[Diagnostic problems in reading chest radiographies of elderly patients at admission. Personal experience].

PURPOSE

More chest radiographs are presently performed in the elderly, especially the hospitalized ones. Reading these images is difficult because of the involutions in the thoracic cage, heart and lungs and the scars or calcifications from different abnormal causes. In the elderly, bronchogenic carcinoma may present as an occasional "coin" lesion and therefore such a finding may be an important diagnostic problem and require some expensive and dangerous examinations next. We investigated the relative frequency of questionable abnormal findings in the daily reading of the chest radiographs of elderly patients, the relative importance of the radiologist's experience and of the examination execution technique; the relative costs were also evaluated.

MATERIAL AND METHODS

Four radiologists, two of them more experienced (FS, PT), read the consecutive chest radiographs of 811 elderly patients (273 men, 538 women) hospitalized May to December, 1997. Four hundred and ten of them were 65-75 years old and 401 over 75 (particularly, 28 were over 90). Five hundred and sixty-five chest radiographs were made with the AMBER technique and 246 with frontal views only. T-MAT G RA Kodak high-contrast films with Kodak Lanex green-transmitting intensifying screens were used in all cases. CT scans were made with conventional (CT Sytec 3000, GE) or spiral (X Vision, Toshiba) scanners.

RESULTS

Seven hundred and fifty-seven radiographs were considered adequate (93%) and 54 inadequate (7%) for diagnosis (25 in patients 65-75 years old, 25 in patients 75-80 and 4 in patients over 90). Thirty-eight of these 54 inadequate radiographs had been made with the AMBER technique and 16 with frontal views only. The more experienced radiologists read 27 (11%) and 19 (10%) of them and the less experienced ones read 4 (2%) and 4 (3%), respectively. The next examinations were other projections and/or radioscopy (4 cases), conventional tomography (7 cases), CT (43 cases), and US (2 cases). "Coin" lesions were the major cause of questionable diagnosis, especially in posterior (10 cases) and peripheral (7 cases) regions, where the differential diagnosis was with vertebral osteophytosis and small rib crowding, respectively.

CONCLUSIONS

More skilled radiologists have more doubts reading the chest radiographs of elderly patients. But the next examinations will likely balance the needless ones after an initial misdiagnosis. The chest of elderly patients remains a complex and very little known subject and the reader's experience plays an important role. The examination execution technique must be as accurate as possible in both optimal and suboptimal settings.