Treating Opportunistic Infections Among HIV-Infected Adults and Adolescents When To Initiate ART in the Setting of an OI       No consensus has been reached about the optimal time to start ART in the presence of a recently diagnosed OI. The decision to start potent ART should take into consideration the availability of effective therapy for the OI, the risk for drug interactions, overlapping drug toxicities, the risk for and consequences of the development of an inflammatory immune reconstitution syndrome, and the willingness and ability of patients to take and adhere to their regimens. In cases of cryptosporidiosis, microsporidiosis, PML, and Kaposi sarcoma, the early benefits of potent ART outweigh any increased risk, and potent ART should be started as soon as possible (AIII). In the setting of TB disease, MAC, PCP, and cryptococcal meningitis, awaiting a response to OI therapy is usually warranted before initiating ART (CIII). When an OI occurs within 12 weeks of starting ART, treatment for the OI should be started, and ART should be continued (AIII). When an OI occurs despite complete virologic suppression (i.e., late OI), therapy for the OI should be initiated, potent ART should be continued, and if the CD4+ T cell response to ART has been suboptimal, modification of the ART regimen may be considered (CIII). When an OI occurs in the setting of virologic failure, OI therapy should be started, antiretroviral resistance testing should be performed, and the ART regimen should be modified if possible to achieve better virologic control (AI). Special Considerations During Pregnancy No large studies have been conducted on the epidemiology or manifestations of HIV-1--associated OIs among pregnant women. No data demonstrate that the spectrum differs from that among nonpregnant women with comparable CD4+ T lymphocyte counts. CD4+ T lymphocyte counts characteristically drop during pregnancy, probably because of dilutional effects of the increased plasma volume. CD4+ T lymphocyte percentages are generally more stable and should be used for determining degree of immune suppression during pregnancy.  Physiologic changes occur during pregnancy that might impact the presentation of acute OIs and the considerations for implementing OI treatment or antiretroviral therapies.: Increased cardiac output by 30%--50% with concomitant increase in glomerular filtration rate, and renal clearance. Increased plasma volume by 45%--50% while red cell mass increases only by 20%--30%, leading to dilutional anemia. Increased tidal volume and pulmonary blood flow, possibly leading to increased absorption of aerosolized medications. Changes in late pregnancy might affect distribution of aerosolized medication. The tidal volume increase of 30%--40% should be considered if ventilatory assistance is required. Placental transfer of drugs, increased renal clearance, altered gastrointestinal absorption, and metabolism by the fetus might affect maternal drug levels. Limited pharmacokinetic data are available about the effects of pregnancy on levels of OI therapy drugs. Use usual adult doses based on current weight, monitor levels if available, and consider the possible need to increase doses if the patient is not responding as expected. Pregnancy also impacts decisions about diagnostic testing. Fetal risk is not increased with cumulative radiation doses below 5 rads. Teratogenesis is observed in animals at doses of 100--200 rads. In humans, the primary risk associated with high dose radiation exposure is growth restriction, microcephaly, and developmental disabilities. The most vulnerable period is 8--15 menstrual weeks of gestation with minimal risk before 8 weeks and after 25 weeks. The apparent threshold for development of mental retardation is 20--40 rads, with risk increasing linearly with increasing exposures above this level. Among children, risk for carcinogenesis might be increased approximately one per 1,000 or less per rad of in utero radiation exposure . The majority of radiographic and nuclear medicine studies result in radiation exposure to the fetus that is much lower than the 5 rad recommended limit; therefore, pregnancy should not preclude usual diagnostic evaluation when an OI is suspected . Abdominal shielding should be used when feasible to further limit radiation exposure to the fetus. Experience with use of magnetic resonance imaging (MRI) in pregnancy is limited. Although no adverse fetal effects have been reported, the National Radiological Protection Board advises against use of MRI in the first trimester.  Other procedures necessary for diagnosis of suspected OIs should be performed in pregnancy as indicated for nonpregnant patients. Pregnant women who are >20 weeks of gestation should not lie flat on their backs but should have the left hip elevated with a wedge to displace the uterus off of the great vessels and prevent supine hypotension. Adequate oxygenation should be maintained. Because of the serious nature of OIs among HIV-1--infected persons, diagnostic procedures and indicated therapy should not be withheld during pregnancy; the therapy with the least potential toxicity should be selected. The predictive value of animal data for effects in humans is unclear. In addition, reproductive studies among animals usually include only one drug at a time, and HIV-1--infected pregnant women might be using multiple antiretroviral, OI, and other drugs concurrently. The potential for enhanced toxicity of combinations of drugs has not been evaluated. For pregnant women who have had an OI diagnosed and are not on ART, immediate initiation of ART with OI therapy should be encouraged (AIII). Decisions about immediate versus delayed initiation of ART in pregnancy should take into account gestational age, maternal HIV-1 RNA levels and clinical condition, and potential toxicities and interactions between ART and OI drugs. Pregnant women with active OIs who receive drugs for which information about their use in pregnancy is limited should have additional evaluation of fetal growth and well-being. After first trimester exposure to agents of uncertain teratogenic potential, a detailed ultrasound examination at 18--20 weeks should be conducted to detect major anomalies, although the ultrasound will not detect all anomalies. For women who receive drugs throughout pregnancy or in the third trimester for which information about their use in pregnancy is limited, an ultrasound should be conducted every 4--6 weeks to assess fetal growth and fluid volume. Pregnant women in the third trimester should be instructed in daily fetal movement counting to detect decreased activity that might indicate fetal compromise. Weekly fetal nonstress testing should be initiated at 32 weeks of gestation unless indicated sooner based on clinical or ultrasound findings .