During the period from early December 1987 to late January 1988, an outbreak of epidemic keratoconjunctivitis (EKG) caused by adenovirus type 8 occurred at a university medical center eye clinic. A telephone survey of patients revealed an attack rate of 17% among patients seen during the epidemic period. A case-control study demonstrated that development of EKC was associated with exposure to pneumotonometry and to three caregivers. After controlling for exposure to pneumo-tonometry, however, only exposure to one caregiver was significantly associated with risk of the disease. On January 19, 1988, after recognition of the outbreak, the eye clinic instituted additional infection control measures. A survey of patients seen the following week showed a substantial decrease in the attack rate (1.9%). Before these control measures were implemented, clinic personnel had followed the manufacturer's recommendations and disinfected pneumotonometer tips with 70% isopropyl alcohol, fearing that other disinfectants would corrode the instrument metal. Isopropyl alcohol, shown to have limited activity against adenovirus in vitro, also was being used to “disinfect” pneumotonometer tips between uses at six other area eye clinics polled by telephone. The results of this study demonstrate the need for changes in the design and manufacture of equipment used in the eye clinic.

One thousand six-hundred and eighteen medical gloves were tested to determine whether, with current increased demands, these gloves are of high quality, i.e., free of leaks. The risk of exposure to potentially infected fluids when using leaky gloves was also estimated. Using a four-stage leak test, no significant difference was found between 64 of 790 (8.1%, range 0% to 44.4%) unsterile latex gloves and 11 of 210 (5.2%, range 1.7% to 21.7%) unsterile vinyl gloves (p=.21). Sterile surgical gloves (7 of 618, 1.13% [range 0% to 3%]) had fewer leaks compared to unsterile latex and vinyl gloves combined p<.OOOl). The safranin test was positive in 27 of 28 (96.4%) leaky gloves tested, indicating a high risk of exposure to potentially infected fluids when leaky gloves are used. Because of these findings, elements of “universal precautions” such as changing gloves after each patient contact and good handwashing after using gloves should be carefully observed. There is a need for the Food and Drug Administration to establish more stringent guidelines for manufacturing gloves and to verify compliance with these guidelines.

Arterial catheters are now commonly used to monitor blood pressure and obtain blood samples for arterial blood gas and other laboratory determinations. Stopcocks inserted into the pressure monitoring circuit have been the primary means of obtaining blood from arterial catheters. However, these stopcock systems have been associated with nosocomial contamination and bacteremias. Because of the problems of bacterial contamination and blood wasting with the stopcock sampling systems, we compared the frequency and extent of contamination of external sampling ports and the monitoring tubing fluid in stopcocks with that of a novel closed needle-sampling system (Lab-Site, Migada Ltd, Rehovot, Israel), incorporated into pressure monitoring tubing (Abbott Laboratories Inc., North Chicago, Illinois). We found that use of the novel sampling system resulted in significantly fewer episodes of internal bacterial contamination of the arterial monitoring line (7%) than did the use of a stopcock system (61%). External contamination of the sampling port was also lower in the novel system (8%) than in the stopcock system (37%). This suggests that the closed system may reduce the risk of nosocomial infections in patients requiring arterial pressure monitoring.

Methicillin-resistant Staphylococcus aureus (MRSA) has been causing a large number of serious hospital infections in some of the major hospitals in Japan. Among the precautions to prevent MRSA infection in hospitals, antisepsis or disinfection is one of the most universal. The in vitro activities of commonly used antiseptics and disinfectants against several strains of MRSA isolated in the University of Tokyo Hospital were evaluated.

Since the advent of antimicrobial therapy, bacterial meningitis has changed from a disease with an almost uniformly fatal course to one in which the majority of patients now survive. The mortality rates for the three most common etiologic agents in bacterial meningitis, Hemophilus influenzae, Neisseria meningitidis and Streptococcus pneumoniae, were 6%, 10.3%, and 26.3%, respectively, in the United States from 1978 to 1981. Morbidity, however, primarily in the form of neurologic sequelae, continues to be a problem in children and adults surviving bacterial meningitis. This article focuses on the principles and practice of therapy in bacterial meningitis, highlighting the experience with new antimicrobial agents and the utility of adjunctive therapy.

Protozoans of genus Cryptosporidium cause infections of the mucosal epithelium of a variety of animals, including humans. Infections by these coccidian parasites were once considered rare in animals. In humans, infections were thought to be the result of an opportunistic occurrence in immunosuppressed persons. Recently however, the concept of this intracellular protozoan parasite has changed markedly in that this parasite is now thought to be an important cause of widespread gastrointestinal (GI) illness commonly seen in several species, including humans. Infections in animals are common in newborn livestock and often are fatal. At least one species of this protozoan genus may cause mild to severe diarrhea in immunocompetent humans; this diarrheal illness may last from several days to a month or more. In immunosuppressed persons seen with acquired immunodeficiency syndrome (AIDS), cryptosporidiosis may cause a prolonged life-threatening diarrheal illness.” Recent descriptions of respiratory and biliary infections by Cryptosporidium species demonstrate that this parasite is not confined to the GI tract of immunosuppressed persons. Unfortunately, antiprotozoan therapy for cryptosporidiosis is ineffective. Persons with normal host defenses or those with restored immune balance are able to control the infection.

Simple random sampling of individuals from a finite population is generally regarded as the most basic of all sample designs. However, it is not commonly used for several reasons. It is often difficult and time consuming to generate a complete listing, or sampling frame, of all individuals in the population. The use of a random number table is also time consuming with a realistically large population, although computer-generated simple random samples greatly reduce this inefficiency.