THE LANCET 1076 Vol 350 • October 11, 1997 baseline concentrations of Hb A1C and serum glucose were 6·69% (1·10) and 102·2 mg/dL (19·9). Comparing the isradipine and diuretic groups, the adjusted HR for experiencing a major vascular event among those below the median for Hb A1C (6·6%) was 1·11 (95% CI 0·43–2·90), but was 2·81 (1·09–7·26) for those above the median. The corresponding HRs for serum glucose (5·3 mmol/L) were 1·48 (0·57–3·82) and 2·22 (0·88–5·59). Patients above the median for Hb A1C had high HRs regardless of the serum glucose concentration: 2·90 for the high-glucose and 2·44 for the low-glucose subgroups. A significant relationship between the risk of a cardiovascular event and Hb A1C was seen in the isradipine group but not in the diuretic group (figure A ). The isradipine/diuretic HR of having a cardiovascular event increased with increasing baseline Hb A1C (figure B). The increased-susceptibility hypothesis is also supported by the observation that in the isradipine group, those with high Hb A1C had a greater blood pressure reduction than those with a low Hb A1C , in spite of similar drug dosing. The longer number of cardiovascular events associated with use of isradipine was found to be among those hypertensive non-diabetic patients with impaired glucose metabolism. These observations, which are biologically plausible, require prospective confirmation. 1 Mason RP, Moisey DM, Shajenko L. Cholesterol alters the binding of Ca 2+ channel blockers in the membrane lipid bilayer. Mol Pharmacol 1992; 41: 315–21. 2 Mazzanti L, Rabini RA, Faloia E, et al. Altered cellular Ca 2+ and Na + transport in diabetes mellitus. Diabetes 1990; 39: 850–54. 3 Borhani NO, Mercuri M, Borhani PA, et al. Final outcome results of the Multicenter Isradipine Diuretic Atherosclerosis Study (MIDAS). A randomized controlled study. JAMA 1996; 276: 785-91. Department of Public Health Sciences, Bowman Gray School of Medicine Winston-Salem, NC 27157-1063, USA (C D Furberg); and Department of Preventive Medicine, University of Tennessee, Memphis, TN these men had urethral inflammation, and 42% (8/19) were LED-test positive. Both of these findings were significantly more frequent than for men without symptoms with no urethral pathogen isolated, of whom 16% (66/419) had urethral inflammation (OR 3·1: 95% CI 1·4–8·2, p=0·02), and 17% (72/419) were LED test-positive (OR 3·5: 95% CI 1·4–9·0, p<0·01). 33% (5/15) of unmixed Ng infection were without symptoms; 60% of symptom-free men with Ng infection had urethral inflammation and 80% were LED positive. 77% (10/13) of men with unmixed Ct infections were without symptoms; 20% of men with symptomless Ct infection had urethral inflammation and 70% were LED positive. Absence of symptoms does not necessarily imply absence of urethral inflammation, especially when Tv or Ct are the infecting pathogens. The association between higher prevalence of Tv infection and older age is unusual for a sexually transmitted disease, and suggests that infection may be of long duration. Tv infection is associated with less severe symptoms than Ng infection and leads fewer men to seek treatment. Even among men with symptoms, guidelines for syndromic management regimens for urethral discharge do not generally include treatment of Tv. So even if a man seeks treatment, he is unlikely to be adequately managed. Tv infection has not generally been included in projections of population-attributable risk fractions for HIV-1 infection associated with various STD pathogens. However, Tv infection is common among women, as evidenced by a rural community study from Uganda which described a Tv infection prevalence of 47%. 3 Tv is probably more common among men than is generally appreciated, and long-standing, low-grade urethral inflammation may lead to an increased risk of male to female transmission of HIV-1. Further study of the role of Tv in HIV-1 transmission is warranted. If Tv does enhance HIV-1 transmission, mass treatment strategies for Tv infection could potentially play a valuable part in HIV-1 prevention in developing countries, as an effective single-dose, low-cost treatment for Tv is available worldwide. 1 Cohen MS, Hoffman IF, Royce RA, et al. Reduction of concentration of HIV-1-1 in semen after treatment of urethritis: implications for prevention of sexual transmission of HIV-1. Lancet 1997; 349: 1868–73. 2 Jackson DJ, Rakwar JP, Chohan B, et al. Urethral infection in a workplace population of east African men: evaluation of strategies for screening and management. J Infect Dis 1997; 175: 833–38. 3 Wawer MJ, McNairn D, Wabwire-Mangen F, Paxton L, Gray RH, Kiwanuka N. Self-administered vaginal swabs for population-based assessment of Trichomonas vaginalis prevalence. Lancet 1995; 345: 130–31. Department of Medical Microbiology, University of Nairobi, Kenya (D J Jackson); Departments of Medical Microbiology, Medicine and Community Health Sciences, University of Manitoba, Canada; Departments of Epidemiology and Medicine, University of Washington, Seattle, USA Urethral Trichomonas vaginalis infection and HIV-1 transmission D J Jackson, J P Rakwar, J J Bwayo, J K Kreiss, S Moses Increased shedding of HIV-1 in semen in the presence of urethral infection has been described by Cohen and colleagues, in a paper which addressed several important issues concerning HIV-1 control in developing countries. 1 Of particular interest to us was their report of four HIV-1- positive men from Malawi with symptomatic urethritis and unmixed Trichomonas vaginalis (Tv) urethral infection who were observed to have significantly increased concentrations of HIV-1 in their semen. We have published a study of urethral infections in a male workplace population in Mombasa, Kenya. 2 In that study, Tv was isolated by culture in 6% (30/504) of men, and was the most commonly isolated urethral pathogen. Age, presence of urethral inflammation on urethral swab, gram stain (five polymorphonuclear cells per high power field), and urine leucocyte esterase dipstick (LED) results of men with unmixed infections with Tv, Neisseria gonorrhoeae (Ng), Chlamydia trachomatis (Ct), or no infection are given in the table. Results from men with and without symptoms are included in the table, but seven men with mixed infections have been excluded. Tv-culture-positive men were older than men with Ng infection (p=0·003). 83% (19/23) of men with unmixed Tv infections were without symptoms. However, 37% (7/19) of Pathogen isolated Number Mean age Urethral Urine (unmixed infections of men (95% CI) inflammation leucocyte only) (n=497) on gram stains esteraie (5 PMNs dipstick per high positive power field) (>trace) Trichomonas vaginalis 23 32·3 (28·4–35·7) 8 (35%) 13 (57%) Neisseria gonorrhoeae 15 24·7 (22·7–27·0) 12 (87%) 13 (87%) Chlamydia trachomatis 13 29·2 (23·8–31·6) 3 (23%) 9 (69%) No pathogen 446 29·9 (29·3–30·5) 73 (16%) 83 (19%) PMN=polymorphonuclear white blood cells. Urethral infection status and age, urethral inflammation, and leucocyte esterase urine test results in trucking workers in Kenya