have the highest chance of isolating an organism. Although this difference was not statistically significant, we did detect a lower culture positivity rate in the patients on antimicrobial therapy before biopsy versus the patients off therapy. Of the patients who received antibiotic therapy within 24 hours of the biopsy, 24% had a positive culture, whereas the patients who did not receive antibiotics had a 42% culture positivity rate. Larger prospective studies are needed to further investigate this finding. Given our results, we have maintained our requests to referring clinicians to discontinue antibiotics for at least 24 hours before biopsy. Third, although one may hypothesize that a larger biopsy needle size would lead to a higher likelihood of culture positivity due to a larger sample volume, our results suggest no difference in culture positivity rates between the two predominant needle sizes used, 14- and 15-gauge. Unfortunately, the two main needles used did not differ much in size. Finally, there was good correlation of both the histologic and microbiologic results from 1532 AJR:188, June 2007

Imaging-Guided Bone Biopsy for Osteomyelitis TABLE 3: Comparison of Histology and Microbiology Results from Imaging-Guided Biopsy and Surgery Patient Surgical Histology Results for Type of Osteomyelitis at Microbiology Results at No. Sex Age (y) Procedure Biopsy Site Biopsy Surgery Biopsy Surgery 1 M 35 Débridement Fibula Acute Acute and chronic 2 F 12 Débridement Pelvis sacrum Acute and chronic Acute and chronic 3 F 13 Débridement Tibia Chronic Chronic 4 F 21 Débridement Tibia Acute and chronic Acute and chronic 5 M 21 Biopsy Sternum Chronic Nondiagnostic Staphylococcus epidermidis 6 M 10 Débridement Femur Acute Acute Fusobacterium organisms 7 M 32 Débridement Humerus Chronic Acute and chronic Staphylococcus aureus 8 F 23 Débridement Tibia Acute Chronic S. aureus 9 M 23 Débridement Tibia Acute and chronic Acute and chronic S. aureus S. aureus 10 M 70 Biopsy Femur Acute Acute S. aureus S. aureus the imaging-guided biopsies and surgical procedures. Aside from one nondiagnostic surgical specimen, all surgical samples were positive for osteomyelitis and were histologically identical to the imaging-guided biopsy results. In no cases did organisms isolated from the imagingguided biopsy differ with those attained at surgery. These results suggest that imaging-guided bone biopsies are accurate procedures, which is similar to the results of past studies [2, 5, 6, 9]. Despite the strengths of our study, a few limitations deserve mention. Our sample size may not be large enough to detect a significant difference in histologic type of osteomyelitis, antibiotic therapy before biopsy, WBC count, ESR, CRP value, or size of biopsy needle in the groups who had positive and negative cultures. Although osteomyelitis is common, requests for imaging-guided biopsies for osteomyelitis are relatively uncommon. Our study consists of 800 bone biopsies over a 7- and an 8-year period at two large tertiary care centers, with only 75 biopsies performed for suspected osteomyelitis. This is only 3 7 cases per year for each institution. Often patients with clear clinical or laboratory data suggesting osteomyelitis are treated with antibiotics or are taken directly to surgery without undergoing an imagingguided biopsy. For instance, a patient with diabetes who has a draining foot ulcer with exposed bone is unlikely to need imagingguided biopsy to diagnosis osteomyelitis. The cases referred for biopsy are often in patients for whom the diagnosis is difficult. Despite the low numbers, the similar culture positivity rates between the two institutions (31% and 36%) suggest that the 34% combined culture positivity rate is accurate. In addition, the 24-hour antibiotic-free interval used in this study is somewhat arbitrary, considering some patients with chronic renal failure in whom the serum level of antibiotics may be high. Nevertheless, we consider this parameter to be acceptable because it is current practice at both institutions is to discontinue antibiotics at least 24 hours before biopsy. Another limitation of the study pertains to using positive histology as the diagnostic standard in the study. Although this may not be optimal, we are likely underdiagnosing, not overdiagnosing, cases of osteomyelitis. Sampling error will reduce the accuracy of histology by increasing the number of falsenegatives; however, the positive cases at histology are likely true cases of osteomyelitis. The histologic definition of acute osteomyelitis is fairly specific; thus, it is unlikely that there is an alternative diagnosis, such as metastasis or myeloma, if the histologic result is osteomyelitis. Our study focuses on these positive histology cases; thus, we believe using positive histology as the diagnostic standard is reasonable. In conclusion, the rate of positive culture of osteomyelitis obtained from imaging-guided core biopsy is low. We found that aspiration of 2 ml of purulent fluid at biopsy is associated with a higher culture positivity rate. Clinicians and radiologists should be aware that the likelihood of isolating an organism with imagingguided biopsies is low, but it is significantly higher in cases of intraosseous abscesses. Acknowledgment We thank A. C. Wu for statistical support and manuscript review. References 1. Dirckx JH. Stedman s concise medical dictionary for the health professions, 3rd ed. Baltimore, MD: Williams and Wilkins, 1997:635 2. Zuluaga AF, Galvis W, Jaimes F, Vesga O. Lack of microbiological concordance between bone and non-bone specimens in chronic osteomyelitis: an observational study. BMC Infect Dis 2002; 2:8 3. Sammak B, Abd El Bagi M, Al Shahed M, et al. Osteomyelitis: a review of currently used imaging techniques. Eur Radiol 1999; 9:894 900 4. Termaat MF, Raijmakers PG, Scholten HJ, Bakker FC, Patka P, Haarman HJ. The accuracy of diagnostic imaging for the assessment of chronic osteomyelitis: a systematic review and meta-analysis. J Bone Joint Surg Am 2005; 87:2464 2471 5. Howard CB, Einhorn M, Dagan R, Yagupski P, Porat S. Fine-needle bone biopsy to diagnose osteomyelitis. J Bone Joint Surg Br 1994; 76:311 314 6. White LM, Schweitzer ME, Deely DM, Gannon F. Study of osteomyelitis: utility of combined histologic and microbiologic evaluation of percutaneous biopsy samples. Radiology 1995; 197:840 842 7. Schweitzer ME, Deely DM, Beavis K, Gannon F. Does the use of lidocaine affect the culture of percutaneous bone biopsy specimens obtained to diagnose osteomyelitis? An in vitro and in vivo study. AJR 1995; 164:1201 1203 8. Han H, Lewis VL Jr, Wiedrich TA, Patel PK. The value of Jamshidi core needle bone biopsy in predicting postoperative osteomyelitis in grade IV pressure ulcer patients. Plast Reconstr Surg 2002; 110:118 122 9. Khatri G, Wagner DK, Sohnle PG. Effect of bone biopsy in guiding antimicrobial therapy for osteomyelitis complicating open wounds. Am J Med Sci 2001; 321:367 371 10. Zuluaga AF, Galvis W, Saldarriaga JG, Agudelo M, Salazar BE, Vesga O. Etiologic diagnosis of chronic AJR:188, June 2007 1533

osteomyelitis: a prospective study. Arch Intern Med 2006; 166:95 100 11. Jeffcoate WJ, Lipsky BA. Controversies in diagnosing and managing osteomyelitis of the foot in diabetes. Clin Infect Dis 2004; 39[suppl 2]:S115 S122 12. Senneville E, Melliez H, Beltrand E, et al. Culture of percutaneous bone biopsy specimens for diagnosis of diabetic foot osteomyelitis: concordance with ulcer swab cultures. Clin Infect Dis 2006; 42:57 62 13. Appelbaum AH, Kamba TT, Cohen AS, Qaisi WG, Amirkhan RH. Effectiveness and safety of imagedirected biopsies: coaxial technique versus conventional fine-needle aspiration. South Med J 2002; 95:212 217 14. Dupuy DE, Rosenberg AE, Punyaratabandhu T, Tan MH, Mankin HJ. Accuracy of CT-guided needle biopsy of musculoskeletal neoplasms. AJR 1998; 171:759 762 15. Hau A, Kim I, Kattapuram S, et al. Accuracy of CTguided biopsies in 359 patients with musculoskeletal lesions. Skeletal Radiol 2002; 31:349 353 16. Chew FS, Kline MJ. Diagnostic yield of CT-guided percutaneous aspiration procedures in suspected spontaneous infectious diskitis. Radiology 2001; 218:211 214 17. Rankine JJ, Barron DA, Robinson P, Millner PA, Dickson RA. Therapeutic impact of percutaneous spinal biopsy in spinal infection. Postgrad Med J 2004; 80:607 609 18. Jelinek JS, Murphey MD, Welker JA, et al. Diagnosis of primary bone tumors with image-guided percutaneous biopsy: experience with 110 tumors. Radiology 2002; 223:731 737 19. Issakov J, Flusser G, Kollender Y, Merimsky O, Lifschitz-Mercer B, Meller I. Computed tomography guided core needle biopsy for bone and soft tissue tumors. Isr Med Assoc J 2003; 5:28 30 20. Ng CS, Salisbury JR, Darby AJ, Gishen P. Radiologically guided bone biopsy: results of 502 biopsies. Cardiovasc Intervent Radiol 1998; 21:122 128 21. Dich VQ, Nelson JD, Haltalin KC. Osteomyelitis in infants and children: a review of 163 cases. Am J Dis Child 1975; 129:1273 1278 1534 AJR:188, June 2007