Search

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.

Loci for resistance to several antibiotics in laboratory-derived strains of Neisseria gonorrhoeae were mapped by genetic transformation. Genes for high-level resistance to streptomycin (str) and spectinomycin (spc) and for low-level resistance to tetracycline (tet) and chloramphenicol (chl) were linked. Also, a locus for high-level resistance to rifampin (rif) was linked to str and tet. The apparent order was rif... str... tet... chl... spc. Loci for resistance to other antibiotics (penicillin, erythromycin) were transferred independently of each other and were not linked to the cluster around str. Similar linkage relationships were found with str, tet, chl, and spc loci obtained from naturally occurring (clinical) isolates of N. gonorrhoeae.

Nosocomial Gram-negative bacillary meningitis and bacteremia occurred in a patient who was receiving intrathecal and intravenous amphotericin B. An epidemiologic investigation found the amphotericin B to be contaminated with Enterobacter agglomerans, Pseudomonas fluorescens, and P aeruginosa. These contaminants were traced to a lot ot sodium phosphate buffer that was added to all intrathecal and intravenous amphotericin B preparations. The phosphate buffer underwent prolonged storage at room temperature and was not subject to terminal sterilization nor sterility testing. This parenteral admixture prepared in the hospital is now steam autoclaved and sterility tested before use.

An increase in the recovery of Aspergillus flavus from respiratory specimens occurred among hospitalized patients during a several-month period in 1977. Notably, 86% of the patients with positive culture results were located in an older hospital section adjacent to a building construction site that created great amounts of dust in the hospital's vicinity. Airborne contamination was suspected and results of air sampling showed A. flavus in 80% of old wing patient rooms (average, 8 A. flavus/positive room) compared with only 23% of the contiguous newer wing patient rooms (average, 1 A. flavus/positive room). Inspection of the main mechanical ventilation units in the two hospital sections disclosed numerous defects in the unit in the old wing and a properly functioning unit in the new wing. After repairing the defective unit, a significant reduction in the recovery of A. flavus from clinical and environmental specimens was noted.

The genetics of low-level resistance to penicillin and other antibiotics in a clinical isolate and a multistep laboratory mutant of Neisseria gonorrhoea was studied by transformation. Mutations at three loci affected sensitivity to penicillin. Mutation at penA resulted in an eightfold increase in resistance to penicillin without affecting response to other antimicrobial agents. Mutation at ery resulted in a two- to fourfold increase in resistance to penicillin and similar increases in resistance to many other antibiotics, dyes, and detergents. Mutation at penB resulted in a fourfold increase in resistance to penicillin and tetracycline, the phenotypic expression of which was dependent on the presence of mutation at ery. The cumulative effect of mutations at penA, ery, and penB was an approximate 128-fold increase in penicillin resistance, to a minimum inhibitory concentration of 1.0 mug/ml. Low-level resistance to tetracycline or chloramphenicol was due to similar additive effects between mutations at the nonspecific ery and penB loci and a locus specific for resistance to each drug (tet and chl, respectively). No evidence was found for penicillinases or other drug-inactivating enzymes.

A brief perioperative course of ampicillin was recommended for high-risk patients undergoing cesarean section at our institution. High-risk patients were defined as those with two or more of the following factors: general anesthesia, obesity, hematocrit less than or equal to 30%, and labor prior to delivery. Fifteen percent of high-risk patients who received prophylactic antibiotics experienced postoperative febrile morbidity compared to 63% for the high-risk untreated group (P less than 0.001). Twenty-five percent of low-risk patients (less than 2 risk factors) developed this complication. Endometritis was the leading cause of postoperative febrile morbidity. Postoperative stay was only slightly decreased in the prophylactically treated group (5.9 vs 6.2 days). Prophylactic amplicillin given perioperatively was an effective agent in reducing post-cesarean-section febrile morbidty among high-risk patients.

Recent studies that emphasize the unpredictability of gentamicin serum concentrations have cast doubt on the accuracy of commonly used dosing nomograms. We therefore developed a digital computer program to evaluate gentamicin pharmacokinetics and have tested its ability to predict correctly gentamicin serum levels under clinical conditions. A significant relation (P less than .001) was found between 130 measured (bioassayed) and computer predicted levels from 40 patients. Serum levels were reliably estimated in patients with unstable renal function and were statistically, but not clinically, affected by changes in hematocrit. The calculation of volume of distribution on the basis of the patient's lean body weight rather than total body weight was an important factor that contributed to the accuracy of the predicted level. A new dosing chart is presented that incorporates the important variables affecting gentamicin pharmacokinetics into an easily used, convenient form.

During an outbreak of infections caused by methicillin-resistant (MR) Staphylococcus aureus in our burn unit, we conducted an extensive 10-week study to define the environmental epidemiology of the organism. The inanimate environment in patient rooms and adjacent areas was examined by using volumetric air samplers and Rodac plates. Airborne and surface level contamination with MR S. aureus was quantitated, and overall, MR S. aureus comprised 16, 31, and 40% of all bacterial growth from air, elevated surfaces, and floor surfaces, respectively. Mean air, elevated surface, and floor surface MR S. aureus contamination in rooms of MR S. aureus-infected burn patients were 1.9 MR S. aureus per ft3 (ca. 0.028 m3), 20 MR S. aureus per Rodac plate and 48 MR S. aureus per Rodac plate, respectively. Peak patient room environmental contamination levels were 6.9 MR S. aureus per ft3 of air, 70 MR S. aureus per Rodac plate per elevated surface and 138 MR S. aureus per Rodac plate per floor surface. Environmental contamination levels in the adjacent work areas were considerably lower than in infected patient rooms. There was ample opportunity for contamination of personnel through the inanimate environment in this unit.

A steam sterilizer (autoclave) was tested to determine the operating parameters that affected sterilization of microbiological waste. Tests involved standardized loads (5, 10 ad 15 lb [ca. 2.27, 4.54, and 6.80 kg, respectively]) contaminated petri plates in autoclave bags placed in polypropylene or stainless steel containers. Thermal and biological data were obtained by using a digital potentiometer and a biological indicator containing spores of Bacillus stearothermophilus, respectively. The transfer of heat was more efficient when smaller loads of microbiological waste were tested and stainless steel rather than polypropylene containers were used. A single bag with the sides rolled down to expose the top layer of petri plates allowed heat to pass better than did a single bag with the top constricted by a twist-tie. The presence of water in the autoclave bag did not significantly improve heat-up time in stainless steel or polypropylene containers. The results of biological tests substantiated the temperature data. When 10 or 15 lb of microbiological waste was exposed to various test conditions, the only condition that ensured the destruction of B. stearothermophilus involved the use of a stainless steel container (with or without water) for 90 min. Autoclaving for 45 min resulted in the destruction of bacteria included in 10 lb (136 +/- 3 plates) or 15 lb (205 +/- 6 plates) of microbiological waste when stainless steel containers with or without water or polypropylene containers with water used, whereas 60 min was required to kill all bacteria if polypropylene containers without water were used.

During 1981, 39 nosocomial infections occurred in 27 of 223 patients undergoing cardiac surgery in the North Carolina Memorial Hospital. The peak attack rate (23.7%) occurred in August and September compared to 10.1% in January through July. A case-control study demonstrated that the only risk factor common to poor and stable health groups compared to controls was duration of stay in the cardiothoracic intensive care unit (CTICU). Microbiologic studies of the environment, personnel and patients showed that colonization or infection of patients occurred 1 to 6 days after admission to the CTICU and that nosocomial pathogens were found: 1) in and around the unit's soiled utility sink, 2) in pooled handwashing cultures of unit personnel, and 3) on contaminated clean hands and air near the soiled utility sink, plus nearby bedside air only while the water in the sink was running. Environmental and personnel hand contamination by soiled utility sink aerosols likely contributed to these infections. Infection control measures, including discontinued use of the soiled utility sink, resulted in a significantly lower infection rate (5.6%).

To assess the scope of infection control programs in extended care facilities, 1-day surveys were conducted in 12 North Carolina facilities over an 8-month period using a standardized questionnaire. All 12 facilities had a designated infection control practitioner (ICP), although none had attended an infection control education course. Eleven had an Infection Control Committee of which 8 (73%) met regularly. The Director of Nurses generally (58%) was the ICP and spent about 2 hr/wk on infection control. Ten (83%) facilities conducted infection surveillance among residents but did not accurately compute nosocomial infection rates. Eleven (92%) facilities had employee health programs that included preemployment and annual tuberculosis screening. None had a comprehensive resident health program. Infection control aspects of patient care practices often varied from facility to facility. Nosocomial infection surveillance among 336 residents in 9 facilities using modified CDC criteria revealed an overall prevalence rate of 5.4%. Additional infections were suspected but not included because of limitations of laboratory data and chart documentation.

An outbreak of nosocomial infections of the urinary tract due to a multiply drug-resistant strain of Serratia marcescens occurred at a community hospital. Acquisition of the epidemic strain was associated with the following factors: (1) exposure to the intensive care unit, (2) presence of an indwelling bladder catheter, (3) treatment with antibiotics, and (4) exposure to devices used from measurements of specific gravity and urine volume. An extensive microbiologic evaluation of the hospital environment failed to reveal the epidemic strain of S. marcescens from any site other than urinometers and urine volume measuring containers. Four of four urinometers and three of seven urine measuring containers tested revealed the epidemic organism. Notably, six of these seven positive cultures were obtained in hospital areas in which no patients infected with S. marcescens were located at the time of sampling. The resistant organism was also recovered from one of three pooled handwashings taken from nursing personnel. Thus, the urinometer and urine measuring container may have served as inanimate reservoirs for the resistant S. marcescens which was subsequently inoculated onto the hands of medical personnel or directly to a catheterized patient. Disinfection procedures were identified which eliminated these items as reservoirs. No additional cases of multiply drug-resistant S. marcescens urinary tract infections have been observed since institution of routine disinfection of the inanimate reservoir.

Amikacin is a new aminoglycoside antibiotic that behaves pharmacokinetically similar to kanamycin, gentamicin, and tobramycin. Our study was designed to test whether a digital computer program could correctly predict amikacin serum concentrations in the clinical setting. A significant relation (P less than 0.0001) was found between 153 measured (bioassayed) and computer-predicted levels from 26 patients. The computer program reliably estimated amikacin serum levels after either intravenous or intramuscular doses. Prediction accuracy was not significantly affected by patient sex, hematocrit, or periods of unstable renal function. Predicted levels were most accurate when based on creatinine clearance corrected to 70 kg body weight or corrected to body surface area. The pharmacokinetic parameters in the computer program were used to generate a new aminoglycoside dosing chart. Ten patients were given amikacin according to this guideline, and the mean peak serum level for the group was 25.6 microgram/ml.

A high incidence of post cesarean section febrile morbidity has caused a number of investigators to recommend using prophylatic antibiotics for patients undergoing this procedure. Recent data suggest that such prophylactic antibiotics should be reserved for high-risk patients. Since previous studies have not adequately defined the high-risk patient, we analyzed 129 patients undergoing cesarean section to identify the factors which predispose to postoperative febrile morbidity. Two different statistical programs identified four factors (general anesthesia, obesity, hematocrit less than or equal to 30%, and labor prior to delivery) which were significantly associated with post cesarean section febrile morbidity (P less than 0.025). Patients with 2 or more of these risk factors were highly likely to experience this postoperative complication (P less than 0.001).

In October 1980, 80% (120/150) of North Carolina (NC) hospitals responded to a comprehensive solid waste questionnaire intended to identify their solid waste disposal practices. NC hospitals generated an average of 13 lb. of solid waste/patient/day and infectious waste comprised 5% to 11% of the total hospital waste. Most hospitals (greater than 66%) considered blood, isolation, laboratory (including microbiology), pathology, and autopsy waste as infectious, while other sources of solid waste were generally considered non-infectious. The infectious wastes were normally incinerated or steam sterilized before discarding. Nearly all hospitals (95%) used a Class A landfill to dispose of the non-infectious waste. Two-thirds of the NC hospitals are discarding blood, microbiological, pathological, and isolation waste in accordance with the CDC recommendations, while the compliance rates for the JCAH and EPA recommendations for these wastes were 94% and 24%, respectively. If the EPA modified its recommendations to include incineration of microbiological waste, which is practiced by 50 responding NC hospitals, the compliance rate would be 95.5% for these wastes and 61.7% overall.

Mutations (env) which resulted in increased sensitivity of gonococci to diverse compounds were studied by transformation. Strains carrying an env mutation were more sensitive than wild-type strains to several antibiotics, dyes, and detergents. The env mutations resulted in complete phenotypic suppression of low-level resistance to these same drugs determined by mutation at ery. Recombination was observed in transformation crosses between various env mutants. The env locus was not linked to the cluster of antibiotic resistance genes near str and spc.

We analyzed all cases of tuberculosis reported in North Carolina between 1966 and 1986, and related the incidence rate of tuberculosis (per 100,000 population) to age (0 to 4 years, 7.59; 5 to 14 years, 3.44; 15 to 24 years, 6.30; 25 to 44 years, 15.92; 45 to 64 years, 33.85; greater than 65 years, 51.54), race (white 9.03, nonwhite 47.40), and gender (male 25.49, female 11.25). Over the 21-year study period the annual number of cases declined from 1,248 to 711 (43%), and the incidence rate from 25.56 to 11.25 (56%). Although the incidence rate of tuberculosis fell for all subgroups, nonwhites continued to have an incidence rate 3.2 to 22.5 times higher than whites, depending on age. The standardized morbidity ratio (SMR) (by age, race, and gender) of tuberculosis in the eastern region of North Carolina was nearly twice that of the western region and unexplainable by its demographics. Between 1983 and 1986 only a small percentage of cases of tuberculosis in North Carolina were accounted for by migrant farm workers (1.7% to 2.7%) and patients with the acquired immunodeficiency syndrome (less than 1%). Tuberculosis is increasingly a disease of the elderly, especially nonwhite men. Tuberculosis is a geographically and demographically focal disease in North Carolina, and preventive strategies should be appropriately targeted.