Diagnosis and treatment of respiratory symptoms of initially unsuspected gastroesophageal reflux in infants.

Gastroesophageal reflux (GER) in infants is most commonly thought of as repeated excessive vomiting and failure to thrive, with most infants responding favorably to medical therapy. However, GER may also manifest exclusively with a variety of respiratory symptoms that, if not detected and treated early, may lead to life-threatening complications. During the period of 1987 to 1992, 39 neonates and infants underwent Nissen fundoplication for the treatment of respiratory symptoms attributed to GER. Symptoms included apnea and bradycardia (64%), pneumonia (31%), cyanosis (28%), cough (18%), and stridor (15%). Most patients were ascribed at least one incorrect diagnosis to explain respiratory symptoms. These include apnea of prematurity (38%), bronchopulmonary dysplasia (31%), asthma (8%), and subglottic stenosis (8%). All patients underwent a variety of investigations and medical treatments without noticeable clinical improvement. These included bronchoscopy, esophagoscopy, and polysomnograms. Treatment such as antibiotics, theophylline, bronchodilators, steroids, and oxygen were directed at presumed primary respiratory disease. On the other hand, H2 blockers, metoclopramide, positioning, and thickened feeds were prescribed to treat GER without objective evidence of disease. Ultimately, GER was demonstrated by upper gastrointestinal series in 64%, pH probe in 61%, and both studies in 38%. All patients underwent Nissen fundoplication after failed attempts at medical therapy. A total of 95% of patients had resolution or substantial improvement of respiratory symptoms postoperatively. Preoperative hospitalization averaged 37.0 days, and postoperative stay averaged only 14.2 days. We present a series of patients with GER, all of whom presented with respiratory symptoms.(ABSTRACT TRUNCATED AT 250 WORDS)

Nosocomial influenza in a general hospital for indigent patients.

Surveillance of influenza virus infections was maintained on the medical and pediatric wards of a general hospital serving indigent patients of Harris County, Texas during the winter of 1987-88. Influenza A/Sichuan/2/87(H3N2) was the predominant virus during the community epidemic. Influenza A(H3N2) virus was isolated from six of 17 adult patients (35%) suspected to have nosocomial infections; two of these patients died of fulminating pneumonia. Another patient, admitted with alcoholism and hematemesis, developed pneumonia after a documented influenza infection, and the hospital stay was extended for two weeks. The paucity of virus positive community acquired infections on the medical wards suggested that the health care team was the probable source of the nosocomial infections. Community-acquired infections with influenza viruses were common on the pediatric service; however, most admissions were for acute conditions requiring brief hospital stays. Therefore, while nosocomial infections were unusual on the pediatric ward, they may have become evident after discharge. Recommendations for reducing the impact of nosocomial influenza virus infections include hospital-centred vaccination programs for both high risk patients and hospital personnel. Amantadine can be an important adjunctive therapy for reducing nosocomial spread during influenza A virus epidemics.

A comparison of plain and fluoroscopically guided radiographs in the assessment of arthroplasty of the knee.

A cadaver model was used to compare the sensitivity of fluoroscopically guided radiographs with that of plain radiographs in revealing lucent lines beneath the tibial component of an uncemented total knee arthroplasty and subsidence of the component. Fluoroscopically guided radiographs allowed accurate measurement of a lucent line that was one millimeter wide. Plain radiographs were inadequate for the detection and measurement of these lucent lines, leading to inaccuracy. Fluoroscopically guided radiographs also allowed measurement of the distance between the tibial component and radiopaque markers in the proximal part of the tibial metaphysis that was reproducible to within one-half millimeter. Plain radiographs did not provide a reproducible measurement of this distance. The relatively simple technique of fluoroscopically guided radiography is recommended to detect the presence and progression of radiolucent lines and the subsidence of uncemented tibial components after total knee arthroplasty.

A ten-year assessment of methicillin-associated side effects.

The exact incidence and importance of side effects associated with methicillin therapy in children is unknown. During the ten-year period from 1964 to 1974 approximately 3,000 Houston children who received methicillin were observed for side effects. The great majority of these patients received the drug for less than ten days; however, side effects were more common in patients receiving long-term therapy. In order to assess these side effects, experiences with 124 children who received methicillin for ten days or longer were reviewed in depth. The average dose of methicillin was 200 mg/kg/day and the average duration was 22 days. In this highly selected group there were no adverse side effects in 54.8% and only eosinophilia in an additional 13%. Leukopenia occurred in 16%, microscopic hematuria in 8%, gross hematuria in 4%, skin rash in 6%, and "drug fever" in 6%. In many instances several of these side effects occurred within a single patient so that the total number of patients with definite side effects was 39 of 124 (31.5%). The true incidence figure of overall side effects is much lower than 31.5%. Corrected incidence rates based on a conservative figure of 3,000 methicillin-treated children are less than 1.5% for all side effects. In some patients the mechanism producing the adverse reactions seemed to be chemical irritation or toxicity and in others hypersensitivity. In nine of the 39 patients follow-up studies were not optimal. In the other 30 patients all side effects were reversible.