An old antibiotic for a new multiple-resistant Enterococcus faecium?

Enterococci have become an important cause of nosocomial infections and may demonstrate high-level resistance to multiple antibiotics. We present the case of a 68-year-old man with a history of small cell lung cancer, who developed bacteremia due to a strain of Enterococcus faecium. The isolate was resistant to multiple antibiotics including vancomycin, ampicillin, aminoglycosides, quinolones, and macrolides. The patient was successfully treated with doxycycline and removal of an infected central venous catheter.

Blepharoplasty.

Cosmetic blepharoplasty is performed in various subspecialties, therefore articles appear in the ophthalmic, plastic surgical, otolaryngologic, and dermatologic literature. We review the past year's articles regarding evaluation, technique, adjunctive procedures, and complications. As the various techniques of upper-eyelid and lower-eyelid blepharoplasty have taken their place in the armamentarium of the aesthetic surgeon, attention has turned toward adjunctive procedures to enhance the surgical result.

Invasive aspergillosis diagnosed by immunohistochemistry with monoclonal and polyclonal reagents.

Therapies differ for invasive infection due to the various filamentous fungi. However, histopathologic identification of Aspergillus, Pseudallescheria, Fusarium, Trichosporon, dematiacious Hyphomycetes, Candida, and Zygomycetes may be confused, and microbiologic isolation is often delayed. To improve diagnosis, we examined the utility of polyclonal and monoclonal antibodies to identify Aspergillus organisms by peroxidase immunohistochemical techniques. Tissues obtained from 68 infected patients were examined. Fungal antigen was detected in all 26 cases of aspergillosis by the monoclonal immunohistochemical test. Hyphae of Aspergillus organisms exhibited strong intracellular and cell wall staining, and extracellular localization was also observed. In contrast, negative tests were observed in tissues from five cases of infection with Pseudallescheria boydii, three with Trichosporon, three with Zygomycetes, six with Candida species, and one with Curvularia, from as well as 20 cases of nonfungal pneumonia. However, cross-reactivity to Coccidioides immitis spherules was noted in three cases. Staining of Candida was observed with the polyclonal antibody but not the monoclonal reagent. This is the first immunohistochemical demonstration of aspergillus antigen with a monoclonal reagent. The study indicates that the monoclonal immunohistochemical technique can distinguish Aspergillus species from other filamentous fungi and may facilitate the clinical diagnosis of invasive aspergillosis.

Microbiologic and serologic studies of Gardnerella vaginalis in intra-amniotic infection.

Our objective was to investigate the role of Gardnerella vaginalis in intra-amniotic infection by use of comparative, quantitative cultures on selective media and by detection of maternal antibody response. Amniotic fluid was collected from patients with intra-amniotic infection and from matched control women. In addition to media for aerobes, anaerobes, and mycoplasmas, we used V agar-selective (Remel, Lenexa, KS) to isolate G vaginalis. Acute and convalescent maternal sera were collected and assayed for antibodies by a microenzyme-linked immunosorbent assay (ELISA) prepared against whole cells of G vaginalis. Gardnerella vaginalis was isolated in the amniotic fluid of 24 (28%) of the 86 patients with intra-amniotic infection, but this was not significantly different from the isolation rate in amniotic fluid of 86 matched controls (21%). No patient exhibited G vaginalis bacteremia. The ELISA performed on paired sera of selected patients showed that 25 had intra-amniotic infection (eight G vaginalis-positive, 17 negative), and 18 were asymptomatic (seven G vaginalis-positive, 11 negative). The amount of G vaginalis antibodies detected by ELISA in acute sera was similar in all four groups. Mean changes during convalescence were small (.053-.084 optical density units) and not significantly different. Although G vaginalis is found commonly in amniotic fluid of patients with intra-amniotic infection, the data do not support a pathogenic role for this organism; however, a facilitating role in polymicrobial infection cannot be excluded.

Mycobacterial antigen detection by immunohistochemistry in pulmonary tuberculosis.

A preliminary diagnosis of tuberculosis can be established by the detection of acid-fast bacilli (AFB) and confirmed by culture of the microorganism. To evaluate an alternative method of diagnosis, the distribution of mycobacterial antigens in lung tissue specimens was characterized by an indirect peroxidase-antiperoxidase method and was compared to the detection of AFB by Ziehl-Neelsen stain. Histologic specimens were obtained from 59 hospital patients. Of nine patients with mycobacterial disease, seven had antigen detected in tissue. In two patients with tuberculous pneumonia, the distribution of mycobacterial antigens was approximately the same as that of AFB. In contrast, in four patients with caseating pulmonary granulomas, clumps of mycobacterial antigens were demonstrated in necrotic areas of the granulomas where there were few or no AFB. In one patient with Mycobacterium intracellulare infection, cross-reactive antigens stained weakly. Antigen was not found in tissue from two patients; one had miliary lung granulomas, and the second had mediastinal lymph node granulomas. Mycobacterial antigens were not detected in specimens from 50 control patients with nonmycobacterial diseases. On the basis of this study of 59 cases, immunohistochemical detection of microbial antigens appears to be useful for establishing the mycobacterial etiology of caseating pulmonary granulomas.

Serodiagnosis of invasive aspergillosis in patients with hematologic malignancy: validation of the Aspergillus fumigatus antigen radioimmunoassay.

Six hundred sixteen sera from 79 hematology patients admitted on 152 occasions were analyzed for validation of the Aspergillus fumigatus antigen radioimmunoassay (RIA). Invasive aspergillosis developed on 24 admissions of 22 patients. Maximal antigenic activity was significantly higher in patients with invasive aspergillosis than in controls (P less than .0005). At the level of antigenic activity selected as the cutoff value, the sensitivity of the RIA was 74%, the specificity 90%, the positive predictive value 82%, and the negative predictive value 85%. Antigen was detected before invasive aspergillosis was suspected during 30% of admissions and before pathological or even preliminary microbiological evidence for disease in 46%. In 17 (77%) of the 22 episodes of pulmonary aspergillosis, the RIA would have been the first positive diagnostic test for aspergillosis or would have confirmed a diagnosis established by other means. Overall, the test would have been of clinical usefulness in diagnosis, management, and prognosis in 80% of 16 fatal cases.

Detection of fungal antigen in body fluids for diagnosis of invasive aspergillosis.

In clinical trials, the diagnostic utility of a radioimmunoassay (RIA) to detect Aspergillus antigen was demonstrated in sera obtained from patients with invasive pulmonary, endothelial, and disseminated aspergillosis, in cerebrospinal fluids from patients with Aspergillus meningoencephalitis, and in bronchoalveolar lavage fluids from patients with invasive pulmonary aspergillosis. The RIA was further evaluated in three blinded, controlled clinical trials. In one, sera were collected prospectively from patients with acute leukemia. Antigenemia was detected in four patients with invasive pulmonary aspergillosis (IPA) due to A. flavus, appeared early in the course of infection of three patients concurrent with the onset and evolution of lung infiltrates, and remitted with antifungal chemotherapy. Antigenemia was not detected in three patients before the onset of IPA, in eight leukemic controls, or in the 24 normals. In contrast, seven bronchoscopies were performed in five patients with IPA; fungi were not isolated in three. In a second study, antigenemia was found in coded sera from two patients with invasive aspergillosis but not in eight controls. In the third controlled study, 616 sera from 79 hematology patients admitted on 152 occasions were analyzed for circulating fungal antigen. The diagnostic utility of the RIA was confirmed and levels of antigenemia correlated with the patients' clinical course. These studies demonstrate the utility of the Aspergillus antigen RIA for diagnosis of invasive aspergillosis in hospitalized high risk patients.

Antibody in host defense against mouse pneumonitis agent (murine Chlamydia trachomatis).

In a murine model of chlamydial pneumonia employing murine Chlamydia trachomatis, immune serum given intranasally delayed death in nude (nu/nu) mice and prevented death in nu/+ mice. Serum-derived immunoglobulin G and A fractions and immune lavage fluid fractions containing immunoglobulins G and A were effective in producing protection when used to opsonize the murine C. trachomatis inoculum. In hyperclean mice (previously made germfree and then colonized with a limited flora nonpathogenic to mice) antibody given intravenously was ineffective. The effects of antibody and nonspecific stimulation of cell-mediated immunity (after previous infection with Histoplasma capsulatum) were additive in increasing host resistance to murine C. trachomatis.

Antigen detection in the diagnosis of invasive aspergillosis. Utility in controlled, blinded trials.

Two blinded, controlled trials were done to evaluate the usefulness of fungal antigen detection for the diagnosis of invasive aspergillosis. Detection of Aspergillus fumigatus carbohydrate by radioimmunoassay was compared with antibody detection by an enzyme-linked immunosorbent assay and with diagnostic microbiologic and histopathologic procedures. In the first trial, antigenemia was detected in 4 of 6 leukemic patients with invasive pulmonary aspergillosis, but not in 8 acute leukemic controls or in 24 normal controls. Fungal antigenemia persisted for 8 to 75 days in 4 patients and seroconversion occurred at the onset of pulmonary infiltrates in 3. Antibody to A. fumigatus was detected in 2 of the 6 patients with aspergillosis, but also in 2 leukemic controls and 6 normal controls. Aspergillus species were identified in four of seven bronchoscopies done in 5 patients with invasive pulmonary aspergillosis. Prospective nasal cultures grew Aspergillus species in 4 of the 6 patients with invasive aspergillosis, but in only 1 patient was this information available before a histologic diagnosis was made. In a second trial, antigenemia was detected in 2 patients with invasive aspergillosis, and in 1 with possible invasive aspergillosis, but not in 9 controls. This study indicates that the radioimmunoassay for A. fumigatus antigen is a highly specific and moderately sensitive serodiagnostic test for invasive pulmonary aspergillosis. Prospective nasal cultures grew Aspergillus species in 4 of the 6 patients with invasive aspergillosis, but in only 1 patient was this information available before a histologic diagnosis was made. In a second trial, antigenemia was detected in 2 patients with invasive aspergillosis, and in 1 with possible invasive aspergillosis, but not in 9 controls. This study indicates that the radioimmunoassay for A. fumigatus antigen is a highly specific and moderately sensitive serodiagnostic test for invasive pulmonary aspergillosis.

Antigenemia detected in human coccidioidomycosis.

To advance the serological diagnosis of coccidioidomycosis, a radioimmunoassay was developed to detect an antigen of Coccidioides immitis and evaluated with serum samples from patients and controls. The assay antigen was fractionated from a culture filtrate of C. immitis cells by Sephacryl S300 gel chromatography. The assay used preliminary acid-heat extraction to dissociate serum-bound antigen and staphylococcal protein A as an absorbent. Antigenemia was detected in five (56%) of nine patients with coccidioidomycosis. Three of four negative tests occurred with sera from patients who had already received antifungal chemotherapy. Antigenemia was not detected in 106 control patients and normal donors. One false-positive reaction was observed in a patient with histoplasmosis. In all nine patients with active coccidioidomycosis, anti-C. immitis antibodies were detected during the initial evaluation or in the follow-up. In summary, an immunoassay for C. immitis antigenemia was developed; antigenemia was detected in five (56%) of nine patients with active coccidioidomycosis and a positive test indicated active or persistent disease.

Candidal antigen detection in pulmonary candidiasis.

To better characterize pulmonary candidiasis, lung tissue samples from 58 hospital subjects were examined by immunohistochemical analysis with antiserums to candidal mannan and cytoplasmic antigens. In nine cases of invasive pulmonary candidiasis, fungal antigens were abundant in lung specimens and were distributed in two immunohistologic patterns. Antigen dissemination beyond the lung was confirmed by detection of antigenemia by mannan radioimmunoassay in serum samples in cases of pulmonary candidiasis of each immunohistologic pattern. In contrast, minimal, focal quantities of fungal antigens were detected in lung tissue specimens from nine subjects with thrush or aspiration but without pulmonary candidiasis; no antigen was detected in 40 additional cases of noncandidal mycotic, bacterial, viral, and parasitic pneumonias. Immunohistochemical criteria that denote invasive pulmonary candidiasis include the abundance of diffuse candidal antigens in bronchiolar and alveolar fluid and cytoplasmic staining of phagocytes. When compared with routine histologic stains, the diagnostic yield in cases of invasive pulmonary candidiasis was significantly increased by detection of candidal antigenic material. This study of 58 well-documented cases shows that immunohistochemical detection of candidal antigens facilitates the diagnosis of candidal pneumonia and distinguishes clinically significant pulmonary candidiasis from noninvasive aspiration, opportunistic colonization, or specimen contamination.

Aspergillus antigen detection in bronchoalveolar lavage fluid from patients with invasive aspergillosis and aspergillomas.

Improved diagnostic techniques have been needed for pulmonary aspergillosis, a common opportunistic fungal infection with a high mortality rate. Radioimmunoassay was used in this study to detect Aspergillus antigen in bronchoalveolar lavage fluid. In four patients with invasive aspergillosis or aspergillomas, Aspergillus antigen was detected in bronchoalveolar lavage fluid. In two patients, results of fungal cultures were negative or delayed. The specificity of antigen detection in bronchoalveolar lavage fluid was 91 percent in 35 control patients with a variety of pulmonary disorders. The technique or radioimmunoassay detection of microbial antigen in bronchoalveolar lavage fluid appears promising for the diagnosis of aspergillosis.

Peritoneal dialysis complicated by Aspergillus flavus peritonitis: a role for fungal antigen serodiagnosis.

A patient is described with aspergillus peritonitis associated with peritoneal dialysis. Although the diagnosis was established by exploratory laparotomy, aspergillus antigenemia was detected retrospectively in sera obtained 15 days prior to the surgical procedure. An approach to the evaluation of a patient with culture negative peritonitis and the potential clinical usefulness of the detection of fungal antigen in serum and peritoneal fluid of patients undergoing peritoneal dialysis is discussed.

Immunodiagnosis of invasive pulmonary aspergillosis in rabbits. Fungal antigen detected by radioimmunoassay in bronchoalveolar lavage fluid.

To improve antemortem diagnosis of invasive pulmonary aspergillosis, we used a radioimmunoassay to detect an Aspergillus fumigatus antigen in bronchoalveolar lavage fluid (BALF) and evaluated this technique in a model of disseminated aspergillosis. Antigen was detected in 91% of BALF samples obtained from 11 rabbits with major pulmonary aspergillosis. In an additional 10 rabbits with only minor pulmonary involvement, antigen was detected in 40% of concentrated lavage samples. In contrast, antigenlike activity was found in only 1 of 17 BALF samples from control animals with systemic candidiasis, in none of 9 control animals with staphylococcal pneumonia and in none of 10 normal control animals. Although antigen was present in the serum of 76% of animals infected with Aspergillus, 27% of those with major pulmonary involvement had antigen detected in BALF alone. We found that an extracellular microbial antigen can be detected in BALF and that this technique in the disseminated aspergillosis model is both sensitive and specific for invasive pulmonary disease.

Immunologic studies of disseminated infection with Aspergillus fumigatus in the nude mouse.

Immunity to infection with Aspergillus fumigatus conidia was studied in the athymic nude (nu/nu) mouse. The intratracheal, intravenous, and intraperitoneal routes of infection were used. Nu/nu mice were not significantly more susceptible than their heterozygous (nu/+) littermates. However, only nu/+ animals could be protected by prior immunization and developed positive, A. fumigatus-specific skin tests. Protection could be transferred to nu/nu recipients by immune glass-adherent cells. Thus, immunity to invasive aspergillus infection is multifactorial and depends on a glass-adherent cell population and T-cells.

Antigenemia detected by radioimmunoassay in systemic aspergillosis.

Because of difficulties in antemortem diagnosis of systemic aspergillosis, a radioimmunoassay to an Aspergillus fumigatus carbohydrate was developed and evaluated in patients with mycotic or bacterial infections. Antigenemia was detected in sera obtained antemortem from four of seven patients with systemic aspergillosis and in pleural fluid from an Aspergillus empyema but not in control sera or pleural fluid from 43 patients or 27 normal donors. When characterized with reference to onset of disease, antigenemia was an early sign of infection. This study shows the usefulness of the Aspergillus antigen radioimmunoassay for early, specific immunodiagnosis of systemic aspergillosis.

Immunodiagnosis of systemic candidiasis: mannan antigenemia detected by radioimmunoassay in experimental and human infections.

A radioimmunoassay (RIA) that detects candida mannan was developed so that immunodiagnosis of systemic candidiasis could be improved. The RIA was evaluated in an animal model of disseminated disease and in a panel of patient sera. Mannan antigenemia was detected with the RIA in 52% of 29 rabbits with systemic candidasis, but not in 60 normal rabbits or 31 rabbits with systemic aspergillosis. In an evaluation of human sera, mannan antigenemia was detected in five of 11 patients with systemic candidiasis, one of three patients with invasive gastrointestinal candidiasis, and one patient with a sustained candidemia associated with an infected intravenous catheter. Mannan was not detected in sera from 11 patients with superficial candida infections, seven patients colonized with Candida, three patients with chronic mucocutaneous candidiasis, eight patients with other systemic mycoses, or 22 normal donors. This study demonstrates the utility of this RIA for early, specific immunodiagnosis of invasive candidiasis.

Immunodiagnosis of systemic aspergillosis. I. Antigenemia detected by radioimmunoassay in experimental infection.

Because systemic aspergillosis is difficult to diagnose ante mortem, a study to improve immunodiagnosis was undertaken in a rabbit model of disseminated infection. We found that the predominant humoral response of infected animals was directed against four Aspergillus antigens identified by crossed immunoelectrophoresis. One of these antigens, a cell-wall carbohydrate, was purified by gel-filtration chromatography and was used to develop a radioimmunoassay. The sensitivity of this assay was increased by testing for serum-bound antigen as well as for free antigen. When the sensitivity of the RIA was evaluated in the animal model, antigenemia was detected in 78% of 51 rabbits with disseminated infection and ante morten in 85% of 42 rabbits with lethal infection. By contrast, with immunoprecipitin analysis only eight of 51 rabbits were positive for antigen, and six of 51 rabbits were positive for Aspergillus antibody. The specificity of the RIA was also tested. Negative controls for antigen included sera from 76 normal rabbits and sera from 25 rabbits with systemic candidiasis. The Candida control group is pertinent because 48% of these rabbits had specific Candida antigenemia detected by a mannan RIA. This study demonstrates that Aspergillus antigenemia occurs during the course of experimental disseminated aspergillosis and illustrates the potential of an Aspergillus antigen RIA for sensitive, specific immunodiagnosis of human infections.

Mannan antigenemia in the diagnosis of invasive Candida infections.

Because it is often difficult to diagnose invasive Candida infections, a sensitive hemagglutination inhibition assay to detect the surface antigen, mannan, was developed. Mannan antigenemia was detected early in the course of infection in 4 of 14 patients with systemic candidiasis and 2 of 5 patients with invasive gastrointestinal candidiasis. Mannan was not detected in 48 patients with noninvasive Candida or other systemic mycotic infections or in 99% of 234 patients in other control groups. Mannan antibodies were almost universally present in both candidiasis and control groups. In four patients with systemic candidiasis, an early period of mannan antigenemia was followed by a rapid rise in mannan antibody titer. These findings suggest that antemortem diagnosis would be improved in one-third of cases of invasive Candida infection detected by the hemagglutination inhibition assay. A positive test for serum mannan would be an early and specific signal of invasive disease.