ObjectivesRecent studies suggest that patients with HIV infection are at increased risk for incident diabetes mellitus (DM). We investigated the incidence and risk factors of DM among HIV-infected patients receiving combination antiretroviral therapy (CART) in Taiwan. MethodsIncident cases of DM were identified among HIV-infected patients at the National Taiwan University Hospital between 1993 and 2006. A retrospective case-control study was conducted after matching cases with controls for sex, age at HIV diagnosis, year of HIV diagnosis, mode of HIV transmission and baseline CD4 lymphocyte count. A multivariate analysis was performed to identify risk factors for incident DM among HIV-infected patients. ResultsIn 824 HIV-infected patients eligible for analysis, 50 cases of incident DM were diagnosed, resulting in an incidence of 13.1 cases per 1000 person-years of follow-up. In total, 100 matched controls were identified. Risk factors for incident DM were a family history of DM [odds ratio (OR) 2.656; 95% confidence interval (CI) 1. 209-5.834], exposure to zidovudine (OR 3.168;) and current use of protease inhibitors (OR 2.528; 95% CI 1.186-5.389). ConclusionsIncident DM was associated with a family history of DM, exposure to zidovudine and current use of protease inhibitors in HIV-infected patients receiving CART in Taiwan.Keywords: antiretroviral therapy, case-control study, diabetes mellitus, HIV infection IntroductionCombination antiretroviral therapy (CART) has greatly improved survival in patients with HIV infection [1,2]. However, as HIV-infected patients exposed to antiretroviral agents experience prolonged survival and ageing, we have observed the emergence of CART-associated metabolic complications that include impaired glucose tolerance, diabetes mellitus (DM) and lipid disorders, which may subsequently result in increased risk for cardiovascular diseases [3][4][5][6].Glucose intolerance and DM are becoming more common in HIV-infected patients in the era of CART [5,[7][8][9]. Previous studies reported a 6-7% risk of incident DM among patients receiving protease inhibitors (PIs) [9]. Cohort studies showed that HIV-infected individuals on CART had a higher incidence of DM than those not on CART or their HIV-uninfected counterparts [10][11][12]. The incidence of DM among patients receiving CART shows wide variation, from 4.4 per 1000 person-years of followup (PYFU) in the Swiss Cohort Study to 34 per 1000 PYFU in the Women's Interagency HIV Study (WIHS) and 47 per 1000 PYFU in the Multicenter AIDS Cohort Study (MACS) [11][12][13][14]. The discrepancies may result from the enrolment of study populations of different race and ethnicity and from differences in the definition of DM [11][12][13][14].With respect to the factors associated with the development of DM in HIV-infected individuals receiving CART, several cohort studies yielded inconsistent results in terms of the relationship between DM and exposure to specific Correspondence: Dr Chien-Ching Hung, Department of Internal Medicine, National Taiwan U...

Although many studies have been carried out on pulmonary diseases in HIV-infected patients, studies specifically investigating the aetiologies of cavitary lung lesions are rare. MethodsHIV-infected patients enrolled in a cohort study who presented with cavitary lung lesions by radiography were identified between June 1994 and March 2008. Medical records and radiological and microbiological data for these patients were retrospectively reviewed using a standardized case collection form. ResultsDuring the 14-year study period, 73 episodes of cavitary lung lesions were diagnosed in 66 of 1790 (3.7%) HIV-infected patients. At the diagnosis of cavitary lung lesions, the median CD4 count was 25 cells/mL (range 1-575 cells/mL). Eighty-one pathogens were considered causative, with fungi being the most common aetiology (42.0%), followed by bacteria (29.6%) and mycobacteria (25.9%). Of the fungal pneumonias, 19 (55.9%) were caused by Penicillium marneffei, 11 (32.4%) by Cryptococcus neoformans, two (5.9%) by Pneumocystis jirovecii, and two (5.9%) by Aspergillus species. During the study period, 11 of 205 patients (5.4%) who were diagnosed as having tuberculosis presented with cavitary lung lesions, compared with 19 of 36 patients (52.8%) with penicilliosis and 11 of 64 patients (17.2%) with cryptococcosis (Po0.0001). The median CD4 count of patients with cavitary lung lesions resulting from tuberculosis (115 cells/mL) was significantly higher than that of patients with cavitary lung lesions resulting from penicilliosis (4 cells/mL) and cryptococcosis (29.5 cells/mL). ConclusionsOur findings suggest that invasive infections attributable to endemic fungi were the leading cause of cavitary lung lesions among patients in the late stage of HIV infection, and were more common than infections attributable to bacteria and mycobacteria.

Having contact and interacting with HIV/AIDS patients has long been recognized as a means for improving AIDS-related knowledge and attitudes among physicians and hence for increasing their intention to provide AIDS care. To investigate the impact of one-month residency training in an AIDS inpatient unit on internal medicine residents, this quasi-experimental, pre-post, two-group study, conducted from April 2000 to April 2001, used questionnaires. At follow-up, residents who received training in the AIDS unit (experimental group) were significantly more knowledgeable about HIV/AIDS, had more positive attitudes and greater intention to care for HIV-infected patients than residents who did not receive this training (control group). Results suggest that a one-month AIDS residency training intervention can effectively enhance residents' HIV-related knowledge, attitudes and intention to care for patients infected with HIV.