714BRITISH MEDICAL JOURNAL 21 SEPTEMBER 1974 g/day when their patients were on Nutrizym (mean fat excretion 42-5 g) compared with when they were on Pancrex V forte (mean fat excretion 57-3 g). Our patients' fat excretion was closely similar with both products (25 3 g on Nutrizym and 24-3 g on Pancrex V forte). Before the start of our trial patients had been taking Pancrex V forte in a dosage which had been established by the parents as being the most effective one for reducing stool frequency and improving stool character. All patients continued with this regimen for the first two weeks of the trial so that dosage and relevant details could be recorded, and the number of tablets was halved (though the enzyme dose for trypsin and lipase remained the same) during the two weeks when Nutrizym was given. Knill-Jones et al. (1970) chose to use suboptimal doses of pancreatic replacement therapy to facilitate the detection of improved digestion with either of the two preparations. Our doses resulted in smaller faecal fat excretions, but these were still sufficiently abnormal, we believe, for any differences in the effectiveness of the tablets to be apparent.In addition we did not find a significant difference in stool volume (mean volume 294 ml/day on Nutrizym and 259 ml/day on Pancrex V forte). Knill-Jones et al. (1970) Though the laboratory investigations did not provide any evidence in favour of either product the final questionnaire completed by the parents showed some patient preference for Nutrizym. There was some discrepancy (unexplained) between the information on stool frequency and character and the occurrence of abdominal pain obtained during the course of the trial, when there was no patient preference, and the information obtained at the end, when some of the parents modified their views in favour of Nutrizym. Clearly it is more convenient, especially for those patients who take a high dose of pancreatic preparation, to be able to reduce the number of tablets.With the exception of the two children whose coefficients of fat absorption were nil (table III) values obtained for the children while on pancreatic replacement therapy were comparable with those in previous reports (Harris et al., 1955;Ross, 1955; Matthew and Spector, 1961), and the mean value (70%) was some 28% less than normal (96% i 40) (Matthews and Spector, 1961).The apparent absence of fat absorption in cases 2 and 5 is difficult to explain. Both children were taking high doses of pancreatic therapy (mean numbers of Pancrex V forte and Nutrizym tablets per week were 133 and 63 respectively for case 2, and 378 and 214 respectively for case 5). One patient (case 2), aged 6-6 years, was on the 75th centile for weight, and the other (case 5), aged 8-7 years, was on the 10th centile.Pancreatic enzyme preparations, even in high dosage, do not eliminate the steatorrhoea of cystic fibrosis and it seems expedient to investigate other mechanisms which may influence fat absorption in this disease.