2 June 2003

SARS: the genetics . . .

As this epidemic continues, scientists in Singapore have sequenced the viral genome of isolates from the local index patient and four contacts.1 Comparing the genotypes of these isolates with those from Canada, Hong Kong, Vietnam and China, it is emerging that they show genetic "signatures", allowing us to track their geographic origins. These findings on the newly dubbed SARS coronavirus are a two-edged sword: its genome is unlikely to change rapidly and the virus may thus not mutate to a benign infection but a vaccine may be more achievable (though still a long way off).2

1. Lancet. May 9, 2003. http://image.thelancet.com/extras/03art4454web.pdf [accessed 12 May 2003]

2. Lancet. May 9, 2003. http://image.thelancet.com/extras/03cmt124web.pdf [accessed 12 May 2003]

. . . and the patients

A prospective study followed 75 SARS patients from the same Hong Kong housing estate for three weeks. Fever and pneumonia initially abated with treatment.1 However, most patients developed recurrent fever, diarrhoea and radiological worsening of their pulmonary abnormalities. A few developed pneumomediastinum and acute respiratory distress syndrome, the latter being associated with increasing age and chronic hepatitis B infection. Respiratory worsening in Week 2 may be due to immunopathological damage rather than viral replication, the authors speculate. Peak nasopharyngeal viral load occurred at Day 10, and viral RNA was detected in faeces two weeks after onset, suggesting that oral-faecal transmission is possible.

Meanwhile, follow-up of the first 10 children treated for SARS in Hong Kong showed that, compared with adults and teenagers, younger children appear to develop milder disease.2

1. Lancet. May 9, 2003. http://image.thelancet.com/extras/03art4432web.pdf [accessed 12 May 2003]

2. Lancet. April 29, 2003. http://image.thelancet.com/extras/03let4127web.pdf [accessed 12 May 2003]

Lowdown on depression

An Australian good news/bad news story: our overall suicide rates remained constant between 1991 and 2000, because increased rates in younger adults were offset by decreases in older adults.1 This analysis of Australian Bureau of Statistics and pharmaceutical sales data showed that the age-group most exposed to antidepressants experienced the largest decline in suicide rates in both men and women. Is this a sign that Australian GPs are getting better at treating depression, ask the authors of this report?

The early 1990s also coincided with the introduction of selective serotonin reuptake inhibitors (SSRIs) in Australia. Studies based in secondary care have shown that SSRIs and their older counterparts, the tricyclic antidepressants, are equally effective. However, the relevance of such studies to primary care (where most patients with depression are treated) has been questioned. A new systematic review, including only studies based in primary care, confirms that the two classes of drugs have similar efficacy and shows that SSRIs are better tolerated. However, the data are limited.2

1. BMJ 2003; 326: 1008-1011

2. BMJ 2003; 326: 1014-1017

Don't get even, get rich

Payback can be a letdown, according to a recent behavioural "experiment". German economics researchers recruited 238 university students from their lunchroom and randomly allocated them to investor/trustee pairs. The investors were each allotted a set amount of money, then asked to decide how much of their money they wanted to transfer to the trustee and how much they wanted in return. Trustees then decided how much they would return. In a subgroup, investors could impose a fine if the return was less than that requested, or refrain from imposing the fine.

The returns were found to be lowest when the investor imposed a fine, higher if a fine was not possible, and highest when the investor refrained from fining. The moral of the study? Generosity pays but penalties don't.

Nature 2003; 422: 137-140

Death after stroke

The QT interval from any ECG lead (except aVR) can be used to predict death from any cause in stroke survivors, according to a recent Scottish study. The QT interval is the time taken for ventricular depolarisation and repolarisation, and varies inversely with heart rate. In this study, the 404 participants had ECGs a median of one year after a stroke. A prolonged heart-rate-corrected QT interval (QTc) in leads V6 and III carried the greatest risk of cardiac death (unadjusted RR = 3.1) over the following six years. Multivariate analysis showed that QTc in lead V6 was a better predictor of cardiac death than traditional risk factors such as ischaemic heart disease, cholesterol and glucose levels, and pulse pressure.

Heart 2003; 89: 377-381