Intended for healthcare professionals



BMJ 1999; 318 doi: (Published 22 May 1999) Cite this as: BMJ 1999;318:1430

Anaesthetists blame all kinds of anatomical barriers for patchy epidural analgesia, including air bubbles, fat, and fibrous bands in the epidural space. None was evident in an imaging study of 20 women receiving brachytherapy for cervical cancer (Anesthesiology1999;90:964-9). Computed tomography showed local anaesthetic solution spreading freely but not uniformly through the space, with a different distribution in almost every patient. The position of the catheter was also extremely variable: in eight women the catheter tip ended up in the posterior epidural space (where it should be), but in the rest the tip wandered into the intervertebral foramina or paravertebral tissues. Badly positioned epidural catheters contribute more to asymmetrical blocks than do anatomical barriers, the authors conclude.

Humans use two types of thinking to solve mathematical problems (Science1999;284:970-4): a visual and spatial sense of quantity is located in the left and right parietal lobes, while the processing of symbols is done in the left frontal lobe. Volunteers in a brain imaging study used their parietal lobes to solve approximate problems like: “Is 53 plus 68 closer to 120 or 159?” They used their left frontal lobe to work out more precise answers. Interestingly, exact arithmetic seems more language dependent than does processing of approximate numbers.

Humans—and some orthopaedic surgeons—are also characterised by their ability to do more than one thing at once, a skill mediated by the anterior prefrontal cortex (Nature1999;399:148-51). By means of functional magnetic resonance imaging, scientists from the US National Institutes of Health mapped multitasking to this region in healthy volunteers. They were looking at a form of multitasking known as “branching,” which allows humans to interrupt a core task and then return to it and take it up where they left off. Any doctor who has ever carried a bleep will be familiar with the technique.

Breastfed infants turn into brighter children than bottle fed infants, even when maternal social class and educational level are accounted for. Could there be something in breast milk that promotes development? Probably not, says a team from the United States which reports that the association between breast feeding and intelligence disappears when maternal intelligence quotient and parenting skills are factored into the analysis ( There are, of course, plenty of other reasons to breast feed. The observed effect on intelligence was always very small and is dwarfed by bigger immunological benefits.

Reye's syndrome has almost disappeared in the United States. Only two cases a year have been reported since 1994 (New England Journal of Medicine1999;340:1377-82). The decline began in the 1980s after reports linked the syndrome to aspirin use in children, and it shows what can be done with timely public education messages and clear warnings on drug packaging. Aspirin is still used against juvenile rheumatoid arthritis and Kawasaki disease, however, so Reye's syndrome is not gone for good. The danger is that doctors will fail to consider it when faced with an encephalopathic child.

Paediatricians should scan the big four general medical journals as well as the specialist press if they want to stay evidence based (Paediatrics 1999;103:941-7). In a study of journal citations by the American Academy of Paediatrics, the Cochrane database of systematic reviews, and the Canadian Paediatric Society seven journals came out on top as sources of best evidence: BMJ, JAMA, Lancet, New England Journal of Medicine, Archives of Diseases in Childhood, Paediatrics, and Journal of Paediatrics. Only one, however, offers full, unfettered access to all articles from its website. (Guess which one.)

Humble devices like walking sticks, zimmer frames, bath benches, and simple home adaptations preserve elderly people's independence and improve their quality of life. They can also cut healthcare costs in half, says a randomised trial in Archives of Family Medicine(1999;8:210-7). Participants who had unlimited help according to need—on average 14 devices each—spent $14 000 on health care over the next 18 months. Controls given “usual care,” which amounted to about 2 devices each, spent over $30 000 during the same period.

Transcranial magnetic stimulation may sound gothic, but recent clinical trials suggest it is a promising alternative to electroconvulsive therapy. Patients with major depression responded well to the treatment in one double blind randomised trial against sham treatment (Archives of General Psychiatry1999;56:3165-20). At the end of 10 daily sessions, half of the 35 patients in the treatment group had improved by over 50% in depression rating scales compared with only a quarter of controls. The next step must be to test magnetic stimulation against electroconvulsive therapy or drugs.

It should be clear by now that inhaled corticosteroids cause systemic side effects. A meta-analysis shows that adverse effects are dose related for all inhaled corticosteroids, but that fluticasone is worse than most (Archives of Internal Medicine 1999;159:941-55). Marked adrenal suppression occurs at doses over 1.5 mg a day for most drugs, but at 0.75 mg a day for fluticasone propionate. Keep doses to a minimum, is the authors' message.


I acquired these lesions one night when I was sleeping under the stars on a sandbank in the middle of the Zambesi river, protected only by a mosquito net. The marks did not appear anywhere else on my body. The affected arm had been resting against the net, which shows that the holes in the mosquito net, although small enough to prevent mosquitoes getting through, were large enough to admit their mouth parts. R H Soper, general practitioner, Victoria Surgery, Bury St Edmunds, Suffolk IP33 3BB

Submissions for this page should include signed consent to publication from the patient.

Hospital mortality from myocardial infarction has dropped as low as 6% in some clinical trials, but things are very different out in the real world. A retrospective look at over 1000 patients admitted to two Irish teaching hospitals showed an in-hospital death rate of 18%and a five year mortality of 50% (Heart1999;81:478-82). Only a quarter of the patients received thrombolytic treatment, and only a third underwent angiography before discharge. Clinical trialists cherry pick patients to suit their needs, say the authors, often excluding elderly patients and those who never make it to a coronary care unit. Their results are simply an indication of what can be done with the best possible patients in the best possible circumstances.

View Abstract

Log in

Log in through your institution


* For online subscription