Twenty‐two patients with intractable malignant ascites who received a peritoneovenous shunt were studied. We found that the peritoneovenous shunt functioned longer in patients whose ascitic fluid was negative for malignant cells. The median shunt survival in alive patients in the negative cytologic group was 140 days compared to 26 days in the positive cytologic group (P = 0.01). The overall survival of these patients was poor, with a median of 32 days. Median survival of patients with positive cytologic results (26 days) was significantly worse than for the cytologically negative group (80 days; P = 0.05). The incidence of tumor emboli, confirmed at autopsy, was estimated to be about 5%. Seventy‐five percent of all complications occurred in the group of patients with a positive cytologic result. We conclude that a positive ascites fluid cytologic finding is a relative contraindication to placement of a peritoneovenous shunt since this is associated with early shunt failure, postoperative coagulopathy, infection, and tumor emboli. However, since the serious complication rate is only 4% and tumor emboli rate 5%, peritoneovenous shunting in symptomatic patients with cytologically negative malignant ascites is a useful palliative procedure.

A prospective pilot study involving preoperative intraarterial chemotherapy with cisplatin in locally advanced oral and oropharyngeal carcinomas was initiated in March 1982. Twenty patients with TNM Stages III and IV disease underwent continuous intraarterial chemotherapy with cisplatin (90 mg/m2, at 1 mg/hour) and intravenous chemotherapy with methotrexate (50 mg/m2 X 1 dose). Arterial access was obtained by retrograde cannulation of the superficial temporal artery. One patient died of sepsis from leukopenia (mortality 5%). Catheter-related complications, most of them avoidable, totalled 30%. The total tumor response rate was 94%. With a median duration of follow-up of 14 months, the median survival of the group was not yet reached at 39 months. Eleven patients subsequently underwent definitive surgery and radiation with curative intent. When compared with matched historical controls, survival benefit was demonstrated: 60% versus 28% alive at 39 months (P = 0.015). Regional chemotherapy of cancers of the head and neck region is a feasible procedure with acceptable and avoidable morbidity. It should continue to be investigated in experimental protocols to improve patient survival.

No abstract