Kathleen A. Steger

Nutritional outcome and pneumonia in critical care patients randomized to gastric versus jejunal tube feedings

ObjectiveTo compare nutritional status, gastric colonization, and rates of nosocomial pneumonia in ICU patients randomized to gastric tube feeding vs. patients fed by an endoscopically placed jejunal tube. DesignRandomized, prospective study. SettingMedical and surgical ICUs at Boston City Hospital; surgical ICU at University Hospital. PatientsOf the 38 study patients, 19 were randomized to gastric tube feeding and 19 were randomized to an endoscopically placed jejunal tube. The two groups were similar in age, sex, race, underlying disease, and type of surgery. ResultsThe two patient groups were similar in number of days fed, duration of ICU stay, duration of mechanical ventilation, days of antibiotic therapy, and days with fever. Compared with the gastric group, the jejunal group had more patients with circulatory shock on admission (79% vs. 68.4%), higher admission Acute Physiology Score (24.0 vs. 21.7), and fewer patients with pneumonia at randomization (26.3% vs. 31.6%). The jejunal group received a significantly higher percentage of their daily goal caloric intake (p = .05), and had greater increases in serum prealbumin concentrations (p <.05) than the patients with gastric tube feeding. Although the jejunal tube group had more days of diarrhea (3.3 ±PT 6.6 vs. 1.8 ±PT 2.9), this difference was not statistically significant. Nosocomial pneumonia was diagnosed clinically in two (10.5%) patients in the gastric tube group and in no patients in the jejunal tube group. ConclusionsPatients fed by jejunal tube received a significantly higher proportion of their daily goal caloric intake, had a significantly greater increase in serum prealbumin concentrations, and had a lower rate of pneumonia than patients fed by continuous gastric tube feeding.

Preventing nosocomial pneumonia: State of the art and perspectives for the 1990s

In the 1980s, nosocomial pneumonia became the second most common nosocomial infection in the United States. Gram-negative bacilli and Staphylococcus aureus were the most frequently isolated bacteria. Methods to improve the diagnostic sensitivity and specificity included transtracheal aspirates and bronchoscopy with protected specimen brush or bronchoalveolar lavage. Multivariate analysis was used to identify independent risk factors for pneumonia and fatality in different subsets of high-risk patients. Gastric pH and colonization were evaluated as risk factors for pneumonia in mechanically ventilated patients. Colonized respiratory therapy equipment and contaminated tubing condensate and in-line medication nebulizers were suggested as possible sources of nosocomial pathogens. Staff education programs, the use of barrier precautions, and selective decontamination of the digestive tract were associated with reduced rates of lower respiratory tract infection. Despite a decade of progress in our understanding of nosocomial pneumonia, progress in the 1990s will undoubtedly include molecular epidemiologic techniques, appropriate application of risk factor data, and the use of new methods for the diagnosis of pneumonia. Prevention strategies should focus on more effective infection control techniques, improved invasive devices/equipment, and the judicious use of antibiotics for treatment and prophylaxis.

Influenza Vaccination of Human Immunodeficiency Virus (HIV)-Infected Adults: Impact on Plasma Levels of HIV Type 1 RNA and Determinants of Antibody Response

We assessed the effect of influenza vaccination on plasma levels of human immunodeficiency virus type 1 (HIV-1) RNA and the impact of age, plasma HIV-1 RNA level, CD4 cell count, and anti-HIV therapy on immune response. Forty-nine adults (mean age, 38.7 years; mean CD4 cell count +/- SD, 190 +/- 169/mL; mean plasma HIV-1 RNA level +/- SD, 154,616 +/- 317,192 copies/mL) were immunized. Elevations of > or = 0.48 log in plasma HIV-1 RNA levels occurred in two (4%) of 49 subjects within 4 weeks of vaccination. A fourfold or greater increase in antibody titer occurred in 13 (45%) of 29 subjects, correlating directly with CD4 cell count (P = .002) and inversely with plasma HIV-1 RNA level (P = .034). By multivariate analysis, CD4 cell count was a stronger predictor of antibody response than was plasma HIV-1 RNA level. We conclude that increases in plasma HIV-1 RNA levels following influenza vaccination are rare and transient and that antibody response is impaired with CD4 cell counts of < 100/mL and plasma HIV-1 RNA levels of > 100,000 copies/mL. Prospective trials are needed to evaluate the impact of highly active therapy on immune response after vaccination.

HIV/AIDS patients’ perspectives on adhering to regimens containing protease inhibitors

AbstractOBJECTIVE: To gather qualitative data regarding HIV/AIDS patients’ perspectives about HIV-1 protease inhibitors (PIs), and about their experiences taking and adhering to regimens containing PIs. DESIGN: Six focus groups of persons under care for HIV were conducted between September and November 1996 regarding participants’ knowledge, awareness, experiences when taking, and adherence to antiretroviral regimens containing PIs. An identical discussion guide was used to facilitate all six groups. Focus group proceedings were audiotaped, transcribed, coded for themes, and analyzed qualitatively. SETTING: HIV/AIDS practices of three teaching hospitals and two community health centers. PATIENTS/PARTICIPANTS: Fifty-six patients with HIV disease: 28 men and 28 women. MEASUREMENTS AND MAIN RESULTS: Knowledge and positive impressions of PIs were prevalent among this diverse group of persons with HIV, and did not differ by race/ethnicity or gender. Most knew that these were new, potent medications for treating HIV/AIDS. Networks of persons with HIV and medical providers were the most important information sources. Those taking PIs were aware that adherence to the regimen is important, and most were using special strategies to maximize their own adherence, but expressed considerable frustration about the central role these medication regimens had assumed in their life. A subset who did not believe they would adhere to these regimens had declined treatment with them. Motivating factors for taking and adhering to these complex regimens were improving CD4 counts and viral loads and the patient-provider relationship. CONCLUSIONS: Among those with HIV/AIDS, awareness of PIs and their effectiveness is substantial, owing to the impact of informal networks and medical providers. This early positive “reputation” of PIs may enhance motivation for adherence. Those who are taking PIs invest substantial effort adhering to these complex regimens, but resent the need to make medications the focus of their lives.

Nosocomial pneumonia : epidemiology and infection control

Elderly, debilitated, or critically ill patients are at high risk for hospital acquired or nosocomial respiratory tract infection. Gram-negative bacilli,Staphylococcus aureus, and anaerobes colonizing the oropharynx are the most frequent etiologic agents. Colonization of the oropharynx may be related to the patients age, underlying disease, nutritional status, prior exposure to antibiotics, supine position, and gastric colonization. Nosocomial pathogens may also be acquired from the hands of hospital personnel, contaminated equipment or fluids. The absence of sensitive and specific methods for accurate diagnosis remain a concern. Despite treatment with appropriate antimicrobial therapy, there is a high mortality and morbidity. Measures for the prevention of nosocomial pneumonia should include compliance with infection control principles, appropriate use of antibiotics, proper patient position, and removal of potential sources of cross colonization.

Provider attitudes regarding participation of women and persons of color in AIDS clinical trials.

Provider attitudes and perceptions that may influence recruitment and enrollment of diverse patients into AIDS clinical trials were examined by conducting a cross-sectional survey of all HIV/AIDS providers at a municipal teaching hospital. Providers were less likely to feel confident explaining trials to non-English-speaking patients (p < .05). Providers also reported being more confident of their ability to give an overview of clinical trials in culturally appropriate terms to white patients than to patients of other races/ethnicities (p < .05). Many providers perceived the interest in clinical trials by African American (25%), Latino (14%), and Haitian patients (30%) to be lower; and primarily cited suspicions about clinical research as the reason. Some providers (13%) perceived that women with HIV/AIDS are less interested in clinical trials. Despite these perceptions, all providers reported that they are just as likely to inform women and African Americans about available clinical trials; a small proportion reported that they were less likely to inform Latinos (6%) and Haitians (11%). None of these findings differed significantly by provider race, gender, HIV experience, languages spoken, or specialty. Underrepresentation of minorities and women in AIDS Clinical Trials may partially result from attitudes and perceptions of providers.

Response of Lymphoepithelial Parotid Cysts to Antiretroviral Treatment in HIV-Infected Adults

Salivary gland enlargement in adults with HIV infection may be caused by viruses, opportunistic pathogens, neoplasms, the Sjogren syndrome, the diffuse infiltrative CD8 lymphocytosis syndrome, or lymphoepithelial parotid cysts [1-14]. The incidence of lymphoepithelial parotid cysts increases in HIV-infected patients, and the cysts are more likely to be chronic, large, bilateral, and multiloculated in these patients than in adults who are not infected with HIV [1, 4-14]. Because most reports of lymphoepithelial cysts in HIV-infected adults have been from retrospective studies of patients who had surgical resection before effective antiretroviral therapy or proper documentation of HIV infection, specific data on the efficacy of current combination antiretroviral therapy are lacking [4, 6-16]. We present data on the antiretroviral treatment of nine HIV-infected patients with chronic lymphoepithelial parotid cysts. Methods Clinical Data Nine HIV-infected patients with chronic lymphoepithelial cysts were followed prospectively between January 1993 to October 1997, with the approval of the human studies committee. All patients had evidence of HIV-1 antibody by enzyme immunoassay and Western blot. CD4 and CD8 lymphocyte counts were measured by flow cytometry, and plasma viral loads were measured by branched-chain DNA assay (Chiron, Emeryville, California). Computed tomography or magnetic resonance imaging was done in eight patients who also had fluid aspirated from at least one parotid cyst. Antiretroviral therapy for HIV infection was administered to all nine patients; five patients also received one or two courses of rapidly tapered prednisone therapy (60 mg daily for 2 days, 40 mg daily for 2 days, 20 mg daily for 2 days, and 10 mg daily for 1 day). Before and after treatment, we clinically estimated cyst size by measuring maximum length and width; the estimated depth of the lesions varied between 5 and 10 cm. Patients with a reduction in gland size of more than 95% and no visible parotid enlargement were considered to have completely responded to medical therapy. Pathologic Evaluation and Immunohistochemical Studies Surgical sections from patient 9 were exposed to lymphocyte cell antigens (CD45), B-cell antigens (CD20, CD79a), T-cell antigens (CD3, CD45Ro), CD68, p24 antigen of HIV (Dako Corp., Carpenteria, California), cytokeratins (AE1/3) and S-100 protein, and cytokeratins (Cam 5.2) and CD1. These antigens were detected by using the strepavidin-horseradish peroxidase detection kit. Results Clinical Data Demographic data, risk behaviors for HIV infection, and CD4 lymphocyte counts are summarized in the Table 1 and Table 2. The patients presented with parotid enlargement that had been present for a mean of 5.4 years (Figure 1, left); four patients also had mild symptoms of dry mouth or dry eyes. In eight patients, computed tomography or magnetic resonance imaging showed bilateral, giant, multiloculated cysts in the intraparotid and periparotid area, with evidence of cervical adenopathy and dense nodular infiltration in and around the parotid gland. Table 1. Characteristics and Management of Nine Patients with Lymphoepithelial Cysts and HIV Infection* Table 2. Table 1. Continued Figure 1. Lymphoepithelial parotid cyst disease in an HIV-infected patient (patient 2) before (left) and after (right) combination antiretroviral therapy. Parotid enlargement was the initial sign of HIV infection in at least seven patients, and all patients reported several previous examinations by health care providers who did not provide a diagnosis, suggest HIV testing, or initiate specific treatment. Related psychosocial reactions, such as anxiety, depression, and reclusiveness, were reported by all patients; six patients who completely responded to medical treatment reported improvement in these problems after therapy. Medical Treatment Although response to medical treatment often varied by the affected side and by patient, all nine patients responded (Table 1 and Table 2). The cysts completely resolved with combination antiretroviral therapy in six patients (Figure 1, right). Of these patients, five received an HIV protease inhibitor and had sustained responses that lasted for a mean of at least 15 months and were associated with increases in CD4 counts and undetectable plasma HIV branched-chain DNA levels. The sixth patient was lost to follow-up 5 months later. Four of the six patients also received one or two short courses of prednisone therapy that produced a rapid and dramatic reduction in parotid gland size without side effects. Eight patients also had cyst aspiration, which was well tolerated but usually provided only temporary and limited improvement. Partial responses followed by relapses were noted in the three patients who were poorly compliant with antiretroviral therapy. Pathologic Evaluation and Immunohistochemical Studies Six months after antiretroviral therapy was stopped, a cystic parotid mass that measured 7 4 3 cm and contained three separate cysts (which ranged from 1.8 to 2.5 cm in diameter) was removed from patient 9. The cysts were distinct from the normal parotid tissue and were lined by both keratinizing and nonkeratinizing squamous epithelium, which was variably infiltrated by small lymphocytes. Adjacent dense lymphoid tissue showed various degrees of follicular hyperplasia, occasional secondary follicle lysis, and multinucleated giant cells. The interfollicular areas were notable for marked proliferation of plasma cells and prominent postcapillary venules. Immunoperoxidase studies of the cyst lining yielded positive results for cytokeratin, which confirmed the cysts epithelial origin. The lymphoid component was positive for CD45, and B cells (CD20, CD79a) were localized in secondary follicles and T cells (CD3, CD45Ro) that stained the paracortical and intrafollicular areas. Lymphocytes that infiltrated the cyst epithelial lining stained predominantly as B cells. Antisera to the HIV p24 antigen localized the virus to germinal centers of secondary follicles and had an interstitial or dendritic pattern. Large multinucleated cells that were present in germinal centers and paracortical areas also stained strongly positive for HIV p24 antigen and expressed CD68 antigen (Kp1); this is consistent with a cell of histiocytic or macrophage lineage. Langerhans cells were occasionally present in paracortical areas. Discussion Our data suggest that lymphoepithelial cysts are not widely recognized as a manifestation of HIV disease. This lack of recognition results in delayed diagnosis and treatment [4, 12]. Patients with benign lymphoepithelial cysts are usually asymptomatic and have CD4 lymphocyte counts that range from 300 to 600 cells/mm3 [4, 9, 11-13]. The cysts are usually chronic, large, multiloculated, and bilateral [4, 12, 13]. Parotid cysts may be diagnosed clinically and confirmed by computed tomography or magnetic resonance imaging [1, 13]. Although lymphoepithelial cysts are highly suggestive of HIV infection, they may be seen with the Sjogren syndrome, the diffuse infiltrative CD8 lymphocytosis syndrome, or tumors or opportunistic infections of the parotid gland [1-3]. In the absence of HIV infection or for patients who have unusual clinical findings or inadequate response to medical therapy, fine-needle aspiration biopsy and surgical resection should be considered. Previous studies of lymphoepithelial cysts often lacked proper documentation of HIV infection, focused on surgical treatment, and provided limited data on the efficacy of antiretroviral therapy [4-14]. Shaha and coworkers [11] reported a complete response in one of two patients treated with zidovudine, and Schiodt and coworkers [13] reported regression of parotid cysts in four patients treated with zidovudine and steroids. Terry and coworkers [4] reported a complete response in one of six patients treated with zidovudine and a partial response in two patients treated with zidovudine and radiation. Low-dose radiation therapy may provide temporary cosmetic palliation of benign lymphoepithelial cysts [17]. Treatment of HIV disease has improved dramatically with the recent increase in our understanding of HIV dynamics and pathogenesis and progress in the use of viral load assays and combination antiretroviral therapy [15, 16, 18]. In our series, complete, sustained resolution of parotid cyst disease was associated with continued compliance with combination antiretroviral therapy, undetectable HIV branched-chain DNA levels, and increases in CD4 lymphocyte counts. Therefore, for HIV-infected patients with lymphoepithelial cysts, we advocate the currently recommended treatment of combination antiretroviral therapy that includes a protease inhibitor [15, 16]. Although corticosteroid therapy (probably because of its anti-inflammatory and lympholytic effects) resulted in rapid, dramatic decreases in the size of parotid cysts, it may be unnecessary for patients receiving highly active antiretroviral combination therapy [13]. Cyst aspiration was helpful in patients with large cysts that did not regress with antiretroviral therapy but, when used alone, it yielded only temporary improvement. Bernier and Bhaskar [5] hypothesized that lymphoepithelial cysts in patients without HIV infection arise from epithelial ductular inclusions in lymph nodes. The increased incidence of lymphoepithelial cysts and the different clinical presentation associated with HIV infection probably reflect the high concentrations and rapid turnover of HIV in hyperplastic lymphoid tissue in and adjacent to the parotid parenchyma [1, 5, 18]. Our immunohistologic studies showed high concentrations of HIV p24 antigen in lymphoid tissue; this confirmed the earlier reports of HIV-infected lymph tissues and the immunohistochemical analysis of a patient with benign lymphoepithelial cysts [19, 20]. We hypothesize that HIV-induced cytokines and lymphoid hyperplasia stimulate the adjacent ductal

Hospital-acquired pneumonia: perspectives for the healthcare epidemiologist.

Nosocomial pneumonia is defined as an infection of lung parenchyma that was neither present nor incubating at the time of the patients admission to the hospital. In the United States, hospital-acquired pneumonia is the second most common nosocomial infection and accounts for the most deaths from nosocomial infection. We describe how infection control personnel can use targeted surveillance to identify clusters of cases and to prevent pneumonia. We also discuss common pathogens that cause nosocomial pneumonia; ventilator-associated pneumonia; and strategies for prevention of hospital-acquired pneumonia.

Occupational exposure to HIV: Experience at a tertiary care center

We examined our hospital-based occupational health clinics experience with combination antiretroviral therapy for postexposure prophylaxis for human immunodeficiency virus (HIV). Over a 12-month period, 68 workers started postexposure prophylaxis: 23 with zidovudine and lamivudine and 45 with zidovudine, lamivudine, and indinavir. Fifty-one (75%) of the 68 workers starting postexposure prophylaxis reported one or more side effects. Side effects were more common among those taking three drugs. Many workers failed to complete the recommended 28-day regimen because of the side effects of the various treatments. The estimated mean cost for evaluations, prophylaxis, and monitoring of exposed workers was

Seroprevalence of human immunodeficiency virus in parturients at Boston City Hospital: implications for public health and obstetric practice.

669 per reported exposure. In our experience, major challenges in carrying out the current HIV postexposure prophylaxis guidelines include expeditious source testing, improved staff education and prevention measures, and scrupulous monitoring of workers taking combination antiretroviral drugs for postexposure prophylaxis, with consideration of alternate regimens for intolerant workers.