Gram-negative nosocomial pneumonia may result from retrograde colonization of the pharynx from the stomach, and this may be more likely when the gastric pH is relatively high. We studied the rate of nosocomial pneumonia among 130 patients given mechanical ventilation in an intensive care unit who were receiving as prophylaxis for stress ulcer either sucralfate (n = 61), which does not raise gastric pH, or conventional treatment with antacids, histamine type 2 (H2) blockers, or both (n = 69). At the time of randomization to treatment, the two groups were similar in age, underlying diseases, and severity of acute illness. Patients in the sucralfate group had a higher proportion of gastric aspirates with a pH less than or equal to 4 (P less than 0.001) and significantly lower concentrations of gram-negative bacilli (P less than 0.05) in gastric aspirates, pharyngeal swabs, and tracheal aspirates than did patients in the antacid-H2-blocker group. The rate of pneumonia was twice as high in the antacid-H2 group as in the sucralfate group (95 percent confidence interval, 0.89 to 4.58; P = 0.11). Gram-negative bacilli were isolated more frequently from the tracheal aspirates of patients with pneumonia who were receiving antacids or H2 blockers. Mortality rates were 1.6 times higher in the antacid-H2 group than in the sucralfate group (95 percent confidence interval, 0.99 to 2.50; P = 0.07). Although our results fell just short of statistical significance when they were analyzed according to intention to treat, they suggest that agents that elevate gastric pH increase the risk of nosocomial pneumonia in patients receiving ventilation by favoring gastric colonization with gram-negative bacilli. We conclude that in patients receiving mechanical ventilation, the use of a prophylactic agent against stress-ulcer bleeding that preserves the natural gastric acid barrier against bacterial overgrowth may be preferable to antacids and H2 blockers.

Twenty-eight health care workers who had a poor antibody response when initially vaccinated with hepatitis B vaccine were revaccinated with three additional 20-microgram doses. Eight of the twenty nonresponders, who had levels of antibody to hepatitis B surface antigen (anti-HBs) of less than 8 estimated radioimmunoassay (RIA) units, and all 8 of the hyporesponders, who had anti-HBs levels of 8 or 16 RIA units, attained anti-HBs levels of 36 RIA units or more after revaccination. Tests for HLA-A, B, C, and DR; for complement proteins C2, C4A, C4B, and BF; and for the erythrocyte enzyme glyoxalase I were done in 17 nonresponders and 3 hyporesponders. Nine (45%) had HLA-DR7 and 8 (40%) had HLA-DR3, compared with an expected rate of 23% in the general population. At least one of two extended haplotypes (B44, DR7, FC31 or B8, DR3, SCO1) were detected in 6 of the 9 who did not respond to revaccination, compared with 2 of 11 who responded to a second course of vaccine. Poor responders to vaccine may benefit from revaccination, and genetic factors may modulate the immune response to vaccination.

In the past 10 years, many researchers have examined relationships between hospital staffing and patients’ risk of health care–associated infection (HAI). To gain understanding of this evidence base, a systematic review was conducted, and 42 articles were audited. The most common infection studied was bloodstream infection (n=18; 43%). The majority of researchers examined nurse staffing (n=38; 90%); of these, only 7 (18%) did not find a statistically significant association between nurse staffing variable(s) and HAI rates. Use of nonpermanent staff was associated with increased rates of HAI in 4 studies (P < .05). Three studies addressed infection control professional staffing with mixed results. Physician staffing was not found to be associated with patients’ HAI risk (n=2). The methods employed and operational definitions used for both staffing and HAI varied; despite this variability, trends were apparent. Research characterizing effective staffing for infection control departments is needed.

We conducted a Phase I open-label trial of 2',3'-dideoxyinosine (ddI) for the treatment of the acquired immunodeficiency syndrome (AIDS) and severe AIDS-related complex. A single daily dose of ddI was administered orally to 34 patients (17 with AIDS and 17 with AIDS-related complex) for a median of 12 weeks (range, 2 to 56). We studied six dose levels from 1.6 to 30.4 mg per kilogram of body weight per day. Of the 17 patients previously treated with zidovudine, 13 had had hematologic side effects. The maximal tolerated dose of oral ddI was estimated to be 20.4 mg per kilogram per day. Pancreatitis and peripheral neuropathy were the major dose-limiting toxic effects. Other toxic effects included elevations in hepatic transaminase levels, abnormalities in cardiac conduction, rash, and asymptomatic elevations in serum urate levels and the creatine kinase fraction from skeletal muscle. Treatment with ddI was associated with an increase in the mean number of CD4 lymphocytes from 125 per cubic millimeter at base line to 182 per cubic millimeter after 10 weeks (P = 0.005). There were also increases after 12 weeks in the mean total lymphocyte count (from 0.8 to 1.2 x 10(9) per liter) and the mean hemoglobin level (from 12.9 to 14.1 g per deciliter) (both P less than 0.01). The amount of human immunodeficiency virus p24 antigen decreased by more than 50 percent in 14 of 19 patients with detectable antigen. No differences in response were observed between patients previously treated with zidovudine and those never treated with the drug. We conclude that ddI has antiretroviral activity in patients with AIDS or AIDS-related complex and that the toxicity of ddI differs from that of zidovudine. However, controlled trials are necessary to evaluate the efficacy of ddI.

HIV remains a complex disease that requires comprehensive, coordinated care to ensure optimal outcomes. In the USA, interdisciplinary models of care have developed over time to optimize treatment outcomes. These models may be increasingly important in an era of healthcare reform in the USA. A qualitative study of nine clinical sites funded by the Ryan White HIV/AIDS Program (RWHAP), the federally funded "safety net" program for uninsured and underinsured people living with HIV, was undertaken to identify components of successful models of interdisciplinary HIV care. Findings suggest that these include: (1) patient-centered, one-stop-shop approaches with integrated or co-located services; (2) diverse teams of clinical and nonclinical providers; (3) a site culture that promotes a stigma reducing environment for clients; (4) the availability of a comprehensive array of medical, behavioral health, and psychosocial services; (5) effective communication strategies, including electronic health records (EHRs); and (6) a focus on quality. The importance of RWHAP funding in sustaining these programs is highlighted.