Outpatient parenteral antimicrobial therapy (OPAT) from an emergency model applied during the COVID-19 pandemic to standard of care: Preliminary lessons from our experience.

OBJECTIVES: We wish to report on our experience of OPAT during the first two years of the COVID19 outbreak. PATIENTS AND METHODS: We recorded data on all patients treated in the OPAT regimen in 2020 and 2021 and compared overall trends, use of carbapenems and saved days of hospitalization. RESULTS: The OPAT model enabled us to ensure the administration of first choice antibiotic therapy to 239 patients with an increase of 21.3% from 2020 to 2021 (108 vs 131). Applying this model, we also recorded a reduction in the use of carbapenems from 33% in 2020 to 26% in 2021 and a total of 3041 recovery days saved in 2021.The clinical cure rate reached 94%. Few adverse events occurred (35/239; 14.6%), and they did not require hospitalization. CONCLUSION: OPAT is a safe, efficacious, and cost-effective model that functioned effectively during the COVID-19 crisis and could become the standard of care for the treatment of selected patients.

Clinical role of lung ultrasound for diagnosis and monitoring of COVID-19 pneumonia in pregnant women.

Lung ultrasound has been suggested recently by the Chinese Critical Care Ultrasound Study Group and Italian Academy of Thoracic Ultrasound as an accurate tool to detect lung involvement in COVID-19. Although chest computed tomography (CT) represents the gold standard to assess lung involvement, with a specificity superior even to that of the nasopharyngeal swab for diagnosis, lung ultrasound examination can be a valid alternative to CT scan, with certain advantages, particularly for pregnant women. Ultrasound can be performed directly at the bed-side by a single operator, reducing the risk of spreading the disease among health professionals. Furthermore, it is a radiation-free exam, making it safer and easier to monitor those patients who require a series of exams. We report on four cases of pregnant women affected by COVID-19 who were monitored with lung ultrasound examination. All patients showed sonographic features indicative of COVID-19 pneumonia at admission: irregular pleural lines and vertical artifacts (B-lines) were observed in all four cases, and patchy areas of white lung were observed in two. Lung ultrasound was more sensitive than was chest X-ray in detecting COVID-19. In three patients, we observed almost complete resolution of lung pathology on ultrasound within 96 h of admission. Two pregnancies were ongoing at the time of writing, and two had undergone Cesarean delivery with no fetal complications. Reverse transcription polymerase chain reaction analysis of cord blood and newborn swabs was negative in both of these cases. Copyright © 2020 ISUOG. Published by John Wiley & Sons Ltd.

SLC22A2 variants and dolutegravir levels correlate with psychiatric symptoms in persons with HIV.

BACKGROUND: Neuropsychiatric symptoms (NPs) have been reported with dolutegravir use. We hypothesized that increasing dolutegravir trough concentrations (Ctrough) and/or polymorphism in the SLC22A2 gene, encoding the organic cation transporter-2 (OCT2), which is involved in monoamine clearance in the CNS and is inhibited by dolutegravir, might be associated with NPs. METHODS: A cross-sectional cohort of HIV-positive patients treated with a dolutegravir-containing regimen underwent determination of allelic discrimination for SLC22A2 808 C → A polymorphism and dolutegravir Ctrough. The Symptom Checklist-90-R [investigating 10 psychiatric dimensions and reporting a general severity index (GSI)], a self-reported questionnaire and the Mini-International Neuropsychiatric Interview were offered to investigate current NPs. The effects of dolutegravir Ctrough and the SLC22A2 gene variant on NPs were explored by multivariable logistic regression. RESULTS: A cohort of 203 patients was analysed: 71.4% were male, with median age 51 years and 11 years of ART exposure. Median time on dolutegravir was 18 months. Dolutegravir was associated with different antiretroviral combinations (mainly lamivudine, 38.9%, and abacavir/lamivudine, 35.5%). SLC22A2 CA genotype was independently associated with an abnormal GSI [adjusted OR (aOR) 2.43; P = 0.072], anxiety (aOR 2.61; P = 0.044), hostility (aOR 3.76; P = 0.012) and with moderate to severe headache (aOR 5.55; P = 0.037), and dolutegravir Ctrough was associated with hostility (fourth versus first quartile aOR 6.70; P = 0.007) and psychoticism (fourth versus first quartile aOR 19.01; P = 0.008). Other NPs were not associated with SLC22A2 polymorphism or dolutegravir Ctrough. CONCLUSIONS: A variant of the OCT2-encoding gene, in addition to or in synergy with higher dolutegravir Ctrough, is associated with a set of NPs observed during dolutegravir therapy.

Atazanavir and darunavir in pregnant women with HIV: evaluation of laboratory and clinical outcomes from an observational national study.

Background: Atazanavir and darunavir represent the main HIV PIs recommended in pregnancy, but comparative data in pregnant women are limited. We assessed the safety and activity profile of these two drugs in pregnancy using data from a national observational study. Methods: Women with atazanavir or darunavir exposure in pregnancy were evaluated for laboratory measures and main pregnancy outcomes (e.g. preterm delivery, low birthweight, non-elective caesarean section and neonatal gestational age-adjusted birthweight Z-score). Results: Final analysis included 500 pregnancies with either atazanavir (n = 409) or darunavir (n = 91) exposure. No differences in pregnancy outcomes, weight gain in pregnancy, drug discontinuations, undetectable HIV-RNA, haemoglobin, ALT, total cholesterol, HDL cholesterol and LDL cholesterol were observed between the two groups. At third trimester, exposure to darunavir was associated with higher levels of plasma triglycerides (median 235.5 versus 179 mg/dL; P = 0.032) and a higher total cholesterol/HDL cholesterol ratio (median 4.03 versus 3.27; P = 0.028) and exposure to atazanavir was associated with higher levels of plasma bilirubin (1.54 versus 0.32 mg/dL; P < 0.001). Conclusions: In this observational study, the two main HIV PIs currently recommended by perinatal guidelines showed similar safety and activity in pregnancy, with no evidence of differences between the two drugs in terms of main pregnancy outcomes. Based on the minor differences observed in laboratory measures, prescribing physicians might prefer either drug in some particular situations where the different impacts of treatment on lipid profile and bilirubin may have clinical relevance.

Pregnant with HIV before age 25: data from a large national study in Italy, 2001-2016.

Young pregnant women with HIV may be at significant risk of unplanned pregnancy, lower treatment coverage, and other adverse pregnancy outcomes. In a large cohort of pregnant women with HIV in Italy, among 2979 pregnancies followed in 2001-2016, 9·0% were in women <25 years, with a significant increase over time (2001-2005: 7·0%; 2006-2010: 9·1%; 2011-2016: 12·2%, P < 0·001). Younger women had a lower rate of planned pregnancy (23·2% vs. 37·7%, odds ratio (OR) 0·50, 95% confidence interval (CI) 0·36-0·69), were more frequently diagnosed with HIV in pregnancy (46·5% vs. 20·9%, OR 3·29, 95% CI 2·54-4·25), and, if already diagnosed with HIV before pregnancy, were less frequently on antiretroviral treatment at conception (<25 years: 56·3%; ⩾25 years: 69·0%, OR 0·58, 95% CI 0·41-0·81). During pregnancy, treatment coverage was almost universal in both age groups (98·5% vs. 99·3%), with no differences in rate of HIV viral suppression at third trimester and adverse pregnancy outcomes. The data show that young women represent a growing proportion of pregnant women with HIV, and are significantly more likely to have unplanned pregnancy, undiagnosed HIV infection, and lower treatment coverage at conception. During pregnancy, antiretroviral treatment, HIV suppression, and pregnancy outcomes are similar compared with older women. Earlier intervention strategies may provide additional benefits in the quality of care for women with HIV.

Amniocentesis and chorionic villus sampling in HIV-infected pregnant women: a multicentre case series.

OBJECTIVES: To assess in pregnant women with HIV the rates of amniocentesis and chorionic villus sampling (CVS), and the outcomes associated with such procedures. DESIGN: Observational study. Data from the Italian National Program on Surveillance on Antiretroviral Treatment in Pregnancy were used. SETTING: University and hospital clinics. POPULATION: Pregnant women with HIV. METHODS: Temporal trends were analysed by analysis of variance and by the Chi-square test for trend. Quantitative variables were compared by Student's t-test and categorical data by the Chi-square test, with odds ratios and 95% confidence intervals calculated. MAIN OUTCOME MEASURES: Rate of invasive testing, intrauterine death, HIV transmission. RESULTS: Between 2001 and 2015, among 2065 pregnancies in women with HIV, 113 (5.5%) had invasive tests performed. The procedures were conducted under antiretroviral treatment in 99 cases (87.6%), with a significant increase over time in the proportion of tests performed under highly active antiretroviral therapy (HAART) (100% in 2011-2015). Three intrauterine deaths were observed (2.6%), and 14 pregnancies were terminated because of fetal anomalies. Among 96 live newborns, eight had no information available on HIV status. Among the remaining 88 cases with either amniocentesis (n = 75), CVS (n = 12), or both (n = 1), two HIV transmissions occurred (2.3%). No HIV transmission occurred among the women who were on HAART at the time of invasive testing, and none after 2005. CONCLUSIONS: The findings reinforce the assumption that invasive prenatal testing does not increase the risk of HIV vertical transmission among pregnant women under suppressive antiretroviral treatment. TWEETABLE ABSTRACT: No HIV transmission occurred among women who underwent amniocentesis or CVS under effective anti-HIV regimens.

Rate, correlates and outcomes of repeat pregnancy in HIV-infected women.

OBJECTIVES: The aim of the study was to assess the rate, determinants, and outcomes of repeat pregnancies in women with HIV infection. METHODS: Data from a national study of pregnant women with HIV infection were used. Main outcomes were preterm delivery, low birth weight, CD4 cell count and HIV plasma viral load. RESULTS: The rate of repeat pregnancy among 3007 women was 16.2%. Women with a repeat pregnancy were on average younger than those with a single pregnancy (median age 30 vs. 33 years, respectively), more recently diagnosed with HIV infection (median time since diagnosis 25 vs. 51 months, respectively), and more frequently of foreign origin [odds ratio (OR) 1.36; 95% confidence interval (CI) 1.10-1.68], diagnosed with HIV infection in the current pregnancy (OR: 1.69; 95% CI: 1.35-2.11), and at their first pregnancy (OR: 1.33; 95% CI: 1.06-1.66). In women with sequential pregnancies, compared with the first pregnancy, several outcomes showed a significant improvement in the second pregnancy, with a higher rate of antiretroviral treatment at conception (39.0 vs. 65.4%, respectively), better median maternal weight at the start of pregnancy (60 vs. 61 kg, respectively), a higher rate of end-of-pregnancy undetectable HIV RNA (60.7 vs. 71.6%, respectively), a higher median birth weight (2815 vs. 2885 g, respectively), lower rates of preterm delivery (23.0 vs. 17.7%, respectively) and of low birth weight (23.4 vs. 15.4%, respectively), and a higher median CD4 cell count (+47 cells/µL), with almost no clinical progression to Centers for Disease Control and Prevention stage C (CDC-C) HIV disease (0.3%). The second pregnancy was significantly more likely to end in voluntary termination than the first pregnancy (11.4 vs. 6.1%, respectively). CONCLUSIONS: Younger and foreign women were more likely to have a repeat pregnancy; in women with sequential pregnancies, the second pregnancy was characterized by a significant improvement in several outcomes, suggesting that women with HIV infection who desire multiple children may proceed safely and confidently with subsequent pregnancies.

HCV-HIV coinfected pregnant women: data from a multicentre study in Italy.

PURPOSE: To provide information about main pregnancy outcomes in HIV-HCV coinfected women and about the possible interactions between HIV and HCV in this particular population. METHODS: Data from a multicenter observational study of pregnant women with HIV, conducted in Italian University and Hospital Clinics between 2001 and 2015, were used. Eligibility criteria for analysis were HCV coinfection and at least one detectable plasma HCV-RNA viral load measured during pregnancy. Qualitative variables were compared using the Chi-square or the Fisher test and quantitative variables using the Mann-Whitney U test. The Spearman's coefficient was used to evaluate correlations between quantitative variables. RESULTS: Among 105 women with positive HCV-RNA, median HCV viral load was substantially identical at the three trimesters (5.68, 5.45, and 5.86 log IU/ml, respectively), and 85.7 % of the women had at least one HCV-RNA value >5 log IU/ml. Rate of preterm delivery was 28.6 % with HCV-RNA <5 log IU/ml and 43.2 % with HCV-RNA >5log (p = 0.309). Compared to women with term delivery, women with preterm delivery had higher median HCV-RNA levels (third trimester: 6.00 vs. 5.62 log IU/ml, p = 0.037). Third trimester HIV-RNA levels were below 50 copies/ml in 47.7 % of the cases. No cases of vertical HIV transmission occurred. Rate of HCV transmission was 9.0 % and occurred only with HCV-RNA levels >5 log IU/ml. CONCLUSIONS: Coinfection with HIV and HCV has relevant consequences in pregnancy: HIV coinfection is associated with high HCV-RNA levels that might favour HCV transmission, and HCV infection might further increase the risk of preterm delivery in women with HIV. HCV/HIV coinfected women should be considered a population at high risk of adverse outcomes.

Rate, predictors, and consequences of late antenatal booking in a national cohort study of pregnant women with HIV in Italy.

OBJECTIVE: To evaluate the prevalence and consequences of late antenatal booking (13 or more weeks gestation) in a national observational study of pregnant women with HIV. METHODS: The clinical and demographic characteristics associated with late booking were evaluated in univariate analyses using the Mann-Whitney U test for quantitative data and the chi-square test for categorical data. The associations that were found were re-evaluated in multivariable logistic regression models. Main outcomes were preterm delivery, low birthweight, nonelective cesarean section, birth defects, undetectable (<50 copies/mL) HIV plasma viral load at third trimester, delivery complications, and gender-adjusted and gestational age-adjusted Z scores for birthweight. RESULTS: Rate of late booking among 1,643 pregnancies was 32.9%. This condition was associated with younger age, African provenance, diagnosis of HIV during pregnancy, and less antiretroviral exposure. Undetectable HIV RNA at third trimester and preterm delivery were significantly more prevalent with earlier booking (67.1% vs 46.3%, P < .001, and 23.2% vs 17.6, P = .010, respectively), whereas complications of delivery were more common with late booking (8.2% vs 5.0%, P = .013). Multivariable analyses confirmed an independent role of late booking in predicting detectable HIV RNA at third trimester (adjusted odds ratio [AOR], 1.7; 95% CI, 1.3-2.3; P < .001) and delivery complications (AOR, 1.8; 95% CI, 1.2-2.8; P = .005). CONCLUSIONS: Late antenatal booking was associated with detectable HIV RNA in late pregnancy and with complications of delivery. Measures should be taken to ensure an earlier entry into antenatal care, particularly for African women, and to facilitate access to counselling and antenatal services. These measures can significantly improve pregnancy management and reduce morbidity and complications in pregnant women with HIV.

Birth defects in a national cohort of pregnant women with HIV infection in Italy, 2001-2011.

OBJECTIVE: We used data from a national study of pregnant women with HIV to evaluate the prevalence of congenital abnormalities in newborns from women with HIV infection. DESIGN: Observational study. SETTING: University and hospital clinics. POPULATION: Pregnant women with HIV exposed to antiretroviral treatment at any time during pregnancy. METHODS: The total prevalence of birth defects was assessed on live births, stillbirths, and elective terminations for fetal anomaly. The associations between potentially predictive variables and the occurrence of birth defects were expressed as odds ratios (ORs) with 95% confidence intervals (95% CIs) for exposed versus unexposed cases, calculated in univariate and multivariate logistic regression analyses. MAIN OUTCOME MEASURES: Birth defects, defined according to the Antiretroviral Pregnancy Registry criteria. RESULTS: A total of 1257 pregnancies with exposure at any time to antiretroviral therapy were evaluated. Forty-two cases with major defects were observed. The total prevalence was 3.2% (95% CI 1.9-4.5) for exposure to any antiretroviral drug during the first trimester (23 cases with defects) and 3.4% (95% CI 1.9-4.9) for no antiretroviral exposure during the first trimester (19 cases). No associations were found between major birth defects and first-trimester exposure to any antiretroviral treatment (OR 0.94, 95% CI 0.51-1.75), main drug classes (nucleoside reverse transcriptase inhibitors, OR 0.95, 95% CI 0.51-1.76; non-nucleoside reverse transcriptase inhibitors, OR 1.20, 95% CI 0.56-2.55; protease inhibitors, OR 0.92, 95% CI 0.43-1.95), and individual drugs, including efavirenz (prevalence for efavirenz, 2.5%). CONCLUSIONS: This study adds further support to the assumption that first-trimester exposure to antiretroviral treatment does not increase the risk of congenital abnormalities.

Cardiovascular risk factors and carotid intima-media thickness are associated with lower cognitive performance in HIV-infected patients.

OBJECTIVES: The aim of the study was to investigate the relationship between metabolic comorbidities, cardiovascular risk factors or common carotid intima-media thickness (cIMT) and cognitive performance in HIV-infected patients. METHODS: Asymptomatic HIV-infected subjects were consecutively enrolled during routine out-patient visits at two clinical centres. All patients underwent an extensive neuropsychological battery and assessment of metabolic comorbidities and cardiovascular risk factors. Moreover, cIMT was assessed by ultrasonography. Cognitive performance was evaluated by calculating a global cognitive impairment (GCI) score obtained by summing scores assigned to each test (0 if normal and 1 if pathological). RESULTS: A total of 245 patients (median age 46 years; 84.1% with HIV RNA < 50 copies/mL; median CD4 count 527 cells/µL) were enrolled in the study. Cardiovascular risk factors were highly prevalent in our population: the most frequent were dyslipidaemia (61.2%), cigarette smoking (54.3%) and hypertension (15.1%). cIMT was abnormal (≥ 0.9mm) in 31.8% of patients. Overall, the median GCI score was 2 [interquartile range (IQR) 1-4]; it was higher in patients with diabetes (P = 0.004), hypertension (P = 0.030) or cIMT ≥ 0.9 mm (P < 0.001). In multivariate analysis, it was confirmed that diabetes (P = 0.007) and cIMT ≥ 0.9 mm (P = 0.044) had an independent association with lower cognitive performance. In an analysis of patients on combination antiretroviral therapy (cART), abacavir use was independently associated with a better cognitive performance (P = 0.011), while no association was observed for other drugs or neuroeffectiveness score. CONCLUSIONS: Diabetes, cardiovascular risk factors and cIMT showed a strong association with lower cognitive performance, suggesting that metabolic comorbidities could play a relevant role in the pathogenesis of HIV-associated neurocognitive disorders in the recent cART era.

Efavirenz associated with cognitive disorders in otherwise asymptomatic HIV-infected patients.

BACKGROUND: Despite the availability of potent antiretroviral regimens (combination antiretroviral therapy [cART]), HIV-associated neurocognitive disorders (HAND) are increasingly recognized. Our aim was to investigate the prevalence and treatment-related correlates of HAND, exploring the potential neurotoxicity of antiretrovirals on cognitive functions. METHODS: We performed a cross-sectional single cohort study by consecutively enrolling asymptomatic HIV+ subjects during routine outpatient visits. Each patient was submitted to a comprehensive neuropsychological battery and was considered cognitively impaired on the basis of results obtained in matched healthy HIV-negative subjects. CNS penetration effectiveness (CPE) rank was calculated for cART regimens according to 2010 CHARTER criteria. Factors associated with cognitive impairment were investigated by linear or logistic regression analysis. RESULTS: A total of 146 patients were enrolled. Of these, 129 (88.4%) were on cART and 59.6% of them were on current regimen from ≥1 year. Sixty-nine patients (47%) were classified as cognitively impaired (35.6% asymptomatic and 11.6% mild neurocognitive impairment). In the multivariate analysis, efavirenz use (odds ratio [OR] = 4.00; p = 0.008) and non-Italian nationality (OR = 3.46; p = 0.035) were associated with increased risk of cognitive impairment, whereas higher education was associated with a lower risk (OR = 0.85; p = 0.002). Furthermore, efavirenz use and age ≥65 years independently predicted worse performance on the double barrage and the Stroop test (time). No association between CPE rank and cognitive impairment was observed. CONCLUSIONS: A high prevalence of HAND was observed in apparently asymptomatic HIV+ individuals. HAND was associated with efavirenz use, suggesting the potential neurotoxicity of this drug. Routine neuropsychological examinations could help clinicians make correct diagnoses and manage mild, but clinically relevant, forms of HAND.

Declining HCV seroprevalence in pregnant women with HIV.

We assessed recent trends in hepatitis C virus (HCV) prevalence in pregnant women with HIV using data from a large national study. Based on 1240 pregnancies, we observed a 3.4-fold decline in HCV seroprevalence in pregnant women with HIV between 2001 (29.3%) and 2008 (8.6%). This decline was the net result of two components: a progressively declining HCV seroprevalence in non-African women (from 35.7% in 2001 to 16.7% in 2008), sustained by a parallel reduction in history of injecting drug use (IDU) in this population, and a significantly growing presence (from 21.2% in 2001 to 48.6% in 2008) of women of African origin, at very low risk of being HCV-infected [average HCV prevalence 1%, adjusted odds ratio (aOR) for HCV 0.09, 95% CI 0.03-0.29]. Previous IDU was the stronger determinant of HCV co-infection in pregnant women with HIV (aOR 30.9, 95% CI 18.8-51.1). The observed trend is expected to translate into a reduced number of cases of vertical HCV transmission.

Women facing HIV. Key question on women with HIV infection: Italian consensus workshop.

A panel of leading Italian specialists in infectious diseases, obstetrics and gynaecology met in a national consensus workshop on women facing HIV to review critical aspects and discuss recommendations for selected key questions on four issues: (1) women and highly active antiretroviral therapy (HAART): access to care and adherence to therapy, side effects and drug-drug interaction; (2) HIV-infected pregnant women: prevention of mother to child transmission; (3) desire for children among women living with HIV: assisted reproduction; (4) sexually transmitted diseases and genital disturbances. The method of a nominal group meeting was used, and recommendations were graded for their strength and quality of evidence using a system based on the one adopted by the Infectious Diseases Society of America. Main conclusions are summarized and critically discussed, and some of the most recent data supporting recommendations are provided.

Genotypic resistance profile and clinical progression of treatment-experienced HIV type 1-infected patients with virological failure.

We explored the relationship between HIV-1 drug resistance in treatment-experienced patients and disease progression in a cohort of patients undergoing resistance testing to guide treatment decisions. A total of 601 treatment-failing individuals tested for genotypic HIV-1 drug resistance between 1998 and 2004 were selected. At genotypic testing, median HIV-1 RNA levels and CD4 counts were 3.8 log copies/ml and 293 cells/mul, respectively; 84% had resistance mutations to nucleoside reverse transcriptase inhibitors (NRTIs), 42% had resistance mutations to non-NRTIs, 51% had major resistance mutations to protease inhibitors (PI), 12% had no major resistance mutations to any drug class, 22% had mutations to one class, 42% had mutations to two classes, and 23% had mutations to three classes. During a follow-up of 714.7 patients/year, 80 patients showed an AIDS-defining event or died. In multivariable models adjusting for prior AIDS, baseline CD4 counts, HIV-1 RNA, and calendar year, viral resistance variables associated with increased hazards of clinical progression were the presence of reverse transcriptase substitution T215F (p = 0.002) and the presence of three or more protease substitutions among L33F/I/V, V82A/F/L/T, I84V, and L90M (p = 0.003). Resistance to three drug classes remained independently predictive of clinical progression only when calendar year was not used as an adjustment factor. Prevention and treatment of multiple drug class resistance are clinical priorities for HIV-infected patients. In recent years, improved treatment options may have helped in reducing part of the resistance-associated clinical progression.

Cardiovascular risk score change in HIV-1-infected patients switched to an atazanavir-based combination antiretroviral regimen.

BACKGROUND: We aimed to establish whether the limited impact of atazanavir on the plasma lipid profile could translate into a reduction in the predicted cardiovascular risk in antiretroviral (ARV)-experienced patients switching to an atazanavir-containing regimen. METHODS: HIV-1-infected treatment-experienced patients, switched to atazanavir for whatever reason and without prior major cardiovascular events, were selected and followed for at least 1 month. An individual cardiovascular risk score (10-year risk of major cardiovascular events) based on validated events and measurable risk factors in Italian cardiovascular cohorts was calculated using software available online. RESULTS: A total of 197 patients were selected for inclusion in the study. After switching to atazanavir, the mean changes from pre-switch to last available measurement were -6.5% (P<0.001) for total cholesterol, -1.7% (P=0.029) for high-density lipoprotein (HDL) cholesterol, -11.3% (P<0.001) for non-HDL cholesterol and -8.6% (P<0.001) for triglycerides. The crude cardiovascular risk score was reduced from 3.43 to 3.38% (P=0.51); the analysis normalized by age showed a reduction from 3.43 to 3.14% (P<0.001). Subsets of patients with high baseline total cholesterol or triglycerides showed more marked reductions. CONCLUSIONS: A treatment switch to atazanavir caused significant reductions in plasma lipids and a modest but significant reduction in the normalized-for-age cardiovascular risk score. Efforts should be made to concomitantly reduce the other preventable cardiovascular risk factors.