- Email: [email protected]
Tc-99m-MDP bone SPECT scintigraphy and computed tomography scan
Evaluation of temporal and facial osteomyelitis by simultaneous In-WBC/Tc-99m-MDP bone SPECT scintigraphy and computed tomography scan PETER C. WEBER, JAMES E. SEABOLD, SCOTT M. GRAHAM, HENRY H. HOFFMANN, TERESAM. SIMONSON, and
BRADH. THOMPSON,Charleston, South Carolina, and Iowa City, Iowa A reliable imaging technique is needed for follow-up of patients with temporal and facial osteomyelitis. Clinical outcome in 20 patients with suspected osteomyelitis of the temporal/mastoid, calvarium, and mandible facial bones was evaluated with 30 combined In-WBC/Tc-99m MDP bone single photon emission computed tomographic [SPECT]scans and 27 computed tomographic scans. Simultaneous dual-tracer 25-minute SPECT scans were acquired 18 to 20 hours after radiotracer injection by use of a three-detector system. Diagnosis of the 20 patients [age range, 3 to 74 years] included 8 with facial osteomyelitis, 6 with malignant otitis externa, 3 with mandibular osteomyelitis, and 3 with calvarial osteomyelitis. Diagnosis was confirmed by biopsy/culture results in 18 patients and by endoscopic and clinical evaluation in 2 patients with initial negative scans. Of the 30 In-WBC/MDP scans, 15 were true-positive, 13 true-negative, I false-negative, and I equivocal. Of a total of 27 CT scans, 9 were true-positive, 5 false-negative, and I equivocal in patients with biopsy-proven osteomyelitis. Three computed tomographic scans were false-positive and I was equivocal in patients without osteomyelitis, because of concurrent postoperative bone abnormalities. Additionally, 8 computed tomographic scans were true-negative. These results suggest that dual In-WBC/Tc-99m MDP bone SPECT scintigraphy provides an accurate imaging modality for diagnosis and follow-up of temporal and facial osteomyelitis when existing clinical or postoperative bone changes make it difficult to detect active osteomyelitis by computed tomographic scan. [OTOLARYNGOLHEADNECKSURG1995;113:36-41 .)
lip
| e m p o r a l / m a s t o i d , facial bone, calvarium, and mandibular osteomyelitis frequently result from complications of otitis externa, otitis media, paranasal sinus infection, periodontal disease, and trauma, as well as after various surgical procedures. Temporal/mastoid 0steomyelitis is usually seen in elderly patients in association with diabetes immunosuppression. 1-5 Facial, calvarial, and mandibular osteomyelitis occur most frequently in previously oper-
From the Department of Otolaryngology,Head and Neck Surgery (Dr. Weber), Medical University of South Carolina; and the Departments of Radiology (Drs. Seabold, Simonson, and Thompson) and Otolaryngology,Head and Neck Surgery(Drs. Graham and Hoffmann), University of Iowa Hospital and Clinics. Received for publication Oct. 4, 1994; accepted Jan. 20, 1995. Reprint requests: Peter C. Weber, MD, Department of Otolaryngology,Head and Neck Surgery, Medical Universityof South Carolina, 171 Ashley Ave., Charleston, SC 29425. Copyright © 1995by the American Academyof OtolaryngologyHead and Neck SurgeryFoundation, Inc. 0194-5998/95/$3.00 + 0 23/1/63541 36
ated on, irradiated, or traumatized patients. First, soft-tissue infections develop near these areas and then progress to osteomyelitis. The etiologic organism is almost always Pseudomonas aeruginosa? ,6,7 Osteomyelitis was associated with a high mortality rate before the use of aminoglycosides and now semisynthetic penicillins, which are effective against P. aeruginosa. The resolution of infection has always been difficult to assess in these patients, particularly in the temporal/mastoid region. Several radiologic imaging studies have been used in an attempt to determine successful therapy, but they often cannot discern between active osteomyelitis, postoperative bony changes, and/or the reparative bony process after successful therapy. Plain-film radiographs demonstrate density changes in both soft tissue and bone. Thus they may be useful in the initial evaluation of patients suspected of having osteomyelitis. Plain films, however, are seldom reliable after treatment of osteomyelitis because there are always bony abnormalities that
Otolaryngology Head and Neck Surgery Volume 113 Number I
cannot be distinguished from findings seen with active osteomyelitis. Computed tomography (CT) is often used for evaluation of patients with suspected cranial osteomyelitis because it affords evaluation of both soft tissue and bone. CT detects earlier changes in bone density, and it has replaced plain film, which takes 7 to 10 days to detect bony changes. 8-14Although CT shows resolution of soft-tissue changes relatively early, it takes much longer for the bone to return to normal after successful treatment of osteomyelitis. ~5'16The remineralization process occurs after successful treatment of osteomyelitis. ~5'16 Technetium-99m methyline diphosphonate (Tc99m-MDP) bone scintigraphy is useful in the initial evaluation of a patient with suspected osteomyelitis because it demonstrates alterations in osteoblastic activity. Increased radiotracer uptake, although nonspecific, is commonly seen in osteomyelitis, a7-23 Tomographic bone scintigraphy has become fairly standard to detect occult sites of osteomyelitis when CT is negative or equivocal. ~7-~°This bone scan will remain positive for a long time after successful treatment of osteomyelitis because of the ongoing process of bone repair? 7 Gallium-67 citrate (Ga-67) scintigraphy has been used to follow the resolution of osteomyelitic disease and to assess when antibiotic therapy can be discontinued. 3'm7 However, it takes 4 to 6 months after successful treatment for a Ga-67 scan to return to normal. 21'22'24 Indium-ill-labeled leukocyte (In-WBC) planar scintigraphy has been reported to give better results for the detection of osteomyelitis than either planar or tomographic Ga-67 and/or Tc-99m-MDP scintigraphy. 24Recently, advances in instrumentation have allowed simultaneous acquisition of single photon emission tomographs (SPECT) using two radiotracers. This technology provides two complementary functional images with exact coregistration. Advances in multidetector SPECT systems have greatly improved the images obtained with lower photon flux radiotracers, such as In-WBCs. The use of three-dimensional SPECT imaging is particularly useful in patients with suspected osteomyelitis of the skull, to determine the site and extent of bony involvement and to distinguish bone from adjacent or overlying soft-tissue inflammation/infection. This article compares the results of simultaneous acquired In-WBC and Tc-99m-MDP bone SPECT scintigraphy with those obtained from CT scans. MEHTODS AND MATERIAL
A blinded retrospective analysis was done with combined Tc-99m-MDP/In-WBC SPECT scintigra-
WEBER et al.
37
phy and CT scans in 20 patients with clinically suspicious postoperative, initial, persistent, or recurrent osteomyelitis. All scans were evaluated by two separate teams of physicians without knowledge of the other diagnostic studies: one team for the nuclear scans and one team for the CT scans. Six patients were suspected of having necrotizing otitis externa, eight facial osteomyelitis associated with sinusitis, three mandibular osteomyelitis, and three calvarial osteomyelitis. Ten were male, and 10 were female, with an age range of 3 to 74 years (average, 46.2 years). Imaging
All CT scans were obtained with either a Siemens (Siemens Medical Systems, Inc., St. Louis, Mo.) or Picker (Picker International, Inc., Wood Dale, Ill.) scanner. These scans included both soft-tissue and bone density windows and were done with and without intravenous contrast. All CT images were interpreted by two radiologists without the knowledge of patients or other diagnostic tests that were performed. In-WBC scans were obtained with autologous leukocytes labeled with In-lll-oxine by use of a modification of the method reported by Thacker et al. 25 The SPECT images were obtained 18 to 24 hours after the intravenous injection of 0.5 mCi (18.5 MBq) of In-WBCs. A three-detector SPECT scanner system equipped with medium energy collimators was used for image acquisition. The images were acquired in a dual-isotope mode with one set of image data originating from a 15% energy window centered on the 140 keV Tc-99m photo peak, and the other set of image data originating from a 15% energy window centered on the 173- and 247-keV In-111 photo peaks. Projection images were acquired for 80 seconds at 5-degree intervals over a complete 360degree rotation. At the completion of the study (and before reconstruction), the projection images were processed with a two-dimensional Wiener (Siemens Nuclear Division, Hoffman Estates, Ill.) smoothing filter, which was optimized for each data set. The processed data were then reconstructed with a ramp reconstruction filter to generate transverse, sagittal, and coronal sections for both the Tc-99mMDP and In-WBC data. The SPECT images were then reinterpreted by two nuclear medicine physicians without knowledge of other diagnostic results. In-WBC-Tc-99m-MDP images were considered positive for osteomyelitis if focal WBC localization greater than adjacent or contralateral background
38
Otolaryngology Head a n d Neck Surgery July 1995
WEBER et al.
Table I. In-WBC/Tc-99m-MDP SPECTand CT
scan results for detection of temporal/mastoid osteomyelitis Patient no.
CT TP TN EQ FP TN TP TP TP FN FN TP FP
In-WBC/I"c-MDP TP TN TP TN TN TP TP TN TP TP FN TP TP TN TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal,
activity was identified that corresponded to abnormal increased bone tracer localization on two or more corresponding tomographic slices, in three planes (coronal, transaxial, and sagittal projections). Data Interpretation
The data that were generated from CT findings, SPECT scan findings, clinical biopsies, and clinical diagnosis were correlated and assessed for positive and negative predictive values of the scans. For calculation purposes, an equivocal scan was not used to assess the predictive value rates. Because of the small patient population, additional statistical analysis was not performed. RESULTS
Among the 20 patients with suspected osteomyelitis, the most common sites were the facial bones (8 patients), the temporal/mastoid bone (6 patients), the calvarium (3 patients), and mandible (3 patients). All patients had undergone CT scan and In-WBC/Tc-99m-MDP scans. Of the 20 patients, 10 were female and 10 were male. The age range was 3 to 74 years, with an average age o f 46.2 years. Necrotizing Otitis Externa
Of the six diabetic patients with suspected necrotizing otitis externa, two were female and four were male. Two of the patients studied with unresolving otitis externa had true-negative In-WBC/Tc-99mMDP scans, which corresponded to two true-negative CT scans (Table 1). Two patients had true-
Table 2. In-WBCfrc-MDP SPECTand CT scan results for detection of osteomyelitis in facial bones/sinuses Patient no.
CT
1 2 3 4 5 6 7 8
TN TN TN TN TN TP FN FP
In-WBC/Tc-MDP TN TN TN TN TN TP TP TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, falsenegative.
positive In-WBC/Tc-99m-MDP and CT scans. One was treated with intravenous antibiotics for 3 months, during which time he had an additional true-positive In-WBC/Tc-99m-MDP scan and then had a negative scan 3 months after treatment. CT scan was not obtained after successful therapy. The second patient is still receiving intravenous antibiotic therapy and has not had a repeat In-WBC/Tc99m-MDP scan. The remaining two patients both had positive In-WBC/Tc-99m-MDP and CT scans initially. After successful therapy, both of these patients had true-negative scans, but their CT scans were interpreted as positive (false-positive). The CT scans could not delineate between postoperative both repair and active osteomyelitis. One patient had false-negative CT and In-WBC/Tc-MDP scans after 2 months of therapy and resolution of osteomyelitis. Facial Bones/sinuses Of the eight patients evaluated, five had truenegative CT and In-WBC/Tc-99m-MDP scans (Table 2). One patient had true-positive CT and InWBC/Tc-99m-MDP scans and is currently receiving intravenous antibiotic therapy. Another patient (case 6), had a false-negative CT scan and a truepositive In-WBC/Tc-99m-MDP scan, which was confirmed at the time of revision maxillectomy when bone cultures were positive for P. aeruginosa. The patient had undergone a decompression of her trigeminal nerve 2 years previously before recurrent left ear and maxillary pain developed. Her CT scan demonstrated postoperative bony changes, but no evidence of osteomyelitis, whereas the In-WBC/Tc99m-MDP scan had demonstrated marked In-WBC localization corresponding to the left maxillary bone and maxillary sinus soft tissues. The last patient had left nasal congestion, left orbital pain, and a history
Otolaryngology Head a n d Neck Surgery Volume 113 Number I
WEBER et al.
Table 3. In-WBC/Tc-99m-MDP SPECT and CT scan results for detection of osteomyelitis in the calvarium Patient no.
CT
3 2
TP TP FN EQ
1
In-WBC/Tc-MDP TP TP TP* TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal. *Patient died.
of several prior sinus infections and surgery. His CT scan demonstrated left maxillary soft-tissue inflammation and left orbital bone erosion in the superior medial wall, consistent with osteomyelitis. The InWBC/Tc-99m-MDP scan was negative, and an endoscopic examination did not reveal any evidence of osteomyelitis or inflammation. His pain subsided within 1 month without treatment. He has been followed up for 1 year and has not had further symptoms. Calvarium
In this group, two were found to have true-positive CT scans and true-positive In-WBC/Tc-99m-MDP scans (Table 3). One patient was a 3-year-old child with purulent drainage from her craniotomy site 1 year after excision of hypothalamic astrocytoma followed by radiation therapy. The CT scan demonstrated erosion of the bone-flap margins compared with prior CT scan 8 months earlier. Her InWBC/Tc-99m-MDP scan showed evidence of osteomyelitis (true-positive) of the bone flap. She began receiving intravenous antibiotics, and cultures of the surgically excised bone flap revealed Pseudomonas. There was no evidence of osteomyelitis during a 7-month follow-up. The second patient was a 55year-old woman who had undergone excision of an olfactory groove meningioma. Initial CT and InWBC/Tc-99m-MDP scans were positive for osteomyelitis. An In-WBC/Tc-99m-MDP scan, after 2 months of intravenous antibiotics, was still positive for osteomyelitis. CT showed healing bone but no evidence of osteomyelitis. Surgical debridement was performed, and histopathologic examination was consistent with osteomyelitis. Before further follow-up could be ascertained, the patient died of a myocardial infarction. The third patient had undergone orbital exoneration and external beam radiation for a rhabdomyosarcoma. He presented with a draining orbital sinus. CT could not differentiate
39
Table 4. In-WBC/Tc-99m-MDP SPECT and CT scan results for detection of osteomyelitis in the mandible Patient no.
CT
2 1 3
TP FN TN
In-WBC/Tc-MDP TP TP EQ
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal,
between postoperative changes and osteomyelitis, whereas In-WBC/Tc-99m-MDP scan was negative. Cultures and histologic findings from debridement were both negative. Mandible
Of the three patients in this group, one patient had true-positive CT and In-WBC/Tc-99m-MDP scans; this finding was confirmed by histology after debridement. This patient was lost to follow-up (Table 4). The second patient had a false-negative CT scan and true-positive In-WBC/Tc-99m-MDP scan. This was a 33-year-old man with pain and swelling 1 year after tooth extraction and drainage of a right temporal space abscess. The CT scan demonstrated soft-tissue edema, but no bony involvement. The In-WBC/Tc-99m-MDP scan showed positive uptake at the right mandibular ramus and adjacent soft tissue. The biopsy specimens were positive for anaerobes and actinomycosis, and the bone was also positive for infection on histopathologic studies. This patient has also been lost to follow-up. The third patient had a true-negative CT scan, whereas the In-WBC/Tc-99m-MDP was indeterminate. The patient was a 53-year-old man who had undergone irradiation therapy for squamous cell carcinoma at the tongue base 5 months previously. He had a buccal ulcer that was associated with mandibular exposure. He underwent excision of this area and debridement of the bone; cultures and histopathology were not obtained. Statistics
The predictive value rates for true-positive and
true-negative In-WBC/Tc-99m-MDP scans appear to be better than those of the CT scan (Table 5). Sensitivity, specificity, and confidence levels were not calculated in this study because the number of patients was too small to obtain significant results. The malignant otitis externa group demonstrated true-
40
Otolaryngology Head and Neck Surgery July 1995
WEBERet al.
Table 5. Predictive value rates in In-WBC/Tc-99m-MDP scans vs. CT scans Site
Positive PV rate of In-WBC/Tc-MDP [%]
Negative PV rate of In-WBC/Tc-MDP [%]
100 100 100 1O0 1O0
86 1O0 1O0
i
Malignant otitis extema Facial bones/sinuses Mandible Calvarium Overall
i
93
Positive PV rate of CT [%)
Negative PV rate of CT [%)
i
71 50
50 83
1O0 75
50 61
PV, Predictive value.
positive and true-negative predictive values of 100% and 86%, respectively, for In-WBC/Tc-99m-MDP scan and 71% and 50% for the CT scan. The facial bone/sinus group demonstrated a positive and negative predictive value rate of 100% for In-WBC/Tc99m-MDP scan and 50% and 83% for the CT scan. The calvarial group also had a positive/negative predictive value of 100% and 0% for the CT scan and a positive and negative predictive value of 100% for the In-WBC/Tc-99m-MDP scan, whereas the mandible demonstrated a positive and negative predictive value rate of 100% and 50%, respectively, for the CT scan, and the In-WBC/Tc-99m-MDP scan had a positive predictive value of 100%. The overall positive and negative predictive value rates were 75% and 61%, respectively, for CT scan and 100% and 93% for In-WBC/Tc-99m-MDP. DISCUSSION
The results of In-WBC/Tc-99m-MDP scintigraphy were better than those of CT in our 20 patients, as indicated by the predictive value rates. InWBC/Tc-99m-MDP scan reverts back to normal sooner than CT scan, as seen in three patients. It appears In-WBC/Tc-99m-MDP is better able to evaluate the status of postoperative patients. Our series showed that in the seven patients misdiagnosed by CT, postoperative bone changes were always the factor for a false-negative/positive interpretation. Bone Tc-99m-MDP scans show increased radiotracer localization at sites of active osteoblastic activity, which can also be caused by infection, inflammation, postoperative bone repair, trauma, or neoplasms. Although a negative bone scan essentially excludes osteomyelitis, the bone scan remains positive for as long as this is active bone reparation, which lags well behind the resolution of infection. The Ga-67 scan has been used to evaluate the resolution of established osteomyelitis. Ga-67 scans, however, often take up to 6 months to return to normal after successful treatment of osteomyelitis,z4
Ga-67 scans were not obtained in our patients, so comparisons could not be made. CT imaging detects changes in bone density, which in turn can be used as an indicator of bone infection. However, noninfected postoperative , postirradiation, or posttraumatic bone density alterations usually cannot be differentiated from bone density changes due to osteomyelitis. In addition, CT findings are very slow to return to normal after successful treatment of active osteomyelitis. Most of the discrepancies between the CT and the InWBC/Tc-99m-MDP interpretations can be attributed to these factors. It would appear that the In-WBC/Tc-99m-MDP scans are superior to CT scans for diagnosis and follow-up of osteomyelitis in most regions of the skull (Table 5). Epstein et al. 24recently reported planar In-WBC scintigraphy as a useful technique for diagnosis and monitoring of osteomyelitis. However, false-negative and false-positive In-WBC scans have been reported in low-grade or chronic infections, at sites with extensive bone marrow (such as the vertebral column), or in immunosuppressed patients, inpatients after long-term intravenous antibiotic therapy, and inpatients with diabetic microangiopathy.22-29By combining SPECT Tc-99m-MDP bone with InWBC SPECT scintigraphy, many of the problems inherent with planar In-WBC scintigraphy can be avoided. All but two In-WBC/Tc-99m-MDP scans correlated with the clinical picture. One was an equivocal, seen in a patient 5 months after external beam radiation thearpy, and who now has a buccal ulcer with exposed mandible. The usual criteria for a positive scan could not be used in this patient because of the decreased true localization of radiotracer at the site of irradiated bone. The falsenegative study was in a patient who had been receiving intravenous antibiotics for 3 months for osteomyelitis of the right temporal bone. Persistent low-grade osteomyelitis was not detected, but a follow-up In-WBC/Tc-99m-MDP scan was marked positive on the opposite temporal bone 3 weeks after
Otolaryngology Head a n d Neck Surgery Volume 113 Number I
intravenous antibiotics were discontinued. This false-negative scan probably occurred because of long-term persistent antibiotics combined with diabetic microangiopathy. This case was recently reported by Redleaf et al. 21 from our institution. It should be emphasized that the clinical manifestations of osteomyelitis, as well as their SPECT scans and CT scans, should all be taken into consideration when deciding on the discontinuation of therapy. CONCLUSION
1. In-WB/Tc-99m-MDP bone SPECT scintigraphy is superior to CT scan in the evaluation of cranial osteomyelitis. 2. In-WBC/Tc-99m-MDP bone SPECT imaging reverts back to normal after successful treatment of osteomyelitis much sooner than CT or Ga-67 scans. 3. In-WBC/Tc-99m-MDP bone SPECT imaging is most useful in the evaluation of postoperative patients and for follow-up after treatment of osteomyelitis of the temporal and facial bones. REFERENCES
1. Bal AJM, Tiwari RM, de Rijcke TBM. Osteomyelitis in the head and neck. J Laryngol Otol 1985;99:1059-65. 2. Noyek AM, Kitsch JC, Greyson ND, Wortzman G, Jazrawy H, Freeman JL. The clinical significance of radionuclide bone and gallium scanning in osteomyelitis of the head and neck. Laryngoscope 1984;94:1-21. 3. Chandler JR, Grobman L, Quencer R, Serafini A. Osteomyelitis of the base of the skull. Laryngoscope 1986;96:245-51. 4. Lloyd GAS, Lund VJ, Scadding GK. CT of the paranasal sinuses and functional endoscopic surgery: a critical analysis of 100 symptomatic patients. J Laryngol Otol 1991;105:181-5. 5. Godofsky EW, Zinreich J, Armstrong M, Leslie JM, Weikel CS. Sinusitis in HIV-infected patients: a clinical and radiographic review. Am J Med 1992;93:163-70. 6. Malone DG, O'Boynick PL, Ziegler DK, Batnitzky S, Hubble JP, Holladay FP. Osteomyelitis of the skull base. Neurosurgery 1992;30:426-31. 7. Chandler JR. Malignant external otitis. Laryngoscope 1968; 78:1257-94. 8. Gherini SJ, Brackman DE, Bradley WB. Magnetic resonance imaging and computerized tomography in malignant external otitis. Laryngoscope 1992;96:542-8. 9. Grobman LR, Ganz W, Casiano R, Goldberg S. Atypical osteomyelitis of the skull base. Laryngoscope 1989;99:671-6. 10. Curtin HD, Wolfe P, May M. Malignant external otitis: CT evaluation. Radiology 1982;145:383-8.
WEBER et al.
4t
11. Mendelson DS, Som PM, Mendelson MH, Parisier SC. Malignant external otitis: the role of computed tomography and radionuclides in evaluation. Radiology 1983;149:745-9. 12. Gold S, Som PM, Lucente FE, Lawson W, Mendelson M, Parisier SC. Radiographic findings in progressive necrotizing "malignant" external otitis. Laryngoscope 1984;94:363-6. 13. Lloyd GAS. CT of the paranasal sinuses: study of a control series in relation to endoscopic sinus surgery. J Laryngol Otol 1990;104:477-81. 14. Rubin J, Curtin HD, Yu VL, Kamerer DB. Malignant external otitis: utility of CT in diagnosis and follow-up. Radiology 1990;174:391-4. 15. Benecke JE. Management of osteomyelitis of the skull base. Laryngoscope 1989;99:1220-3. 16. Rubin J, Yu VL. Malignant external otitis: insights into pathogenesis, clinical manifestations, diagnosis and therapy. Am J Med 1988;85:391-8. 17. Parisier SC, Lucente FE, Som PM, Hirschman SZ, Arnold LM, Roffman JD. Nuclear scanning in necrotizing progressive "malignant" external otitis. Laryngoscope 1982;92: 1016-9. 18. Strashun AM, Nejatheim M, Goldsmith SJ. Malignant external otitis: early scintigraphic detection. Radiology 1984; 150:541-5. 19. Strauss M, Kaufman RA, Baum S. Osteomyelitis of the head and neck: sequential radionuclide scanning in diagnosis and therapy. Laryngoscope 1985;95:81-4. 20. Uri N, Gips S, Front A, Meyer SW, Hardoff R. Quantitative bone and 67Ga scintigraphy in the differentiation of necrotizing external otitis from severe external otitis. Arch Otolaryngol Head Neck Surg 1991;117:623-6. 21. Redleaf MI, Angelis I, McCabe GF. Indium ill-labeled white blood cell scintigraphy as an unreliable indicator of malignant otitis resolution. Ann Otol Rhinol Laryngol 1994; 103:444-8. 22. A1-Sheikh W, Sfakianakis N, Mnaynmeh W. Subacute and chronic bone infection. Radiology 1985;155:501-6. 23. Larcos G, Brown ML, Suaon RT. Diagnosis of osteomyelitis of the foot in diabetic patients. Am J of Radiol 1991;157: 527-31. 24. Epstein JS, Ganz WI, Lizak M, Grobman L, Goodwin WJ, Dewanjee MK. Indium-Ill-labeled leukocyte scintigraphy in evaluating head and neck infections. Ann Otol Rhinol Laryngol 1992;101:961-8. 25. Thakur ML, Coleman RE, Mayhill CG. Preparation and evaluation of In-III-labeled leukocytes as in abscess imaging agent doses. Radiology 1976;119:731-2. 26. Scharwecker DS. Osteomyelitis: diagnosis with indium-Illlabeled leukocytes. Radiology 1986;96:542-8. 27. McAfee JG, Samin A. In-111 labeled leukocytes. Radiology 1985; 155:221-9. 28. Kipper MS, Williams RJ. Indium-ill white blood cell imaging. Clin Nucl Med 1983;8:449-55. 29. Date FL, Thorne DA. Effect of chronicity of infection on the sensitivity of In-ill-labeled leukocyte scan. Am J Radiol 1986;142:809-12.
BRADH. THOMPSON,Charleston, South Carolina, and Iowa City, Iowa A reliable imaging technique is needed for follow-up of patients with temporal and facial osteomyelitis. Clinical outcome in 20 patients with suspected osteomyelitis of the temporal/mastoid, calvarium, and mandible facial bones was evaluated with 30 combined In-WBC/Tc-99m MDP bone single photon emission computed tomographic [SPECT]scans and 27 computed tomographic scans. Simultaneous dual-tracer 25-minute SPECT scans were acquired 18 to 20 hours after radiotracer injection by use of a three-detector system. Diagnosis of the 20 patients [age range, 3 to 74 years] included 8 with facial osteomyelitis, 6 with malignant otitis externa, 3 with mandibular osteomyelitis, and 3 with calvarial osteomyelitis. Diagnosis was confirmed by biopsy/culture results in 18 patients and by endoscopic and clinical evaluation in 2 patients with initial negative scans. Of the 30 In-WBC/MDP scans, 15 were true-positive, 13 true-negative, I false-negative, and I equivocal. Of a total of 27 CT scans, 9 were true-positive, 5 false-negative, and I equivocal in patients with biopsy-proven osteomyelitis. Three computed tomographic scans were false-positive and I was equivocal in patients without osteomyelitis, because of concurrent postoperative bone abnormalities. Additionally, 8 computed tomographic scans were true-negative. These results suggest that dual In-WBC/Tc-99m MDP bone SPECT scintigraphy provides an accurate imaging modality for diagnosis and follow-up of temporal and facial osteomyelitis when existing clinical or postoperative bone changes make it difficult to detect active osteomyelitis by computed tomographic scan. [OTOLARYNGOLHEADNECKSURG1995;113:36-41 .)
lip
| e m p o r a l / m a s t o i d , facial bone, calvarium, and mandibular osteomyelitis frequently result from complications of otitis externa, otitis media, paranasal sinus infection, periodontal disease, and trauma, as well as after various surgical procedures. Temporal/mastoid 0steomyelitis is usually seen in elderly patients in association with diabetes immunosuppression. 1-5 Facial, calvarial, and mandibular osteomyelitis occur most frequently in previously oper-
From the Department of Otolaryngology,Head and Neck Surgery (Dr. Weber), Medical University of South Carolina; and the Departments of Radiology (Drs. Seabold, Simonson, and Thompson) and Otolaryngology,Head and Neck Surgery(Drs. Graham and Hoffmann), University of Iowa Hospital and Clinics. Received for publication Oct. 4, 1994; accepted Jan. 20, 1995. Reprint requests: Peter C. Weber, MD, Department of Otolaryngology,Head and Neck Surgery, Medical Universityof South Carolina, 171 Ashley Ave., Charleston, SC 29425. Copyright © 1995by the American Academyof OtolaryngologyHead and Neck SurgeryFoundation, Inc. 0194-5998/95/$3.00 + 0 23/1/63541 36
ated on, irradiated, or traumatized patients. First, soft-tissue infections develop near these areas and then progress to osteomyelitis. The etiologic organism is almost always Pseudomonas aeruginosa? ,6,7 Osteomyelitis was associated with a high mortality rate before the use of aminoglycosides and now semisynthetic penicillins, which are effective against P. aeruginosa. The resolution of infection has always been difficult to assess in these patients, particularly in the temporal/mastoid region. Several radiologic imaging studies have been used in an attempt to determine successful therapy, but they often cannot discern between active osteomyelitis, postoperative bony changes, and/or the reparative bony process after successful therapy. Plain-film radiographs demonstrate density changes in both soft tissue and bone. Thus they may be useful in the initial evaluation of patients suspected of having osteomyelitis. Plain films, however, are seldom reliable after treatment of osteomyelitis because there are always bony abnormalities that
Otolaryngology Head and Neck Surgery Volume 113 Number I
cannot be distinguished from findings seen with active osteomyelitis. Computed tomography (CT) is often used for evaluation of patients with suspected cranial osteomyelitis because it affords evaluation of both soft tissue and bone. CT detects earlier changes in bone density, and it has replaced plain film, which takes 7 to 10 days to detect bony changes. 8-14Although CT shows resolution of soft-tissue changes relatively early, it takes much longer for the bone to return to normal after successful treatment of osteomyelitis. ~5'16The remineralization process occurs after successful treatment of osteomyelitis. ~5'16 Technetium-99m methyline diphosphonate (Tc99m-MDP) bone scintigraphy is useful in the initial evaluation of a patient with suspected osteomyelitis because it demonstrates alterations in osteoblastic activity. Increased radiotracer uptake, although nonspecific, is commonly seen in osteomyelitis, a7-23 Tomographic bone scintigraphy has become fairly standard to detect occult sites of osteomyelitis when CT is negative or equivocal. ~7-~°This bone scan will remain positive for a long time after successful treatment of osteomyelitis because of the ongoing process of bone repair? 7 Gallium-67 citrate (Ga-67) scintigraphy has been used to follow the resolution of osteomyelitic disease and to assess when antibiotic therapy can be discontinued. 3'm7 However, it takes 4 to 6 months after successful treatment for a Ga-67 scan to return to normal. 21'22'24 Indium-ill-labeled leukocyte (In-WBC) planar scintigraphy has been reported to give better results for the detection of osteomyelitis than either planar or tomographic Ga-67 and/or Tc-99m-MDP scintigraphy. 24Recently, advances in instrumentation have allowed simultaneous acquisition of single photon emission tomographs (SPECT) using two radiotracers. This technology provides two complementary functional images with exact coregistration. Advances in multidetector SPECT systems have greatly improved the images obtained with lower photon flux radiotracers, such as In-WBCs. The use of three-dimensional SPECT imaging is particularly useful in patients with suspected osteomyelitis of the skull, to determine the site and extent of bony involvement and to distinguish bone from adjacent or overlying soft-tissue inflammation/infection. This article compares the results of simultaneous acquired In-WBC and Tc-99m-MDP bone SPECT scintigraphy with those obtained from CT scans. MEHTODS AND MATERIAL
A blinded retrospective analysis was done with combined Tc-99m-MDP/In-WBC SPECT scintigra-
WEBER et al.
37
phy and CT scans in 20 patients with clinically suspicious postoperative, initial, persistent, or recurrent osteomyelitis. All scans were evaluated by two separate teams of physicians without knowledge of the other diagnostic studies: one team for the nuclear scans and one team for the CT scans. Six patients were suspected of having necrotizing otitis externa, eight facial osteomyelitis associated with sinusitis, three mandibular osteomyelitis, and three calvarial osteomyelitis. Ten were male, and 10 were female, with an age range of 3 to 74 years (average, 46.2 years). Imaging
All CT scans were obtained with either a Siemens (Siemens Medical Systems, Inc., St. Louis, Mo.) or Picker (Picker International, Inc., Wood Dale, Ill.) scanner. These scans included both soft-tissue and bone density windows and were done with and without intravenous contrast. All CT images were interpreted by two radiologists without the knowledge of patients or other diagnostic tests that were performed. In-WBC scans were obtained with autologous leukocytes labeled with In-lll-oxine by use of a modification of the method reported by Thacker et al. 25 The SPECT images were obtained 18 to 24 hours after the intravenous injection of 0.5 mCi (18.5 MBq) of In-WBCs. A three-detector SPECT scanner system equipped with medium energy collimators was used for image acquisition. The images were acquired in a dual-isotope mode with one set of image data originating from a 15% energy window centered on the 140 keV Tc-99m photo peak, and the other set of image data originating from a 15% energy window centered on the 173- and 247-keV In-111 photo peaks. Projection images were acquired for 80 seconds at 5-degree intervals over a complete 360degree rotation. At the completion of the study (and before reconstruction), the projection images were processed with a two-dimensional Wiener (Siemens Nuclear Division, Hoffman Estates, Ill.) smoothing filter, which was optimized for each data set. The processed data were then reconstructed with a ramp reconstruction filter to generate transverse, sagittal, and coronal sections for both the Tc-99mMDP and In-WBC data. The SPECT images were then reinterpreted by two nuclear medicine physicians without knowledge of other diagnostic results. In-WBC-Tc-99m-MDP images were considered positive for osteomyelitis if focal WBC localization greater than adjacent or contralateral background
38
Otolaryngology Head a n d Neck Surgery July 1995
WEBER et al.
Table I. In-WBC/Tc-99m-MDP SPECTand CT
scan results for detection of temporal/mastoid osteomyelitis Patient no.
CT TP TN EQ FP TN TP TP TP FN FN TP FP
In-WBC/I"c-MDP TP TN TP TN TN TP TP TN TP TP FN TP TP TN TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal,
activity was identified that corresponded to abnormal increased bone tracer localization on two or more corresponding tomographic slices, in three planes (coronal, transaxial, and sagittal projections). Data Interpretation
The data that were generated from CT findings, SPECT scan findings, clinical biopsies, and clinical diagnosis were correlated and assessed for positive and negative predictive values of the scans. For calculation purposes, an equivocal scan was not used to assess the predictive value rates. Because of the small patient population, additional statistical analysis was not performed. RESULTS
Among the 20 patients with suspected osteomyelitis, the most common sites were the facial bones (8 patients), the temporal/mastoid bone (6 patients), the calvarium (3 patients), and mandible (3 patients). All patients had undergone CT scan and In-WBC/Tc-99m-MDP scans. Of the 20 patients, 10 were female and 10 were male. The age range was 3 to 74 years, with an average age o f 46.2 years. Necrotizing Otitis Externa
Of the six diabetic patients with suspected necrotizing otitis externa, two were female and four were male. Two of the patients studied with unresolving otitis externa had true-negative In-WBC/Tc-99mMDP scans, which corresponded to two true-negative CT scans (Table 1). Two patients had true-
Table 2. In-WBCfrc-MDP SPECTand CT scan results for detection of osteomyelitis in facial bones/sinuses Patient no.
CT
1 2 3 4 5 6 7 8
TN TN TN TN TN TP FN FP
In-WBC/Tc-MDP TN TN TN TN TN TP TP TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, falsenegative.
positive In-WBC/Tc-99m-MDP and CT scans. One was treated with intravenous antibiotics for 3 months, during which time he had an additional true-positive In-WBC/Tc-99m-MDP scan and then had a negative scan 3 months after treatment. CT scan was not obtained after successful therapy. The second patient is still receiving intravenous antibiotic therapy and has not had a repeat In-WBC/Tc99m-MDP scan. The remaining two patients both had positive In-WBC/Tc-99m-MDP and CT scans initially. After successful therapy, both of these patients had true-negative scans, but their CT scans were interpreted as positive (false-positive). The CT scans could not delineate between postoperative both repair and active osteomyelitis. One patient had false-negative CT and In-WBC/Tc-MDP scans after 2 months of therapy and resolution of osteomyelitis. Facial Bones/sinuses Of the eight patients evaluated, five had truenegative CT and In-WBC/Tc-99m-MDP scans (Table 2). One patient had true-positive CT and InWBC/Tc-99m-MDP scans and is currently receiving intravenous antibiotic therapy. Another patient (case 6), had a false-negative CT scan and a truepositive In-WBC/Tc-99m-MDP scan, which was confirmed at the time of revision maxillectomy when bone cultures were positive for P. aeruginosa. The patient had undergone a decompression of her trigeminal nerve 2 years previously before recurrent left ear and maxillary pain developed. Her CT scan demonstrated postoperative bony changes, but no evidence of osteomyelitis, whereas the In-WBC/Tc99m-MDP scan had demonstrated marked In-WBC localization corresponding to the left maxillary bone and maxillary sinus soft tissues. The last patient had left nasal congestion, left orbital pain, and a history
Otolaryngology Head a n d Neck Surgery Volume 113 Number I
WEBER et al.
Table 3. In-WBC/Tc-99m-MDP SPECT and CT scan results for detection of osteomyelitis in the calvarium Patient no.
CT
3 2
TP TP FN EQ
1
In-WBC/Tc-MDP TP TP TP* TN
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal. *Patient died.
of several prior sinus infections and surgery. His CT scan demonstrated left maxillary soft-tissue inflammation and left orbital bone erosion in the superior medial wall, consistent with osteomyelitis. The InWBC/Tc-99m-MDP scan was negative, and an endoscopic examination did not reveal any evidence of osteomyelitis or inflammation. His pain subsided within 1 month without treatment. He has been followed up for 1 year and has not had further symptoms. Calvarium
In this group, two were found to have true-positive CT scans and true-positive In-WBC/Tc-99m-MDP scans (Table 3). One patient was a 3-year-old child with purulent drainage from her craniotomy site 1 year after excision of hypothalamic astrocytoma followed by radiation therapy. The CT scan demonstrated erosion of the bone-flap margins compared with prior CT scan 8 months earlier. Her InWBC/Tc-99m-MDP scan showed evidence of osteomyelitis (true-positive) of the bone flap. She began receiving intravenous antibiotics, and cultures of the surgically excised bone flap revealed Pseudomonas. There was no evidence of osteomyelitis during a 7-month follow-up. The second patient was a 55year-old woman who had undergone excision of an olfactory groove meningioma. Initial CT and InWBC/Tc-99m-MDP scans were positive for osteomyelitis. An In-WBC/Tc-99m-MDP scan, after 2 months of intravenous antibiotics, was still positive for osteomyelitis. CT showed healing bone but no evidence of osteomyelitis. Surgical debridement was performed, and histopathologic examination was consistent with osteomyelitis. Before further follow-up could be ascertained, the patient died of a myocardial infarction. The third patient had undergone orbital exoneration and external beam radiation for a rhabdomyosarcoma. He presented with a draining orbital sinus. CT could not differentiate
39
Table 4. In-WBC/Tc-99m-MDP SPECT and CT scan results for detection of osteomyelitis in the mandible Patient no.
CT
2 1 3
TP FN TN
In-WBC/Tc-MDP TP TP EQ
TP, True-positive; TN, true-negative; FP, false-positive; FN, false-negative; EQ, equivocal,
between postoperative changes and osteomyelitis, whereas In-WBC/Tc-99m-MDP scan was negative. Cultures and histologic findings from debridement were both negative. Mandible
Of the three patients in this group, one patient had true-positive CT and In-WBC/Tc-99m-MDP scans; this finding was confirmed by histology after debridement. This patient was lost to follow-up (Table 4). The second patient had a false-negative CT scan and true-positive In-WBC/Tc-99m-MDP scan. This was a 33-year-old man with pain and swelling 1 year after tooth extraction and drainage of a right temporal space abscess. The CT scan demonstrated soft-tissue edema, but no bony involvement. The In-WBC/Tc-99m-MDP scan showed positive uptake at the right mandibular ramus and adjacent soft tissue. The biopsy specimens were positive for anaerobes and actinomycosis, and the bone was also positive for infection on histopathologic studies. This patient has also been lost to follow-up. The third patient had a true-negative CT scan, whereas the In-WBC/Tc-99m-MDP was indeterminate. The patient was a 53-year-old man who had undergone irradiation therapy for squamous cell carcinoma at the tongue base 5 months previously. He had a buccal ulcer that was associated with mandibular exposure. He underwent excision of this area and debridement of the bone; cultures and histopathology were not obtained. Statistics
The predictive value rates for true-positive and
true-negative In-WBC/Tc-99m-MDP scans appear to be better than those of the CT scan (Table 5). Sensitivity, specificity, and confidence levels were not calculated in this study because the number of patients was too small to obtain significant results. The malignant otitis externa group demonstrated true-
40
Otolaryngology Head and Neck Surgery July 1995
WEBERet al.
Table 5. Predictive value rates in In-WBC/Tc-99m-MDP scans vs. CT scans Site
Positive PV rate of In-WBC/Tc-MDP [%]
Negative PV rate of In-WBC/Tc-MDP [%]
100 100 100 1O0 1O0
86 1O0 1O0
i
Malignant otitis extema Facial bones/sinuses Mandible Calvarium Overall
i
93
Positive PV rate of CT [%)
Negative PV rate of CT [%)
i
71 50
50 83
1O0 75
50 61
PV, Predictive value.
positive and true-negative predictive values of 100% and 86%, respectively, for In-WBC/Tc-99m-MDP scan and 71% and 50% for the CT scan. The facial bone/sinus group demonstrated a positive and negative predictive value rate of 100% for In-WBC/Tc99m-MDP scan and 50% and 83% for the CT scan. The calvarial group also had a positive/negative predictive value of 100% and 0% for the CT scan and a positive and negative predictive value of 100% for the In-WBC/Tc-99m-MDP scan, whereas the mandible demonstrated a positive and negative predictive value rate of 100% and 50%, respectively, for the CT scan, and the In-WBC/Tc-99m-MDP scan had a positive predictive value of 100%. The overall positive and negative predictive value rates were 75% and 61%, respectively, for CT scan and 100% and 93% for In-WBC/Tc-99m-MDP. DISCUSSION
The results of In-WBC/Tc-99m-MDP scintigraphy were better than those of CT in our 20 patients, as indicated by the predictive value rates. InWBC/Tc-99m-MDP scan reverts back to normal sooner than CT scan, as seen in three patients. It appears In-WBC/Tc-99m-MDP is better able to evaluate the status of postoperative patients. Our series showed that in the seven patients misdiagnosed by CT, postoperative bone changes were always the factor for a false-negative/positive interpretation. Bone Tc-99m-MDP scans show increased radiotracer localization at sites of active osteoblastic activity, which can also be caused by infection, inflammation, postoperative bone repair, trauma, or neoplasms. Although a negative bone scan essentially excludes osteomyelitis, the bone scan remains positive for as long as this is active bone reparation, which lags well behind the resolution of infection. The Ga-67 scan has been used to evaluate the resolution of established osteomyelitis. Ga-67 scans, however, often take up to 6 months to return to normal after successful treatment of osteomyelitis,z4
Ga-67 scans were not obtained in our patients, so comparisons could not be made. CT imaging detects changes in bone density, which in turn can be used as an indicator of bone infection. However, noninfected postoperative , postirradiation, or posttraumatic bone density alterations usually cannot be differentiated from bone density changes due to osteomyelitis. In addition, CT findings are very slow to return to normal after successful treatment of active osteomyelitis. Most of the discrepancies between the CT and the InWBC/Tc-99m-MDP interpretations can be attributed to these factors. It would appear that the In-WBC/Tc-99m-MDP scans are superior to CT scans for diagnosis and follow-up of osteomyelitis in most regions of the skull (Table 5). Epstein et al. 24recently reported planar In-WBC scintigraphy as a useful technique for diagnosis and monitoring of osteomyelitis. However, false-negative and false-positive In-WBC scans have been reported in low-grade or chronic infections, at sites with extensive bone marrow (such as the vertebral column), or in immunosuppressed patients, inpatients after long-term intravenous antibiotic therapy, and inpatients with diabetic microangiopathy.22-29By combining SPECT Tc-99m-MDP bone with InWBC SPECT scintigraphy, many of the problems inherent with planar In-WBC scintigraphy can be avoided. All but two In-WBC/Tc-99m-MDP scans correlated with the clinical picture. One was an equivocal, seen in a patient 5 months after external beam radiation thearpy, and who now has a buccal ulcer with exposed mandible. The usual criteria for a positive scan could not be used in this patient because of the decreased true localization of radiotracer at the site of irradiated bone. The falsenegative study was in a patient who had been receiving intravenous antibiotics for 3 months for osteomyelitis of the right temporal bone. Persistent low-grade osteomyelitis was not detected, but a follow-up In-WBC/Tc-99m-MDP scan was marked positive on the opposite temporal bone 3 weeks after
Otolaryngology Head a n d Neck Surgery Volume 113 Number I
intravenous antibiotics were discontinued. This false-negative scan probably occurred because of long-term persistent antibiotics combined with diabetic microangiopathy. This case was recently reported by Redleaf et al. 21 from our institution. It should be emphasized that the clinical manifestations of osteomyelitis, as well as their SPECT scans and CT scans, should all be taken into consideration when deciding on the discontinuation of therapy. CONCLUSION
1. In-WB/Tc-99m-MDP bone SPECT scintigraphy is superior to CT scan in the evaluation of cranial osteomyelitis. 2. In-WBC/Tc-99m-MDP bone SPECT imaging reverts back to normal after successful treatment of osteomyelitis much sooner than CT or Ga-67 scans. 3. In-WBC/Tc-99m-MDP bone SPECT imaging is most useful in the evaluation of postoperative patients and for follow-up after treatment of osteomyelitis of the temporal and facial bones. REFERENCES
1. Bal AJM, Tiwari RM, de Rijcke TBM. Osteomyelitis in the head and neck. J Laryngol Otol 1985;99:1059-65. 2. Noyek AM, Kitsch JC, Greyson ND, Wortzman G, Jazrawy H, Freeman JL. The clinical significance of radionuclide bone and gallium scanning in osteomyelitis of the head and neck. Laryngoscope 1984;94:1-21. 3. Chandler JR, Grobman L, Quencer R, Serafini A. Osteomyelitis of the base of the skull. Laryngoscope 1986;96:245-51. 4. Lloyd GAS, Lund VJ, Scadding GK. CT of the paranasal sinuses and functional endoscopic surgery: a critical analysis of 100 symptomatic patients. J Laryngol Otol 1991;105:181-5. 5. Godofsky EW, Zinreich J, Armstrong M, Leslie JM, Weikel CS. Sinusitis in HIV-infected patients: a clinical and radiographic review. Am J Med 1992;93:163-70. 6. Malone DG, O'Boynick PL, Ziegler DK, Batnitzky S, Hubble JP, Holladay FP. Osteomyelitis of the skull base. Neurosurgery 1992;30:426-31. 7. Chandler JR. Malignant external otitis. Laryngoscope 1968; 78:1257-94. 8. Gherini SJ, Brackman DE, Bradley WB. Magnetic resonance imaging and computerized tomography in malignant external otitis. Laryngoscope 1992;96:542-8. 9. Grobman LR, Ganz W, Casiano R, Goldberg S. Atypical osteomyelitis of the skull base. Laryngoscope 1989;99:671-6. 10. Curtin HD, Wolfe P, May M. Malignant external otitis: CT evaluation. Radiology 1982;145:383-8.
WEBER et al.
4t
11. Mendelson DS, Som PM, Mendelson MH, Parisier SC. Malignant external otitis: the role of computed tomography and radionuclides in evaluation. Radiology 1983;149:745-9. 12. Gold S, Som PM, Lucente FE, Lawson W, Mendelson M, Parisier SC. Radiographic findings in progressive necrotizing "malignant" external otitis. Laryngoscope 1984;94:363-6. 13. Lloyd GAS. CT of the paranasal sinuses: study of a control series in relation to endoscopic sinus surgery. J Laryngol Otol 1990;104:477-81. 14. Rubin J, Curtin HD, Yu VL, Kamerer DB. Malignant external otitis: utility of CT in diagnosis and follow-up. Radiology 1990;174:391-4. 15. Benecke JE. Management of osteomyelitis of the skull base. Laryngoscope 1989;99:1220-3. 16. Rubin J, Yu VL. Malignant external otitis: insights into pathogenesis, clinical manifestations, diagnosis and therapy. Am J Med 1988;85:391-8. 17. Parisier SC, Lucente FE, Som PM, Hirschman SZ, Arnold LM, Roffman JD. Nuclear scanning in necrotizing progressive "malignant" external otitis. Laryngoscope 1982;92: 1016-9. 18. Strashun AM, Nejatheim M, Goldsmith SJ. Malignant external otitis: early scintigraphic detection. Radiology 1984; 150:541-5. 19. Strauss M, Kaufman RA, Baum S. Osteomyelitis of the head and neck: sequential radionuclide scanning in diagnosis and therapy. Laryngoscope 1985;95:81-4. 20. Uri N, Gips S, Front A, Meyer SW, Hardoff R. Quantitative bone and 67Ga scintigraphy in the differentiation of necrotizing external otitis from severe external otitis. Arch Otolaryngol Head Neck Surg 1991;117:623-6. 21. Redleaf MI, Angelis I, McCabe GF. Indium ill-labeled white blood cell scintigraphy as an unreliable indicator of malignant otitis resolution. Ann Otol Rhinol Laryngol 1994; 103:444-8. 22. A1-Sheikh W, Sfakianakis N, Mnaynmeh W. Subacute and chronic bone infection. Radiology 1985;155:501-6. 23. Larcos G, Brown ML, Suaon RT. Diagnosis of osteomyelitis of the foot in diabetic patients. Am J of Radiol 1991;157: 527-31. 24. Epstein JS, Ganz WI, Lizak M, Grobman L, Goodwin WJ, Dewanjee MK. Indium-Ill-labeled leukocyte scintigraphy in evaluating head and neck infections. Ann Otol Rhinol Laryngol 1992;101:961-8. 25. Thakur ML, Coleman RE, Mayhill CG. Preparation and evaluation of In-III-labeled leukocytes as in abscess imaging agent doses. Radiology 1976;119:731-2. 26. Scharwecker DS. Osteomyelitis: diagnosis with indium-Illlabeled leukocytes. Radiology 1986;96:542-8. 27. McAfee JG, Samin A. In-111 labeled leukocytes. Radiology 1985; 155:221-9. 28. Kipper MS, Williams RJ. Indium-ill white blood cell imaging. Clin Nucl Med 1983;8:449-55. 29. Date FL, Thorne DA. Effect of chronicity of infection on the sensitivity of In-ill-labeled leukocyte scan. Am J Radiol 1986;142:809-12.