In Other Journals

HABITS trial terminated

Scandinavian researchers have pulled the plug on their own trial, HABITS (hormonal replacement therapy after breast cancer — is it safe?).1 An interim safety analysis found that breast cancer survivors allocated to receive 2 years of hormone replacement therapy (HRT) for menopausal symptoms were experiencing an "unacceptably high risk" of breast cancer compared with those allocated to best symptomatic treatment without hormones (26/174 v 7/171; relative hazard 3.5). Any participants currently receiving HRT were advised to discontinue. An accompanying commentary said that the HABITS finding was consistent with those from the Women’s Health Initiative (WHI) trial.2

1. Lancet 2004; 363: 453-455 2. Lancet 2004; 363: 410-411

Don't drink plenty of fluids

"Pushing fluids" beyond satisfying thirst in patients with respiratory infections may cause serious harm, caution Australian authors. Their systematic review found that, although no randomised controlled trials have compared increased and restricted fluid intake in patients with respiratory infection, reports from two prevalence studies and several case series suggest that an increased fluid intake could lead to hyponatraemia, especially in lower respiratory tract infection. Long-held concerns about the risks of fluid loss from fever and respiratory tract evaporation may be unfounded, thanks to a compensatory increase in antidiuretic hormone secretion.

BMJ 2004; 328: 499-500

When two betters one . . .

. . . in HIV survival

The US Multicenter AIDS Cohort Study has found that long-term persistent co-infection with the non-pathogenic GB virus C (GBV-C) — an RNA flavivirus closely related to hepatitis C — is associated with longer survival in men with HIV. The link between GBV-C viraemia and survival held at 5 to 6 years, but not at 12 to 18 months, after HIV seroconversion. The protective effect of co-infection is lost, and the clinical prognosis worse, once GBV-C is cleared.

N Engl J Med 2004; 350: 981-990

. . . in rheumatoid remission

Combined treatment of active, adult-onset rheumatoid arthritis with a subcutaneous tumour necrosis factor antagonist and an oral antimetabolite leads to better disease control than using either agent alone, according to an international study involving nearly 700 patients. In the double-blind, randomised controlled trial of etanercept and methotrexate with radiographic patient outcomes (TEMPO), patients receiving both agents concurrently were much more likely to achieve clinical remission at 1 year (35% v 16% for etanercept and 13% for methotrexate). Further, the scoring of serial joint x-rays not only found that joint damage could be slowed but also suggested that repair of structural damage may be achievable.

Lancet 2004; 363: 675-681

How to tell if it’s bacterial

A simple, sensitive assay can safely reduce unnecessary antibiotic use in lower respiratory tract infection (LRTI), say Swiss researchers. Levels of circulating calcitonin precursors, including procalcitonin, are raised in severe bacterial infections but stay fairly low in viral infections and non-specific inflammatory diseases. In a randomised study, they allocated 243 patients admitted with a suspected LRTI to either standard care or procalcitonin-guided treatment. Clinicians were advised about the likelihood of bacterial infection according to the serum procalcitonin level, ranging from "unlikely" if ≤ 0.1g/L to "suggestive" if ≥ 0.5g/L.

The procalcitonin patient group were about half as likely to receive antibiotics as those receiving standard care, equating to 39 fewer courses per 100 patients with LRTI, with no difference in outcomes.

The researchers tempered their enthusiasm for the procalcitonin assay by acknowledging that circulating levels can also be enhanced in non-infectious disorders, and may remain low even in sepsis. As always, consider all test results in light of the clinical context.

Lancet 2004; 363: 600-607

Bright sparks burn longer

Good cognitive ability in both childhood and as a mature adult may protect against mid-life cognitive decline, according to a longitudinal study. Richards et al analysed full sets of data from 2058 men and women, all born in England, Scotland or Wales in the same week in 1946; all had completed various cognitive tests at ages 15, 43 and 53 years. Childhood ability was inversely linked with the rate of decline in tasks of memory, speed and concentration as measured in mid-life. Adult ability was independently and similarly linked, leading to the researchers' suggestion that adult training and education may protect against cognitive decline in later life.

BMJ 2004; 328: 552-556

— Dr Ann Gregory, MJA