The effect of combination antiretroviral therapy on CD5 B- cells, B-cell activation and hypergammaglobulinaemia in HIV-1-infected patients.

OBJECTIVES: This study assessed B-cell activation, CD5 B-cells and circulating immunoglobulin levels in HIV-infected patients treated with combination antiretroviral therapy (CART). METHODS: Measurement of plasma immunoglobulin levels and electrophoresis of plasma proteins, and analyses of total numbers of B-cells and B-cells expressing CD 38 and CD5 in whole blood, were undertaken in 47 consecutive HIV-1-infected patients attending an out-patient clinic. RESULTS: All HIV-infected patients had similar percentages and numbers of B-cells. Proportions of CD5 B-cells in all HIV-infected patients were significantly lower than those in HIV-negative controls. Aviraemic HIV-infected patients on CART had lower percentages of CD5, CD 38 and CD5 CD 38 B-cell subsets and lower plasma levels of immunoglobulin G (IgG) and immunoglobulin A (IgA) than viraemic HIV-infected patients (untreated or on CART). However, 33-37% of aviraemic HIV-infected patients had IgG and IgA levels above the 95th percentile of the normal range defined in HIV-seronegative donors. In aviraemic HIV-infected patients, plasma IgA levels correlated only with proportions of activated (CD 38) B-cells. IgG levels did not correlate with the proportions of B-cell subsets or any marker of HIV disease activity. Monoclonal immunoglobulins were not detected in any plasma sample. CONCLUSIONS: Aviraemic HIV-infected patients on CART have lower plasma levels of IgG and IgA than viraemic HIV-infected patients, but levels are often above the normal range. CD5 B-cell numbers are depressed, so these cells are unlikely to contribute to hypergammaglobulinaemia in HIV-infected patients.

Chronic hyperlactatemia in HIV-infected patients taking antiretroviral therapy.

OBJECTIVE: To determine the prevalence, course and risk factors for hyperlactatemia in HIV-infected patients. DESIGN: A prospective, longitudinal study of venous lactate concentrations over an 18-month period in 349 participants of the Western Australian HIV Cohort Study. RESULTS: In 516 patient-years of observation, two patients experienced severe fulminant lactic acidosis (lactate > 5 mmol/l) and hepatic steatosis attributable to nucleoside analogue reverse transcriptase inhibitors (NRTI). A further five patients with lesser elevations of lactate (2.8-4.1 mmol/l) but with symptoms of nausea or abdominal discomfort and evidence of hepatic steatosis had NRTI therapy revised, with relief of symptoms and a fall in lactate levels. Most remaining patients on highly active antiretroviral therapy (HAART) had mild, chronic, asymptomatic hyperlactatemia, with mean lactate level between 1.5 mmol/l and 3.5 mmol/l most commonly. Longitudinal data was analysed in a non-linear mixed effects growth model which indicated that average lactate levels rose after the start of HAART but tended to stabilise at low-grade elevation, with an average 0.23 mmol/l greater long term level in stavudine users compared with zidovudine users (p < 0.01). A multiple linear regression model showed that the association between stavudine and higher lactate level was not confounded by longer duration of total NRTI exposure. Risk of hyperlactatemia was not significantly associated with use of other NRTIs, protease inhibitors, non-nucleoside analogue reverse transcriptase inhibitors or multiple immunological and virological factors in multivariate analyses. CONCLUSIONS: Chronic, compensated, asymptomatic hyperlactatemia is common in patients taking HAART. Decompensated, life-threatening lactic acidosis/hepatic steatosis is rare. Treatment with stavudine appears to be the predominant risk factor for development of chronic hyperlactatemia.

Cluster analysis of radionuclide concentrations in beach sand.

This paper presents a method in which natural radionuclide concentrations of beach sand minerals are traced along a stretch of coast by cluster analysis. This analysis yields two groups of mineral deposit with different origins. The method deviates from standard methods of following dispersal of radionuclides in the environment, which are usually based on the construction of lines of equal concentrations. The paper focuses on the methodology of quantitatively correlating activity concentrations of natural radionuclides in two groups of minerals. The methodology is widely applicable, but is demonstrated for natural radioactivity in beach sands along the coast of South West Australia.

Contribution of nucleoside analogue reverse transcriptase inhibitors to subcutaneous fat wasting in patients with HIV infection.

BACKGROUND: Progressive subcutaneous fat wasting, fat accumulation, dyslipidaemia and insulin resistance in HIV-infected patients on antiretroviral therapy has been attributed to the long-term toxicity of HIV protease inhibitors (PI). More recently, fat wasting has been observed in patients who have never taken a PI, implicating an independent effect of nucleoside analogue reverse transcriptase inhibitor (NRTI) therapy. OBJECTIVES: To determine the relative contribution of NRTI and PI, as well as any other factors, to fat wasting in HIV-infected patients. DESIGN: Longitudinal cohort study involving 277 participants of the Western Australian HIV Cohort Study. METHODS: The time to onset of clinically apparent fat wasting in patients receiving different antiretroviral regimens was compared using standardized clinical criteria. Regional fat measured by dual energy X-ray absorptiometry (DEXA) in 161 patients was also compared. The average rate of percentage fat reduction was estimated in 70 patients who had consecutive DEXA scans at approximately 6-monthly intervals. Multiple confounding factors were considered in the analyses. RESULTS: Progressive subcutaneous fat wasting, indistinguishable from that described in PI-treated patients, does occur in PI-naive, NRTI-treated patients. In patients taking triple combination antiretroviral therapy, age (relative risk = 1.052 per year; P < 0.0001), white race (relative risk = 3.9; P = 0.023), longer duration of dual NRTI therapy prior to addition of PI (relative risk = 1.021 per month; P = 0.0046) and increased cumulative time on stavudine-containing regimens compared with time on zidovudine-containing regimens (relative risk = 1.085 per month; P < 0.0001) are associated with increased risk of fat wasting. Stavudine increases the risk of fat wasting by 265% per year compared with zidovudine. However PI therapy is associated with faster progression to clinically apparent wasting compared with dual NRTI therapy without PI. The results of DEXA scanning supports these clinical data and suggest a non-linear decline in fat over time. CONCLUSIONS: NRTIs do have an independent contribution to fat wasting, but PI are the predominant influence and may act synergistically with NRTIs. NRTIs appear to predispose individuals to slowly progressive fat loss, which is markedly accelerated when a PI and NRTIs are combined. Of the NRTIs, stavudine leads to an earlier onset of clinically apparent fat wasting compared with zidovudine. Fat wasting associated with NRTI use may be a manifestation of mitochondrial toxicity, which may be exacerbated by PI use.

Immune restoration disease after the treatment of immunodeficient HIV-infected patients with highly active antiretroviral therapy.

BACKGROUND: To determine if infectious disease events in HIV-infected patients treated with highly active antiretroviral therapy (HAART) are a consequence of the restoration of pathogen-specific immune responses, a single-centre retrospective study of all HIV-infected patients commencing HAART prior to 1 July 1997 was undertaken to determine the incidence, characteristics and time of onset of disease episodes in HAART responders (decrease in plasma HIV RNA of > 1 log10 copies/mL). METHODS: Baseline and post-therapy changes in CD4 T-cell counts and HIV RNA were compared in patients with and without disease and delayed-type hypersensitivity responses to mycobacterial antigens were measured in selected patients. RESULTS: Thirty-three of 132 HAART responders (25%) exhibited one or more disease episodes after HAART, related to a pre-existent or subclinical infection by an opportunistic pathogen. Disease episodes were most often related to infections by mycobacteria or herpesviruses but hepatitis C virus (HCV), molluscum contagiosum virus and human papilloma virus were also implicated. They were most common in patients with a baseline CD4 T-cell count of < 50/uL and occurred most often during the first 2 months of therapy and when CD4 T-cell counts were increasing. Mycobacteria- and HCV-related diseases were associated with restoration of pathogen-specific immune responses. CONCLUSIONS: We conclude that improved immune function in immunodeficient patients treated with HAART may restore pathogen-specific immune responses and cause inflammation in tissues infected by those pathogens.

Dependence estimation for marginal models of multivariate survival data.

We have previously (Segal and Neuhaus, 1993) devised methods for obtaining marginal regression coefficients and associated variance estimates for multivariate survival data, using a synthesis of the Poisson regression formulation for univariate censored survival analysis and generalized estimating equations (GEE's). The method is parametric in that a baseline survival distribution is specified. Analogous semiparametric models, with unspecified baseline survival, have also been developed (Wei, Lin and Weissfeld, 1989; Lin, 1994). Common to both these approaches is the provision of robust variances for the regression parameters. However, none of this work has addressed the more difficult area of dependence estimation. While GEE approaches ostensibly provide such estimates, we show that there are problems adopting these with multivariate survival data. Further, we demonstrate that these problems can affect estimation of the regression coefficients themselves. An alternate, ad hoc approach to dependence estimation, based on design effects, is proposed and evaluated via simulation and illustrative examples.

Detection of subclinical Mycobacterium avium intracellulare complex infection in immunodeficient HIV-infected patients treated with zidovudine.

OBJECTIVE: To test the hypothesis that subclinical Mycobacterium avium intracellulare complex (MAC) infection may result in the development of a tuberculin response in immunodeficient HIV-infected individuals treated with zidovudine. DESIGN: Longitudinal, observational study. SETTING: The Western Australian HIV Cohort Study; a prospective, single centre, population-based observational study of the natural history of HIV disease. PATIENTS: Forty-nine patients with impaired delayed-type hypersensitivity (DTH) responses and negative tuberculin responses in whom DTH responses were augmented within 6 months of starting zidovudine therapy. OUTCOME MEASURES: Progression to disseminated MAC infection stratified according to the presence or absence of a tuberculin response in the first 6 months of zidovudine therapy. RESULTS: Twenty-nine of the patients developed a post-zidovudine tuberculin response. None of the tuberculin non-responders developed disseminated MAC infection during the study period; the Kaplan-Meier probability estimate of disseminated MAC infection was 50% at 24 months and reached 100% 40 months after zidovudine was commenced in tuberculin responders. All patients with disseminated MAC infection had become anergic to all antigens, including tuberculin, before diagnosis. The probability of a post-zidovudine tuberculin response was related to the severity of peripheral blood CD4+ T-cell depletion, rising from an estimated 20% at 20% CD4+ T cells to 100% at < or = 1% CD4+ T cells. CONCLUSIONS: The restoration of a cellular immune response against subclinical MAC infection can be demonstrated by measuring the DTH response to tuberculin in patients with impaired DTH augmented by zidovudine therapy. The findings suggest that MAC infection is almost inevitable, but often asymptomatic, in profoundly immunodeficient HIV-infected patients and that a prolonged subclinical phase of MAC infection is usual.

Severity and outcome of Pneumocystis carinii pneumonia (PCP) in patients of known and unknown HIV status.

Of 170 Western Australian patients who had their first AIDS-defining illness between 1 January 1983 and 31 December 1991, 61 (36%) were of unknown HIV antibody status (AIDS presenters), while 109 (64%) were of known HIV antibody status (HIV presenters). Pneumocystis carinii pneumonia (PCP) was less common as the AIDS-defining illness in HIV presenters (41% versus 62%, p = 0.005). In this study of 70 patients with PCP as the index AIDS diagnosis, 36 were HIV presenters and 34 were AIDS presenters. Ten HIV presenters were taking prophylaxis at the time PCP manifested. The duration of symptoms of cough or dyspnea before the diagnosis of PCP was shorter, and the arterial PO2 measurement on admission was higher in those on prophylaxis, and a lower proportion of patients on prophylaxis required hospital admission (p < or = 0.05 for all comparisons). Furthermore, the CD4 counts at diagnosis of PCP were lower in patients taking PCP prophylaxis (mean 26 x 10(6)/L) than in patients who were not (mean 94 x 10(6)/L, p = 0.007). Of seven patients who died of PCP, none were receiving treatment for HIV disease before AIDS presentation. These findings suggest that PCP is prevented or deferred in patients receiving care for HIV disease and is less severe as a result of early diagnosis and treatment.

Relationship between urinary cotinine level and diagnosis in children admitted to hospital.

The reported association between passive smoking and respiratory illness in children has been based on the parents' assessment of their own level of smoking. To more critically evaluate a causal relationship between passive smoking and childhood ill health, we used urinary cotinine, which is the major metabolite of nicotine and has a long half-life, to objectively quantitate the level of passive smoking in children. Urine was collected from 609 children (median age 3.8 yr, range 1 month to 17 yr) on admission to hospital; cotinine levels were obtained in 491 of these samples, and a comprehensive respiratory questionnaire was completed for 468 children. Statistical analysis was carried out on transformed data using both parametric and nonparametric statistics. Cotinine levels in the children correlated with the parents' current smoking (p less than 0.001). Elevated levels were found in the 41 children admitted with bronchiolitis compared with a group of a similarly aged children with nonrespiratory illnesses (p less than 0.02). Elevated levels were not found for any other diagnosis. We conclude that the urinary cotinine approach has provided objective evidence linking passive smoking to hospital admission for bronchiolitis in infants.

Circadian changes in salivary constituents and conductivity in women and men.

Circadian rhythms in salivary [glucose], [Na+], [K+] and conductivity were measured in 2 age groups of men (men A, 20-45 years and men B, 46-60 years) and 8 different states of fertility in women (normally menstruating, taking oral contraceptives, pregnant, lactational amenorrhea, lactational amenorrhea and taking oral contraceptives, lactating and menstruating, menopausal, and post-menopausal). Unstimulated whole saliva (2-3 ml) was collected every 3 h over a 48 h span. Analysis of Spearman Rank Correlations indicated significant circadian rhythms (significant positive coefficients) for all groups of [Na+] (mean = 0.577 +/- 0.040) and conductivity (mean = 0.410 +/- 0.050). There was no evidence of differences in prominence of rhythm across groups for [Na+] and conductivity. [K+] showed less evidence of rhythms and much greater variability between groups (mean correlation coefficient = 0.198 +/- 0.055). Rhythms in [glucose] (mean correlation coefficient = 0.409 +/- 0.051) were evident in all groups except men B (0.016), menopausal women (0.151) and post-menopausal women (0.310). Model analysis of the data showed no discernible rhythmic trend with age for either conductivity, [Na+] or [K+], where any differences were explainable by the group characteristics. The rhythm in [glucose] showed a significant weakening with age over all groups (F-ratio = 7.46**), and was different between men A and men B (F-ratio = 6.95**). It was concluded that circadian rhythms were present in whole unstimulated saliva for conductivity and [Na+] and that these rhythms were independent of reproductive state, whereas circadian rhythms in [K+] were dependent on reproductive state. Circadian rhythms for [glucose] were dependent on age. The loss of a rhythm in [glucose] with age indicates that glucose, Na+ and K+ are not linked in their entry into saliva. The influence of entry and reabsorption on the final concentrations of glucose, Na+ and K+ in saliva is discussed.

An evaluation of the efficacy of Doppler flow velocity waveform analysis as a screening test in pregnancy.

In a prospective double-blind study of 535 medium-risk pregnancies, growth data obtained by ultrasonography and Doppler flow velocity waveform systolic to diastolic ratios were recorded at 18, 24, 28, and 34 weeks' gestation. A significant association was observed between uteroplacental systolic to diastolic ratios at 24 weeks' gestation and subsequent fetal hypoxia with a sensitivity of 24.0% and a specificity of 93.9%. However, 70% of abnormal results were not followed by fetal hypoxia. Umbilical artery systolic to diastolic ratios at 24, 28, and 34 weeks' gestation were found to be predictive of intrauterine growth retardation. This predictive capability was enhanced in those growth-retarded fetuses in which hypoxia developed, but was weak when umbilical artery systolic to diastolic ratios were evaluated as primary screening tests for fetal hypoxia. The results confirm a role for Doppler systolic to diastolic ratios in the evaluation of high-risk pregnancies but do not support a role for their use as primary screening tests in low-risk obstetric populations.

An improved method of analysis of observer variation between pathologists.

An improved method of analysing interobserver variation in histopathological studies is described and illustrated, by use of data from a congruence survey of malignant melanoma. The method provides, between any number of pathologists, an assessment of overall agreement and of agreement on each individual category of a classification system. Adjustment for differences in chance agreement due to varying numbers of categories or an altered composition of cases is included in the analysis. A generalization of the procedure designed to measure the strength of associations between different categories is formulated and explained with the use of an example.

On a method of mortality analysis incorporating age--year interaction, with application to prostate cancer mortality.

A model used in the analysis of tabular mortality data (or incidence data), in which deaths are classified by age and year of occurrence, is shown to arise from the assumption that effects due to epoch of birth (the cohort effects) act multiplicatively on an underlying hazard function which may change with calendar year. The model disentangles the cohort effects and the effects due to epoch of death (year effects) and age at death (age effects), and incorporates an age--year interaction which has an interpretation in terms of changing the shape of the underlying hazard function. In addition, the derivation of the hazard function suggests meaningful interpretation of certain combinations of the model parameters. As an illustration, application of the model to prostate cancer mortality in England and Wales indicates, among other things, that after allowing the changing cohort effects, the risk for young men has increased relative to that for older men in recent years.

Recent trends in mortality from prostate cancer in male populations of Australia and England and Wales.

Mortality rates from cancer of the prostate in successive periods from 1908 to 1978 in Australia, and 1911 to 1977 in England and Wales, have been examined for trends with time and birth cohort. Age-specific rates and a proportional hazards model, designed to isolate the effect of birth cohort from those of calendar year and age, were used in the analysis. During the period of study, age-standardized mortality rose more than 5-fold in Australian men compared to just over 3-fold in men in England and Wales. In both countries the increases occurred almost entirely before 1960, with relative stability in age-standardized rates since then. The trends in mortality with year of birth were similar in the two sets of data. The risk of death from prostate cancer increased with successive birth cohorts to reach a peak in men born around 1865-1880 in Australia and men born around 1876-1896 in England and Wales. Males born later experienced a continuing reduction in rates, with the exception of age groups between 50 and 69 in which a further increase has appeared, starting with cohorts born after 1910. On the basis of current knowledge of the aetiology of prostate cancer, possible relationships between changes in sexual practices and prostate-cancer risk in successive generations have been explored. It is suggested that lowered sexual activity during the Great Depression may account for the recent cohort-based increases in mortality in middle-aged men.

Occurrence of breast cancer in Australian women.

No data are available on the incidence of breast cancer for the whole of Australia. Review of published incidence data from 1972 to 1978 from New South Wales, South Australia, and Tasmania shows that the incidence of breast cancer in women aged 25 years and over ranged between 97.8/100 000 and 106.5/100 000 of these women. These rates are similar to those in other countries with high rates. Comparison of mortality rates in Australian States between 1968 and 1978 for women aged 25 years and over showed rates from 28.3/100 000 in Victoria and 44.1/100 000 in the Australian Capital Territory. The rates for Victoria and the Australian Capital Territory were significantly higher than those in the other States. Trends in mortality in Australia for women aged 25 years and over were studied in the period from 1907 to 1977. Initially, the mortality rate was 28.5/100 000 and increased to reach a peak of 41.5/100 000 in 1940-1944. Thereafter, rates fell to 37.1/100 000 in 1960-1964, but have begun to rise again since 1970. The 1940-1944 peak was largely confined to women over the age of 50 years, and further analysis of the age-specific mortality rates showed the peak to be cross-sectional in type (that is, affecting each age group in the same calendar period). The rise after 1970 occurred mainly in women aged between 30 and 44 years and 60 and 79 years.

An analysis of trends in mortality from malignant melanoma of the skin in Australia.

Australian mortality rates from cutaneous malignant melanoma in successive periods from 1931 to 1977 have been examined with respect to geographic variation and trend with time and birth cohort. The age-standardized rates rose from 0.8/100,000 in males and 0.6/100,000 in females in 1931-34 to 4.2/100,000 and 2.5/100,000 in 1975-77. Mortality rates were highest in Queensland in the north of Australia and diminished on a gradient from the north to south of the country. An analysis designed to separate effects due to calendar year, birth cohort and age showed that virtually all the secular trend in rates could be explained by increases in successive birth cohorts, beginning as early as 1865 and stabilizing with the cohorts born around 1925 in women and 1935 in men. It is suggested that the cohort-based increase in mortality resulted from life-style changes occurring with successive generations. Its stabilization in recent birth cohorts, if persistent, suggests that the secular trend towards increasing total mortality from melanoma will also stabilize over the next 40 years.

Estimation of the mixing proportion in a mixture of two normal distributions from simple, rapid measurements.

A mixture of two or more normal distributions often provides an adequate model for the distribution of a population consisting of varying proportions of component subpopulations. We consider here the problem of estimating the mixing proportion in a mixture of two normal distributions, the parameters of which can be assumed known. Very large samples may be needed if reasonably precise estimates are to be obtained, thus bringing into consideration the cost or time involved in obtaining large numbers of exact measurements and computing the estimates from them. Simple estimators based on simple, rapidly obtained measurements may then be attractive alternatives provided efficiency losses are not too great. Three such estimators studied here are based on (a) the number of observations less than a fixed point r, (b) the nembers less than s and greater than t, and (c) the sample mean. Optimal choices of the points r, s and t are considered, and the efficiencies of the estimators relative to maximum likelihood estimators (MLE) using the full data are obtained. The simple estimators often perform sufficiently well to make the collection of full data not worthwhile in practice.