How reassuring is a normal breast ultrasound in assessment of a screen-detected mammographic abnormality? A review of interval cancers after assessment that included ultrasound evaluation.

AIM: To review factors resulting in a false-negative outcome or delayed cancer diagnosis in women recalled for further evaluation, including ultrasound, after an abnormal screening mammogram. MATERIALS AND METHODS: Of 646,692 screening mammograms performed between 1 January 1995 and 31 December 2004, 34,533 women were recalled for further assessment. Nine hundred and sixty-four interval cancers were reported in this period. Forty-six of these women had been recalled for further assessment, which specifically included ultrasound evaluation in the preceding 24 months, and therefore, met the inclusion criteria for this study. Screening mammograms, further mammographic views, ultrasound scans, clinical findings, and histopathology results were retrospectively reviewed by two consultant breast radiologists. RESULTS: The interval cancer developed in the contralateral breast (n=9), ipsilateral breast, but different site (n=6), and ipsilateral breast at the same site (n=31) as the abnormality for which they had recently been recalled. In the latter group, 10 were retrospectively classified as a false-negative outcome, nine had a delay in obtaining a biopsy, and 12 had a delay due to a non-diagnostic initial biopsy. Various factors relating to these outcomes are discussed. CONCLUSION: Out of 34,533 women who attended for an assessment visit and the 46 women who subsequently developed an interval breast cancer, 15 were true interval cancers, 10 had a false-negative assessment outcome, and 21 had a delay to cancer diagnosis on the basis of a number of factors. When there is discrepancy between the imaging and histopathology results, a repeat biopsy rather than early follow-up would have avoided a delay in some cases. A normal ultrasound examination should not deter the radiologist from proceeding to stereotactic biopsy, if the index mammographic lesion is suspicious of malignancy.

Investigation of acute lower gastrointestinal bleeding with 16- and 64-slice multidetector CT.

We evaluated the usefulness of 16- and 64-slice multidetector CT (MDCT) in the detection of a bleeding site in acute lower gastrointestinal tract (GIT) haemorrhage by conducting a retrospective study of cases of presumed acute lower GIT haemorrhage imaged with CT in two teaching hospitals in an 11-month period. The patients underwent contrast enhanced CT using either a 16 or 64 MDCT. No oral contrast was used. One hundred milliliters of non-ionic intravenous contrast agent was injected at 4.5 mL/s, followed by a 60 mL saline flush at 4 mL/s through a dual head injector. Images were acquired in arterial phase with or without non-contrast and portal phase imaging with 16 x 1.5 mm or 64 x 0.625 mm collimation. Active bleeding was diagnosed by the presence of iodinated contrast extravasation into the bowel lumen on arterial phase images with attenuation greater than and distinct from the normal mucosal enhancement or focal pooling of increased attenuation contrast material within a bowel segment on portal-venous images. Further management and final diagnosis was recorded. Fourteen patients and 15 studies were reviewed. CT detected and localized a presumed bleeding site or potential causative pathology in 12 (80%) of the patients. Seven of these were supported by other investigations or surgery, while five were not demonstrated by other modalities. Eight patients had mesenteric angiography, of which only four corroborated the site of bleeding. CT did not detect the bleeding site in three patients, of which two required further investigation and definitive treatment. We propose that MDCT serves a useful role as the initial rapid investigation to triage patients presenting with lower GIT bleeding for further investigation and management.

MRI of bilateral internal carotid artery agenesis.

Bilateral internal carotid artery (ICA) agenesis is rare. A patient presented with symptoms of a transient ischaemic attack. Magnetic resonance imaging with magnetic resonance angiography showed multiple foci of chronic cerebral ischaemic change and bilaterally absent ICA. Cerebral supply was derived from the posterior circulation. Differentiation has to be made between congenital agenesis/hypoplasia and acquired ICA occlusion. Computed tomography through the skull base showed the absence of the carotid canals bilaterally confirming congenital ICA agenesis.

Femoro-distal bypass surgery at Groote Schuur hospital--4-year retrospective study.

UNLABELLED: An analysis of the demographics and outcome of femoro-distal bypass in patients presenting with defined critical limb ischaemia at Groote Schuur Hospital, Cape Town, is presented. MATERIALS AND METHODS: A retrospective review was conducted between January 1998 and December 2001. During this period, 65 patients underwent femoro-distal bypass. Of these, 57 patients were analysed and 8 patients were excluded from the study because of incomplete medical records RESULTS: The median age of the patients in this study was 62 years, with a male-to-female ratio of 34:23. Twenty-eight patients (49%) were diabetic. An overall 2-year mortality of 19.2% was recorded. Reversed saphenous vein graft (RSVG) was used in 29 legs (50.9%) and in situ vein graft (ISVG) in 19 legs (33.3%). There was no statistically significant difference in the graft patency between the two methods (p = 0.39); the 2-year cumulative patency was 40% for the RSVG and 39% for the ISVG. In our unit only 2 factors influenced the outcome of femoro-distal bypass--local sepsis in the foot and an increased early postoperative ankle-brachial index (p < 0.05). Diabetes mellitus, gender, age and race had no influence on the outcome. This study showed that the site of the distal anastomosis had no effect on graft patency or limb salvage. At 2 years, the cumulative patency and limb salvage rates were 40% and 56% respectively. Of the 11 grafts with stenotic lesions requiring intervention, 6 were detected between 18 and 24 months. CONCLUSION: Management of critical limb ischaemia is a major part of the workload in our unit, with most patients undergoing primary amputation. The surgical outcome of femoro-distal bypass was largely influenced by local sepsis and early postoperative ABI. We found that more than half the stenotic lesions detected during graft surveillance occurred beyond 18 months postoperatively. This suggests that a graft surveillance programme should continue beyond 18 months.

Haemobilia after penetrating and blunt liver injury: treatment with selective hepatic artery embolisation.

Although traumatic haemobilia is uncommon and occurs in less than 3% of liver injuries, the magnitude of the bleeding may result in life-threatening complications. This study evaluated the efficacy of selective hepatic artery embolisation (HAE) in the control of bleeding in patients with traumatic haemobilia. The demographic, clinical and angiographic data on all patients with traumatic haemobilia were obtained from a prospectively documented database of patients undergoing visceral angiography for liver haemorrhage between 1967 and 2002. During the 36-year period under review, 30 patients were found to have haemobilia on selective hepatic angiography. Ten of these 30 patients had haemobilia due to accidental non-iatrogenic trauma and form the basis of this study. In 8 of the 10 patients haemobilia resulted from penetrating liver injuries and two patients had blunt trauma. The mean delay between the initial injury and the diagnosis of haemobilia was 23.5 (range 1-120) days. The mean blood loss before angiography was 8 (range 3-19) units. Six patients were treated successfully with selective hepatic arterial embolisation, three required surgery and one resolved without any intervention. There were no deaths and no complications resulting in long term sequelae. Traumatic haemobilia is an uncommon but life-threatening complication of liver injury. Selective arterial embolisation is the initial treatment of choice with a substantial rate of success and a low incidence of serious complications.