Flavimonas oryzihabitans (Pseudomonas oryzihabitans; CDC group Ve-2) bacteremia in the immunocompromised host.

Flavimonas oryzihabitans, known previously as Pseudomonas oryzihabitans, and a member of the Centers for Disease Control group Ve-2, is a gram-negative organism that has rarely been implicated as a human pathogen. Flavimonas oryzihabitans appears to be a soil and saprophytic organism that survives in moist environments and is indigenous to rice paddles. To our knowledge, only seven cases of human infection caused by this organism have been reported; they involved four patients with bacteremia and three patients with peritonitis who were receiving continuous ambulatory peritoneal dialysis. In this report, we describe three immunocompromised patients with catheter-associated bacteremia: a patient with cancer, a patient with acquired immunodeficiency syndrome, and a patient with sickle cell disease. There is emerging clinical evidence that F oryzihabitans should be recognized as an organism that is capable of causing human disease, particularly in immunocompromised patients and with the increased usage of permanent catheters.

Incidence of Candida parapsilosis colonization in an intensive care nursery population and its association with invasive fungal disease.

We have studied 82 consecutive intensive care nursery admissions to determine rates of colonization and incidence of fungal sepsis. Cultures were obtained from stool, gastric aspirate and skin at three different times. Infants studied ranged in gestational age from 23 to 38 weeks (mean +/- SEM 29 +/- 0.4 weeks). Nineteen percent of all infants were colonized with Candida sp.; stools were more frequently culture-positive than skin or gastric aspirates. Colonized infants began enteral feeds at a later time compared with noncolonized neonates. Five of the study infants developed fungal sepsis. One had congenital Candida albicans sepsis and died at 10 days of age; the other four had Candida parapsilosis sepsis and survived. The development of C. parapsilosis sepsis was significantly associated with gastrointestinal colonization. Our results suggest that early initiation of enteral feeds decreases gastrointestinal colonization with C. parapsilosis. Gastrointestinal colonization was strongly associated with the subsequent development of C. parapsilosis sepsis in this group of high risk neonates.

Serum gentamicin assay by a radiometric procedure.

A new radiometric (BACTEC) microbiologic procedure, using a 14C-urea substrate and a Proteus species culture, was compared with three microbiologic assays and a radioimmunoassay (RIA) method for determination of gentamicin levels in serum. The non-radiometric microbiologic assays did not differ significantly from the RIA assay, but the BACTEC method showed significant differences with specimens containing greater than 4 microgram/ml gentamicin. Overall, the BACTEC assay was found to be simple to run, fast and reproducible. At concentrations of gentamicin less than 4 microgram/ml, it was just as accurate as were the microbiologic assays. However, at concentrations greater than 4 microgram/ml, the BACTEC assay read consistently less than RIA and the other assays. Because of the BACTEC assay's high cost per single test, it cannot approach the utility of the rapid, easy, reliable, and comparatively inexpensive microbiologic assays. The BACTEC assay is, however, a useful alternative to the RIA method in laboratories that already have radiometric equipment and in which batching of serum samples for gentamicin assay is necessary.

Bacterial contaminants of collected and frozen human milk used in an intensive care nursery.

BACKGROUND: Use of human milk for preterm and high-risk neonates conveys many potential benefits but also poses practical difficulties. This prospective study examined the prevalence and degree of bacterial contamination of human milk used in the intensive care nursery. METHODS: One hundred eight milk samples collected from 40 mothers were tested for contamination. Samples from mothers whose milk showed a high degree of contamination were retested after counseling on collection methods. RESULTS: Only 12.5% of the samples showed no bacterial growth. Of the contaminated samples, 38% contained > 30,000 colony-forming units/ml. The most common contaminants were Staphylococcus epidermidis (82%) and Acinetobacter (9%), but other contaminants were also encountered. CONCLUSIONS: There were not statistically identifiable common characteristics of mothers whose milk showed abundant bacterial contamination. Only 30% of these mothers showed improvement in the degree of contamination after counseling regarding techniques of milk collection.

Aerobes isolated in fecal microflora of infants in the intensive care nursery: relationship to human milk use and systemic sepsis.

BACKGROUND: Staphylococcus epidermidis is a leading cause of nosocomial sepsis in the intensive care nursery. The relationship between rates of gastrointestinal colonization and the incidence of systemic sepsis with S. epidermidis in hospitalized neonates is under investigation. METHODS: In this study, we enrolled 46 infants consecutively admitted to the intensive care nursery (mean +/- standard deviation, birth weight 1300 +/- 337 gm, gestational age 29.4 +/- 2.2 weeks). At the time of enrollment, infants had been fed enterally for at least 1 week (28 were fed formula and 18 received their own mothers' frozen milk). Stool samples were collected when infants were 2 to 3 weeks of age (16.3 +/- 7.4 days). RESULTS: Aerobic stool flora were present in 65% of all patients. Human milk use was associated with a significant increase in the presence of aerobic stool flora (78% vs 46%, p = 0.035), as well as more frequent isolation of S. epidermidis. The incidence of S. epidermidis sepsis was 33% in those infants whose stool specimens grew S. epidermidis and 3.5% in those whose stool specimens did not (p < 0.01). CONCLUSIONS: These findings suggest the gastrointestinal tract as a possible site of entry for S. epidermidis in the hospitalized preterm infant. In addition, frozen human milk may be a vehicle for gastrointestinal S. epidermidis colonization.

Comparison of three techniques for concentrating positive BACTEC 13A bottles for mycobacterial DNA probe analysis.

Three methods of concentrating 1 ml aliquots from BACTEC 13A bottles containing patient blood samples were evaluated for testing with the Gen-Probe Rapid Diagnostic System for Mycobacteria avium complex: 1. using no reagents, 2. using both lysing and wash reagents; and 3. using lysing reagent only. Aliquots from 13As containing human blood and seeded with eight mycobacterial species were also concentrated directly and using both reagents. Results for samples containing M. avium were as follows: 1. using the direct concentration technique 34 of 47 samples (72%) gave unequivocally positive results; 2. 43 of 47 samples (92%) concentrated using both reagents gave positive results; 3. the technique using lysing reagent only was not found useful. There were no false positives with any of the seeded specimens. We were also able to define the minimum Growth Index necessary to ensure un-equivocally positive results for each concentration technique. For those samples containing M. avium these values were 42 for the technique using both reagents and 86 for the direct technique. Direct or reagent concentration of 13A aliquots for testing with Gen-Probe DNA probes provides a rapid, sensitive, and specific means for the identification of M. avium complex bacteremia.

Mycobacteremia caused by simultaneous infection with Mycobacterium avium and Mycobacterium intracellulare detected by analysis of a BACTEC 13A bottle with the Gen-Probe kit.

Simultaneous infection with Mycobacterium avium and Mycobacterium intracellulare in an AIDS patient was suspected after direct analysis of two BACTEC 13A blood cultures with the Gen-Probe kit for M. avium complex. A mixed infection was confirmed by evaluating isolated colonies. The Gen-Probe kit may provide a simple technique for detecting mixed M. avium-M. intracellulare infections.

Optimization for detection of cytomegalovirus by the polymerase chain reaction (PCR) in clinical samples.

PCR is 100 times more sensitive than traditional tube culture for detecting cytomegalovirus (CMV) but may require up to 12 reactions per specimen (Sandin et al., 1991). In order to make the assay practical for use in a clinical laboratory the procedure used to detect CMV must be simplified. In this study, the effect of reducing the number of reactions per specimen on sensitivity and specificity of the PCR assay was evaluated. 53 residual samples from specimens processed for CMV by shell vial assay/routine tube tissue culture (SVA/TTC) were analyzed by PCR. The residual samples were separated into a supernatant and pellet fractions, then tested for CMV with primers to the immediate early (IEP) and late protein (LP) genes using a nested procedure. To exclude false negatives due to the presence of inhibitors in the sample fractions, all fractions were tested for the presence of the human myosin heavy chain gene also using a nested procedure. SVA/TTC had a sensitivity and specificity of 52/96% in comparison to PCR when data from all 12 PCR reactions was considered. However, high sensitivity and specificity were retained when only the data of the IEP primers with two samples were considered. The results from examining only the 1:10 dilution of pellet and the undiluted supernatant by PCR provided a 60% increase in sensitivity over SVA/TTC, high specificity and a clinically feasible assay.

Comparison of sensitivity for human cytomegalovirus of the polymerase chain reaction, traditional tube culture and shell vial assay by sequential dilutions of infected cell lines.

Although traditional tube culture (TTC) is still considered by many as the 'gold standard' for the laboratory diagnosis of human cytomegalovirus (HCMV), the shell vial assay (SVA) offers greater speed of detection. This technique utilizes immunofluorescence (IF) to detect early or immediate early nuclear antigens (IEA). The detection capabilities of these two tests were compared with the polymerase chain reaction (PCR), a technique that amplifies enzymatically selected DNA target sequences. Serial dilutions of crude culture harvests from 2 HCMV strains, Towne and a clinical urine isolate, were made up to 1:1 000,000. Ten-microliters aliquots of the original sample and each dilution were tested by PCR, TTC and SVA. For PCR, the nested-primer approach was used. Outer primers delimited a 721-bp sequence contained within the 2nd to 4th exons of the immediate-early protein. Inner nest primers delimited a 167-bp sequence in the third exon, detected by a 32P-labelled probe. The results show that: (1) control samples which contained all PCR reagents but no DNA were uniformly negative; (2) radiolabelled-probe detection (RPD) of PCR products is, on average, 100 x more sensitive than detection by ethidium bromide; (3) PCR is, on average, 100 x more sensitive than evaluation of cytopathic effect (CPE) in the TTC; (4) the predictive value of a negative SVA result is low compared to PCR.

Use of human milk in the intensive care nursery decreases the incidence of nosocomial sepsis.

OBJECTIVES: This study compares stool colonization and incidence of sepsis in human milk-fed (HM) and formula-fed (FF) intensive care nursery (ICN) patients. STUDY DESIGN: Infants recruited prospectively were fed HM based on the decision of their mothers (59 HM and 114 FF). The incidence of sepsis was determined during the following three intervals: period 1, first 10 days of life; period 2, 11 to 24 days; and period 3, 25 to 38 days. RESULTS: Frequency of Escherichia coli and Enterococcus sp. colonization was increased in HM infants. The incidence of sepsis was 9.5% in period 1 (5% in HM vs 10% in FF), 17.2% in period 2 (9% in HM vs 20% in FF), and 12.5% in period 3 (0% in HM vs 15% in FF). The odds ratio for sepsis in HM infants was 0.4, the 95% limits 0.15 to 0.95, p = 0.04. CONCLUSIONS: HM feeding in the ICN has a protective effect against nosocomial sepsis, which is unrelated to its influence on gastrointestinal (GI) flora.

Antibiotic prophylaxis in low-risk cesarean section.

This prospective study was undertaken in an effort to evaluate the role of systemic antibiotic prophylaxis in elective abdominal delivery. Eighty-two patients undergoing elective cesarean section who were not in labor and who did not have ruptured membranes were assigned on a randomized, double-blind basis to receive a three-dose perioperative course of either placebo or ampicillin. Postoperatively, patients were evaluated for the development of infection-related complications. Patients in the antibiotic group experienced less febrile morbidity, had lower fever indices and developed fewer operative-site infections than did patients in the control group. No patient in either group, however, developed a potentially life-threatening infection, and all infected patients responded promptly to parenteral antibiotic therapy. Because of the limited morbidity associated with elective cesarean section in this patient population, it is concluded that the theoretical risks of antibiotic prophylaxis outweigh the expected benefits.

A comparative study of two antibiotic regimens for the treatment of operative site infections.

This prospective study was designed to compare the relative efficacy of two antibiotic regimens for the treatment of operative site infections subsequent to pelvic operations. Patients with endomyoparametritis after delivery or pelvic cellulitis subsequent to hysterectomy were randomized to treatment with the combination of penicillin-gentamicin or the single agent cefoxitin. Seventeen of the 26 patients (65%) with endomyoparametritis who were treated with penicillin-gentamicin were cured by antibiotic therapy alone, in comparison to 15 of 23 (65%) patients treated with cefoxitin. Fifty-eight percent of the patients with pelvic cellulitis who were treated with penicillin-gentamicin responded favorably, in comparison to 50% of the patients treated with cefoxitin. None of these differences was statistically significant. In this study, neither antibiotic regimen provided satisfactory initial treatment for surgically induced soft tissue pelvic infection. Moreover, 11 of the 28 patients with treatment failures (40%) developed serious sequelae of their primary infection.

Critical pathways: design, implementation, and evaluation.

As David M. Eddy, M.D., Ph.D., Senior Advisor for Health Policy and Management to Southern California Kaiser Permanente, discusses in his excellent book, Clinical Decision Making: From Theory to Practice (1), we are now in a time where we must rethink what we are doing and how we are doing it. Substantial variations among physicians in almost every aspect of the diagnostic process have been documented repeatedly, and these variations appear to cause patients to be treated differently. Eddy says these variations are not the fault of physicians or anyone else because of the complexity of the medical decision process. Nonetheless, the cost and quality of health care have suffered as a result. Numerous articles and individuals such as Jay McDonald, M.D., Professor and Chair of the Department of Pathology at the University at the University of Alabama at Birmingham Medical Center, also have highlighted these variables in practice patterns and their consequences (2). Dr. Eddy, Dr. McDonald, Michael G. Bissell, M.D., Ph.D., Director, Clinical Pathology, Allegheny General Hospital, Pittsburgh, Pennsylvania, and other leaders in the field have stressed the need for more standardization of health care; clinical decisions concerning diagnostic testing and therapeutic choices must be based on scientific evidence that demonstrates the practice being used is truly effective (1-6). This evidence, as well as other parameters discussed below, are known as outcomes. As expressed by Dr. McDonald, "there is a transition that is going on from doing what seems best to doing what one knows is best" (2). Practice guidelines and critical pathways now are seen by many as one solution to providing more standardization of health care and to meeting the demands of the rapidly changing medical environment for simultaneously increasing the quality of care while decreasing the costs.

Prevalence and toxigenicity of Clostridium difficile isolates in fecal microflora of preterm infants in the intensive care nursery.

Fecal isolates of Clostridium difficile and its toxin B were followed prospectively in 50 preterm intensive care nursery (ICN) patients. The first stool specimen was obtained after 1 week of enteral feeding, at 15 +/- 1 days of life, and 2 more specimens were collected at 2-week intervals, 24 +/- 1 and 32 +/- 2 days of life. The stools were cultured for C. difficile, and tested for C. difficile toxin B. In the first specimen 15% of stools grew C. difficile. In the second specimen C. difficile isolation rates increased to 33% and plateaued. Toxin B was detected in 71, 93 and 100% of culture-positive stools in the first, second, and third specimens, respectively. C. difficile colonization was not associated with a higher incidence of necrotizing enterocolitis or diarrhea, and using precollected, frozen human milk did not protect from C. difficile colonization.

Staphylococcus epidermidis sepsis in the intensive care nursery: a characterization of risk associations in infants < 1,000 g.

We undertook to determine Staphylococcus epidermidis colonization patterns and risks of sepsis in a cohort of 82 consecutive intensive care nursery admissions (birth weight 1,285 +/- 57 g), with 24 infants weighing < 1,000 g at birth. Colonization was determined by skin and stool cultures collected at three time points. Multiple neonatal variables were classified into three intervals preceding the time of sample collection including the occurrence of S. epidermidis sepsis. 16 infants (20%) developed S. epidermidis sepsis. 81% of these episodes occurred in infants < 1,000 g. Skin colonization was nearly universal at all sampling points. Rectal colonization was 63.6% initially (10 +/- 0.4 days), then declined to 32% by the third sample (37 +/- 0.4 days). Neither prevalence of skin nor rectal colonization influenced the incidence of sepsis significantly. Statistically significant risk associations for sepsis for the entire intensive care nursery population included: low birth weight, gestational age, presence of a central line, and delayed feeding. For infants < 1,000 g the occurrence of sepsis during the second study time period (54% of the episodes) was associated with preceding steroid exposure. During the third study time period, birth weight and delayed attainment of full enteral feeds showed a statistically significant association with sepsis. We conclude that infants < 1,000 g are at an increased risk of S. epidermidis sepsis. Extreme immaturity, steroid therapy, and prolonged hyperalimentation are all significant risk associations.

Comparison of BACTEC 13A medium and Du Pont isolator for detection of mycobacteremia.

BACTEC 13A medium (Johnston Laboratories, Towson, Md.) was compared with Isolator (Du Pont Co., Wilmington, Del.) concentrate for sensitivity, speed, and technical ease of isolation of mycobacteria from paired patient blood samples. Of 72 positive cultures, 63 were positive by both systems. Five positive cultures were detected by BACTEC 13A medium alone, and four were detected by Isolator alone. The median numbers of days to positivity were 12 for BACTEC 13A medium and 14 for Isolator concentrate. BACTEC 13A medium has an advantage over the Isolator in requiring less laboratory manipulation of the specimen but has the disadvantages of not providing isolated colonies or quantitation of organisms. Some technical problems with contamination in both systems are also discussed.

Nested polymerase chain reaction assay for the detection of cytomegalovirus overcomes false positives caused by contamination with fragmented DNA.

The polymerase chain reaction (PCR) technique offers a promising alternative to tissue culture for the rapid and sensitive detection of cytomegalovirus (CMV) infection. However, high levels of background amplification detected in samples containing water but no DNA make interpretation of borderline positive samples extremely difficult and reduce the sensitivity of the assay. The signal from amplification of water or positive samples can be eliminated by DNase treatment, but not by filtration through anisotropic membrane, autoclaving, or ultraviolet irradiation. A lag time of 10 to 12 cycles is observed before the reactions with water will show product formation by liquid hybridization detection. The use of nested PCR eliminates the background and, in serial dilutions of a positive sample, shows a 500- to 1000-fold increase in sensitivity by liquid hybridization detection. We suggest that the background signal is arising from small fragments of DNA, which may be produced by autoclaving viral culture material. Such fragments would escape filtration, and overlapping fragments of DNA can prime one another to form complete mosaic sequences that will then amplify. Nested PCR, appropriately controlled for the number of cycles at each step, should successfully overcome such false positives caused by fragmented DNA, no matter if the contamination occurs at the collection site, in processing, or at the facility performing the test.