Prevalence and predictors of transmitted drug resistance (TDR), defined as the presence of at least one WHO surveillance drug resistance mutation (SDRM), were investigated in antiretroviralnaїve HIV-1-infected patients, with a genotypic resistance test (GRT) performed ≤6 months before starting cART between 2000 and 2010. 3163 HIV-1 sequences were selected (69% subtype B). Overall, the prevalence of TDR was 12% (13.2% subtype B, 9% non-B). TDR significantly declined overall and for the single drug classes. Older age independently predicted increased odds of TDR, whereas a more recent GRT, a higher HIV-RNA and C vs. B subtype predicted lower odds of TDR.

We analysed trends of human immunodeficiency virus type 1 (HIV-1) drug resistance during 2007-2009 in the Italian national HIV drug resistance database 'ARCA'. Prevalence of resistance in each year was examined on the basis of the presence of major International AIDS Society-2009 mutations. Predictors of resistance were analysed by multivariable logistic regression. Nine hundred and sixty-six patients were selected. Resistance to nucleoside reverse transcriptase inhibitors and protease inhibitors showed a significant decline with respect to previous surveys. Resistance to any class of drug and three drug classes remained stable. Independent predictors of three-class resistance were the number of treatment regimens experienced, prior suboptimal nucleoside reverse transcriptase inhibitor therapy and the current use of ritonavir-boosted protease inhibitors.

We investigated, in patients tested between 1991 and 2004, the patterns of mutually exclusive human immunodeficiency virus-1 thymidine-analogue mutations (TAMs) in 4039 reverse-transcriptase sequences with > or = 1 TAM. TAM pattern 1, which included M41L and L210W and excluded K70R and is coupled with more-extensive cross-resistance to drugs, became the most frequent pattern after 1996. In 1465 genotypes from 684 patients in whom highly active antiretroviral therapy (HAART) was unsuccessful, predictors of this pattern were the number of previous HAART regimens undergone (adjusted odds ratio [OR], 1.09 [95% confidence interval {CI}, 1.02-1.16]), use of stavudine/lamivudine (adjusted OR, 1.42 [95% CI, 1.05-1.99]), use of nevirapine (adjusted OR, 1.60 [95% CI, 1.14-2.24]), use of efavirenz (adjusted OR, 1.56 [95% CI, 1.08-2.27]), and use of ritonavir (adjusted OR, 1.35 [95% CI, 1.04-1.75]).

To the Editor: Lopinavir, coformulated with a boosting dose of ritonavir, (LPV/r) has been recently licensed as a new protease inhibitor (PI) with excellent pharmacokinetic properties (1). Due to the achievement of high blood levels, there is a high genetic barrier to clinical resistance to LPV/r. Indeed, as many as 11 mutations at HIV-1 protease codons 10, 20, 24, 46, 53, 54, 63, 71, 82, 84 and 90 have been reported to contribute to LPV/r resistance, and at least 8 of these appear to be required for significant clinical resistance (2). Although preliminary LPV/r efficacy studies have yielded promising results, especially on drug-naive subjects (3–5), the role of LPV/r as a salvage therapy for heavily experienced patients in clinical practice has not yet been fully elucidated. We report an observational study of the first 41 patients shifted to an LPV/r-including regimen and reaching at least 16 weeks (mean ± SD, 23 ± 5 weeks) of LPV/r therapy in our area. The subjects were pretreated with a median of 5 (range 4–6) nucleoside reverse transcriptase inhibitors (NRTIs), 1 (range 1–2) nonnucleoside reverse transcriptase inhibitors (NNRTIs) and 4 (range 2–5) PIs. The median number of different treatment regimens used before commencing LPV/r therapy was 8 (range 4–13), including dual NRTI treatments, and the median HIV-1 RNA load and CD4 counts at baseline were 5.29 (range 3.79–7.00) log10 copies/mL and 138 (range 4–689) cells/ L, respectively. A total of 26 (63.4%) and 30 (73.2%) subjects had >10 HIV-1 RNA copies/mL and <200 CD4 cells/ L, respectively. Genotypic antiretroviral resistance analysis (6) at baseline revealed a median number of 6 (range 0–10) NRTI resistance mutations, 1 (range 0–4) NNRTI resistance mutations, and 2 (range 0–4) and 5 (range 1–8) primary and accessory PI resistance mutations, respectively. There was only one case without resistance mutations, possibly resulting from multiple treatment interruptions before starting LPV/r. The number of patients harboring virus with at least one primary NRTI, NNRTI, and PI resistance mutations was 38 (92.7%), 28 (68.3%), and 35 (85.4%), respectively. Five (12.2%) subjects harbored virus with a 69S-XX insertion or Q151M complex NRTI class resistance, and more than two primary PI resistance mutations were present in virus from 20 (48.8%) individuals. Except for two subjects in whom LPV/r was associated with one NRTI and one NNRTI (efavirenz), the new regimen combined LPV/r with two NRTIs, mainly stavudine/didanosine (16 cases) and stavudine/lamivudine (8 cases). However, based on baseline RT genotype, only 9 (22.0%) and 18 (49.3%) patients were administered 2 and 1 reverse transcriptase inhibitors expected to retain activity against their viral isolate, respectively, according to the on-line drug resistance interpretation system available at the Stanford University web site (http://hivdb. stanford.edu/). Based on a standardized questionnaire, the selfreported adherence to treatment was optimal in this cohort. Overall, there was a significant response to LPV/r therapy both in terms of HIV-1 RNA levels (5.25 ± 0.68 log10 at baseline vs. 3.94 ± 1.27 log10 at the end of follow-up; p < .001, paired t test) and in terms of CD4 cell counts (166 ± 153 vs. 225 ± 193; p .009). Interestingly, changes in HIV-1 RNA levels and CD4 cell counts were comparable in subjects with <5 log10 HIV-1 RNA copies/mL (−1.30 log10 copies/mL and +45 cells/ L) and >5 log10 HIV-1 RNA copies/mL (−1.32 log10 copies/mL and +67 cells/ L), respectively. HIV-1 RNA levels were suppressed to <500 copies/mL and <50 copies/mL in 13 (31.7%) and 9 (21.9%) patients, respectively. Table 1 shows the evolution of PI resistance mutations under the selective pressure of LPV/r (GenBank accession numbers AF493336 to AF493415). There was a statistically significant increase in the prevalence of total PI resistance mutations (me-

To define factors predictive of failure to respond to nucleoside reverse-transcriptase inhibitors in human immunodeficiency virus type-1 (HIV-1)-infected subjects pretreated with zidovudine (ZDV), three groups of subjects shifted to double therapy with ZDV plus didanosine (ddI, n = 13), zalcitabine (ddC, n = 14), or lamivudine (3TC, n = 12) were retrospectively evaluated, with respect to addition of the second NRTI, at week 0 and week 24. Factors considered included duration of ZDV pretreatment, CD4(+) cell counts, plasma HIV-1 RNA load, peripheral blood mononuclear cell HIV-1 DNA load, and HIV-1 DNA genotypic resistance to nucleoside reverse-transcriptase inhibitors. The three groups were well matched for baseline characteristics and did not differ significantly in virological and immunological response to the different combination treatments. Drug-specific resistance mutations were selected in more than half the cases by 3TC, but not by ddI and ddC. Low-level and substantial genotypic resistance to ZDV was detected 13 (33.3%) and in 19 (48.7%) patients at baseline, respectively, and evolved through week 24 in several patients. When subjects were divided into responders and nonresponders to the second nucleoside reverse-transcriptase inhibitor on the basis of a decrease of more than 0.5 log(10) (n = 15) or less than 0.5 log(10) (n = 21) in HIV-1 RNA, respectively, baseline genotypic ZDV resistance was the only independent predictor of failure in a logistic regression model (P = 0.003 or P = 0.024, depending on whether low-level resistance was considered or not, respectively). Thus, selection of ZDV resistance mutations may impair subsequent use of different nucleoside reverse-transcriptase inhibitor compounds.

In a preliminary cross-sectional analysis of 109 human immunodeficiency virus type 1 (HIV-1)-infected subjects the presence of 2-long terminal repeat (LTR) unintegrated circular HIV-1 DNA in peripheral blood mononuclear cells (PBMC) was found to be associated with both symptomatic infection (P = 0.0037) and low CD4 counts (P = 0.0004). To investigate the prognostic significance of the presence of 2-LTR HIV-1 DNA, a subset of 23 2-LTR-negative and 25 2-LTR-positive asymptomatic individuals were followed up for 12-24 months. The two groups did not differ in terms of baseline CD4 counts, zidovudine (ZDV) therapy, and duration of HIV-1 infection. Longitudinal analysis of CD4 values did not indicate a significantly different CD4 outcome between the two groups. However, when only ZDV-treated subjects were considered, a significant (P = 0.042) decrease in CD4 counts was found at month 24 with respect to baseline in 2-LTR-positive (n = 12) but not in 2-LTR-negative (n = 11) patients. Moreover, when >40% CD4 loss from baseline and/or development of CDC stage B or C symptoms were considered as indicators of disease progression, there was a significantly higher number of events in the whole 2-LTR-positive group than in the whole 2-LTR-negative group (P = 0.0197 at month 12, P = 0.0299 at month 18, P = 0.0373 at month 24). Thus, the presence of 2-LTR HIV-1 DNA in PBMC merits further investigation as a simple, qualitative, molecular predictor of disease progression and decreased response to antiretroviral therapy. (C) 1997 Wiley-Liss, Inc.

To determine the sensitivity of a nested PCR procedure for detecting human immunodeficiency virus type 1 DNA in clinical specimens, 553 peripheral blood mononuclear cell samples obtained from 268 human immunodeficiency virus type 1-seropositive subjects were assayed by use of two independent primer sets for each sample. Overall, 1,088 of 1,106 (98.37%) reactions were positive. Investigation of the negative reactions showed that a low viral burden in some infected subjects, rather than primer-template mismatches, was the primary cause for the false-negative PCR results.

A highly sensitive nested polymerase chain reaction (PCR) protocol was used to detect human immunodeficiency virus type 1 (HIV-1) DNA in peripheral blood mononuclear cells from 271 HIV-1-seropositive patients, 240 HIV-1-seronegative subjects at increased risk for HIV-1 infection, 51 serologically indeterminate individuals, and 120 healthy blood donors. PCR was carried out in a multiplex nested configuration with pol and env region primer sets. HIV-1 DNA was detected in all of the HIV-1 seropositive patients. In contrast, HIV-1 DNA was not detected in any of the either seronegative or serologically indeterminate subjects. Only one of 37 seronegative regular sexual partners of HIV-1-infected patients who were followed longitudinally was found to seroconvert to HIV-1. However, HIV-1 DNA and antibody results were concordant in the four samples obtained from this subject prior to and after seroconversion. These results show an excellent concordance between HIV-1 DNA and antibody detection for diagnosis of HIV-1 infection and suggest that long-term HIV-1 infection in the absence of detectable antibody is likely to occur at a very low frequency.

Between January 1980 and December 1983, 332 consecutive cases of acute hepatitis were observed in adult patients admitted to the Department of Infectious Disease and Gastroenterology of Siena. Sex and age of the patients, the presence of jaundice, the maximum value of the serum-glutamate-pyruvate-transaminase (SGPT) were considered. Serum specimens were tested for hepatitis B surface antigen (HBsAg), antibody against hepatitis B core antigen (anti-HBc) of the IgM class, antibody against hepatitis A virus (anti-HAV) of the IgM class, antibody against cytomegalovirus (CMV) of the IgM class, Paul Bunnel Davidshon reaction. Hepatitis A was diagnosed in 25 cases (7.5%). Hepatitis B in 167 (50.3%). Hepatitis due to CMV in 2 cases (0.6%). And, by exclusion, hepatitis non A, non B in 138 cases (41.6%). Male patients were affected with significantly higher frequency than female (p less than 0.01); the same was seen for young patients (14-30 years) compared to the older ones (31-50 years, and over 50 years) (p less than 0.01 in both). Biochemical investigation showed that hepatitis A and B had a significantly higher, maximum SGPT value than hepatitis non A non B (p less than 0.01 in both). Icteric patients were significantly more frequently observed among hepatitis A and B cases than hepatitis non A non B cases (p less than 0.01 in both).

The results are reported of an epidemiological assessment of 94 patients hospitalised with diagnosed acute viral hepatitis in the Infectious Disease Division of the S. Spirito Hospital, Casale Monferrato in the period march 1 1981, april 30 1982. Hepatitis was found to occur most frequently among young (under 30) males. There were more positive than negative HBsAg cases, most of the latter arising in heroin addicts.