ICAAC/IDSA/ASTMH 2003

Conference Coverage

The following summary of selected abstracts from 3 meetings will be published in multiple parts. The 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC) met in Chicago September 14-17, 2003. The Infectious Disease Society of America (IDSA) met in San Diego October 9-12, 2003. The American Society of Tropical Medicine and Hygiene met in Philadelphia December 3-7, 2003. — Stan Deresinski, MD, FACP

Viral Infection

Viral Hepatitis

A case control study found that the risk of HCV transmission to health care workers after percutaneous exposure increases with the volume of blood injected and, probably, with higher titer of HCV in the blood of the source patient (ICAAC V-772).

Approximately one-fifth of 248 saliva samples from 12 HCV-infected subjects contained HCV RNA. Salivary shedding was strongly associated with serum HCV RNA concentration and may be associated with the presence of periodontal disease. Mother-to-child transmission of HCV at birth occurred at an incidence of 4.1% and was not increased in the presence of maternal coinfection with HIV (ICAAC V-773, V-774).

Of 209 Alaskan natives with untreated chronic HCV infection who were followed for a mean of 9.4 years, only 7 (3%) spontaneously cleared their viremia after an average of 7.4 years (0.45% per year). Patients who experienced spontaneous resolution of viremia had significantly lower viral loads than those who did not (IDSA 508).

The prevalence of diabetes mellitus is increased in patients with chronic HCV infection, regardless of the presence or absence of HIV coinfection (IDSA 594).

The median duration of survival in 43 HIV-infected patients with HCV coinfection after the onset of ascites was only 123 days (range, 10-2355 days) (ICAAC V-782).

CD4 counts are reduced in HIV-seronegative patients with hepatic cirrhosis, but CD4 percentage is preserved (IDSA 603).

A prospective study found that serum HCV RNA decay after 4 weeks of combination therapy correlated with outcomes at 6 months, regardless of the presence or absence of HIV coinfection. None of 40 HIV/HCV-coinfected patients treated with ribavirin and interferon alpha who failed to achieve HCV RNA undetectability at 24 weeks had sustained virological responses. Fifty percent of those with such a response at 24 weeks did have a sustained response (ICAAC V-777, V-1726).

Only 1 of 11 HIV/HCV-coinfected patients, most with unfavorable HCV genotype, had a sustained virologic response after treatment with pegylated interferon alpha, but all 10 with initial and follow-up liver biopsies demonstrated histologic and biochemical improvement (ICAAC V-1724).

Renal dysfunction was observed in 3 patients undergoing treatment with interferon alpha and ribavirin for chronic HCV infection (IDSA 593).

Seventy percent of HIV/HCV-coinfected patients had serological evidence of prior HBV infection, and 36% of those had isolated antibody to HBc. Separately, one-half of HIV-infected patients had isolated antibody to HBc, and that was much more commonly observed in those with HCV coinfection. However, only 1 of 48 HIV-infected individuals with isolated antibody to HBc had HBV DNA detectable in plasma (IDSA 600, 601).

Eighty-four HIV/HBV-coinfected patients who received tenofovir as part of their antiretroviral therapy were followed for a median duration of 9 months. At baseline, although all but 4 patients were receiving lamivudine, HBV DNA was detectable in 84.5%. However, HBV DNA became undetectable in 37.1% after initiation of tenofovir. This observation was confirmed in a study of 32 HIV/HBV-coinfected patients who received tenofovir as part of an antiretroviral regimen. Administration of tenofovir in this cohort was associated with rapid decreases in plasma HBV DNA regardless of lamivudine therapy and the presence of YMDD mutations (ICAAC V-783, V-784).

Respiratory Tract Infections

Viral Respiratory Infection

Epidemiology

In households containing at least 1 child aged 6 months to 5 years enrolled in childcare, the use of alcohol-based gels for hand hygiene was associated with lower rates of reported secondary respiratory illness (IDSA LB-8). A randomized trial using a variety of commercial products, however, has failed to demonstrate benefit (Ann Intern Med. 2004;40:321).

Diagnosis

Nasopharyngeal washes from 90 pediatric inpatients were tested by commercial antigen detection kits for influenza A and B, RSV, and parainfluenza virus, as well as with a DNA microarray bearing the most conserved sequences of each of 934 viruses. The DNA microarray was as sensitive and specific for detection of these viruses but also detected viruses not tested for by the commercial kits, including rhinovirus, coronavirus, and metapneumovirus (IDSA LB-3).

Influenza

Increasing evidence demonstrates the value of vaccination against influenza in broad swaths of the population. For instance, vaccination of children against influenza was associated with a 43-67% reduction in influenza incidence in household members. A primary target group, the elderly often have impaired serological responses to vaccination. Increasing the dose of standard inactivated influenza virus vaccine was associated with enhanced antibody responses in ambulatory elderly subjects (IDSA 497, 530).

Another very important target of vaccination is health care workers, but a survey of nursing home health care workers in southern California found that only 34% received the influenza vaccine. Furthermore, only 54% of the workers received paid sick leave, providing incentive for the other 46% to work while ill, thus increasing the risk of influenza transmission (IDSA 532).

Clinical differentiation of influenza from other respiratory illness is an important, but difficult, task. Multivariate analysis of data from a prospective study of almost 40,000 patients found that temperature > 100.4° F with combined complaints of fever chills and sweating, headache, and cough were of predictive value in differentiating influenza-like illness from influenza itself. However, even many patients given a diagnosis of influenza are inappropriately prescribed antibiotics. Antibacterial agents were prescribed to 40% of patients aged 5-49 years with a diagnosis of influenza in the United States from 1997 to 2000. This represented 2.4 million prescriptions (ICAAC V-792, A-1367a).

A fatal case of infection caused by avian influenza virus H7N7 in a Dutch veterinarian occurring in association with an outbreak in chickens is described (ICAAC V-480).

Paramyxoviruses

A prospective study of children younger than 3 years hospitalized in Quebec because of acute respiratory tract infection in the winter/spring of 2002-2003 found that 5.8% were due to human metapneumovirus, 21.6% to influenza A, and 51% to respiratory syncytial virus. Two-thirds of children with metapneumovirus infection, which tended to occur later in the season than RSV infection, had bronchiolitis, and 16.7% had pneumonia. Metapneumovirus exposure during childhood appears to be almost universal. Eighty percent of children tested in Israel had antibody to human metapneumovirus by age 2 years (ICAAC V-478, V-479).

RSV and human metapneumovirus infections are not restricted to infants and children. A PCR assay detected RSV or, at lesser frequency, human metapneumovirus, in approximately one-fifth of adults with acute exacerbations of chronic obstructive lung disease. Metapneumovirus was detected as frequently as was influenza A virus (ICAAC L-1582).

The relative roles of direct cytotoxicity and the immune response in causing respiratory impairment in infants with RSV infection remains incompletely defined. In one study, the RSV load in respiratory secretions was shown to correlate with disease severity in previously healthy infants (ICAAC V-788).

An evaluation of 3 kits for rapid detection of RSV (Directigen RSV, Directigen EZ RSV, and Now RSV) found that all gave comparable results and performed well when compared to shell vial culture results (ICAAC V-790).

Hantavirus

A randomized trial of intravenously administered ribavirin to patients with hantavirus cardiopulmonary syndrome was discontinued because of slow enrollment. The investigators concluded, however, that ribavirin was probably ineffective in the treatment of this infection (ICAAC V-482).

Follow-up of survivors of hantavirus pulmonary syndrome found that one-third had significant sequelae. These included mild-to-moderate changes in pulmonary function. Eleven of 33 had developed late-onset progressive proteinuria. IgM antibody to hantaviruses disappeared by 12 months (IDSA 843).

Mimivirus

Mimivirus, an intra-amoebal organism, is a giant virus that can resemble small Gram-positive cocci. Its detection in bronchoalveolar lavage fluid of a patient with pneumonia, as well as serological studies in patients with pneumonia, implicate a mimivirus as a possible pathogen (IDSA 102).

Picornavirus

In a prospective study, 7 of 9 upper respiratory tract infections caused by picornaviruses were associated with an exacerbation of multiple sclerosis. This was true of only 2 of 12 (P = .01) infections in which picornavirus was not detected (IDSA 735).

SARS

Prolonged duration of exposure and proximity of contact increase the risk to health care workers of developing SARS. In Toronto, 8 of 32 (25%) nurses who entered a room with a SARS patient developed SARS; none died. The probability of a nurse developing SARS was 6% per shift worked. Being present at activities related to intubation increased the risk, and the use of a mask, particularly N95, was protective. Very few infections of health care workers were asymptomatic. A serological study of a large number of exposed health care workers found asymptomatic SARS CoV infection in < 2%. In one setting, however, none of 110 health care workers with a median 3 exposures each to 1 of 6 patients with confirmed SARS in the United States had serological evidence of SARS infection, despite the fact that a large proportion of the exposures were unprotected (ICAAC K-1314, K-1315a, K-1315c, IDSA LB-17).

Molecular testing of respiratory specimens from 117 California patients with pneumonia of unknown etiology who met CDC criteria for SARS during the time of the 2003 outbreak with pneumonia of unknown etiology yielded evidence of a pathogen in 24%. The vast majority of pathogens identified were influenza A virus. Influenza B was detected in 1, RSV in 4, and parainfluenza virus in 1. None had evidence of infection with the SARS coronavirus (IDSA 101).

In an evaluation of 117 hospitalized patients with SARS in the Toronto outbreak, SARS CoV RT-PCR on specimens from multiple sites was positive in only 54%, with the  result more likely to be positive on days 9-11 than earlier. Stool and lower respiratory specimens were each positive in 63-65%. Convalescent serology became positive in 35 of 39 (90%) at a mean time of 26 days. A separate Toronto study reported that stool specimens were positive as early as 3 days after onset of illness and persisted so for up to 44 days (ICAAC V-488a, A-1367a, V-485b).

In the absence of the availability of rapid diagnostic testing for SARS, a strategy was frequently adopted involving rapid testing for alternative causes of respiratory infection and assuming the absence of SARS if one of those tests was negative. Two studies that found a high frequency of coinfection in SARS patients make this strategy no longer tenable. Thirty-four percent of patients with laboratory-confirmed SARS studied in Toronto had evidence of coinfection with another respiratory pathogen. In a separate report, 32 of 111 (29%) Toronto patients with laboratory-diagnosed SARS had evidence of coinfection with another respiratory pathogen. This included 19% DNA or serology positive for C pneumoniae, 7% for M pneumoniae, and 1.8% with additional viral infection (IDSA LB-16, ICAAC K-1315d).

Nine Toronto SARS patients treated with Interferon Alfacon-1, a consensus interferon, and corticosteroids were compared to 13 controls given steroids alone in a nonrandomized manner. The interferon recipients had a shorter time to 50% resolution of chest x-ray abnormalities (4.0 vs 11.5 days) and shorter duration of needed supplemental oxygen therapy (ICAAC K-1315e; JAMA. 2003;290:3222).

Postmortem examination of 19 Toronto patients with SARS dying < 60 days after onset of illness found RT-PCR evidence of SARS CoV in the lungs of all. Also virologically positive were the liver (41%), spleen (53%), bowels (73%), lymph nodes (69%), and kidney (38%). All samples were negative in 1 patient who died > 100 days after onset. There was a significant relationship between the viral load in the lungs and the presence of multiorgan dissemination. There was no correlation between viral loads and the antemortem use of ribavirin or corticosteroids (ICAAC K-1315b).

Post-traumatic stress disorder was frequently observed in both patients who had recovered from SARS and their caregivers (ICAAC K-750a, V-796a).

Varicella Virus

In the United States, an estimated average of 2.6 million to 4.3 million cases of varicella occurred annually from 1970 to 1994. The estimated annual average incidence of herpes zoster over the same period, prior to the availability of the varicella vaccine beginning in 1995, was 263,724-448,331 (IDSA 899).

The estimated efficacy of varicella vaccine over 10 years of follow-up was 94.4% for those who received a single injection and 98.3% after 2 injections (P < .001). In settings of household exposure, varicella vaccination was > 95% effective in preventing moderate and severe disease and about 80% effective in preventing all disease. Varicella introduction by a vaccinated child was only approximately one-half as likely to be transmitted as that introduced by an unvaccinated child. Similarly, a case-control study combined with active surveillance found that the overall efficacy of varicella vaccine was 87% but decreased from 97% in the first year to 72% in the sixth year after vaccination. Disease was mild in 94% of vaccines and 64% of unvaccinated children (IDSA 895, 900, 891).

While breakthrough infections do occur in some varicella-vaccinated individuals, they are generally mild and limited. Follow-up of 7461 children 5 years after varicella vaccination given at 12-24 months of age detected a total of 1079 cases of breakthrough varicella infection, for an average annual incidence of 2.3 per 100 person-years. The average incidence of cases with > 50 lesions was only 0.47 per 100 person-years (IDSA 892).

Bacterial Infections

Paranasal Sinusitis

Treatment of patients with acute sinusitis with azithromycin 500 mg daily for either 3 or 6 days yielded similar rates of clinical efficacy. The end-of-treatment clinical response rates among patients with pretreatment sinus puncture cultures yielding S pneumoniae were 89% and 93%, respectively (ICAAC L-1380).

Eighty-seven percent of 71 patients with acute maxillary sinusitis or community-acquired pneumonia (CAP) from whom erythromycin-resistant S pneumoniae was isolated experienced clinical cure after treatment with telithromycin. The cure rate for all S pneumoniae infections was 93%. Similar cure rates were achieved regardless of the mechanism of macrolide resistance and included 3 of 3 patients with isolates containing ermB plus mefA (ICAAC L-467).

A pathogen was detected in sinus aspirates of 52% of 775 patients enrolled in a noncomparative open trial of treatment of acute bacterial sinusitis in adults with amoxicillin/clavulanate 2000 mg/125 mg given b.i.d. for 10 days. One-third (133 patients) of those with an identified pretherapy pathogen were infected with S pneumoniae, 19 of which had penicillin MICs of 2 to > 32 mg/mL. Amoxicillin MICs of 4 and 8 mg/mL were found for 4 and 2 isolates, respectively. The success rates at end of treatment and at follow-up for patients with pneumococcal infections were 93% and 96%, respectively (IDSA 300).

Treatment of patients compiled from 4 studies with acute sinusitis due to penicillin nonsusceptible S pneumoniae with amoxicillin/clavulanate 2000 mg/125 mg twice daily was associated with clinical success in 46 of 48 (96%), including 12 of 14 (86%) amoxicillin MICs of 4-8 mg/mL (ICAAC L-1382).

Streptococcal Pharyngitis

Fifteen of 17 (88%) children with acute uvulitis as evidenced by the presence of a beefy red swollen uvula, were found to be infected with Group A streptococci (IDSA 783).

Two separate meta-analyses of randomized trials found that the likelihood of bacteriologic eradication and of clinical cure was approximately twice as high in children, as well as in adults and adolescents, with Group A streptococcal tonsillopharyngitis treated with an orally administered cephalosporin compared with penicillin treatment. Nationwide, 5.4% of S pyogenes isolates were macrolide resistant in 2002-2003 (ICAAC G-1546. L-1383, IDSA 210).

Acute Exacerbations of Chronic Bronchitis/ Cystic Fibrosis

"Atypical pathogen" DNA was identified by PCR in the sputum of 14 of 161 (8.7%) patients with acute exacerbation of chronic bronchitis, including 7 with Legionella pneumophila, 4 with Chlamydiophila pneumoniae, and 2 with Mycoplasma pneumoniae. Separately, RSV or human meta-pneumovirus were detected in approximately one-fifth of patients with acute exacerbations of obstructive lung disease (ICAAC L-1585, L-1582).

In a randomized trial, amoxicillin/clavulanate 2000 mg/125 mg b.i.d. for 5 days was as effective as 875 mg/125 mg b.i.d. for 7 days in the treatment of patients with acute exacerbations of chronic bronchitis. Five-day treatment with moxifloxacin was superior to comparator agents (amoxicillin, clarithromycin, cefuroxime axetil) in patients with severe acute exacerbations of obstructive lung disease when the results of several randomized trials were combined for analysis (ICAAC L-1592, L-1593).

In a randomized trial, both meropenem/tobramycin and ceftazidime/tobramycin improved pulmonary status and reduced bacterial burden in patients with acute pulmonary exacerbations of cystic fibrosis. Significantly more children had improved airflow at 7 days in the meropenem/tobramycin group (ICAAC L-1598).