Excimer laser corneal surgery: new strategies and old enemies

This study aimed at investigating the accuracy of seven forensic odontologists (FO) and seven police officers (PO) in determining person identity using comparable intraoral radiographs from implant treated patients, and evaluating the different characteristics used for identification. The investigation was based on 34 edentulous patients, from 26 of whom a matched pair of radiographs was constructed in such a way that one, taken after insertion of an implant supported prosthesis in the anterior part of the mandible, was regarded as the antemortem radiograph, while another picture from a later follow-up examination served as the postmortem x-ray. From each of the remaining eight patients similar radiographs were selected so that four antemortem and four postmortem ones were obtained. These eight radiographs were also paired but did not match. There were thus 30 pairs of radiographs. A classification of the 26 matched pairs regarding degree of ease (easy, moderate, difficult) in combining the radiographs, using as parameters the design of the implants, shape of the abutments, shape of the bridges and bony anatomy of the jaws was established. The total number of errors made by the FO were higher (26) than those made by the PO (18) (one PO combined all 26 matching radiographs correctly) and 12 of the 26 matchable cases were correctly paired by all observers. The design of the fixed prostheses was the most often used characteristic in the exercise.     

Radiographic evaluation of osseous displacement following intra-articular fractures of the distal radius: reliability of plain radiography versus computed tomography

This study evaluated the reliability of plain radiography versus computed tomography (CT) for the measurement of small (< 5 mm) intra-articular displacements of distal radius fracture fragments. The plain radiographs and CT scans of 19 acute intra-articular distal radius fractures were used by 5 independent observers, using 2 standardized techniques, to quantify incongruity of the articular surface in a blinded and randomized fashion. Repeat measurements were performed by the same observers 2-4 weeks later, allowing determination of intraclass correlation coefficients (ICC) as a measure of intraobserver and interobserver agreement. The average maximum gap displacement on plain radiographs was 2.1 mm (range, 0.0-15.0 mm, lateral view) and on CT images was 4.9 mm (range, 0.7-17.3 mm, axial view). The average maximum step displacement on plain radiographs was 0.9 mm (range, 0.0-6.4 mm, lateral view) and on CT images was 1.2 mm (range, 0.0-6.0 mm, sagittal view). More reproducible values determining step and gap displacement were obtained when the arc method of measurement was used on CT scans (ICC values, .69-.97) as compared to the longitudinal axis method for plain radiographs (ICC values, .30-.50). For measured displacements of 2 mm or more, our data demonstrated poor correlation between measurements made on CT images and those made on plain radiographs (gap or step displacement > 2 mm, K = 0.21; step displacement > 2 mm, K = 0.21). Thirty percent of measurements from plain radiographs significantly underestimated or overestimated displacement compared to CT scan measurements. From these data, we conclude that CT scanning data, using the arc method of measurement, are more reliable for quantifying articular surface incongruities of the distal radius than are plain radiography measurements.     

Assessing the quality of radiographs accompanying biopsy specimens

According to a long-established protocol, biopsy specimens of bony lesions sent to the University of Washington Histopathology Service are accompanied by relevant radiographs. In this study, the authors evaluated the quality of 114 radiographs, both intraoral and panoramic, from 89 consecutive cases. The radiographs were judged on their clinical usefulness, rather than on criteria for technically ideal images, and on the severity, rather than the number, of errors. Only 33 percent of the panoramic radiographs were of acceptable quality, while the quality of the intraoral radiographs was markedly better. The most striking feature was the lack of three-dimensional radiographic coverage of the site of surgical exploration.     

The reliability of apical x-ray pictures in the diagnosis of mandibular bone lesions. A review of the literature and in-vitro study

Using 6 macerated human anatomical preparations of mandibular jaw regions with either front teeth or premolars and molars, the radiological detectability of artificial periapical bone lesions was evaluated in relation to the size of the lesions and to the angle of x-ray projection. The artificial lesions were enlarged stepwise until distinct radiological visibility was attained. For each step of enlargement, x-ray pictures were taken with orthoradial as well as with at 25 degrees mesially and distally excentered projections. All x-ray pictures were evaluated by 6 dentists. The following results were obtained: The size of a periapical bone lesion, at which it is becoming radiologically detectable, varies between the different regions of the lower jaw. Isolated spongiosa lesions being larger than 3 mm in diameter are most often detectable at mandibular front teeth and premolars. Isolated spongiosa lesions at mandibular molars are generally non-detectable. Atypical lesions, e.g. discontinuities of bony structures are particularly difficult to detect radiologically. There were no statistically significant differences in lesion detectability between x-rays of different angle projections.     

Systemic arterial supply to normal basal segments of the left lung: characteristic features on chest radiography and CT

OBJECTIVE: The purpose of this report is to present the characteristic features on chest radiography and CT of systemic arterial supply to normal basal segments of the left lung in three cases. CONCLUSION: Accurate interpretation of chest radiographs and CT scans is important in making the correct diagnosis and avoiding life-threatening lung biopsy.     

Effects of a high NaCl diet on eating and drinking patterns in pygmy goats

PURPOSE: Our goal was to determine the spectrum of 2-[18F]fluoro-2-deoxy-D-glucose (FDG) PET findings in patients with round atelectasis (RA). METHOD: All patients from 1992 to 1997 with radiologic features of RA and FDG-PET scans were evaluated. There were nine men ranging in age from 52 to 75 years (mean 65 years). All had chest radiographs and CT scans that were correlated with FDG-PET. FDG-PET was considered positive if lesion activity was greater than mediastinal activity and negative if lesion activity was the same as or less than mediastinal activity. RESULTS: Nine patients had 10 lesions, ranging in size from 1.2 to 5.0 cm (mean 3.1 cm). Lesion locations were right lower lobe (n = 5), left lower lobe (n = 4), and lingula (n = 1). All lesions were homogeneous and of soft tissue attenuation on CT. None contained air bronchograms or calcification. All had in-curving vessels and bronchi (comet tail sign), adjacent pleural thickening, and volume loss on CT. All lesions were negative on FDG-PET. Four lesions were percutaneously biopsied and showed chronic inflammation consistent with RA. Two lesions were unchanged on 2 and 3 year follow-up CT and were presumed to be RA as were four other lesions with characteristic CT features and negative FDG-PET. CONCLUSION: Our experience suggest that RA in not metabolically active on FDG-PET imaging. Thus, FDG-PET scans can play a role in differentiating RA from malignancy when there are few or atypical features of RA on chest radiographs and CT.     

High-resolution CT of the lung: determination of the usefulness of CT scans obtained with the patient prone based on plain radiographic findings

OBJECTIVE: We assessed the usefulness of chest radiographs for predicting whether high-resolution CT scans obtained with the patient prone would be valuable in assessing suspected diffuse lung disease. MATERIALS AND METHODS: In 100 consecutive patients undergoing high-resolution CT, findings on plain chest radiographs were classified as normal, possibly abnormal, or abnormal. CT scans obtained with the patient supine were assessed for the presence and distribution of lung abnormalities without knowledge of the plain radiographic classification. A second review of the CT scans was done with equal numbers of scans obtained with the patient prone and with the patient supine. The usefulness of the CT scans obtained with the patient prone for detecting lung disease was determined and related to the plain radiographic classifications. RESULTS: High-resolution CT scans obtained with patients prone were helpful in excluding or confirming posterior lung abnormalities in 10 (28%) of 36 patients who had normal findings on chest radiographs, five (28%) of 18 patients who had possibly abnormal findings on chest radiographs, and only two (4%) of 46 patients who had abnormal findings on chest radiographs. The proportion of patients who benefited from high-resolution CT scans obtained with the patient prone was significantly lower among the patients with abnormal findings on chest radiographs than among the patients with normal (p = .008) or possibly abnormal (p = .02) findings on chest radiographs. The two patients with abnormal findings on radiographs in whom CT scans obtained with the patient prone were helpful had minimal radiographic abnormalities. CONCLUSION: In patients with suspected diffuse lung disease, obtaining high-resolution CT scans with the patient prone may be useful when chest radiographs show normal findings, possibly abnormal findings, or minimal abnormalities indicative of diffuse lung disease. However, such scans are of little value in patients whose radiographs show abnormalities indicative of diffuse lung disease.     

Implant reconstruction of the jaws and craniofacial skeleton

Full oral rehabilitation with a high degree of success is now possible with osseointegrated implants. Osseointegration is a direct connection between living bone and the titanium implant at the level of the light microscope. Osseointegrated implants are currently used to replace single teeth, support fixed bridges and stabilize full dentures. These implants can also be placed extraorally for attachment of facial prosthesis. The surgical technique used to place implants intraorally into jaws or facial skeleton is performed in two stages using a local anesthetic and/or conscious sedation. During stage I surgery, holes are placed into the jaw using a series of gradually larger diameter burs until the desired diameter and depth of the bony preparation is achieved. The implant is then placed. The implant must remain undisturbed for 4 months for osseointegration to take place. Stage II surgery is then required to remove the mucosa over the implant and place the transmucosal abutment. After 1-2 weeks of healing, the restorative dentist can take an impression and fabricate the prosthesis. On occasion, it is necessary to augment the height and width of the atrophic jaw with autogenous or allogeneic bone grafts prior to implant placement. Bone grafts are sometimes placed on the floor of the nose or the floor of the maxillary sinus. Guided tissue regeneration is a technique used to generate bone within bony defects adjacent to implants. With long-term rates of success (5 years) of 99% for implants placed in the mandible and 95% for those placed in the maxilla, reconstruction of the jaws and cranial facial skeleton with osseointegrated implants has become the treatment of choice.     

A 3-year experience with guided tissue regeneration procedures

The Guided Tissue Regeneration (GTR) procedures are promoting a clinically and radiologically as well as histologically verifiably periodontal attachment gain. The objective of the study was to evaluate the clinical efficacy of these GTR techniques. In the past four years different barrier membranes (Gore-tex, Resolut and Guidor) were used around 318 teeth of 196 patients. 169 periodontal defects of 140 patients were followed up at least for two years. 54 patient had chronic adult type periodontitis, 67 suffered with rapidly progressing periodontitis and 15 had different severe mucogingival lesions. 111 vertical bony defects, 43 Class II-III furcation lesions and 15 mucogingival lesions were surgically corrected. The average preoperative probing depth (PD) and the clinical attachment loss (CAL) of the vertical bony defects were 5.3 +/- 1.7 mm and 6.2 +/- 1.9 mm respectively. The PD of the deepest Class III furcation lesion was 11 mm. The average gingival recession of the mucogingival lesions was 4.5 +/- 1.1 mm. The GTR technique provided the best results in the Class II-III furcation lesions, where an average 2.4 +/- 0.9 clinical attachment gain was observed one year postoperatively. The GTR techniques provided an average 1.8 +/- 1.2 mm attachment gain in the vertical bony crater cases. In both groups of cases a marked gingival recession followed the healing and the periodontal regeneration. In this way the average reduction in the probing depth exceeded the average attachment gain by more that 1.5 mm. 1 year after the operation the average radiologic bone fill was about 0.9-1.2 mm. The resorbable barrier membranes resulted in clinically significant root coverage and an average 3.5 +/- 1.7 mm gain in the width of keratinized gingiva. The success or failure of our cases were mainly determined by the patient's compliance, the level of the postoperative professional and individual oral hygiene and the number of periodontal recalls. These findings are also underlining the importance of the high standard of oral hygiene in the postoperative periodontal regeneration.     

Omega-3 fatty acids in the prevention-management of cardiovascular disease

Eight cervical spines were used to evaluate the relation of the screw tip to the nerve root in the intervertebral foramen. The specimens were divided into two groups: (a) lateral placement without contact with the nerve root, and (b) lateral placement with penetration of the nerve root. Six screws were used for each specimen. After screw placement, oblique radiographs and axial computed tomography (CT) scans were taken. The results on oblique radiographs showed that 23 (95.8%) of 24 screws without contact with the nerve root were found in the upper zone or the junction between the upper and lower zones of the intervertebral foramen. Twenty (83.3%) of 24 screws with penetration of the nerve root were located in the junction between the lower zone of the intervertebral foramen and the pedicle zone. No definite diagnosis of screw penetration of the nerve root could be made on axial CT scans, although scans can show that the screw is violating the foramen. Whether or not a screw violating the intertransverse foramen and affects the nerve root depends on its position on the oblique radiograph. It may be not necessary to remove or change the screw immediately if a longer screw is found in the upper portion of the intervertebral foramen on the oblique view and angled laterally on axial CT scan in a patient without radicular symptoms.     

Effect of decalcified freeze-dried bone allograft on the healing of jaw defects after cyst enucleation

PURPOSE: This study radiographically evaluated the changes that occur in jaw defects after enucleation of cysts and either grafting with demineralized freeze-dried bone allograft (DFDBA) or packing with absorbable gelatine sponge (Gelfoam). PATIENTS AND METHODS: Thirty-two patients (14 males, 18 females) with cystic lesions of the jaws were divided into two groups. Group A underwent enucleation and bone grafting with DFDBA. Group B underwent enucleation and packing with Gelfoam. Both groups were evaluated radiographically before surgery, immediately postoperatively, and 6, 12, 24, and 36 months postoperatively. The density of the area was measured with a digital densitometer. RESULTS: The density in group A was significantly (P < .05 to .01) greater than in group B at 6 and 12 months postoperatively. The difference at 24 months was not significant. Group A reached density levels at 6 months that were reached by group B at 12 months. CONCLUSION: The findings demonstrate that decalcified freeze-dried bone allograft enhances bone healing in jaw defects after cyst removal.     

Usefulness of computed tomography and scintigraphy in diagnosis of emphysematous bullae in the lung

The x-ray examinations usually do not reveal morbid changes after lung expansion in the treatment of spontaneous pneumothorax. In our observation computed tomography (CT) and scintigraphy enable not only the exact determination of the extent of changes but also they disclose bullae invisible in conventional chest radiographs. 15 patients with cured spontaneous pneumothorax and 10 patients with radiographic evidence of a bulla or bullae were examined. CT scans showed bullae from 3 to 18 cm in diameter involving predominantly the upper lobes. 4 patients had additionally subpleural or intraparenchymal bullae of various degrees. In all patients with cured spontaneous pneumothorax, CT scans revealed intraparenchymal bullae, and in 6 cases bilateral intraparenchymal bullae were revealed. Only in sites of large bullae, no isotopic marker or its low elimination was shown in perfusion and inhalation scintigraphy. 10 patients with giant bullous emphysema were operated; in 6 patients enucleation of bullae, in 3 lobectomy and in one patient bullectomy were performed. CONCLUSION: CT is a method of choice in the diagnosis of lung emphysematous bullae and it enables the detection of the changes undetectable in chest radiographs. Perfusion and inhalation scintigraphy is helpful in the diagnosis of large emphysematous bullae and postoperative follow-up examination.     

99mTc-HMPAO SPECT of the brain in mild to moderate traumatic brain injury patients: compared with CT--a prospective study

Single photon emission computed tomography (SPECT) with Technetium-99m hexamethyl propylenamine oxime (Tc-99m-HMPAO) was used in 20 patients with mild to moderate traumatic brain injury (TBI) to evaluate the effects of brain trauma on regional cerebral blood flow (rCBF). SPECT scan was compared with CT scan in 16 patients. SPECT showed intraparenchymal differences in rCBF more often than lesions diagnosed with CT scans (87.5% vs. 37.5%). In five of six patients with lesions in both modalities, the area of involvement was relatively larger on SPECT scans than on CT scans. Contrecoup changes were seen in five patients on SPECT alone, two patients with CT alone and one patient had contrecoup lesions on CT and SPECT. Of the eight patients (50%) with skull fractures, seven (43.7%) had rCBF findings on SPECT scan and five (31.3%) demonstrated decrease in rCBF in brain underlying the fracture. All these patients with fractures had normal brain on CT scans. Conversely, extra-axial lesions and fractures evident on CT did not visualize on SPECT, but SPECT demonstrated associated changes in rCBF. Although there is still lack of clinical and pathological correlation, SPECT appears to be a promising method for a more sensitive evaluation of axial lesions in patients with mild to moderate TBI.     

Prevalence and duration of postoperative pneumoperitoneum: sensitivity of CT vs left lateral decubitus radiography

OBJECTIVE: The purpose of this study was to evaluate the prevalence, location, and duration of pneumoperitoneum in postoperative patients and to compare the sensitivities of CT and left lateral decubitus radiography in the detection of postoperative pneumoperitoneum. SUBJECTS AND METHODS: Twenty-seven CT scans and 27 abdominal radiographs with the patient in the left lateral decubitus position were obtained prospectively in 17 patients after uncomplicated abdominal surgery. Fifteen patients were examined 3 days after surgery and 12 were examined 6 days after surgery. The studies were evaluated in a blinded fashion for the presence, location, and volume of free air. The presence of air on the radiographs and the presence and quantity of air on the CT scans were correlated with each subject's surgical procedure, age, sex, and body habitus. RESULTS: Pneumoperitoneum was seen on 13 (87%) of 15 CT scans and eight (53%) of 15 radiographs obtained 3 days after surgery and on six (50%) of 12 CT scans and one (8%) of 12 radiographs obtained 6 days after surgery. The calculated volume of free air seen on the CT scans ranged from 0.3 to 5.8 ml. Sixty-two percent of collections by volume were located in the midline/parahepatic space, 22% in the pelvis, and 16% in the mesentery. Radiographs showed pneumoperitoneum in only nine (47%) of 19 examinations in which the corresponding CT scans showed free air. Findings on radiographs were false-negative in seven (87%) of eight obese patients in whom pneumoperitoneum was detected on CT scans. CONCLUSION: The prevalence of pneumoperitoneum in the postoperative period based on CT findings is greater than that previously reported. Small amounts of pneumoperitoneum frequently collect along the anterior abdominal wall in two preferential spaces, the pararectus and midrectus recesses. The results of this study show that CT is significantly more sensitive than plain radiography for detecting small amounts of free intraperitoneal air in postoperative patients. Radiography is particularly insensitive for imaging obese and heavy patients.     

Exercise and cellular innate immune function

PURPOSE: To determine, in three-dimensions, the difference between prostate delineation in magnetic resonance (MR) and computer tomography (CT) images for radiotherapy treatment planning. PATIENTS AND METHODS: Three radiation oncologists, considered experts in the field, outlined the prostate without seminal vesicles both on CT, and axial, coronal, and sagittal MR images for 18 patients. To compare the resulting delineated prostates, the CT and MR scans were matched in three-dimensions using chamfer matching on bony structures. The volumes were measured and the interscan and interobserver variation was determined. The spatial difference between delineation in CT and MR (interscan variation) as well as the interobserver variation were quantified and mapped three-dimensionally (3D) using polar coordinates. A urethrogram was performed and the location of the tip of the dye column was compared with the apex delineated in CT and MR images. RESULTS: Interscan variation: CT volumes were larger than the axial MR volumes in 52 of 54 delineations. The average ratio between the CT and MR volumes was 1.4 (standard error of mean, SE: 0.04) which was significantly different from 1 (p < 0.005). Only small differences were observed between the volumes outlined in the various MR scans, although the coronal MR volumes were smallest. The CT derived prostate was 8 mm (standard deviation, SD: 6 mm) larger at the base of the seminal vesicles and 6 mm (SD 4 mm) larger at the apex of the prostate than the axial MRI. Similar figures were obtained for the CT and the other MRI scans. Interobserver variation: The average ratio between the volume derived by one observer for a particular scan and patient and the average volume was 0.95, 0.97, and 1.08 (SE 0.01) for the three observers, respectively. The 3D pattern of the overall observer variation (1 SD) for CT and axial MRI was similar and equal to 3.5 to 2.8 mm at the base of the seminal vesicles and 3 mm at the apex. CONCLUSION: CT-derived prostate volumes are larger than MR derived volumes, especially toward the seminal vesicles and the apex of the prostate. This interscan variation was found to be larger than the interobserver variation. Using MRI for delineation of the prostate reduces the amount of irradiated rectal wall, and could reduce rectal and urological complications.     

Alveolar bone levels in AIDS and HIV seropositive patients and in control subjects

Alveolar bone levels were studied from intraoral radiographs of 24 non-hospitalized patients with AIDS, 17 HIV seropositive subjects, and 39 matched control subjects. The AIDS/HIV subjects were seeking dental care in a faculty practice. The matched control subjects came from those non-HIV-infected patients seeking dental care at the University of Washington. Magnified intraoral radiographs were used to assess the distance between the cemento-enamel junction (CEJ) and the alveolar bone level (BL). The extent of vertical defects and furcation invasions was also assessed. The mean age of the AIDS, HIV, and control subjects was 38.9 +/- 6.6 years, 37.1 +/- 7.6, and 39.9 +/- 5.6, respectively, and was not statistically different. Among the AIDS patients, 75% were smokers, while 88.2% of the HIV subjects were cigarette smokers. Therefore, the matched control subjects were also smokers to the same extent. The mean difference in distance CEJ-BL was 0.1 mm (mesial) and 0.3 mm (distal) and greater in the HIV/AIDS group than in the control group, but not statistically different. No vertical defects > or = 3.0 mm were found in 69.2% of the control subjects and in 58.5% of the combined HIV/AIDS group. None of the HIV/AIDS subjects had more than 5 defects > 3.0 mm, while 7.6% of the control subjects had such defects. Significant associations were found between smoking and extent of alveolar bone loss (distance) (P < 0.001) as well as the number and extent of vertical defects (P < 0.01), but were not associated with HIV status. The extent of furcation invasions, as read radiographically, did not differ between groups. In conclusion, smoking but not HIV status was the primary factor for alveolar bone loss.     

Invasive pulmonary aspergillosis in AIDS: radiographic, CT, and pathologic findings

PURPOSE: To review the radiographic and computed tomographic (CT) manifestations of invasive pulmonary aspergillosis and to correlate the imaging and pathologic findings in patients with acquired immunodeficiency syndrome (AIDS). MATERIALS AND METHODS: Chest radiographs, CT scans, and pathologic specimens were reviewed retrospectively in 10 AIDS patients with proved invasive pulmonary aspergillosis. RESULTS: The most common radiographic finding was the presence of thick-walled cavitary lesions. Less common findings included nodules, consolidation, and pleural effusion. CT depicted more nodules and cavities than did radiography. The predominant pathologic abnormalities consisted of tissue invasion and abscess formation and angioinvasion with or without infarction. All patients had infection with Aspergillus fumigatus as well as other pathogens, the most common being cytomegalovirus and Pseudomonas aeruginosa. CONCLUSION: Thick-walled cavitary lesions are the most common radiologic manifestation of invasive pulmonary aspergillosis in AIDS. The findings are more numerous and better defined on CT scans. The radiologic findings reflect a spectrum of pathologic abnormalities.     

Use of chest computed tomography in the staging of pediatric Wilms' tumor: interobserver variability and prognostic significance

PURPOSE: To determine the specificity and prognostic significance of computed tomography (CT) of the chest in pediatric Wilms' tumor. PATIENTS AND METHODS: Patients treated for newly diagnosed Wilms' tumor at St Jude Children's Research Hospital between December 1978 and July 1995 were included in the study if an initial chest radiograph and CT were available and if pulmonary involvement (determined by chest radiographs) was absent. For the 202 patients studied, radiographs and CT scans were reviewed blindly and independently by three experienced radiologists for the presence of pulmonary nodules. Outcome variables consisted of intraobserver variability (in a subsample of 40 cases) and concordance between ratings on radiographs and CT scans (both by McNemar's test), interrater variability (by logistic regression), and the cumulative incidence of pulmonary relapse for patients with and without positive CT scans, by reviewer. RESULTS: As expected, ratings of pulmonary involvement on radiographs were discordant with CT ratings. There was marked variability among reviewers in CT ratings (P = .0001). Of 202 CT scans, 78 were read as positive by at least one reviewer, 41 were rated positive by only one reviewer, 18 by two reviewers, and 19 by all three. Intrarater variability on repeat reviews was not significant. Patients with nodules identified on CT had a significantly higher pulmonary relapse rate when analyzed separately by reviewer. However, for the 14 patients who had pulmonary relapse, CT scans were rated positive by all three reviewers in only five cases and as negative by all three in another five cases. CONCLUSION: The variability in interpretation of chest CT scans in patients with Wilms' tumor limits the predictive utility of these studies. Optimal, standardized techniques and central review are essential if chest CT is to be used for staging in cooperative studies.     

Early detection of pneumonia in febrile neutropenic patients: use of thin-section CT

OBJECTIVE: The purpose of this study was to evaluate the usefulness of thin-section CT for early detection of pneumonia in neutropenic patients with an unknown site of infection and normal or nonspecific findings on chest radiographs. SUBJECTS AND METHODS: Eighty-seven patients with febrile neutropenia that persisted for more than 2 days despite empiric antibiotic treatment underwent 146 prospective examinations. If findings on chest radiographs were normal (n = 126) or nonspecific (n = 20), thin-section CT (1-mm collimation, 10-mm increment) was done. If thin-section CT scans showed opacities, bronchoalveolar lavage was recommended. RESULTS: Findings on chest radiographs were nonspecific for pneumonia in 20 (14%) of 146 cases, and CT findings in those cases were suggestive of pneumonia. Microorganisms were detected in 11 of those 20 cases. Seven of the 11 cases were not optimally treated before CT diagnosis, the other four were sufficiently treated. Findings on chest radiographs and thin-section CT scans were normal in 56 (38%) of 146 cases. In 70 (48%) of 146 cases, findings on chest radiographs were normal, whereas findings on thin-section CT scans were suggestive of pneumonia. Microorganisms were detected in 30 of the 70 cases. Nineteen of 30 cases were not optimally treated before CT, whereas the other 11 cases were sufficiently treated before CT. In 22 (31%) of these 70 cases, an opacity was observed on the chest radiograph during the 7 days after the CT study. Only three (5%) of 56 pneumonias occurred during the first 7 days after thin-section CT studies with normal findings (p < .005). Additional risk factors for pneumonia occurring later that were detectable on chest radiographs were poorly defined nodules (p < .05), consolidation (p < .05), and younger age (p < .05). CONCLUSION: Thin-section CT scans show findings suggestive of pneumonia about 5 days earlier than chest radiographs show suggestive findings. When thin-section CT scans show findings suggestive of pneumonia, the probability of pneumonia being detected on chest radiographs during the 7-day follow-up is 31%, whereas the probability is only 5% when the findings on the prior thin-section CT scan were normal (p < .005). All neutropenic patients with fever of unknown origin and normal findings on chest radiographs should be examined with thin-section CT.