Opiate agonists and enterotoxins.

the other two agents in improving stool consistency and relieving urgency (Palmer et al., 1980). Side effects. Diphenoxylate causes more side effects than either codeine phosphate or loperamide. Both codeine and diphenoxylate cross the blood-brain barrier and may cause central-nervous-system side effects such as nausea, dizziness, drowsiness and depression. Loperamide does not usually cross the blood-brain barrier and central-nervoussystem side effects are therefore uncommon. The commonest cause for stopping loperamide in one study was abdominal pain and constipation! (Palmer et al., 1980). Safety. The margin of safety when using a drug is usually assessed from the LDso: EDso ratio, which is the relationship between the dose required to kill half the animals to the dose required to cure half of the animals. Greater ratios are associated with safer drugs. In tests of castor oil-induced diarrhoea in rats, the ratio is only 20: 1 with codeine, 85 : 1 with diphenoxylate, and over 1000: 1 with loperamide, i.e. the dose of loperamide required to kill half the animals is over loo0 times greater than the dose required to cure half of the animals of their diarrhoea. In man, lethal overdoses of codeine and diphenoxylate have both been reported on a number of occasions, and with codeine phosphate only two or three 30mg tablets ingested by a small child would be enough to exceed the usually lethal dose of 5 mg/kg. No fatal overdoses of loperamide have been reported but several overdosed childen have exhibited central-nervous-system depression requiring naloxone treatment. It may be that loperamide can cross an immature blood-brain barrier and this area requires further study. It would seem that loperamide, although the most expensive of the three opiates, is the safest choice. Although the opiates have proved successful in treating acute and chronic diarrhoea in the developed world and appear safe in therapeutic doses, several anxieties exist about their use in third world countries for severe acute diarrhoeas. Several potential problems should be noted. Toxicity may be greater than hitherto appreciated as the doses required for severe secretory diarrhoeas may be higher than those usually recommended. The antimotility effect of these drugs may predispose to paralytic ileus, especially in hypokalaemic individuals, and pooling of fluid in the gut could lead to death from dehydration despite the apparent cessation of diarrhoea. Opiates have been incriminated in precipitating toxic megacolon in acute ulcerative colitis, although this complication can occur in patients with ulcerative colitis on no treatment, and therefore cannot definitely be attributed to the opiates. In one volunteer study, diphenoxylate enhanced symptoms and prolonged faecal excretion of Shigella organisms (Dupont & Hornick, 1973). This single, small study requires confirmation but it raises the possibility that opiate therapy could exacerbate infection with invasive organisms. Loperamide is undergoing field trials at the moment in a variety of doses and it remains to be seen whether the opiates prove as useful in controlling diarrhoea in third world countries as they have been in the developed world.