Radiography of perforating centerfire rifle wounds of the trunk.

All deaths resulting from perforating centerfire rifle wounds of the chest and abdomen, investigated by the Office of the Chief Medical Examiner for the Province of Alberta from 1988 to 1995. were reviewed retrospectively to determine whether the radiographic distribution of bullet fragments in such cases is a useful predictor of bullet trajectory. Study cases were limited to single gunshot wounds without surgical intervention or intermediate targets, and for which adequate radiography was available. Three pathologists individually viewed the radiographs on two separate occasions; wound locations were provided for the second viewing (Group 2). Differences in opinion regarding direction of fire were resolved by consensus review. A trauma radiologist independently made two sets of interpretations in the same way. Comparisons of these groups of interpretations were made with the actual bullet direction determined at autopsy. Of 21 cases included in the study, only three (14.3%) did not require consensus resolution in either group. Accuracy of pathologists' interpretation improved from 38.1% (8/21) to 76.2% (16/21) with provision of wound locations (p = 0.012). The radiologist achieved similar improvement, from 28.6% (6/21) to 47.6% (10/21). The rate of agreement between radiologist and pathologists increased from 42.9% (9/21) to 61.9% (13/21) between Groups 1 and 2. Both the pathologists and radiologist interpreted several cases the same way in both groups; of those cases interpreted differently, the second interpretation was occasionally incorrect after correct interpretation in Group 1. We conclude that bullet direction for perforating centerfire rifle wounds cannot be accurately determined from postmortem radiographs. When wound location is known, the ability to predict bullet direction improves but is still subject to error, including a lack of consistency between observers.

Fine-needle aspiration cytology of sarcomatoid carcinoma of the breast: a cytologically overlooked neoplasm.

Sarcomatoid carcinoma of the breast is a very uncommon neoplasm. Fine-needle aspiration findings have been rarely reported. We report a case of sarcomatoid carcinoma of the breast that was diagnosed as a typical ductal carcinoma cytologically. The patient was a 45-year-old female who had a right breast mass for a short duration. Fine-needle aspiration (FNA) showed smears predominantly composed of single cells exhibiting plasmacytoid features with moderate degree of pleomorphism. A granular bluish background was noted on the Papanicolaou-stained smears. This background was thought to represent tumor diathesis, and the diagnosis of grade II ductal carcinoma of the breast was rendered. The patient underwent a subsequent lumpectomy with axillary node dissection. Histological examination of the lumpectomy revealed a neoplasm predominantly composed of a grade II chondrosarcoma with increased cellularity. A small (10%) component showed grade II invasive ductal carcinoma of the breast. No in situ component was seen. The sarcomatoid component was positive for S-100 protein and vimentin and negative for keratin, whereas the carcinomatous component was positive for vimentin, epithelial membrane antigen, and low molecular weight keratin. Ultrastructurally, the tumor showed features of a poorly differentiated epithelial neoplasm without evidence of chondrosarcoma. Metastases were seen in seven of ten axillary lymph nodes with exclusive epithelial component. Retrospective evaluation of the FNA smears revealed an extensive metachromatic stroma of the DIFF QUIK-stained smears. Our findings and review of the literature suggest that sarcomatoid carcinoma of the breast will be very frequently overlooked or misdiagnosed either because of the subtlety of the stroma or the predominance of the mesenchymal component. Our findings also provide supportive evidence that the use of the DIFF QUIK stain can be very helpful in the identification of the stroma in this neoplasm. The predominance of epithelium in the metastases in our case is further proof that tumors with dual differentiation such as sarcomatoid carcinomas and mixed muellerian tumors are truly carcinomas with metaplastic features.

Influence of Intralipid infusion on the elimination of phenytoin.

On the ground of clinical observations the hypothesis was tested that a changed disposition of phenytoin could bring about a shortened half-life. In vitro it appeared that the amount of phenytoin bound to the Intralipid fraction was more than twice the amount present in the plasma fraction. In an in vivo rat model it was shown that the half-life of phenytoin was indeed influenced by Intralipid administration, but contrary to clinical observations phenytoin half-life was prolonged. Some possible explanations for this discrepancy are discussed.