Hospital-based directly observed therapy for HIV-infected children and adolescents to assess adherence to antiretroviral medications.

BACKGROUND: The introduction of highly active antiretroviral therapy for HIV led to significant declines in HIV-associated morbidity and mortality in children. Nonadherence to antiretroviral therapy is the leading cause of treatment failure in HIV-infected patients. The ability to recognize nonadherence is suboptimal, and differentiating it from other causes of inadequate viral suppression may be difficult. OBJECTIVES: The purpose of this work was to examine the efficacy of hospital-based directly observed therapy in assessing adherence to antiretroviral medications in HIV-infected children and adolescents suspected of nonadherence and failing other interventions. METHODS: The medical charts of all HIV-infected patients admitted to the University of Chicago Comer Children's Hospital for directly observed therapy from July 2004 to June 2006 were reviewed. Patients were hospitalized for 7 days. Data collected included demographics, clinical and immune class category, previous and current antiretroviral medications, viral resistance tests, HIV-1 RNA viral load, and CD4+ T-cell number and percentage before and after directly observed therapy. RESULTS: There were 9 perinatally infected patients with a total of 13 admissions. The median age was 13 years, and 8 had been treated with multiple antiretroviral regimens. Three common patterns of changes in the viral load over time were observed. In the first, the viral load dropped at the end of the directly observed therapy period and stayed low thereafter. In the second, the drop in the viral load seen at the end of the period was not sustained. In the third, there was no change in the viral load during or after the directly observed therapy period. Compared with the viral load at admission, the viral load at the end of directly observed therapy was lower in 8 patients with a mean +/- SD decrease of 0.8 +/- 0.55 log10 copies per mL. CONCLUSIONS: Short, hospital-based directly observed therapy was helpful in confirming nonadherence to antiretroviral medications, therefore impacting future therapeutic decisions in HIV-infected children and adolescents. Short, hospital-based directly observed therapy should be considered in patients with poor virological control for whom outpatient interventions have failed.

Clindamycin treatment of methicillin-resistant Staphylococcus aureus infections in children.

BACKGROUND: Methicillin-resistant Staphylococcus aureus (MRSA) with a narrower antibiotic resistance pattern have emerged. There is a risk for the appearance of resistance during clindamycin therapy of erythromycin-resistant MRSA infections because of the linked resistance mechanisms. METHODS: We analyzed clindamycin-susceptible MRSA organisms from children (1987 to 2000) along with clinical data. Antibiotic susceptibilities of organisms were tested, pulsed field gel electrophoresis (PFGE) was done and the linked resistance mechanism was detected by the D test. RESULTS: An average of 11 clindamycin-susceptible MRSA per year were obtained from children since 1993. Of 88 isolates 33 (38%) were erythromycin-resistant. The latter were less often community-acquired (45% vs. 69%), more often from infants <1 month of age (24% vs. 4%) and less likely to be in the community acquisition-associated PFGE Group 1 (62% vs. 87%) than those that were susceptible. The D test was positive in 31 of 33 erythromycin-resistant isolates. A 9-month-old boy with pneumonia/empyema caused by a clindamycin-susceptible, erythromycin-resistant, D test-positive MRSA developed a PFGE-identical clindamycin-resistant isolate and clinical relapse during clindamycin treatment. In contrast a 12-year-old girl with abscesses caused by a similar MRSA developed another abscess after clindamycin therapy, but the organism was unchanged in susceptibility. CONCLUSIONS: Erythromycin resistance was present in 38% of clindamycin-susceptible MRSA in children, and clindamycin resistance was detected during treatment in one child. Clindamycin remains a treatment option if the clinician is notified of the risk by the microbiology laboratory and the clinical situation is suitable.