Invasive eye infections due to Pseudomonas frequently lead to bacteremia in infancy, but the relationship between Pseudomonas conjunctivitis and subsequent systemic complications such as bacteremia is less well defined. To examine the importance of systemic complications of Pseudomonas conjunctivitis in patients hospitalized in a NBSCU, the courses of 18 infants in the NBSCU at Yale-New Haven Hospital (YNHH) with Pseudomonas conjunctivitis from 1986 to 1996 were reviewed. An additional 21 well-described, nonepidemic cases of Pseudomonas conjunctivitis in infants reported in the literature were also reviewed, providing 39 cases for analysis. P. aeruginosa (37) and P. cepacia(2) were identified as the responsible pathogens. The M:F ratio was 1.7:1. Birth weight ranged from 630 to 3450g (mean ± standard error; 1465± 138g) and gestational age ranged from 24 to 40wks (29.9 ± 0.8wks). Twenty-seven (69%) of the cases occurred in infants weighing < 1500g. Thirty-two infants were premature; 17 (44%) ≤ 28 wks gestation. The age of onset of conjunctivitis ranged from 2 to 77 days (17 ± 3 days). The infection was unilateral in 22 cases and bilateral in 15. Systemic complications occurred in 16 (41%) of infants and included bacteremia (13) meningitis (3), brain abscess (1) pneumonia (1), otitis externa (1), suppurative otitis media (1) and death (6). Infants who developed systemic complications associated with Pseudomonas conjunctivitis were of lower mean birth weight (1097 ± 159g vs. 1710g ± 182g, P = 0.01), lower mean gestational age (28 ± 0.9 wks vs. 31 ± 1 wks, P = 0.03) and presented at an earlier mean age of onset of Pseudomonas conjunctivitis (13± 2 days vs. 24 ± 4 days, P = 0.09) than infants who did not develop systemic complications. There were no significant differences in the presence of factors predisposing to bacteremia such as the requirement for mechanical ventilation or central vascular catheterization, or in the presence of associated complications of prematurity such as RDS, PDA, NEC or IVH. There were 6 fatalities directly related to Pseudomonas infection; all patients who died were bacteremic. The case fatality rate was 40% for infants < 1000g and 0 for infants > 1500g (P = 0.006). These data emphasize the need for early detection and systemic treatment of this infection in this population.
