Endophthalmitis results from hematogenous seeding of the eye in preterm infants with candidiasis. Early systemic therapy decreases the frequency of eye involvement, now only about 6% in infants with systemic candidiasis. Eye disease can occur on the first day, but is more likely with prolonged candidemia. Candida albicans and C tropicalis predominate in newborns. Indirect ophthalmoscopy will identify the characteristic unilateral or bilateral, yellow-white, fluffy, retinal or vitreal balls with frequent hemorrhage or inflammatory vitreous haze. Systemic antifungal therapy alone usually results in good visual outcome, but unresponsive or recurrent lesions are reported. Lens abscesses presenting as cataracts may be seeded before regression of the tunica vasculosa lentis, leaving an infection nidus in a subsequently avascular structure that is poorly reached by antifungal drugs. Lensectomy may be required to clear the visual field and the infection. Candidemia is also implicated in progression of retinopathy of prematurity to stage 3 or beyond, possibly due to induction of cytokines or angiogenic factors.

Objective. To determine the safety and efficacy of monthly prophylaxis with respiratory syncytial virus immune globulin, intravenous (RSV-IGIV) for reduction of the incidence of RSV-associated hospitalization.Methods. A randomized, double-blind, placebo-controlled clinical trial was conducted at 54 centers in the United States during the 1994 to 1995 RSV season. A total of 510 children with bronchopulmonary dyspIasia and/or a history of prematurity were randomized to receive either 750 mg/kg RSV-IGIV (n = 250) or placebo (1% albumin; n = 260) intravenously every 30 days. Randomized groups were well balanced at entry for demographics, RSV risk factors, and birth characteristics. Children were monitored for adverse events and for RSV-associated hospitalization from randomization through 30 days after the last infusion visit; serious adverse events were monitored for an additional 30 days. For children hospitalized with RSV, data were collected regarding the total days of RSV stay, total days of increased oxygen requirement, total days with a moderate or severe lower respiratory tract illness, and frequency and duration of intensive care unit stay and mechanical ventilation. Ninety-five percent of participants completed the protocol and 85% received a complete course of infusions.Results. The incidence of RSV hospitalization was reduced by 41% in children receiving RSV-IGIV prophylaxis; 35 (13.5%) of the children in the placebo group were hospitalized for RSV, compared with 20 (8.0%) RSV-IGIV recipients. RSV-IGIV recipients had a 53% reduction in the total number of RSV hospital days per 100 children, a 60% reduction in the number of RSV days with increased oxygen requirement, and a 54% reduction in the number of RSV hospital days with a moderate or severe lower respiratory tract illness. In addition, children receiving RSV-IGIV had a 38% reduction in hospitalization for respiratory illness of any cause and a 46% reduction in total hospital days for respiratory illness per 100 children. RSV-IGIV was safe and well tolerated, with a safety profile similar to other IGIV preparations. Between 1% to 3% of children had medically significant adverse events related to RSV-IGIV administration.Conclusions. Monthly administration of 750 mg/kg of RSV-IGIV was safe and well tolerated and was effective in reducing the incidence and total days of both RSV hospitalization and overall respiratory hospitalization in infants with a history of prematurity or bronchopulmonary dysplasia or both.

To investigate the pathogenicity of Ureaplasma urealyticum and Mycoplasma hominis in preterm infants, we conducted a study to determine (1) frequency of isolation from cerebrospinal fluid and tracheal aspirate specimens and (2) clinical outcomes and effect of erythromycin treatment in ureaplasma-colonized infants. From the cerebrospinal fluid of 920 infants, U. urealyticum was isolated from 2 (0.2%) and M. hominis from none. From tracheal aspirate specimens from 224 infants, U. urealyticum was recovered from 37 (17%) and M. hominis from 4 (2%). Demographic characteristics and clinical outcomes were compared in very low birth weight infants (< 1500 gm) who were culture-positive or -negative for U. urealyticum. Although infants with positive results were less mature than their cohorts with negative results, there were no substantive differences in clinical outcomes between the two groups. Initiation of erythromycin treatment of infants with positive ureaplasma culture results at a mean age of 16.4 days did not appear to alter the clinical outcome. We conclude that in preterm infants (1) infection of the cerebrospinal fluid by U. urealyticum is infrequent, (2) ureaplasma organisms are frequently present in tracheal aspirate specimens but do not appear to be related to the presence or the subsequent development of respiratory disease, and (3) initiation of erythromycin treatment at 1 to 3 weeks of age does not alter the clinical course.

The current challenge of neonatal candidiasis arises from the increased survival of very low birthweight infants. It is important to understand those factors affecting normal colonization of the neonate, concomitant with those factors that predispose to fungal invasiveness. Disseminated candidiasis may present with subtle signs and symptoms, but has the potential for a wide variety of organ system involvement. Early initiation of aggressive therapy, with careful monitoring, can lead to a successful outcome.

To determine the pharmacokinetics of amphotericin B and 5-fluorocytosine in neonates, we measured serum concentrations at first dose and after 5 days of therapy by high-performance liquid chromatography in 13 neonates (mean birth weight 1.2 +/- 0.8 kg). The dose of amphotericin B was serially increased from 0.1 to 0.5 mg/kg/day in 10 infants but was decreased from 0.8 to 1.0 to 0.5 mg/kg/day in three infants. Amphotericin B concentrations were not detectable in infants receiving 0.1 mg/kg/day. Amphotericin B cerebrospinal fluid concentrations were 40% to 90% of serum values obtained simultaneously. Serum concentrations after oral administration of 5-fluorocytosine (dose 25 to 100 mg/kg/day) were detectable in all infants. We found extreme interindividual variability for the half-life, volume of distribution, and clearance for both drugs. Four infants had minimal elimination for both drugs between doses, a finding that correlates with rises in serum creatinine (greater than 0.4 mg/dl, 40 mumol/L) and blood urea nitrogen (greater than 10 mg/dl, 3.6 mmol/L). We recommend that the dose of amphotericin B given on the first day of treatment be greater than the usual testing dose of 0.1 mg/kg/day. We also recommend an initial 24-hour dosing interval for amphotericin B and 5-fluorocytosine. Serum drug concentrations may need to be monitored in high-risk, low birth weight infants.

Granulocyte transfusions are increasingly being used as therapy for newborns with sepsis and neutropenia. We injected either group B Streptococcus or phosphate-buffered saline solution intraperitoneally into adult and newborn rats. Human granulocytes, labeled with chromium 51, were transfused seven hours later. When the newborn rats were killed 13 to 19 hours after injection, they had 10(2) to 10(6) cfu/gm Streptococcus organisms in both lung and brain. Only one third of the adult rats had 10(2) to 10(4) cfu/gm Streptococcus organisms in either lung or brain. A greater proportion of the transfused granulocytes was present in lung and brain tissue of newborn rats, compared with adult rats (p less than 0.05), irrespective of infection. Granulocyte transfusion did not change the peripheral blood leukocyte count in adult rats but increased the count in newborn rats (p less than 0.05). The immature myeloid pool in the bone marrow of adult rats increased significantly with either infection or transfusion (p less than 0.01). The immature pool in newborn rats increased significantly only with infection (p greater than 0.001), although the combination of infection and transfusion also had a significant effect on the pool (p less than 0.01). Infection and both infection and transfusion, but not transfusion alone, significantly affected the mature myeloid bone marrow pool in adult and newborn rats (p less than 0.001). The depletion of the mature myeloid elements of the bone marrow in response to infection was dramatic in neonatal rats, compared with that in adult rats. Both transfused granulocytes and hematogenously spread streptococci lodge in the brains and lungs of neonatal rats more effectively than in those of adult rats.

In the fetus, hematopoietic stem cells originate in the yolk sac and are believed to be transferred to all other hematopoietic organs via the circulation. In humans, the time course of this transfer has not been systematically evaluated in viable premature infants. We examined the cord blood of 13 preterm (25 to 36 weeks of gestation) and 10 term (38 to 42 weeks of gestation) infants for pluripotent (mixed colony-forming unit-granulocyte, erythrocyte, macrophage, megakaryocyte), erythroid (burst-forming unit-erythroid, colony-forming unit-erythroid) and myeloid (colony-forming unit-granulocyte, macrophage) stem cells. A gestational age-dependent decrease in all lineages of circulating hematopoietic stem cells was noted (p less than 0.001). By 34 weeks of gestation, preterm infant cord blood had a similar concentration of circulating stem cells compared with that of term infants. This gestational age-dependent decrease in hematopoietic stem cells of all lineages supports the hypothesis of a blood-borne transfer of hematopoiesis that appears largely complete by 34 weeks of gestation. Infants born after less than 32 weeks of gestation have high levels of circulating hematopoietic stem cells that may reflect the active transfer of hematopoiesis from liver to bone marrow.

To evaluate the use of intravenously administered immune globulin (IVIG) for prevention of sepsis in preterm infants, we administered IVIG in a protocol designed to maintain a therapeutic serum "target level" of 700 mg/dl. The 200 patients who were eligible for the study (600 to 2000 gm birth weight) were monitored throughout their initial hospitalization. Of these, 115 patients were randomly assigned in a double-blind, controlled trial to treatment and placebo groups. The remaining 85 infants were not randomly assigned to a group, by parental request, but were followed and analyzed separately. In one patient who received IVIG, transient tachycardia and a decrease in blood pressure developed during an infusion; resolution occurred promptly after the infusion was discontinued. No persistent hepatic or renal abnormalities were noted in either the IVIG- or the placebo-treated group. There were seven episodes of sepsis in the placebo group and nine in the group whose parents refused consent to the study. No infant who received IVIG acquired nosocomial sepsis (p less than 0.01). All patients in the placebo group in whom sepsis developed had serum IgG levels less than 400 mg/dl at the time sepsis developed. Serum IgG levels were maintained near 700 mg/dl in patients who received IVIG. These data indicate that administration of sufficient IVIG to maintain target serum IgG levels throughout hospitalization may decrease the incidence of nosocomial sepsis in preterm infants.

Neutropenia, defined as an absolute neutrophil count that falls below 2.0 x 10(9)/L, is being identified more frequently in the newborn intensive care unit and significantly influences clinical decisions regarding therapy. We prospectively identified 119 episodes of neutropenia in 87 infants (6% of admissions). Less than half of the episodes could be attributed to infections. The majority of noninfectious neutropenia episodes were related to specific perinatal events or were of unknown cause. Infants weighing less than 2500 g were more likely to have neutropenia than term infants (13% vs 3%, respectively) and less likely to have neutropenia related to bacterial infections. Short-term survival (89% vs 95%) and long-term survival (74% vs 77%) were not different in infants with infectious diseases compared with those with noninfectious diseases. Mortality was highly correlated with the need for assisted ventilation (20%) or with an absolute neutrophil count of 0.5 x 10(9)/L (24%). We conclude that the cause of neutropenia and the clinical condition must be carefully evaluated before instituting aggressive therapy for infection.

The cutaneous manifestations of 18 infants treated for systemic candidiasis during a 3 3/4-year period were examined. Eight infants, with a mean birth weight of 712 +/- 161 g, had a diffuse burn-like dermatitis, usually within the first three days of life. Candida pseudohyphae were identifiable on skin scrapings. A history of a maternal cerclage or intrauterine device complicated by chorioamnionitis was common. A delay in diagnosis or therapy resulted in mortality, whereas promptly treated infants survived. Nine additional infants had monilial diaper rashes, which spread to the trunk and extremities in four infants. These infants were older at the onset of the dermatitis, and all survived the systemic infection. Systemic candidiasis without any cutaneous involvement developed in only one infant. Candidiasis should be more frequently considered, and prompt systemic therapy should be instituted when cutaneous candidiasis occurs within the first few days of life in infants who weigh less than 1,500 g.

There is tremendous potential for immunotherapy of neonatal sepsis, considering the continued high mortality of the disease. Current studies, particularly of intravenous immunoglobulin, have had encouraging outcomes. However, much more data need to be accumulated before clinical application.

In Reply.— The comments raised by Baley et al are well taken and bring up one of the most important variables in the role of granulocyte transfusion in neonatal sepsis. Most of the studies to date have involved small numbers of patients in single institutions. This has prevented us from accumulating a large enough population of patients to adequately assess the role of either buffy coat transfusions or leukapheresed transfusions in this clinical setting. My comments in the commentary suggested that, with only eight patients in a group of preterm infants with presumed sepsis and only five patients with presumed sepsis with birth weights greater than 1,500 g, the numbers are essentially too small to determine any statistical significance between the role of buffy coat transfusions or supportive care.

ABSTRACT: Because increased complement receptor expression is necessary for optimal function of adult neutrophils, we tested the hypothesis that the increased susceptibility of neonates to infection might be due to an impaired ability of neonatal neutrophils to increase expression of complement receptors in response to chemotactic stimuli. We used monoclonal antibodies and flow cytometry to compare surface expression of the receptors for the complement components C3b (CR1) and C3bi (CR3) on adult and neonatal cord blood neutrophils (PMNs). We also compared receptor expression on PMNs from infants delivered by cesarean section without labor versus infants delivered vaginally. Expression of both CR1 and CR3 was minimal on resting adult and neonatal PMNs maintained at 0° C. There was a modest increase in expression of both receptors when PMNs were warmed to 37° C. This increase was similar on adult and neonatal cells, both unfractionated in whole blood and after isolation with Percoll density centrifugation, with one exception. Expression of CR1 was greater on isolated PMNs from vaginally delivered infants versus adults when the cells were warmed to 37° C. This difference was not observed with cells from infants delivered by cesarean section without labor, suggesting this modest increase in receptor expression may be due to factors associated with labor. When isolated cells were stimulated with either N-formyl-methionyl-Ieucyl-phenylalanine or zymosan-activated serum, expression of CR1 increased to the same extent in both neonatal and adult PMNs. In contrast, maximal CR3 expression on cord PMNs stimulated with N-formyl-methionyl-leucyl-phenylalanine or zymosan-activated serum was only 75% of the adult values. This was significantly less than CR3 expression in adult PMNs across a broad range of N-formylmethionyl- Ieucyl-phenylalanine concentrations. Maximal stimulated expression of both receptors was the same in PMNs from infants delivered by cesarean section without labor and from infants delivered vaginally. The diminished CR3 expression on cord PMNs was not due to decreased responsiveness to N-formyl-methionyl-leucyl-phenylalanine per se since deficient CR3 expression was observed at all concentrations tested and also when zymosan-activated serum was the stimulus. Our observations of impaired expression of CR3, which is necessary for adherence and directed migration, may explain some of the impairment of these functions in neonatal neutrophils.

Neonatal sepsis, accompanied by neutropenia, is associated with a high mortality. To determine whether granulocyte transfusions improve the survival of critically ill neutropenic infants, we prospectively randomized 25 infants to transfusion and nontransfusion groups, matching for birth weight (less than or equal to 1,500 g or greater than 1,500 g). Infants with necrotizing enterocolitis were randomized separately. Neutropenia was established by two successive absolute neutrophil counts less than or equal to 1,500 cells prior to randomization. The transfusion (n = 12) and nontransfusion (n = 13) groups did not differ with respect to clinical or hematologic characteristics. In 23 of 25, bone marrow aspirations were performed to determine the percentage of neutrophil storage pool. Granulocyte transfusions of buffy coats from single units of whole blood (0.1 to 0.9 X 10(9) polymorphonuclear leukocytes per kilogram) were given daily until the absolute neutrophil count increased to more than 1,500/microL. Only five infants, mostly those with necrotizing enterocolitis, required more than one transfusion. A circulating immature to total neutrophil ratio (I:T) greater than or equal to 0.80 was not predictive of an infant with a neutrophil storage pool less than or equal to 7%, and neither an I:T less than 0.80 nor a neutrophil storage pool greater than 7% were predictive of survival. Granulocyte transfusions did not improve survival when either comparing the whole group, those 17 infants with cultures positive for bacteria or viruses, the 19 infants with a circulating I:T greater than or equal to 0.80, or the nine infants with a neutrophil storage pool less than or equal to 7%.(ABSTRACT TRUNCATED AT 250 WORDS)

In the neonate, fungal infections result in significant morbidity and mortality. For very low birth weight (less than 1,500 g) infants, we prospectively determined the fungal colonization rate to be 26.7%. In one third of infants with fungal colonies, mucocutaneous candidiasis developed, and in 7.7%, systemic disease developed. Two thirds of the infants had colonies in the first week of life. This colonization was probably acquired during labor and delivery, because those infants who had colonization were more often delivered vaginally than by cesarean section. Early colonization, commonly from the gastrointestinal or respiratory tract, featured Candida albicans and Candida tropicalis. Late colonization, occurring after 2 weeks of life (15.0% of patients), was more likely to be cutaneous and was associated with either Candida parapsilosis or such poor growth that the organism could not be identified. Infants with colonization only rarely had budding yeasts (6.1%), whereas more than half of the infants with either a urinalysis showing budding yeasts or a urine culture growing fungi had invasive disease. Fungal contamination was not found on either thoracotomy tubes or catheter tips. In the low birth weight infant, fungal colonization represents a significant risk factor for cutaneous or systemic candidiasis in these infants.