Jul 11, 1981 - 400 g (14 oz) cabbage as well as the 50 g packet of peanuts he mentions in ... the opinion, like Roger Bacon, that "neither the voice of authority...
140
BRITISH MEDICAL JOURNAL
in temperature and blood pressure with paracetamol is interesting and must be kept in mind when treating patients with Hodgkin's disease who might require antipyretic drugs. My patient developed hypothermia and hypotension which seemed to be unrelated, as I mentioned, to any drug ingestion. The patient did not receive paracetamol, salicylate, or any similar drug prior to the drop of temperature and blood pressure, though 10 days earlier he had received a five-day course of ethambutol and isoniazid because a provisional diagnosis of sarcoidosis had been made. To my knowledge those two drugs have no thermoregulatory side effect, either alone or when given in combination.
OSAMA M KORIECH Department of Oncology, Ministry of Defence and Aviation Medical Services Department, Riyadh-al-Kharj Hospital Programme, Riyadh, Saudi Arabia
annually. In the 50 years that the low-residue diet was fashionable no such results were reported. The authors gave a set amount of fibre in the form of biscuits or ispaghula husk, so that their study would be "controlled." This will not give the best results. Patients and their eating habits vary. Hence they will require varying amounts of fibre supplements. As man has eaten bran for centuries, particularly in the winter months, it is logical to add fibre in the form of bran. There is no "standard" dose of bran any more than there is of insulin. Sufficient bran should be taken to open the bowels once or twice a day without straining. This amount must be found by trial and error. Only then will the best results be obtained. It would be a pity if this trial of a bran biscuit and a plant gum were to be interpreted as having any bearing on the proved effects of miller's bran. NEIL S PAINTER Manor House Hospital, London NW1 1 7HX
Occupational hazard of infection in the operating theatre
SIR,-Among the accidental injuries quoted by Mr J E deB Norman and Mr M Eagleton (6 June, p 1875) are lacerations of hand or finger received during attempts to remove contaminated scalpel blades from handles. To avoid just this type of hazard we designed a simple instrument which allows blades to be removed quickly and safely.' It is manufactured under an agreement with the Greater Glasgow Health Board by Eschmann Bros and Walsh Ltd and is retailed at under £5. J McKIE A SHAw West of Scotland Health Boards Department of Clinical Physics and Bioengineering, Glasgow G4 9LF 1 McKie J, Shaw A. Lancet 1975;ii:395.
Are fibre supplements really necessary in diverticular disease? SIR,-The paper "Are fibre supplements really necessary in diverticular disease ?" by Mr M H Ornstein and his colleagues (25 April, p 1353) concludes that extra fibre is unnecessary unless the patient's symptoms are those of constipation. Surely this has always been the argument in favour of giving bran? The disease is found in populations who are constipated compared with those eating highfibre diets. The symptoms of diverticular disease are caused by the colonic muscle struggling with stiff stools, not by the diverticula. This is obvious as bran abolishes symptoms without removing the diverticula. The same applies in the symptoms of constipation and the irritable bowel syndrome. What evidence is there for the statement that the low-residue diet relieves symptoms as well as does a high-fibre diet ? All the evidence points to the contrary. After resection or myotomy performed for symptoms, only 60% of patients remained symptom free if they continued to eat the same foods. However, if they changed to a high-fibre diet over 900o lost their symptoms. Before the introduction of bran 17 elective colonic resections were carried out annually at the Royal Berkshire Hospital, Reading. Since the introduction of the high-fibre diet only one is performed
***We sent this letter to the authors, who reply to it and to earlier correspondence below.-ED, BMJ. SIR,-We are pleased to find such an interest in our paper, but it is nLWoc1e that the correspondence has been entirely from doctors who "understand" dietary fibre. It is a pity therefore that despite their prolific studies in this field they have failed to teach the large majority of us the "correct" way to give dietary fibre, nor have they been able to persuade us that patients will continue conscientiously taking diets rich in fibre after the first few months. The usual practice remains to advise patients to "take two tablespoonfuls of bran daily." It is remarkable that our study has excited such strong criticisms while Brodribb's1 produced not a single letter. Is this merely because our conclusions are at variance with popular clinical thinking? Attention has been focused on the "dose" of fibre administered in these two trials; in fact, they were almost identical. The apparent discrepancy has arisen because Dr D A T Southgate's analysis of the placebo biscuits (1-6 g dietary fibre daily)-which we used-differs from the figure quoted by Brodribb (0 6 g daily). If he used Southgate's figures Brodribb would find that, like us, he had a difference of only 4 8 g fibre daily between his active and control treatments-equivalent to 800 g (1 lb 2 oz) potato or 400 g (14 oz) cabbage as well as the 50 g packet of peanuts he mentions in his letter (30 May, p 1792). Our paper has at least highlighted an important problem in relation to the correct dose and method of administration of dietary fibre. Mr Brodribb now suggests that 40 g of added dietary fibre is needed to treat this condition; why did he not test this dose in his Oxford trial'? Perhaps he, like us, was concerned lest his patients (only nine) defaulted at such a high dose. The need to give a fixed dose applies to controlled studies of any therapeutic agent and is always open to criticism -but it is only by such studies that the optimal dose can be ascertained. We began this trial without any preconceived ideas about the outcome and have attempted to interpret our findings in the light of current scientific evidence. Like your correspondents, we were surprised by the size of the placebo response. We are just completing a trial using different doses of bran tablets, which we hope will go some way towards answering our critics; so far we have only confirmed the difficulty of persuading patients to take the high dose long enough to complete the trial. Mr Painter has taken our point that the usefulness of fibre in treating diverticular disease is due to its benefit in relieving constipation. However,
VOLUME 283
11 JULY 1981
he goes on to say that we found that a "lowresidue diet relieves the symptoms as well as does a high-fibre diet." We did not say this. As we understand it, the "normal British diet" is very different from one low in residue-recently defined as containing less than 10 g dietary fibre daily.2 He cites a now almost extinct operation in support of his argument in favour of bran but forgets that there are many reasons why surgical fashions change; one likely possibility is that the previously fashionable low-fibre treatment of diverticular disease3 was positively harmful and led to the frequent need for surgery. A normal diet is what is required. We are, of course, aware of the potential pitfalls in the design of our trial. Because of the possible carry-over effect mentioned by Mr Brodribb and the problem of between-patient variation alluded to by Sir Francis Avery-Jones (30 May, p 1972), we chose treatment periods of four months and aimed at studying 60 patients. A disease with many inter-patient variables can be studied only with very large numbers (hundreds) of patients or by using a cross-over design so that each patient acts as his own control. We have carefully scrutinised our results using analysis of variance and methods described by Hills and Armitage4 and have confirmed the very large inter-patient variation; but we have been unable to find any differences between the six possible treatment orders. Mr John Northover's suggestion (30 May, p 1792) of a long-term trial to study the progression to its more serious complications of uncomplicated diverticular disease would make a good sequel to our study. Perhaps the committee of the Medical Research Council which agreed the protocol of our trial would consider it. Mr Northover complains that "over-interpretation by readers" of original papers will lead to misquotation and then proceeds to do just that himself by confusing prevention of diverticular disease with treatment. We did not mention prevention because we had no facts either for or against it. We are sorry that this trial, approved by a committee of which he was a member, did not match Sir Francis Avery-Jones's expectations. His strong condemnation of the trial was because the results do not confirm "clinical impressions over a number of years"; but 60 years ago Alvarez's clinical impression was very different: ". . . bran and other coarse foods upset digestion. I am sure it is preferable, unless the patient has the digestion of an ostrich, to give a mild chemical stimulus rather
than a rough mechanical one."5 We remain of the opinion, like Roger Bacon, that "neither the voice of authority nor the weight of reason and argument are as significant as experiment, for thence comes quiet to the mind."
M H ORNSTEIN Ashford Hospital, Ashford, Middx TW15 3AA
RODNEY LITTLEWOOD I MCLEAN BAIRD West Middlesex University Hospital, Isleworth, Middx TW7 6AF
A G Cox Northwick Park Hospital, Harrow, Middx HAI 3UJ IBrodribb AJM. Lancet 1977;i:664-6. 2 Bingham S. I Hum Nutr 1979;33:5-16. 3Alvarez WC. The mechanics of the digestive tract. New York: P Hoeber, 1922:110. 4Hills M, Armitage P. Br J7 Clin Pharmacol 1979;8: 7-20. 5Alvarez WC. JAMA 1919;72:8-13.
Late consequences of abortion
SIR,-I appreciated the reassessment of the long-term sequelae of induced abortion in your leading article (16 May, p 1564) and the
