Health care worker exposure to HIV-1 and HTLV I-II in critically ill, resuscitated emergency department patients

STUDY OBJECTIVE

Exposure to HIV-1 is of profound concern to health care workers. HTLV-I and HTLV-II, retroviruses with similar modes of transmission as HIV-1, also cause disease in human beings. Emergency department resuscitations are high-risk situations for such exposure. The purpose was to determine the seroprevalence of HIV-1 and HTLV I-II in patients undergoing ED resuscitations, the magnitude of health care worker exposure, and risk factors associated with infection.

DESIGN

Prospective identity-unlinked seroepidemiologic study.

SETTING

ED of a 950-bed private inner-city teaching hospital. Participants included 370 patients undergoing ED resuscitations.

MEASUREMENTS

Serum was tested for antibodies to HIV-1 and HTLV I-II. Questionnaires were completed by the physician in charge of the ED resuscitations.

RESULTS

Fifteen (4.1%) (95% confidence interval [CI], 2.1% to 6.1%) patients were HIV-1 seropositive, and seven (1.9%) (95% CI, 0.7% to 3.1%) were HTLV I-II positive. Eleven (5.6%) (95% CI, 2.4% to 8.8%) of 197 trauma patients and 11 (6.4%) (95% CI, 2.8% to 10.0%) of 173 medical patients were infected with one of these viruses. Health care workers had direct cutaneous contact with patient blood during 114 (31%) ED resuscitations and with infected patient blood during 11 (3%) ED resuscitations. An additional 11 ED resuscitations involved parenteral exposures, one to HIV-1-infected blood. No factors could be identified that would quickly and reliably predict infection.

CONCLUSION

Health care workers must protect themselves in such high-risk situations by strict compliance to mandatory universal precautions.

Comparison of five commercial western blot kits for detection of human immunodeficiency virus 1

Twenty-two human immunodeficiency virus 1 (HIV-1) enzyme immunoassay (EIA) reactive and two non-reactive patient specimens were analyzed using five commercially available HIV-1 Western blot kits. The percentage of HIV-1 bands detected by each kit was recorded. The differences between pairs of kits were not found to be statistically significant at the 0.05 level. All EIA reactive specimens were reconfirmed as reactive by each Western blot kit tested.

Reduction in Legionella pneumophila through heat flushing followed by continuous supplemental chlorination of hospital hot water

An increase in endemic rate of nosocomial Legionella pneumophila pneumonia prompted an investigation that revealed 16.2% (12/74) of patient care hot-water sites surveyed were culture-positive for L. pneumophila. No positive cultures were recovered from cooling towers, air intakes, or construction areas. Heat flushing of hospital hot-water outlets to temperatures greater than 60 degrees C for 30 min achieved a 66% reduction in positive Legionella cultures. After 4 1/2 months, different serotypes recurred in previously eradicated areas and there were new positive cultures. Continuous supplemental chlorination of the hot-water system (2 parts per million [ppm]) significantly reduced the number of culture-positive samples from 37.4% (43/115) to 7.0% (8/115) after 6 weeks (P less than .005). Of 30 sites surveyed 6 months after hot-water chlorination, 67% (20) were still culture-negative. Of those positive, 70% had less than or equal to 150 L. pneumophila/ml and 90% were from bathtubs. Adverse effects of chlorination on users and plumbing have not been seen. There have been no definite cases of nosocomial L. pneumophila in areas served by supplemental chlorine during the first 17 months of the chlorination project. Technology allowing tighter regulation of chlorine and use of silicates to control corrosion have made continuous hot-water chlorination a safe and effective option in Legionella control.

Comparison of ciprofloxacin and rifampicin in experimental Legionella pneumophila pneumonia

We evaluated intraperitoneal ciprofloxacin and rifampicin alone and as combination therapy in experimentally induced Legionella pneumophila pneumonia in guinea pigs. Intraperitoneal treatment began 48 h after intratracheal inoculation of 3 X 10(6) L. pneumophila and consisted of sterile saline (0.3 ml bid), ciprofloxacin (30 mg/kg bid), rifampicin (10 mg/kg/bid), or ciprofloxacin plus rifampicin (same doses). Animals were treated for five days and survivors killed after 11 days. Quantitative lung cultures were done post mortem. Respective mean and median days of animal survival were increased by treatment with ciprofloxacin plus rifampicin (8.4 and 9.5 days), ciprofloxacin (8.2 and 7.5 days), or rifampicin (8.3 and 7.5 days), compared with controls (5.5 and 5.0 days). Compared with control animals (log rank test) survival was improved by treatment with ciprofloxacin plus rifampicin (P less than or equal to 0.047) ciprofloxacin (P less than or equal to 0.047) or rifampicin (P less than or equal to 0.047). Quantitative lung cultures (cfu/g) were also decreased by treatment with ciprofloxacin plus rifampicin (2.0 X 10(4)), ciprofloxacin (5.4 X 10(4)), or rifampicin (1.7 X 10(4)) compared with controls (3.2 X 10(8)). No differences in survival, quantitative lung cultures, or animal weights were noted between treatment groups. This study demonstrates that ciprofloxacin is as effective as rifampicin in the treatment of experimentally induced L. pneumophila pneumonia and that the combination of ciprofloxacin plus rifampicin has no advantages over single agent therapy in this model.

Effect of quinolones and other antimicrobial agents on cell-associated Legionella pneumophila

We evaluated the in vitro susceptibility of Legionella pneumophila ATCC 33152 (serogroup I) to 13 antibiotics alone and in combination with rifampin (0.1 mg/liter) by three methods. Extracellular susceptibility was determined by MIC determinations and time kill curves in buffered yeast extract broth, while intracellular susceptibility was determined by peripheral human monocytes in RPMI 1640 culture medium. Antibiotic concentrations equal to or greater than the broth dilution MIC inhibited or killed L. pneumophila by the time kill method, except this was not the case for trimethoprim-sulfamethoxazole. Antibiotic concentrations below the broth dilution MIC did not inhibit Legionella growth. The only antibiotic-rifampin combinations which produced improved killing of L. pneumophila by the time kill method were those in which the logarithmic growth of L. pneumophila occurred during the experiment (rosoxacin, amifloxacin, cinoxacin, trimethoprim-sulfamethoxazole, clindamycin, and doxycycline). Neither direct MICs nor time kill curve assays accurately predicted intracellular L. pneumophila susceptibility. Rifampin, erythromycin, ciprofloxacin, rosoxacin, enoxacin, amifloxacin, gentamicin, clindamycin, and doxycycline all inhibited intracellular L. pneumophila growth at readily achievable concentrations in serum. Cefoxitin and thienamycin showed no inhibition of growth, although they were present extracellularly at concentrations that were 20 to 1,000 times their broth dilution MICs. Clindamycin was the only antibiotic that was able to inhibit intracellular L. pneumophila growth at an extracellular concentration below its MIC. The gentamicin (5 mg/liter)-rifampin combination was the only antibiotic-rifampin combination which demonstrated decreased cell-associated Legionella survival in this model of in vitro susceptibility.

The in-vitro and in-vivo efficacy of cisplatin and analogues in the treatment of herpes simplex virus-II infections

The antitumour effect of cisplatin results from cross-linking and disruption of DNA when it binds to DNA bases, especially cytosine and guanine. Since herpes simplex virus (HSV) has a high cytosine and guanine content, cisplatin might be expected to have an antiviral effect against HSV. The 50% inhibitory concentration of cisplatin for HSV-II was 0.06 mg/l. Six of ten platinum analogues had 50% inhibition of plaques at less than or equal to 10 mg/l. We evaluated the in-vivo activity of cisplatin against the MS strain of HSV-II in the mouse genital HSV model. Mice were treated either intraperitoneally or intravaginally beginning at 3 or 51 h after inoculation. In the intraperitoneally treated groups infection rates were lower, but not significantly; 4 of 15 in the 3-h and 7 of 15-h group, compared to 9 of 15 in the untreated control group (P greater than 0.18, chi-square test). Intravaginal cisplatin demonstrated a significant reduction of the infection rate from 10 of 15 untreated controls, compared to 5 of 18 in the 3-h and 5 of 17 in the 51-h group (P less than 0.05, chi-square test). No toxic effects of intravaginal cisplatin were seen in uninfected mice. These studies suggest that platinum containing drugs warrant further evaluation as a new class of antiviral agents with activity against HSV.

Failure of therapy in pseudomonas endocarditis: selection of resistant mutants

Despite optimal use of available antibacterial agents, endocarditis due to Pseudomonas aeruginosa is commonly associated with poor response to medical treatment. Two patients are described in whom emergence of resistance to beta-lactam antibiotics was associated with clinical failure. A subpopulation of resistant mutants (10(-7)) was found within the initial, apparently sensitive population of bacteria. These resistant mutants were similar to posttherapy isolates in their increased production of beta-lactamase and in their identical pattern of resistance to beta-lactam antibiotics. Moreover, the only beta-lactamase produced was type Id, and this enhanced production proved to be constitutive. A relatively large inoculum (10(6) colony-forming units/g of tissue) of bacteria was found postoperatively in the heart valves of both patients. The failure to respond is postulated to be due to the selection of these producers of high levels of beta-lactamase in a large bacterial inoculum.

Single-dose cephalexin therapy for acute bacterial urinary tract infections and acute urethral syndrome with bladder bacteriuria

The efficacy of single-dose therapy with 3 g of cephalexin was evaluated in 129 women with symptoms of acute uncomplicated lower urinary tract infections. Of 91 patients with significant bacteriuria, 61 (67%) were cured of their original infection; this was similar to the 54 to 79% cure rates reported in unselected populations of women of a wide age range treated for acute uncomplicated urinary tract infections with a single dose of amoxicillin or trimethoprim-sulfamethoxazole (J. Rosenstock, L. P. Smith, M. Gurney, K. Lee, W. G. Weinberg, J. N. Longfield, W. B. Tauber, and W. W. Karney, Antimicrob. Agents Chemother. 27:652-654, 1985; N. E. Tolkoff-Rubin, M. E. Wilson, P. Zuromskis, I. Jacoby, A. R. Martin, and R. H. Rubin, Antimicrob. Agents Chemother. 25:626-629, 1984). The cure rates of (87%) for our younger patients, those less than 25 years of age, was better than that (46%) for our patients over 40 years of age (P less than 0.001). Patients with infections that were negative in an antibody-coated bacteria test were cured at a significantly higher rate than those with infections that were positive in an antibody-coated bacteria test (71 versus 19%; P = 0.003). Those patients with infections caused by cephalexin-susceptible organisms were cured at a rate similar to that for patients with infections caused by cephalexin-resistant organisms (68 versus 50%; P = 0.62). The cure rate for suburban patients was 90%, versus 45% for inner-city patients (P = 0.008). Of the 28 women with acute urethral syndrome due to low-level bacteriuria, 27 were cured.

Evaluation of the sterility and stability of insulin from multidose vials used for prolonged periods

Diabetic patients frequently use their insulin vials for prolonged periods, even though antibacterial preservatives used in multidose insulin vials are not required to be effective beyond 28 days. For this reason, we evaluated the antibacterial activity present in multidose insulin vials for up to 50 days. Multidose lente insulin vials were inoculated with S. aureus and P. aeruginosa. Vials incubated at room temperature (21 degrees C) were sterile by 48 hours, whereas when they were incubated at refrigerator temperatures (4 degrees C), S. aureus contamination persisted up to the 17th day and P. aeruginosa were killed after 10 days. The same vials were serially contaminated on days 17, 30, and 50, and a similar antibacterial effect was maintained. Sixty-nine multidose insulin vials used for an average of 53 days were cultured. Eight vials demonstrated bacterial contamination with 1 cfu/ml of S. epidermidis or Propionibacterium acnes. No endotoxin was detected in the multidose vials used for more than 28 days. Insulin assays on 40 multidose insulin vials used for more than 28 days showed an average insulin content of 101.6 +/- 1.9 units/ml. This study did not demonstrate significant bacterial contamination, endotoxin activity, or insulin degradation of multidose insulin in vials used for periods longer than 28 days. In addition, antibacterial preservatives were more effective at room temperature than at refrigerator temperature; thus, the practice of patients not refrigerating insulin vials is sensible.

An outbreak of gentamicin-resistant Klebsiella pneumoniae: analysis of control measures

In April 1978, a strain of gentamicin-resistant Klebsiella pneumoniae (GRK) was introduced into the neonatal intensive care unit of Henry Ford Hospital. An additional ten cases of GRK occurred over the subsequent 16 months and intestinal colonization occurred in up to 91% of admissions per month. All GRK were susceptible to amikacin and were capsular serotype 19. Though hand contamination of hospital personnel with GRK was documented, increased handwashing practices did not reduce colonization rates of neonates with the epidemic strain. Intestinal carriage persisted for up to ten months and could not be eradicated by administering oral colistin sulfate. Discontinuation of gentamicin and utilization of amikacin were associated with a significant reduction in colonization with GRK (p less than 0.05). However, the only control measure that prevented both new cases and colonization with the epidemic strain was the utilization of a strict cohort system.

Cephapirin therapy in infective endocarditis

Ten patients, eight of them heroin addicts, with infective endocarditis were treated with cephapirin. Causative organisms included Staphylococcus aureus in six patients and viridans streptococci, S bovis, and S pneumoniae in one patient each. In one of the patients with endocarditis, both S aureus and an organism of the beta-Streptococcus group were simultaneously isolated. All patients except two, both with monomicrobial S aureus endocarditis, were cured. All isolates were inhibited by less than or equal to 0.5 micrograms/ml of cephapirin. Peak serum bactericidal activity greater than or equal to 32 times the minimal inhibitory concentration (MIC) of the causative organism was achieved in all cases after giving 2 gm IV every four hours. After using large inoculums of staphylococci for in vitro tests, we found that twofold to eightfold increases in MIC occurred. Cephapirin was well tolerated intravenously and high levels in serum were achieved without toxicity or accumulation of the drug.

Methicillin-resistant Staphylococcus aureus. Epidemiologic observations during a community-acquired outbreak

Infection with strains of methicillin-resistant Staphylococcus aureus occurred in 40 patients at time of admission to a large urban hospital from March to December 1980. Community-acquired methicillin-resistant S. aureus infections occurred in 24 drug abusers and 16 nonabusers. Patients with infections had a longer mean hospitalization and previously had received antimicrobial therapy more frequently than control subjects. Drug abusers with infections had been treated with cephalosporins more often than control subjects (P less than 0.05). Phage typing of 32 isolates showed that 21 were linked by a common phage type (29/52/80/95). Transmission of methicillin-resistant S. aureus from community-acquired cases occurred in the hospital. By January 1981, methicillin-resistant S. aureus accounted for 30.6% of nosocomial S. aureus infections at Henry Ford Hospital. Methicillin-resistant S. aureus infection may arise in the community as well as in the hospital and has the potential to disseminate in both settings.

Ceforanide: in vitro and clinical evaluation

Ceforanide, a new cephalosporin antibiotic with a long half-life (3 h), can be administered twice daily. We evaluated its antimicrobial activity, pharmacology, and clinical efficacy. Twenty-seven patients with infections due to susceptible organisms received ceforanide, 0.5, 1, or 2 g, intramuscularly or intravenously every 12 h for 6 to 28 days. In vitro studies with the clinical isolates from 27 patients treated plus 263 additional isolates showed that ceforanide was active against cephalothin-susceptible gram-positive and gram-negative microorganisms. In addition, ceforanide inhibited 65% of cephalothin-resistant Escherichia coli and 65% of Enterobacter spp. at </=12.5 mug/ml. After a single 1-g intramuscular dose, the mean peak plasma concentration at 1 h was 48.9 mug/ml and that at 12 h was 4.7 mug/ml. Plasma accumulation occurred in some patients. The infections included 10 pneumonias, 3 with bacteremia and 1 with empyema; 11 soft tissue infections, 4 with abscesses and 3 with sepsis; and 3 urinary tract infections. One case each of endocarditis, osteomyelitis, and septic thrombophlebitis, all due to Staphylococcus aureus, were treated. Clinical response was satisfactory in all patients; bacteriological response was satisfactory in 26 of 27 patients. Ceforanide was well tolerated. Three patients developed mild increases in liver enzymes, and one developed slight eosinophilia. In another case, the antibiotic was discontinued because of a fivefold rise in serum glutamic-oxalacetic transaminase (aspartate aminotransferase) and serum glutamic-pyruvic transaminase (alanine aminotransferase) and a twofold rise in lactic acid dehydrogenase and alkaline phosphatase.

Effect of renal failure and dialysis on the serum concentration of the aminoglycoside amikacin

Serum and dialysate levels of amikacin were determined at appropriate intervals after a 300-mg intravenous dose as a continuous infusion in six patients with end-stage renal failure undergoing hemodialysis and in three patients on peritoneal dialysis. The mean serum half-life of amikacin was 3.75 h during (or after) hemodialysis and 29 h during (or after) peritoneal dialysis. Although not on hemodialysis in the same six patients, the serum half-life was 28 h. The results indicate that the maintenance dose of amikacin should be markedly decreased in patients with severe renal failure even if they are treated with peritoneal dialysis, and that serial serum antibiotic concentrations are essential to prevent cumulative toxicity of the drug.

In vitro and clinical studies of cefatrizine, a new semisynthetic cephalosporin

Cefatrizine, a new oral semisynthetic cephalosporin, was evaluated in vitro and in the treatment of 18 patients with acute urinary tract infection, pneumonia, and soft tissue infection. In vitro, it was more active than cephalexin for gram-positive and gram-negative bacteria. It was also more active than cephalothin, cefazolin, and cephapirin against most of the gram-negative bacteria but less active against the gram-positive bacteria. Of the patients treated with cefatrizine, only one failed to respond. This patient had pneumococcal conjunctivitis and hypogammaglobulinemia and neutropenia. The mean peak serum level after multiple 6-hourly doses of 500 mg was 6.2 mug/ml. The serum levels of cefatrizine necessary for inhibition of most susceptible organisms were well within the achievable range. The drug was well tolerated, and no renal, hepatic, or hematological toxicity was detected.

Effects of renal failure and dialysis on cefazolin pharmacokinetics

Serum and urinary levels of cefazolin were determined after a 500-mg parenteral dose in eight azotemic volunteers. The mean peak serum concentration was 1.5 to 5 times the levels obtained in nonazotemic patients. The serum half-life of cefazolin was increased significantly. In patients on dialysis, the mean serum half-life of cefazolin was 4.05 h during (or after) hemodialysis, and 32.1 h during (or after) peritoneal dialysis. There was a significant decrease in cefazolin removal when dialysate flow or membrane surface area of the dialyzer were decreased. It was also shown that one circuit through the dialysis unit caused measurable decrease in cefazolin concentration. These data and previously published reports suggest: (i) the maintenance dose of cefazolin can be decreased in azotemic patients; (ii) patients on hemodialysis will require an additional half dose after dialysis because of efficient removal during hemodialysis; and (iii) patients on peritoneal dialysis do not require an extra dose.

Clinical studies of cefazolin and comparison with other cephalosporins

Cefazolin, a new cephalosporin derivative, was studied in the treatment of 105 hospitalized patients with a variety of infections including endocarditis, pneumonia, and urinary and soft tissue infections, and was found to be effective in 104 patients. Cefazolin was also tested in vitro and shown to be effective against staphylococci, pneumococci, Escherichia coli, Klebsiella sp., and Proteus mirabilis by agar dilution method. It was shown to produce high serum levels when administered in a 250- to 1,000-mg intramuscular dose and was well tolerated and free from renal toxicity. Comparison of the results of this study with those from our prior studies on cephaloridine revealed equivalent antibiotic potency, good tolerance to both the agents when given intramuscularly, superior, average blood levels with cefazolin, equal clinical efficacy, and absence of renal toxicity with cefazolin (unlike cephaloridine). Similarly, the results of treatment of pneumococcal pneumonia with intramuscular cefazolin were found to be superior to those for oral cephalexin.