Surgical selection of infants with gastroesophageal reflux.

Gastroesophageal reflux (GER) can produce debilitating and even fatal disease in infants and children. GER is common in infancy, and most GER subsides with time or with postural and dietary therapy. Operation is justified only to control effectively by nonoperative methods. The clinical history and the patient's response to medical treatment remain the most important factors in our decision for or against operation. As in adults, esophagoscopy and esophageal biopsy are useful to document the presence of reflux in older children, but they seldom contribute to the decision for operation in infants. In our hands, gastric scintiscan has provided useful qualitative data on reflux-induced respiratory symptoms and quantitative data on gastric emptying. Esophageal pH monitoring is more quantitative for reflux evaluation and allows analysis and clinical correlations with reflux patterns. We have not used the reflux patterns to predict outcome or response to treatment in the individual patient. A prolonged average duration of reflux during sleep does appear to increase the probability that a patient with respiratory symptoms will respond to operative control of reflux. Our presently imprecise technique for patient selection, however, must not cause us to withhold operation from needy infants. For the infant who does not grow or who aspirates on a reflux board, or who requires prolonged hospitalization and for the preemie who aspirates at every extubation attempts or requires long-term nasojejunal feedings to prevent aspiration, we think antireflux surgery is appropriate, humane, and cost-effective. We see no excuse for persisting with ineffective management of a process that may result in stunting, chronic illness, persistent pain, esophageal scarring, or even respiratory death.

Splenectomy for immune thrombocytopenic purpura.

Of 481 splenectomies performed at the University of Utah, Salt Lake City, 78 (16.2%) were for immune (idiopathic) thrombocytopenic purpura (ITP). The mean platelet count prior to therapy was 13,800/cu mm (range, 500 to 80,000/cu mm). All but two patients were initially treated with corticosteroids, and 58.2% responded with an increase in platelets (mean, 77,900/cu mm). The indications for splenectomy included (1) failure to respond to steroids (33.8%); (2) inability to taper steroids (52.1%); (3) recurrent ITP (5.6%); and (4) miscellaneous (8.5%). There was one death following splenectomy, and the postoperative morbidity was 14.1%. Complete remission occurred in 77.3%, and 84.8% were judged to have benefited from the procedure. The remission rate following splenectomy in patients who responded to preoperative steroids was 93.2%, whereas improvement fell to 68.3% in patients who failed to respond to steroids. This study confirms the benefit of splenectomy for ITP and demonstrates a predictive correlation with response to preoperative preparation with corticosteroids in these patients.