universally accepted we need to know how many times a syringe can be used, how patients can keep count of use, how syringes are to be stored between injections, and whether a combined needle and syringe is better than separate components.
Regional Research and Development Laboratory, Central Pharmacy,
M. C. ALLWOOD
Cambridge CB2 2QQ INHALER REFILLS
StR,-The launch of’Ventolin’ (salbutamol, albuterol) inhalers in the USA raises some interesting questions about the marketing of metered dose aerosols in the UK. Glaxo have given prominence in their US advertisements to the refill canisters of ventolin (and ’Beclovent’, beclomethasone). These are described as "more economical to help patients save money". However, no such refill canisters are marketed by Glaxo (Allen & Hanburys) in the UK. At one time it would appear that the Department of Health and Social Security discouraged manufacturers from making such refills available; but now they positively encourage the practice. There is only one 02-stimulant standard metered dose inhaler for which a refill is available (’Alupent’; Boehringer). This refill costs 17% less than the complete inhaler. If the total aerosol market in the UK is estimated at f20 million, the widespread prescribing of refills might save at least 1.5million every year. If more refills for metered dose aerosols were marketed in the UK and if doctors prescribed them more often, the National Health Service could enjoy the financial savings currently offered to the individual in the USA. London Hospital, London E1 1BB
I. D. COX
CLUSTER OF CASES OF ARTHROGRYPOSIS
SIR,-A 22-year-old multigravida, after a pregnancy complicated only by polyhydramnios, went into spontaneous labour at 35 weeks’ gestation and on Sept 9, 1982, was delivered of an infant weighing 217 kg. Effective ventilation could not be established and the infant died
6 hours. Clinical examination revealed
postural deformities, with flexion contractures of the fingers, wrists, elbows, hips, and knees, bilateral talipes equinovarus, and limitation of jaw movement. Additional dysmorphic features included a small mouth, abnormal dermatoglyphics, and hypoplastic genitalia. This syndrome of arthrogryposis and failure to establish respiration in premature infants is not unfamiliar on our special care baby unit; we saw four similar cases in the 10 months March to October, 1982. Karyotyping and complete necropsy with histology of brain, spinal cord, and muscle have been unhelpful. Inquiry into antenatal history has revealed no obvious common factor; in particular there was no suggestion of oligohydramnios or chronic amniotic fluid leak. The term "arthrogryposis" encompasses a range of clinical entities from severe arthrogryposis multiplex congenita of neurogenic or myogenic origin to milder forms of joint contracture without evident cause. The Pena-Shokeir i syndrome (described in Pediatrics 1974; 85: 373) consists of neurogenic arthrogryposis, pulmonary hypoplasia, and hypertelorism, and is thought to have an autosomal recessive inheritance with an estimated frequency of 1 in 12 000 total births. Our 5 cases fit this syndrome. However, all forms of arthrogryposis appear to be uncommon, with incidences of between 1 in 5000 and 1 in 50 000; the 5 cases we have seen represent an incidence of about 1 in 1000 over one year. We appreciate that this cluster may reflect only a fortuitous collection of cases or our increased awareness of arthrogryposis in premature infants, but we would be interested in the experience of other neonatal units. The incidence of arthrogryposis (live and stillborn) in the register of congenital abnormalities that is kept by the Office of Population Censuses and Surveys was 1 in 40 000 in 1980 and 1 in 55 000 in 1981. S. WYATT Departmentof Paediatrics R. C. BEACH and Child Health, C. STUART St Mary’s Hospital, Portsmouth P03 6AD R. J. HALLETT
INTERPRETATION OF FAECAL INDICATOR COUNTS FROM TROPICAL WATERS
has suggested that the application to drinking developing countries of World Health Organisation international standards might be inappropriate. Most untreated water supplies in these countries would fail to meet the standard of less than 1 faecal coliform per decilitre (< 1 FC/dl). However, since water with any degree of faecal contamination may contain faecally derived pathogens, W.H.O. could hardly suggest the acceptability of such water. The third edition of International Standards for Drinking Water Quality2 pays insufficient attention to the realities of water quality in the developing world, and I hope that this will be remedied in future revisions-indeed, this is already indicated by the proposal to use the word "guidelines" instead of "standards". If current W.H.O. standards for small, untreated water supplies are too stringent3 and alternative guidelines are used,1,4 the results of water analysis, particularly in tropical countries, will need to be interpreted carefully.
A small settlement of, say, fewer than a hundred inhabitants which had been using a discrete water supply, such as a well, spring, or rainwater catchment, containing an average of 200 FC/dl, could be advised to change to a non-discrete supply (river, stream, swamp, or lake) if this were shown to have a lower average FC count such as 50/dl. However, the traditional supply is exposed to contamination only by inhabitants of the settlement using it, whereas the proposed new supply may be contaminated by other settlements. In this situation the water with the lower FC content could be a more serious health hazard than the water with the higher FC count. An "improved" supply may be acceptable to a public health official but those who are to use it will take into account other factors, such as-how far away is the new supply and how much work is required to abstract the water? The faecal indicator counts used in decision making must be average figures of sufficient samples, taken over an extended period (to include seasonal variations, which are usually extreme in the tropics) and involving some form of indicator confirmation. This rarely happens in developing countries, which lack sufficient laboratories and staff to do the work. 1. Feachem R. Bacterial standards for drinking water quality in developing countries. Lancet 1980; ii: 255-56. 2. World Health Organisation. International standards for drinking water quality, 3rd ed. Geneva: WHO, 1971. 3. Feachem R. Water supplies for low income communities: Resource allocation, planning and design for a crisis situation. In: Feachem R, McGarry M, Mara D, eds. Water, wastes and health in hot climates. Chichester. Wiley, 1977: 75-95. 4. Lloyd B. Water quality surveillance. Waterlines 1982; 1: 19-22.
Relative survival of E. coli and FC other than E. coli in well Sierra Leone.
Six isolates of each category of organism were obtained from traditional supplies, then individually reintroduced into fresh samples of well water to give a concentration of about 10’ cells/ml and counted, at the times indicated, by a selective membrane-filtration procedure.9 water