Note: The following summaries represent a selection of papers from among those presented at the 37th Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC), held September 28-October 1, 1997, in Toronto and the 35th Annual Meeting of the Infectious Disease Society of America (IDSA), held September 13-16, 1997, in San Francisco. Although these two meetings were held separately for the first time, my contrary nature causes me to combine them here. The abstract designations which begin with "IDSA" indicate that they were presented at that meeting; the remainder were presented at ICAAC. It is important to recognize that many of these summaries are extracted only from the published abstracts, and it is possible that some of the material presented at the conference may have differed. SD
Pharyngitis, sinusitis, otitis, and bronchitis
The reasons for persistence or recurrence of pharyngeal infection or colonization with the Group A beta streptococcus remain obscure, despite much speculation. In one study, toothbrushes and dental braces were identified as possible reservoirs of Streptococcus pyogenes of persistence after children completed a 10-day course of penicillin. (IDSA-635.)
In a study performed in Sweden, Chlamydia pneumoniae DNA was detected by PCR of throat swabs of 38 (45%) of 85 children with respiratory tract illness and in five (5.7%) of 86 healthy control children. A number of the C. pneumoniae-positive children had middle ear infection; the organism was detected by PCR in middle ear fluid in one of four children who underwent myringotomy tube placement. (K-97.)
Pertussis is an increasingly recognized problem in adolescents and adults. A U.S. survey found that there was an increase in reported pertussis cases of 28% from 1994 to 1996 relative to the previous three-year period. While a small decreased incidence was noted in those up to 4 years of age, there was a 45% increase in those 5-9 years old, 113% in those 10-19 years old, and 96% in those more than 20 years of age. (K-164.)
Outpatient management of otitis media is becoming problematic because of the increasing prevalence of penicillin-resistant pneumococci. Five hundred thirteen children ages 4-30 months in an area of a high prevalence of penicillin-resistant pneumococci with acute otitis media were randomized to receive either a single IM dose (50 mg/kg) of ceftriaxone or 10 days of oral treatment with amoxicillin-clavulanic acid given tid. Treatment outcomes did not differ significantly between the two groups. (LM-35.)
Xylitol has been demonstrated to inhibit the growth of nasopharyngeal bacteria in vitro. Eight hundred fifty-eight children were randomized to receive sucrose chewing gum, xylitol chewing gum, lozenge, xylitol "mixture," or a control syrup. The incidence of acute otitis media in the xylitol group was reduced by 31.8% (P = 0.013) relative to the control and by 45.1% (P = 0.005) relative to those receiving sucrose chewing gum. (LM-36.)
Meningitis
Some physicians routinely perform a brain imaging study prior to lumbar puncture. One hundred fifty-nine adults with suspected meningitis underwent CT scanning of the brain prior to planned lumbar puncture. While imaging abnormalities were noted in 23 (14.5%), the results precluded LP performance in only three (1.9%). Comorbidity, history of a previous CNS lesion, recent seizure, and focal motor examination were independent predictors of CT abnormality. No CT abnormality was noted among the 61 immunocompetent patients with absent history of CNS lesion, a normal neurological examination, and absent clinical evidence of intracranial hypertension (normal systemic blood pressure and no papilledema). Thus, at least 38% of CT scans could be safely avoided in adults with suspected meningitis. (K-1223.)
Community-acquired pneumonia
Retrospective analyses of clinical trial experience with azithromycin therapy of community-acquired pneumonia (CAP) due to Chlamydia pneumoniae found that this infection accounted for 13%, 15%, and 11% of CAP in, respectively, adult outpatients, pediatric outpatients, and hospitalized patients. Satisfactory responses were observed in 83-91% of those treated with azithromycin and 79-100% of those treated with various comparator regimens (most containing erythromycin). (K-138.)
A similar analysis found that Mycoplasma pneumoniae accounted for 15% of CAP in adult outpatients, 28% in pediatric outpatients, and 4.6% among hospitalized patients. Of those participating, 90-95% responded to azithromycin and 86-93% responded to comparator regimens. (K-139.)
One hundred forty-eight patients with community-acquired pneumonia were randomized to receive either IV followed by oral azithromycin (500 mg daily) or IV followed by oral cefuroxime with addition of erythromycin if an atypical pathogen was suspected. Clinical response was observed in 91% of each group. (IDSA-410.)
Four hundred forty-three adults with community-acquired pneumonia were randomized to receive either IV alatrovafloxacin (a trovafloxacin prodrug) followed by oral trovafloxacin 200 mg qd or IV ceftriaxone followed by oral cefpodoxime with optional IV/oral erythromycin. The clinical success rates at the end of treatment were, respectively, 90% and 87%. (LM-72.)
One patient developed Legionella infection whose source appeared to be a malfunctioning automobile air conditioner. (IDSA-570.) Flooding in a cocktail bar was associated with three additional cases. (IDSA-287.)
Diabetic foot infections
A total of 225 patients with moderate to severe diabetic foot infections were randomized to receive oral ciprofloxacin with or without added metronidazole or IV ticarcillin/clavulanic acid (TC) followed by amoxicillin/clavulanic acid. The ciprofloxacin regimen was significantly less effective in bacterial eradication than the TC regimen, especially among those with osteomyelitis and especially with regard to streptococci. Although overall clinical outcomes were similar among the three arms, 24% of ciprofloxacin recipients, 14% of ciprofloxacin/metronidazole recipients, and only 9% of TC recipients underwent amputation (P = 0.011). The authors conclude that "oral ciprofloxacin is not optimal empirical therapy for diabetic foot infection, particularly if streptococci and anaerobes are present." (IDSA-413.) Perhaps combining clindamycin with ciprofloxacin would have proven more efficacious.
Streptococcus pneumoniae
Empiric treatment of community acquired pneumonia commonly consists of the administration of both a beta-lactam antibiotic and a macrolide. However, significant antagonism was demonstrated in vitro between penicillin and erythromycin for five of five pneumococcal isolates. Studies with two of these isolates in a murine model of pneumococcal infection also demonstrated antagonism. (A-29.)
A nationwide (U.S.) study of 4000 pneumococcal isolates found that 25% of penicillin-susceptible isolates were of intermediate susceptibility or resistant to cefaclor. While the NCCLS guidelines indicate that strains susceptible to penicillin can be considered susceptible to cefaclor, cefixime, cefpodoxime, and cefprozil, this study concludes that each cephalosporin must be tested individually. (E-48.)
Neisseria meningitidis
The prevalence of moderate penicillin resistance in meningococcal isolates in La Coruna, Spain, increased from 0% in 1993 to 30.9% in 1996 and to 92.8% in 1997. This was associated predominantly with Group C meningococci, which increased in frequency relative to Group B during this time. (9C-22.) In an another Spanish study, it was found that patients with meningococcemia due to moderately penicillin-resistant strains who were treated with penicillin did not exhibit excess mortality. However, there was significant excess mortality in patients with meningitis due to such strains. (C-34.)
Because of its low cost, chloramphenicol remains the optimal drug for treating meningococcal meningitis in developing countries. Unfortunately, high-level resistance to this drug has been identified in 12 CSF isolates from Vietnam and France. (LB-19.)
A single, 500 mg dose of azithromycin was as effective as the administration of 600 mg rifampin bid for two days in the short-term elimination of asymptomatic nasopharyngeal meningococcal colonization. (LM-4.)
Some of the antibacterial cationic proteins under study appear quite exciting. Among these is bactericidal/permeability increasing protein (BPI). Twenty-five (96%) of 26 children with severe meningococcemia and a predicted mortality of 20-50% who were given rBPI21 survived. (IDSA-414.)
Enterobacteriaciae
In the United States, Escherichia coli remains largely susceptible to fluoroquinolones, a situation likely to change soon if the current experience in Spain is relevant. The proportion of monomicrobial bacteremias caused by E. coli in neutropenic patients with hematologic malignancies increased from 5% in 1989-1990 to 21% in 1996 at Hospital LaFe in Valencia, Spain. During that interval, 28 (70%) of 40 episodes of E. coli bacteremia were caused by quinolone-resistant isolates. Ciprofloxacin prophylaxis was the factor most strongly associated with the emergence of quinolone-resistant E. coli bacteremia. (C-16.)
The prevalence of ciprofloxacin resistance in E. coli urine isolates in a general practice patient population in San Sebastian, Spain, increased from less than 1% in 1989 to 15% in 1995. (C-17.) In Barcelona, 24% of uninfected adults and 20.4% of uninfected children studied had ciprofloxacin-resistant E. coli in their feces. (C-18.)
Convincing evidence of plasmid mediated resistance to quinolones was demonstrated when it was found that the transfer of pMG252, a multiresistance plasmid from a urinary isolate of Klebsiella pneumoniae, into a porin deficient clinical isolate of K. pneumoniae led to an increase in MIC to ciprofloxacin to 32 mcg/mL. The plasmid has a broad host range, indicating the potential to spread to many other bacteria. (LB-20.) The ability of quinolone resistance to spread via plasmids suggests that the days of quinolone efficacy may be numbered.
Resistance to third-generation cephalosporins as the result of the spread of extended spectrum beta-lactamases (ESBL) is also of major concern. A retrospective analysis of 35 infections due to E. coli or K. pneumoniae producing ESBL found that the response rate to therapy was higher for the former than the latter (15/22 [68%] vs 4/13 [31%]; P = 0.03).
None of five patients with K. pneumoniae infection responded to monotherapy, while 67% (8 of 12; P = 0.013) with E. coli infection did so. There was a non-significant trend toward higher response rates to non-cephalosporin than to cephalosporin-containing regimens. The authors conclude that non-cephalosporin therapy may be preferred for infections due to ESBL-producing organisms and that, while monotherapy may be effective against ESBL-E. coli, combination therapy may be preferred against ESBL-K. pneumoniae. (J-8.)
In a prospective, observational study of 216 cases of bacteremia due to K. pneumoniae, of which 32 (15%) were due to ESBL-producing strains, 31% with ESBL-K. pneumoniae and 3% with non-ESBL-K. pneumoniae had received a third-generation cephalosporin in the preceding 14 days (P < 0.01). Among non-ICU patients, the mortality rate was 40% for infections with ESBL strains and 18% for infections with non-ESBLstrains (P = 0.08). The mortality of ESBL-K. pneumoniae bacteremia was 75% if inappropriate initial empiric therapy was used and 28% if appropriate therapy was administered ( P = 0.022). If imipenem was used, the mortality associated with infection with the ESBL-producing organisms was 23%, but was 42% when other antibiotics were used (P = 0.07). Subsequent infection with fungi or multiresistant bacteria was significantly more common in those who had been infected with ESBL-K. pneumoniae. (J-211.)
The lesson here is to do everything you can to prevent ESBL-producing organisms from becoming prevalent in your institution.
Campylobacter
Quinolone resistance is also increasing in Campylobacter. Of 722 human C. jejuni isolates in Minnesota, 37 (5%) were resistant to both nalidixic acid and ciprofloxacin. Resistance was associated with recent foreign travel, particularly to Mexico, but not with recent antimicrobial use. (C-20.)
The prevalence of C. jejuni and C. coli isolated from humans in the Basque region of Spain resistant to nalidixic acid increased from approximately 1% prior to 1990 to 29% in 1991. This was coincident with the licensing in Spain of a fluoroquinolone, enrofloxacin, for use in animals. Testing of ciprofloxacin started in 1993 and by 1996, 534 (81.6%) of 654 isolates were resistant to this drug. (C-21.)
These observations suggest that the common strategy of using a fluoroquinolone for empiric therapy of suspected bacterial diarrhea may soon prove ineffective.
Clostridium difficile
In 192 asymptomatic subjects with Clostridium difficile in their stool, the incidence of developing colitis was 1.0% within a mean of 1.5 weeks of observation, while 3.6% of 618 initially culture negative subjects did so within 1.7 weeks (P = 0.021). Among those who received antibiotics, the incidences were, respectively, 1.1% and 4.5% (P = 0.024). These data suggest that primary asymptomatic colonization with C. difficile may be protective against the development of associated colitis. (IDSA-41.)
DNA fingerprinting of fecal isolates of Clostridium difficile from 27 patients with a second clinical episode of colitis occurring five days to two months after the first found that at least 56% were due to reinfection (i.e., had a different DNA type) rather than relapse. (J-152.) Similar results were found in another, much smaller, study. (IDSA-643.)
Pseudomonas aeruginosa
A rapid increase in resistance to imipenem among P. aeruginosa isolates at the University of Maryland was demonstrated to be primarily the result of nosocomial transmission, rather than selective pressure from the use of imipenem. (C-27.)
Exposure of P. aeruginosa to an eluate from siliconized latex urinary catheters resulted in the development of resistance to imipenem which was associated with alterations in its outer membrane proteins. There was no evidence of degradation of the carbapenem or of beta lactamase induction. (C-40.)
Enterococcus
A prospective multicenter study of 375 cases of enterococcal bacteremia found that prior receipt of corticosteroids or vancomycin were independent risk factors. The 14-day mortality was 20%, and multiple underlying problems were independent risks for death. However, vancomycin resistance was also an independent risk for mortality (OR, 2.3; 95% CI 1.2-4.3; P = 0.007). (J-6.) A nationwide (U.S.) case control study found that, in addition to multiple underlying medical conditions, prior use of vancomycin or anti-anaerobe agents were important risk factors for VRE bacteremia and a poor clinical outcome. (J-30.) Modeling of VRE infection rate and antimicrobial use found that controlling the use of vancomycin and anti-anaerobe drugs would yield the greatest reduction in VRE infections. (J-32.)
A retrospective study including 100 patients with vancomycin-resistant enterococcal bacteremia found that the mortality was 48% among patients who received specific treatment for this infection and 48% in those who did not. (J-7.)
Resistance to vancomycin in Enterococcus faecium is almost invariably associated with high-level resistance to penicillins. The reason for this observation has now been demonstrated to be that penicillin resistance, caused by mutations and/or overexpression of the pbp 5 gene, is transferable and is closely linked to the vanB operon. (LB-17.)
A 44-year-old woman with endocarditis due to a strain of E. faecalis with high level resistance to aminoglycosides was successfully treated with a combination of vancomycin, imipenem, and ampicillin. (IDSA-572.)
Stenotrophomonas
Of 36 strains of Stenotrophomonas maltophilia, 100% were susceptible to minocycline (MIC90 = 1.0 mcg/mL); only 28% were susceptible to doxycycline. (E-142.)
Staphylococcus
Prospective analysis of 222 patients with community-acquired S. aureus bacteremia found that 42 (19%) had endocarditis. Mortality did not differ, even among those with endocarditis, between those infected with methicillin-susceptible or resistant strains. However, patients with methicillin-susceptible S. aureus bacteremia treated with nafcillin had no recurrences, while there was a 14% recurrence rate in those treated with vancomycin. (IDSA-439.)
VISA (vancomycin intermediate Staphylococcus aureus) was the focus of several papers. Such strains are characterized by an MIC of 8.0 mcg/mL. (LB-14, 15, 16.)
Brucella
Streptomycin is ordinarily recommended as the aminoglycoside of choice in the treatment of brucellosis. However, assays for monitoring blood levels are not widely available, and streptomycin may cause significant vestibular toxicity. Thus, the knowledge that a more accessible aminoglycoside, such as gentamicin, is also effective would be of great value. Forty patients with brucellosis (25 blood culture positive, 15 diagnosed serologically) were randomized to receive doxycycline 100 mg bid for 30-45 days (90 days if spondylitis was present) plus either streptomycin (1 g IM daily for 14 days) or gentamicin (240 mg IM daily for 7 days). Patients with endocarditis or central nervous system infection were excluded. There were two (of 22) treatment failures and two relapses in the streptomycin group and no treatment failures (of 18 patients) and one relapse in the gentamicin group. (K-65.)
Helicobacter
The Enterotest ("string test") proved a useful, nonendoscopic method to detect Helicobacter pylori in gastric contents. (D-147.)
Saccharomyces
"Saccharomyces boulardii," which has been administered for the treatment of diarrhea, including that due to Clostridium difficile, is actually an asporogenous strain of Saccharomyces cerevisiae and is moderately virulent in murine models of infection, suggesting the possible need for caution in its use. (B-39.)
Candida
The predictive value of antifungal susceptibility testing remains a matter of investigation and debate. The results of Candida susceptibility testing using the E-test were superior to the recommended standard macrobroth dilution test in predicting response to therapy. A poor response to amphotericin B administration was observed in 56% whose MIC was greater than 0.25 mcg/mL and in only 15% with a lower MIC. (IDSA-14.)
Susceptibility testing using a microtiter method correctly predicted response in 69% and failure in 71% of 114 patients treated for significant fungal infections. (IDSA-145.)
At one institution using the E-test routinely for antifungal susceptibility testing of Candida isolated from the blood stream, however, there was no difference in overall or attributable mortality between infections with susceptible or resistant Candida regardless of antifungal therapy. (IDSA-141.)
The outcome of treatment of 105 patients with candidemia with amphotericin B was correlated with the results of susceptibility tests. Of patients whose isolate had an MIC greater than 1 mcg/mL, 100% had microbiologic failure (persistence of candidemia despite at least 3 days of therapy), while only 28% of those with a lower MIC failed (P = 0.003). The specificity for failure of an MIC greater than 1 mcg/mL in predicting failure was 100%, as was the positive predictive value, while the sensitivity was only 15%. (E-81.)
Thirty-one patients with candidemia, 55% of whom were candidemic for more than two days, underwent serial ophthalmological examinations. Chorioretinitis was present in five (16%) on initial examination and in an additional three (18%) of 17 after two weeks. No additional cases were detected during follow-up at weeks four, 12, or 24; overall, eight (26%) of 31 patients developed Candida chorioretinitis. Thus, ophthalmologic follow-up should continue for at least two weeks in individuals with initially negative examinations. (IDSA-451.)
One hundred twenty-eight patients with Candida esophagitis (78% HIV infected) were randomized to intravenous therapy with either the echinocandin MK-991 (50 mg or 70 mg daily) or amphotericin B (0.5 mg/kg/day). The echinocandin was better tolerated than amphotericin B and was associated with a response rate of 85.1% compared to 66.7% in the latter group. (LB-33.)
Three hundred sixteen hospitalized patients with asymptomatic candiduria were randomized to receive fluconazole 200 mg daily or placebo for 14 days. On day 14, candiduria was absent in 50% of fluconazole recipients and 29% of placebo recipients (P < 0.001). Better results were obtained when catheters were removed. However, two weeks later, there was no difference in the prevalence of candiduria between the two groups. (IDSA-12.)
For five days, 36 hospitalized patients with indwelling bladder catheters and candiduria were randomized to receive either fluconazole 100 mg daily or continuous amphotericin B irrigation (50 mg/1000 mL sterile water). The eradication rates 14 days after therapy were, respectively, 75% and 64.7% (P =NS). (IDSA-140.)
Aspergillus and Fusarium
Aerosolized amphotericin B was unsuccessful in preventing invasive aspergillosis in neutropenic patients. (LM-87.) Fifty-five corticosteroid dependent patients with allergic bronchopulmonary aspergillosis were randomized to receive either itraconazole (200 mg po bid) or placebo while steroid reduction was attempted. Overall, responses were observed in 46% of itraconazole and 19% of placebo recipients (P = 0.04). (LB-32.)
In vitro studies found that azithromycin significantly enhances the antifungal activity of amphotericin B against Aspergillus and Fusarium. (IDSA-723.)
Fungi were recovered from 14/27 (52%) of water samples, 7/87 (8%) of water-related surfaces, and 12/22 (55%) of air samples obtained at two Little Rock hospitals. Among the fungi recovered from water were Fusarium spp., Alternaria spp., and Bipolaris spp. Fusarium and Zygomycetes were recovered from water-related surfaces, while air samples yielded, among others, Aspergillus (not fumigatus) and Phialophora spp. (J-93.) Fusarium was recovered from water storage, distribution, and drainage systems, as well as water obtained from patient showers, sinks, and shower drains, water faucet taps and shower heads in a tertiary care cancer center in the Houston area. Airborne dissemination of F. oxysporum was documented after showering. (J-94.)
Coccidioides immitis
A case control study found that migration to Arizona at an advanced age was an independent risk factor for developing coccidioidomycosis, as was the presence of underlying illness. (IDSA-135.)
An analysis of 389 patients with coccidioidomycosis, including 77 with severe pulmonary disease, 33 with dissemination, and 270 with mild infection not requiring hospitalization, found that the following were independent risk factors for severe pulmonary disease: diabetes mellitus, cigarette smoking in the previous six months, income below $15,000 annually, and older age. Oral antifungal therapy before admission was protective. Independent risk factors for disseminated disease were black race, low income, and pregnancy. This suggests that individuals with these risk factors should be targeted for early antifungal intervention. (IDSA-637.)
Antifungal Agents
Six hundred eighty-six neutropenic patients with persistent fever, despite 96 hours of antibacterial therapy, were randomized to receive either standard or liposomal amphotericin B. The latter was better tolerated and associated with a lower frequency of proven "treatment-emergent" (i.e., first appearing during antifungal therapy) fungal infections (5% vs 9%; P = 0.021). There was no significant difference in survival. (LM-90.)
Burn patients were given 400 mg fluconazole daily. Compared to published data, the renal clearance of the drug was similar, but the steady state volume of distribution was approximately two times greater and the total clearance 2-3 times higher. Thus, burn patients may require higher doses of fluconazole than other patients (A-89). CSF concentrations of amphotericin B after intravenous administration of four different preparations of this polyene for seven days to rabbits were undetectable. Amphotericin was, however, measurable in the brain and the concentration found after liposomal amphotericin B (Ambisome; 5 mg/kg/d) administration was 4-7 times higher than after administration of the deoxycholate preparation (1 mg/kg/d), the colloidal dispersion (5 mg/kg/d), or the lipid complex (5 mg/kg/d) (A-90).
Concomitant administration of terbinafine significantly reduces clearance of theophylline. (A-94.)
A study of PPD skin testing and boosting in employees of a hospital in an area of high TB prevalence could find no difference in mean reaction size between those remotely vaccinated with BCG (13.5 + 4.2 mm) and those not (14.4 + 4.6 mm). A significant booster effect was observed in BCG vaccinated subjects. (J-138.) This data further complicates the interpretation of PPD results in BCG-vaccinated individuals.
The sensitivity and specificity of the Amplicor PCR assay for Mycobacterium tuberculosis when applied to pleural fluid from 75 patients, 31 of whom had tuberculous pleuritis were, 27% and 98%, respectively. These values were similar to those of standard methods. Thus, the assay provided no diagnostic advantage except with regard to the speed of providing an answer (D-86). When applied to cerebrospinal fluid for the diagnosis of tuberculous meningitis, "despite being highly specific, the sensitivity . . . is limited by the low CSF mycobacterial load." (D-90.) With direct examination of sputum, the Amplicor had a sensitivity of 84.7% and specificity of 94.4% relative to culture. (D-91.)
Genetic fingerprinting studies of tuberculosis transmission found that, while smear-negative cases are 3-7 times less infectious than smear-positive cases, they are responsible for 10-20% of TB transmissions in San Francisco. (IDSA-311.) This observation raises important questions with regard to hospital policies concerning isolation of patients with suspected pulmonary tuberculous.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content