Non-gonococcal ophthalmitis associated with erythromycin ointment prophylaxis of gonococcal ophthalmia neonatorum.

Substitution of erythromycin ointment for silver nitrate in the prophylaxis of gonococcal ophthalmia neonatorum (GON) was accompanied by eight infections in 749 (1.1%) well-born and 21 infections in 285 (7.4%) intensive care infants during an eight-month period. This was significantly higher than previous rates of ophthalmitis during the use of silver nitrate, 0.3% (5/1877) and 2.1% (19/904) for well and intensive care infants, (P less than 0.01). Multiple bacteria were isolated, polymicrobial infection occurred frequently and the pattern of bacterial isolates did not favor cross-contamination between infants. Observation suggested the frequent unintentional introduction of ungloved fingers into neonatal eyes during attempts to insert erythromycin ointment. Replacement of ophthalmic ointment with a liquid tetracycline preparation resulted in a decrease in non-gonococcal ophthalmitis to rates similar to the baseline period 8/997 (0.8%) for both nurseries. The increased number of infections appear related to the ointment vehicle, difficulty in its application and the mechanical introduction of bacteria. It is inferred that appropriate application of liquid medication also reduced the risk of inadequate prophylaxis.

An evaluation of daily bacteriologic monitoring to identify preventable episodes of catheter-associated urinary tract infection.

We evaluated the efficacy of a daily bacteriologic monitoring program for preventing symptomatic urinary tract infections (UTI) in hospitalized patients with temporary indwelling urethral catheters. We identified 99 instances in which bacteriuria was present at the time of catheter insertion among 1,140 catheterizations. Of those, 62 patients were asymptomatic and 37 patients had fever or symptoms attributable to UTI. Of the 37 symptomatic episodes, only 14 developed symptoms 24 hours or more after the first culture and might be considered potentially preventable. We also identified 76 episodes of acquired bacteriuria among 608 catheterizations that were initially nonbacteriuric in which at least two cultures were available. Of these 76 patients, 51 (67%) remained asymptomatic throughout their period of hospitalization. Of the 25 patients who developed symptomatic infections, only ten were potentially preventable. In all, only 24 symptomatic episodes among 1,140 catheterizations (2%) occurred 24 hours or more after colonization was first detected and might be considered potentially preventable. Our data suggest that routine daily bacteriologic monitoring of urine from all catheterized patients is not an efficient way to decrease the incidence of symptomatic, catheter-associated UTI.