Nosocomial respiratory tract infections have occasionally been associated with contaminated respiratory therapy devices and techniques. In the past two years, our hospital purchased disposable saline squeeze vials for use during suctioning of intubated patients. These vials have a cap which must be flipped or twisted-off by the user before the contents can be instilled into the patient's respiratory tract. We observed use of this item in our intensive care units (ICU) and studied the potential for contamination. ICU nurses use ungloved hand(s) to flip or twist-off squeeze vial cups resulting in skin contact with the vial opening. Cultures of nurses' hands and squeeze vial contents were obtained in the ICU where 24 nurses opened 92 vials in the manner previously described. Twenty-three percent of vial contents were contaminated and cultures revealed S. epidermidis, S. aureus, Streptococcus viridans and enterococcus. On six occasions, organisms recovered from vial contents were believed to be identical to organisms isolated from the hands of the nurse who opened the vial. When nurses used care in removing the saline vial cap, cultures of vial contents were sterile.

The effect of daily treatments with 70% ethanol and with 0.5% Chlorhexidine in 70% ethanol, respectively, on navel colonization in newborns has been studied in 624 infants in two maternity wards during a four-month period. Staphylococcus aureus, group B and G streptococci, E. coli and anaerobes were significantly less often isolated from newborns whose umbilical cord was treated with Chlorhexidine in ethanol than from those treated with ethanol only. Staphylococcus epidermidis and enterococci, on the other hand, were significantly more often isolated from the chlorhexidine-ethanol than from the ethanol group. More infants without colonization of the umbilicus on their fourth day of life were found in the Chlorhexidine in ethanol than in the ethanol group. The infants in the ethanol group were colonized with significantly more bacterial species than the infants in the other group. Signs of dissemination of group B and group G streptococci between babies were seen more often in the group of infants treated with ethanol than in the one treated with chlorhexidine-ethanol.

In the neonatal intensive care unit (NICU) of one hospital, 16 infants became colonized or infected with multiply-resistant Escherichia coli (MR-E.coli) over an 8-month period. Isolates were obtained from blood, urine, and sputum of three patients and from rectal surveillance cultures of 13 patients. The one patient with the blood isolate died. A matched case-control study identified continuous feeding (nine of 16 cases vs. one of 16 controls, p ≤ 0.001) and receipt of aminoglycosides (p ≤ 0.03) as risk factors. For case-babies not exposed to continuous feeding, duration of bolus feeding was significantly greater than for their controls (cases, 22 days; controls, 7 days; p ≤ 0.02). All 16 isolates were the same serotype and were resistant to amikacin, tobramycin, kanamycin, and gentamicin. The epidemiologic investigation suggested that MR-E. coli may have spread from person-to-person on the hands of personnel and that MR-E. coli persisted in the NICU for 8 months until effective control measures were instituted.

A retrospective review of microbiology records revealed 19 documented episodes of M. morganii bacteremia in 18 patients at a Veterans Administration hospital during a 5.5 year period. Thirteen of 19 bacteremias were related to nosocomial infections; 11 of the 13 nosocomial bacteremias occurred in surgical patients. Nine of the 13 patients with nosocomial bacteremia had received recent therapy with a beta-lactam antibiotic. The most common source of bacteremia was a postoperative wound infection (seven episodes). Only one episode was related to a urinary tract infection.

Retrospective analysis showed that clusters of cases of M. morganii bacteremia had occurred almost yearly. This finding prompted a six-month period of prospective monitoring of all cultures for M. morganii to identify human reservoirs in our institution. Sixty percent of all cultures growing M. morganii came from urine cultures, 18% came from wound cultures, and the remaining 22% came from a variety of body fluids or tube drainage. Thirty-one percent of patients harboring M. morganii were on the Surgical Service.

M. morganii bacteremia most commonly occurs in postoperative patients who receive beta-lactam antibiotics. From the data in this study, M. morganii is an infrequent cause of bacteremia, and its presence in blood cultures may be an indicator of an environment conducive for an outbreak of nosocomial infection.

Data related to risk factors for catheter-acquired bacteriuria were collected prospectively on 112 patients consecutively catheterized for >24 hours at the Hadassah University Hospital. Logistic regression analysis indicated that factors independently associated (p≤0.05) with a higher risk of catheter-acquired bacteriuria were as follows: hospitalization in orthopedics or urology, ethnic origin (Arabs > Jews), insertion of a catheter after the sixth day of hospitalization, catheterization outside the operating theaters, lack of administration of systemic antibiotics, unsatisfactory catheter care, and prolonged duration (≥7 days) of catheterization before infection occurred. The risk associated with catheterization outside the operating theater could be explained by its correlate, that is, catheterization for incontinence/obstruction as opposed to output measurement. Life-table analyses demonstrated that the daily risk for acquiring bacteriuria during the first six days of catheterization was higher among patients ultimately catheterized for ≥7 days than among those ultimately catheterized for < 7 days (P<0.05).

During a ten-month period from September 1981 to July 1982 three episodes of pseudobacteremia due to Bacillus species occurred at this 550-bed institution. The first involved eight isolates, the second 11, and the third seven isolates of the organism, all with the same antibiogram.

The patients involved did not exhibit clinical signs of septicemia, and in only one case was more than one specimen per patient positive when multiple blood samples were obtained. Occasional blood cultures of Bacillus species identified in between clusters revealed a different antibiogram.

Extensive epidemiologic investigation of patient locations, phlebotomists, and time of cultures yielded no common source. Components involved in the transport and processing of blood cultures, including the radiometric blood culture processor, were also sampled but without recovery of the organism. After the last episode, a layer of dust was noted inside the machine, and culture of this dust grew Bacillus spp. with the same antibiogram as those found in the blood cultures. The filter from an air conditioning unit in close proximity to the machine grew several species of Bacillus.

It is presumed that Bacillus spores in the dust were introduced into the blood culture bottles following the heat sterilization of the gas sampling (inoculation/removal) needles.

Modification of the cover of the machine was undertaken to prevent access of dust bearing microbes to the inside of the machine. In addition, maintenance now includes regular disinfection/cleaning of the “floor” of the machine, and more frequent changes of the air conditioner filter.

A prospective study was undertaken to examine the clinical presentation of peritonitis in patients maintained on intermittent peritoneal dialysis and to determine the value of qualitative and quantitative dialysate cultures, gram stain, neutrophil counts, and a semiquantitative leukocyte test strip for case detection. Seven cases of peritonitis developed among 30 patients who underwent 553 dialyses. In most cases, neutrophil counts, cultures, and leukocyte test strip determinations were done within 48 hours prior to the clinical onset of peritonitis and in all instances failed to provide clues for incipient infection. Peritonitis was associated with a dialysate neutrophil count of >500/mm3 and leukocyte test strips were highly sensitive and specific for the detection of this quantity of neutrophils. A total of 16 dialysate cultures was positive in asymptomatic patients who did not have peritonitis. None of these patients subsequently developed peritonitis with the same organism. Dialysate gram stains, cultures, neutrophil counts or leukocyte test strips did not provide an early diagnosis of peritonitis and their use in the absence of symptoms is therefore not recommended.

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<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.

Seventeen episodes of persistent Staphylococcus epidermidis bacteremia (one to nine days) occurred in 16 patients with vascular catheters during a 26-month period. Cases were statistically more likely to have a longer hospitalization (54 v 7.6 days, p < .0005), longer duration of antibiotic therapy (22 v 2.5 days, p = .002), presence of a central venous pressure (CVP) catheter (14 v 2, p < 3 × 10-8), and presence of an arterial catheter (4 v 1, p = 0.037) than randomly selected hospitalized patients matched for age, sex, and date of admission. However, when cases were compared with similarly matched non-bacteremic patients having CVP catheters, these characteristics were not significantly different in the two groups. Furthermore, exposure to total parenteral nutrition (TPN) and duration of TPN were not significantly different between cases and controls. Hence, the presence of a CVP catheter appeared to be the major risk factor for 5. epidermidis bacteremia. In 16 episodes, patients had temperature > 38.6°C without another identifiable cause, and the average white cell count for the case group was 19,400/mm. Seven patients also had diaphoresis, confusion, hypotension, or oliguria. Temperatures returned to normal in 13 within 24 hours after catheter removal, and all patients were afebrile and symptom-free within 72 hours. Thus, vascular catheter-associated S. epidermidis bacteremia was an important case of febrile morbidity in these patients.