The EuroSIDA study: 25 years of scientific achievements.

The EuroSIDA study was initiated in 1994 and follows adult people living with HIV (PLHIV) in 100 collaborating clinics across 35 countries covering all European regions, Israel and Argentina. The study aims to study the long-term virological, immunological and clinical outcomes of PLHIV and to monitor temporal changes and regional differences in outcomes across Europe. Annually collected data include basic demographic characteristics, information on AIDS- and non-AIDS-related clinical events, and details about antiretroviral therapy (ART), hepatitis C treatment and other medications, in addition to a range of laboratory values. The summer 2016 data set held data from a total of 23 071 individuals contributing 174 481 person-years of follow-up, while EuroSIDA's unique plasma repository held over 160 000 samples. Over the past 25 years, close to 300 articles have been published in peer-reviewed journals (h-index 52), covering a range of scientific focus areas, including monitoring of clinical and virological outcomes, ART uptake, efficacy and adverse events, the influence of hepatitis virus coinfection, variation in the quality of HIV care and management across settings and regions, and biomarker research. Recognizing that there remain unresolved issues in the clinical care and management of PLHIV in Europe, EuroSIDA was one of the cohorts to found The International Cohort Consortium of Infectious Disease (RESPOND) cohort consortium on infectious diseases in 2017. In celebration of the EuroSIDA study's 25th anniversary, this article aims to summarize key scientific findings and outline current and future scientific focus areas.

Establishing a hepatitis C continuum of care among HIV/hepatitis C virus-coinfected individuals in EuroSIDA.

OBJECTIVES: The aim of the study was to establish a methodology for evaluating the hepatitis C continuum of care in HIV/hepatitis C virus (HCV)-coinfected individuals and to characterize the continuum in Europe on 1 January 2015, prior to widespread access to direct-acting antiviral (DAA) therapy. METHODS: Stages included in the continuum were as follows: anti-HCV antibody positive, HCV RNA tested, currently HCV RNA positive, ever HCV RNA positive, ever received HCV treatment, completed HCV treatment, follow-up HCV RNA test, and cure. Sustained virological response (SVR) could only be assessed for those with a follow-up HCV RNA test and was defined as a negative HCV RNA result measured > 12 or 24 weeks after stopping treatment. RESULTS: Numbers and percentages for the stages of the HCV continuum of care were as follows: anti-HCV positive (n = 5173), HCV RNA tested (4207 of 5173; 81.3%), currently HCV RNA positive (3179 of 5173; 61.5%), ever HCV RNA positive (n = 3876), initiated HCV treatment (1693 of 3876; 43.7%), completed HCV treatment (1598 of 3876; 41.2%), follow-up HCV RNA test to allow SVR assessment (1195 of 3876; 30.8%), and cure (629 of 3876; 16.2%). The proportion that achieved SVR was 52.6% (629 of 1195). There were significant differences between regions at each stage of the continuum (P < 0.0001). CONCLUSIONS: In the proposed HCV continuum of care for HIV/HCV-coinfected individuals, we found major gaps at all stages, with almost 20% of anti-HCV-positive individuals having no documented HCV RNA test and a low proportion achieving SVR, in the pre-DAA era.

Comprehensive nationwide analysis of mother-to-child HIV transmission in Finland from 1983 to 2013.

HIV-positive children are still born in Europe despite low mother-to-child transmission (MTCT) rates. We aimed to clarify the remaining barriers to the prevention of MTCT. By combining the national registers, we identified all women living with HIV delivering at least one child during 1983-2013. Of the 212 women delivering after HIV diagnosis, 46% were diagnosed during the pregnancy. In multivariate analysis, age >30 years (P = 0.001), sexual transmission (P = 0.012), living outside of the metropolitan area (P = 0.001) and Eastern European origin (P = 0.043) were risk factors for missed diagnosis before pregnancy. The proportion of immigrants increased from 18% before 1999 to 75% during 2011-2013 (P < 0.001). They were diagnosed during the pregnancy equally to natives and achieved similar, good treatment results. No MTCT occurred when the mother was diagnosed before the delivery. In addition, 12 women had delivered in 2 years prior their HIV diagnosis, most before implementation of the national screening of pregnant women. Three of these children were infected, the last one in 2000. Our data demonstrate that complete elimination of MTCT is feasible in a high-income, low-prevalence country. This requires ongoing universal screening in early pregnancy and easy access to antiretroviral therapy to all HIV-positive people.

The extent of B-cell activation and dysfunction preceding lymphoma development in HIV-positive people.

OBJECTIVES: B-cell dysfunction and activation are thought to contribute to lymphoma development in HIV-positive people; however, the mechanisms are not well understood. We investigated levels of several markers of B-cell dysfunction [free light chain (FLC)-κ, FLC-λ, immunoglobulin G (IgG), IgA, IgM and IgD] prior to lymphoma diagnosis in HIV-positive people. METHODS: A nested matched case-control study was carried out within the EuroSIDA cohort, including 73 HIV-positive people with lymphoma and 143 HIV-positive lymphoma-free controls. Markers of B-cell dysfunction were measured in prospectively stored serial plasma samples collected before the diagnosis of lymphoma (or selection date in controls). Marker levels ≤ 2 and > 2 years prior to diagnosis were investigated. RESULTS: Two-fold higher levels of FLC-κ [odds ratio (OR) 1.84; 95% confidence interval (CI) 1.19, 2.84], FLC-λ (OR 2.15; 95% CI 1.34, 3.46), IgG (OR 3.05; 95% CI 1.41, 6.59) and IgM (OR 1.46; 95% CI 1.01, 2.11) were associated with increased risk of lymphoma > 2 years prior to diagnosis, but not ≤ 2 years prior. Despite significant associations > 2 years prior to diagnosis, the predictive accuracy of each marker was poor, with FLC-λ emerging as the strongest candidate with a c-statistic of 0.67 (95% CI 0.58, 0.76). CONCLUSIONS: FLC-κ, FLC-λ and IgG levels were higher > 2 years before lymphoma diagnosis, suggesting that B-cell dysfunction occurs many years prior to lymphoma development. However, the predictive value of each marker was low and they are unlikely candidates for risk assessment for targeted intervention.

Incidence of cancer and overall risk of mortality in individuals treated with raltegravir-based and non-raltegravir-based combination antiretroviral therapy regimens.

OBJECTIVES: There are currently few data on the long-term risk of cancer and death in individuals taking raltegravir (RAL). The aim of this analysis was to evaluate whether there is evidence for an association. METHODS: The EuroSIDA cohort was divided into three groups: those starting RAL-based combination antiretroviral therapy (cART) on or after 21 December 2007 (RAL); a historical cohort (HIST) of individuals adding a new antiretroviral (ARV) drug (not RAL) to their cART between 1 January 2005 and 20 December 2007, and a concurrent cohort (CONC) of individuals adding a new ARV drug (not RAL) to their cART on or after 21 December 2007. Baseline characteristics were compared using logistic regression. The incidences of newly diagnosed malignancies and death were compared using Poisson regression. RESULTS: The RAL cohort included 1470 individuals [with 4058 person-years of follow-up (PYFU)] compared with 3787 (4472 PYFU) and 4467 (10 691 PYFU) in the HIST and CONC cohorts, respectively. The prevalence of non-AIDS-related malignancies prior to baseline tended to be higher in the RAL cohort vs. the HIST cohort [adjusted odds ratio (aOR) 1.31; 95% confidence interval (CI) 0.95-1.80] and vs. the CONC cohort (aOR 1.89; 95% CI 1.37-2.61). In intention-to-treat (ITT) analysis (events: RAL, 50; HIST, 45; CONC, 127), the incidence of all new malignancies was 1.11 (95% CI 0.84-1.46) per 100 PYFU in the RAL cohort vs. 1.20 (95% CI 0.90-1.61) and 0.83 (95% CI 0.70-0.99) in the HIST and CONC cohorts, respectively. After adjustment, there was no evidence for a difference in the risk of malignancies [adjusted rate ratio (RR) 0.73; 95% CI 0.47-1.14 for RALvs. HIST; RR 0.95; 95% CI 0.65-1.39 for RALvs. CONC] or mortality (adjusted RR 0.87; 95% CI 0.53-1.43 for RALvs. HIST; RR 1.14; 95% CI 0.76-1.72 for RALvs. CONC). CONCLUSIONS: We found no evidence for an oncogenic risk or poorer survival associated with using RAL compared with control groups.

Three-decade neurological and neurocognitive follow-up of HIV-1-infected patients on best-available antiretroviral therapy in Finland.

OBJECTIVES: Is it possible to live without neurocognitive or neurological symptoms after being infected with HIV for a very long time? These study patients with decades-long HIV infection in Finland were observed in this follow-up study during three time periods: 1986-1990, in 1997 and in 2013. SETTING: Patients from greater Helsinki area were selected from outpatient's unit of infectious diseases. PARTICIPANTS: The study included 80 HIV patients. Patients with heavy alcohol consumption, central nervous system disorder or psychiatric disease were excluded. PRIMARY AND SECONDARY OUTCOME MEASURES: The patients underwent neurological and neuropsychological examinations, MRI of the brain and laboratory tests, including blood CD4 cells and plasma HIV-1 RNA. Neuropsychological examination included several measures: subtests of Wechsler Adult Intelligence Scale, Wechsler Memory Scale-Revised, list learning, Stroop and Trail-Making-B test. The Beck Depression Inventory and Fatigue Severity Scale were also carried out. The obtained data from the three time periods were compared with each other. RESULTS: Owing to high mortality among the original 80 patients, eventually, 17 participated in all three examinations performed between 1986 and 2013. The time from the HIV diagnosis was 27 (23-30) years. Blood CD4 cells at the diagnosis were 610 (29-870) cells/mm(3), and the nadir CD4 168 (4-408) cells/mm(3). The time on combined antiretroviral treatment was 13 (5-17) years. 9 patients suffered from fatigue, 5 had polyneuropathy and 3 had lacunar cerebral infarcts. There was a subtle increase of brain atrophy in 2 patients. Mild depressive symptoms were common. The neuropsychological follow-up showed typical age-related cognitive changes. No HIV-associated dementia features were detected. CONCLUSIONS: Polyneuropathy, fatigue and mild depression were common, but more severe neurological abnormalities were absent. These long-term surviving HIV-seropositive patients, while on best-available treatment, showed no evidence of HIV-associated neurocognitive disorder in neuropsychological and neuroradiological evaluations.

Podocyte apoptosis is prevented by blocking the Toll-like receptor pathway.

High serum lipopolysaccharide (LPS) activity in normoalbuminuric patients with type 1 diabetes (T1D) predicts the progression of diabetic nephropathy (DN), but the mechanisms behind this remain unclear. We observed that treatment of cultured human podocytes with sera from normoalbuminuric T1D patients with high LPS activity downregulated 3-phosphoinositide-dependent kinase-1 (PDK1), an activator of the Akt cell survival pathway, and induced apoptosis. Knockdown of PDK1 in cultured human podocytes inhibited antiapoptotic Akt pathway, stimulated proapoptotic p38 MAPK pathway, and increased apoptosis demonstrating an antiapoptotic role for PDK1 in podocytes. Interestingly, PDK1 was downregulated in the glomeruli of diabetic rats and patients with type 2 diabetes before the onset of proteinuria, further suggesting that reduced expression of PDK1 associates with podocyte injury and development of DN. Treatment of podocytes in vitro and mice in vivo with LPS reduced PDK1 expression and induced apoptosis, which were prevented by inhibiting the Toll-like receptor (TLR) signaling pathway with the immunomodulatory agent GIT27. Our data show that LPS downregulates the cell survival factor PDK1 and induces podocyte apoptosis, and that blocking the TLR pathway with GIT27 may provide a non-nephrotoxic means to prevent the progression of DN.

Clinical and demographic factors associated with low viral load in early untreated HIV infection in the INSIGHT Strategic Timing of AntiRetroviral Treatment (START) trial.

OBJECTIVES: A small subset of HIV-positive adults have low HIV RNA in the absence of therapy, sometimes for years. Clinical factors associated with low HIV RNA in early infection have not been well defined. METHODS: We assessed factors associated with low plasma HIV RNA level at study entry in the Strategic Timing of AntiRetroviral Treatment (START) trial. All START participants had a baseline HIV RNA assessment within 60 days prior to randomization. The key covariables considered for this analysis were race, and hepatitis B virus (HBV) and hepatitis C virus (HCV) status. We assessed factors associated with HIV RNA ≤ 50 and ≤ 400 HIV-1 RNA copies/mL using logistic regression. Because of the strong association between region of randomization and baseline low HIV RNA, analyses were stratified by region. RESULTS: We found that, of 4676 eligible participants randomized in START with a baseline HIV RNA assessment, 113 (2.4%) had HIV RNA ≤ 50 copies/mL at baseline, and a further 257 (5.5%) between 51 and 400 copies/mL. We found that HIV exposure routes other than male homosexual contact, higher high-density lipoprotein (HDL) cholesterol levels, higher CD4 cell counts, and higher CD4:CD8 ratio were associated with increased odds of low HIV RNA. HCV antibody positivity was borderline statistically significantly associated with low HIV RNA. Race and HBV surface antigen positivity were not significantly associated with low HIV RNA. CONCLUSIONS: In a modern cohort of individuals with early untreated HIV infection, we found that HIV exposure routes other than male homosexual contact and higher HDL cholesterol were associated with increased odds of low HIV RNA.

A survey of ATRIPLA use in clinical practice as first-line therapy in HIV-positive persons in Europe.

ATRIPLA is licensed for use only in HIV-positive persons whose viral loads <50 for ≥ 3 months. We investigated the use of ATRIPLA as first-line antiretroviral therapy (ART) in EuroSIDA using a web-based survey performed in Autumn 2012. 96/112 clinics (85.7 %) completed the survey. Recommendations when initiating first-line ART was TRUVADA plus efavirenz in 36 (37.5 %), ATRIPLA in 35 (36.5 %), a different first-line regimen in 12 clinics (12.5 %), and no recommendation in 7 clinics (7.3 %). ATRIPLA was commonest in Northern (15/21 clinics; 71.4 %), and least common in Eastern Europe (2/31 clinics; 6.5 %; p < 0.0001). Over one-third of the participating clinics in this survey were using ATRIPLA as first-line antiretroviral therapy, despite EMA recommendations.

Severe bacterial non-aids infections in HIV-positive persons: incidence rates and risk factors.

OBJECTIVES: This study aimed to determine incidence rates (IR) and identify risk factors for severe bacterial non-AIDS infections (SBnAI) requiring hospital admission. METHODS: Data from the prospective EuroSIDA cohort were utilized to determine IRs of first diagnosis of the following SBnAI requiring hospital admission: bacteremia, endocarditis, meningitis, peritonitis, pneumonia, osteitis, and pyolonephritis. Incidence rate-ratios (IRRs) and risk factors were assessed by Poisson regression. RESULTS: During 35,839 person-years of follow-up (PYFU), 275 patients were diagnosed with SBnAI (IR = 7.67 per 1000 PYFU, 95% confidence interval: 6.79-8.64). The most frequent infections were pneumonia (IR = 5.36, 4.63-6.17), bacteremia (IR = 1.14, 0.82-1.55), and pyelonephritis (IR = 0.67, 0.43-1.00). A strong risk factor for SBnAI was reduced estimated glomerular filtration rate [eGFR] (adjusted IRR = 5.07, 2.12-12.1 and IRR = 2.73, 1.63-4.56 for eGFR ≤ 60 and 60.1-90 compared to eGFR > 90, respectively). No current combined antiretroviral therapy (cART) compared with current cART use increased the risk of SBnAI (adjusted IRR = 2.96, 2.03-4.32). Other risk factors for SBnAI included current CD4+ count <350 cells/µL, female gender, age, infection with HIV through IDU, prior AIDS diagnosis, and anaemia. CONCLUSIONS: Enhanced attention directed towards people with comorbidity is warranted to limit the burden of these infections.

Longer prior exposure to zidovudine/lamivudine-containing combination antiretroviral therapy, age, and male gender are each associated with reduced subcutaneous adipose tissue.

OBJECTIVES: Whether zidovudine (AZT)-associated lipoatrophy occurrence differs by concomitant exposure to protease (PIs) or non-nucleoside reverse transcriptase inhibitors (NNRTIs) remains unclear. Baseline body composition data from a randomized trial in subjects stable on first-line AZT-based therapy were used to explore this issue. METHODS: In this substudy of the PREPARE trial, centrally read baseline whole-body dual energy x-ray aborptiometry (DXA) and single-slice abdominal CT scans were analyzed with respect to duration and type of prior AZT/lamivudine (3TC) combination antiretroviral therapy (cART), including by multivariate linear regression adjusted for age, gender, ethnicity, body mass index (BMI), and nadir CD4. RESULTS: DXA and CT, from 134 and 136 patients, respectively [87% male; 82% Caucasian; mean (SD) age, 45.6 years (10); BMI, 24.3 kg/m² (3.2)], were analyzed. Prior AZT/3TC cART exposure was 5.5 (2.2) years. Seventy-eight and 27 patients had concomitantly and exclusively used NNRTIs and PIs, respectively. AZT/3TC cART, AZT/3TC/NNRTI, and AZT/3TC/PI, respectively, were associated with the presence of a mean (95% CI) of 247 g (-438 to -56; P = .012), 267 g (-467 to -66; P = .010), and 216 g (-430 to -1.7; P = .048) less baseline limb fat per additional year of prior exposure. Although abdominal subcutaneous (SAT) adipose tissue was likewise less with longer AZT/3TC cART, this was only significant for AZT/3TC/ NNRTI but not AZT/3TC/PI. Visceral adipose tissue (VAT) amount was not clearly associated to prior treatment. Increased age and male gender were independently associated with lower limb fat and SAT, but more VAT. CONCLUSIONS: Longer exposure to AZT/3TC, regardless of whether in combination with PI or NNRTI, as well as increased age and male gender are independently associated with lower limb fat mass.

Determinants of late HIV diagnosis among different transmission groups in Finland from 1985 to 2005.

OBJECTIVES: To study determinants of late HIV diagnosis in a low-HIV-prevalence (<0.1%) country where HIV spread among men who have sex with men (MSM) and heterosexuals in the 1980s, and among injecting drug users (IDUs) in the late 1990s. METHODS: Newly diagnosed HIV cases referred to the Helsinki University Central Hospital between 1985 and 2005 were reviewed to identify determinants of late HIV diagnosis, defined as diagnosis when the first CD4 count was <200 cells/microL, or when AIDS occurred within 3 months of HIV diagnosis. Determinants of late diagnosis were analysed using multivariate logistic regression. RESULTS: Among 934 HIV cases, 211 (23%) were diagnosed late. In the first 4-year interval of each sub-epidemic (1985-1989 for MSM and heterosexuals, 1998-2001 for IDUs), rates of late HIV diagnosis were 13%, 18% and 6%, respectively, but increased thereafter to 29%, 27% and 37%. Late diagnosis was associated with non-Finnish ethnicity, older age, male gender, lack of earlier HIV testing, diagnosis at health care settings and later stage of the sub-epidemic. CONCLUSIONS: The lower rate of late diagnosis in the first 4-year interval of each HIV sub-epidemic suggests that the early stages of the HIV epidemic in Finland were detected early. This factor may have contributed to the low prevalence of HIV infection in Finland. The stage and age of the epidemic should be taken into account when interpreting the data on late HIV diagnosis, especially in cross-country comparisons.

Control of methicillin-resistant Staphylococcus aureus outbreak involving several hospitals.

This population-based, retrospective, cohort study describes a large methicillin-resistant Staphylococcus aureus (MRSA) epidemic caused by one strain (E1) in the greater Helsinki region. The epidemic comprised 210 cases at several hospitals, but was finally controlled. The study period ranged from June 1991 to December 2000. The epidemic peaked in 1993-1995 with 143 cases (68% of total cases). From August 1993, all MRSA-positive cases at the eight municipal hospitals were isolated and barrier nursed. Contacts were cohorted and screened for MRSA colonization. Decolonization treatment was administered to some chronic carriers. MRSA cases and contacts were identified in the joint patient register of the municipal hospitals from August 1993. The annual incidence of MRSA E1 in Helsinki City area per 100,000 inhabitants rose from 0.2 in 1991 to 13.6 in 1994. It decreased from 1995, reaching 0.7 per 100,000 in 2000. A jointly agreed policy on MRSA and timely co-operation between all units were essential for control of this epidemic.

Sources of disseminated Mycobacterium avium infection in AIDS.

OBJECTIVES: To identify the sources of disseminated Mycobacterium avium complex (MAC) infection in AIDS. METHODS: HIV positive subjects with CD4 counts <100/mm(3) in Atlanta, Boston, New Hampshire and Finland were entered in a prospective cohort study. Subjects were interviewed about potential MAC exposures, had phlebotomy performed for determination of antibody to mycobacterial lipoarabinomannin and for culture. Patient-directed water samples were collected from places of residence, work and recreation. Patients were followed for the development of disseminated MAC. Univariate and multivariate risk factors for MAC were analyzed. RESULTS: Disseminated MAC was identified in 31 (9%) subjects. Significant risks in univariate analysis included prior Pneumocystis carinii pneumonia (PCP) (hazard ratio 1.821), consumption of spring water (4.909), consumption of raw seafood (34.3), gastrointestinal endoscopy (2.894), and showering outside the home (0.388). PCP, showering and endoscopy remained significant in a Cox proportional hazards model. There was no association between M. avium colonization of home water and risk of MAC. In patients with CD4<25, median OD antibody levels to lipoarabinomannin at baseline were 0.054 among patients who did not develop MAC and 0.021 among patients who did develop MAC (P=0.077). CONCLUSIONS: MAC infection results from diverse and likely undetectable environmental and nosocomial exposures. Mycobacterial infection before HIV infection may confer protection against disseminated MAC in advanced AIDS.

Neutrophil free oxygen radical production and blood total antioxidant capacity in patients with coronary heart disease using various medications.

Despite convincing results of studies in vitro, less is known about the effects of antioxidants on in vivo redox balance in humans. We developed a novel parameter of in vivo redox balance, and studied it and its relation to dental infections in 51 patients on medication for coronary heart disease (CHD) and 39 random controls matched for age group, sex, social class and locality. In vivo redox balance was the ratio of plasma antioxidant capacity, as measured with radical-trapping assay, to neutrophil respiratory burst capacity, as measured with whole blood chemiluminescence assay. Dental infections were quantitated with four rating scales. CHD patients had higher values than controls. Patients on acetosalicylic acid (ASA), diuretics or beta blockers, but not the ones on calcium channel blocker, had significantly higher redox balance than non-users. Combination of calcium channel blockers and ASA was associated with redox balance similar to taking beta blockers or diuretics. Diuretics and ASA were independent determinants of redox balance in multivariate analyses. Redox balance did not correlate with severity of dental infections (Spearman's r 0.06 to 0.11). The results contrast experimental data indicating that calcium channel blockers are as antioxidants superior to other cardiovascular drugs. Total antioxidant capacity in parallel with oxygen species production capacity should be considered in attempts to solve the antioxidant paradox.

High rates of disseminated infection due to non-tuberculous mycobacteria among AIDS patients in Finland.

OBJECTIVE: to determine the rate of disseminated infection due to non-tuberculous mycobacteria (NTM) among Finnish AIDS patients, and to analyse the epidemiology of these infections. METHODS: in a prospective cohort study HIV-infected patients with CD4 counts < 200 x 10(6)/l were interviewed, and had mycobacterial blood cultures performed at baseline and at 6 months, then subsequently for clinical indications; autopsies were performed on patients who died. The cohort was followed at least for 24 months or to death. Water samples were collected from the homes of patients and from the environment and cultured for organisms of the Myobacterium avium complex (MAC). Environmental and clinical isolates were compared using pulsed field gel electrophoresis (PFGE). RESULTS: NTM infection occurred in 22 (43%) of 51, 19 isolates were Mycobacterium avium, two M. genavense and one M. intracellulare. Multivariate analysis identified urban residence (P=0.04) and eating raw fish (P=0.04) as independent risk factors. Molecular analysis revealed two clusters of related isolates (three M. avium, two M. genavense) among urban residents. CONCLUSION: AIDS patients in Finland have high rates of disseminated infection due to NTM. Clusters of identical organisms and association with urban residence suggests that these are newly acquired infections in advanced AIDS.

Isolation of mycobacteria from indoor swimming pools in Finland.

The presence of mycobacteria in seven indoor pools in Finland was evaluated by multiple culture methods. Replicate samples, with and without inactivation of chlorine by sodium thiosulfate, were cultured in two laboratories. Laboratory I used two methods: (A) no decontamination and (B) cetylpyridinium chloride (0.005%, 20 min); and Laboratory II two methods: (C) cetylpyridinium chloride (0.005%, 18 h) and (D) oxalic acid (5%, 15 min). Samples processed by methods (A) and (B) were cultured on different egg media of pH 6.3 or 5.8; by method (C) on Middlebrook and Cohn 7H10 (+OADC) agar of pH 5.5; and by method (D) on Middlebrook and Cohn 7H10 agar (+OADC) with cycloheximide (500 microg/ml). Mycobacteria were recovered from five (71%) of seven pools. Detection of mycobacteria depended on the method used. High isolation rates (36-46% of the samples) were obtained by methods (A), (B) and (D). Contamination was a problem only with method (A). Inactivation of chlorine had a variable impact on mycobacterial detection. Isolates included M. kansasii, M. gordonae, M. fortuitum complex, M. sphagni, and M. vaccae, as well as M. simiae-like and M. chubuense-like organisms. In addition, a group of slowly growing and a group of rapidly growing isolates with previously unknown fatty acid and alcohol composition were isolated. No M. avium was detected. Mycobacterial counts were highest in a small pool with high temperature, low pH, and low content of free available chlorine.