Retinal periphlebitis in patients with acquired immunodeficiency syndrome with cytomegalovirus retinitis mimics acute frosted retinal periphlebitis.

Acute frosted retinal periphlebitis is an inflammatory condition of unknown origin characterized by marked perivenular infiltration in otherwise healthy patients. We encountered seven patients with acquired immunodeficiency syndrome who exhibited visual loss associated with an unusual diffuse retinal periphlebitis very similar in appearance to acute frosted retinal periphlebitis. Each patient developed a thick inflammatory infiltrate surrounding the retinal venules, creating a frosted appearance. Two cases were bilateral. All patients had areas of more typical cytomegalovirus retinitis in their involved eye(s). Five of six patients treated with ganciclovir sodium showed improvement not only of the cytomegalovirus retinitis but also of the periphlebitis. Although we do not have histopathologic evidence that cytomegalovirus was the cause of these cases of periphlebitis, we believe that periphlebitis may be a previously unrecognized finding of cytomegalovirus retinitis in patients with acquired immunodeficiency syndrome. So far, there is no evidence implicating cytomegalovirus as the cause of acute frosted retinal periphlebitis in healthy patients.

Clinical use of the insulin infusion pump in 100 patients with type I diabetes.

We investigated the efficacy of insulin-pump therapy in insulin-dependent diabetics, aged 18 to 69 years, by comparing the metabolic control achieved in 100 patients using this technique with that previously obtained by conventional insulin therapy. Patients were followed during pump therapy for as long as 15 months. Fasting and nonfasting blood glucose levels (mean +/- S.E.M.) decreased from 201 +/- 6 and 213 +/- 6 mg per deciliter (11.2 +/- 0.3 and 11.8 +/- 0.3 mmol per liter), respectively, to 158 +/- 5 and 145 +/- 3 mg per deciliter (8.77 +/- 0.3 and 8.05 +/- 0.2 mmol per liter) after one month of pump therapy (P less than 0.001). Ninety-three patients had improved blood sugar control; 71 per cent had a mean blood sugar concentration of 150 mg per deciliter (8.3 mmol per liter) or less after six months. Glycosylated hemoglobin values became normal in 44 per cent of 88 patients who had follow-up determinations. In over 500 patient-months there were four episodes of ketoacidosis and five episodes of serious hypoglycemia. Three patients abandoned pump therapy. We conclude that insulin-pump therapy is acceptable to patients and that it can be successfully applied to clinical practice and large-scale research studies.