Biofilm has been shown to be important in prosthetic infections, dental plaque and cystic fibrosis; it has now been found in an animal model of otitis media (OM). Researchers in the United States injected H. influenzae into the middle ears of chinchillas. Starting treatment with ampicillin 72 hours later rendered any effusion culturally sterile. Pairs of chinchillas were killed at 10 intervals, three hours to three weeks after being infected, and their eardrums removed. Scanning electron microscopy showed biofilm formation in all of the animals which had developed middle ear effusions. Microcolony formation was evident after 24 hours, and by five days mature biofilm was present, with tower-type structures consisting of many layers of bacteria in an exopolysaccharide matrix. Frozen specimens were stained (so that, with confocal laser scanning microscopy, live bacterial cells appeared green and dead cells red), showing viable bacteria within the biofilm.

JAMA 2002; 287: 1710-1715

Overseas research supports the Australian approach of not screening infants for neuroblastoma. Screening of urine for catecholamine metabolites was offered to 476 654 children in Quebec, Canada,¹ between May 1989 and April 1994; 92% were screened at age three weeks and/or six months. Neuroblastoma was detected in 43 infants, who were all still alive in April 2000. However, 22 children in the target population died from neuroblastoma during the 6–11 years’ follow-up (diagnosed aged < 3 weeks, 3; screened negative, 18; not screened, 1). Based on comparisons with other populations (eg, Ontario, Minnesota, Florida), rates of death due to neuroblastoma in children under eight years had not been reduced.

In Germany,² urine screening was offered to 2.5 million children in 6 out of 16 states from 1995 to 2000 (1 475 773 children were tested). The screening detected 149 cases; to date three children have died, all from complications of treatment. By June 2001, 55 children with negative screening results had presented with neuroblastoma, 14 of whom had died. Children in the screened group and in the control states had similar rates of both stage 4 and fatal neuroblastoma.

The Canadian group¹ note that a body of evidence points to at least two clinical and biological entities. Very few cases of neuroblastoma detected by screening have unfavourable biological features, and there is a high rate of spontaneous regression or maturation into benign ganglioneuromas. On the other hand, disease with an unfavourable prognosis is rarely detectable by screening.

1. N Engl J Med 2002; 346: 1041-1046
2. N Engl J Med 2002; 346: 1047-1053

In a Swiss study, training young physicians to identify patients’ readiness to stop smoking, then implement stage-specific strategies, resulted in improved rates of smoking cessation among their outpatients. The eight hours of training included video demonstrations of motivational interviewing, role-playing, practice interviews and written materials. Doctors were told that a survey of cardiovascular risk factors was being done; they were not aware that they had been randomly assigned to training in smoking cessation (n =17) or a talk on managing dyslipidaemia (controls, n =18). Both the research assistants and the patients (aged 36614 years) whom they interviewed after their outpatient clinic appointment were blind to key elements of the study. At one-year follow-up, 15 of 115 patients treated by intervention group doctors reported not having smoked in the previous week, compared to 7 of 136 control group patients (13% [95% CI, 7%–21%] v 5% [95% CI, 1%–9%]; P = 0.005).

Ann Intern Med 2002; 136: 429-437

The results of a review of antidepressant trials in outpatients with major depressive disorder confirm the necessity of establishing the efficacy of any new antidepressant against placebo. A systematic search identified 75 randomised, placebo-controlled trials published between 1981 and 2000. Entry to most studies required > 2 weeks of symptoms and a defined minimum score on the Hamilton Rating Scale for Depression (HRSD), and response was a reduction of ≥ 50% in HRSD score. Overall, response to placebo was 29.7% (SD, 8.3%; range, 12.5%–51.8%), and to antidepressant 50.1% (SD, 9%; range, 31.6%–70.4%). Although both response rates increased with year of publication, the association for placebo response rates was stronger. The researchers, from Columbia University, could not identify other factors that accounted for the increase in placebo response.

JAMA 2002; 287: 1840-1847

Home | Issues | eMJA shop | My account | Classifieds | More... | Contact | Topics | Search

The Medical Journal of Australia    eMJA