129 and then increased rather abruptly to an average of 8.2% during the second week. From week 1 onwards, the intensity of pain also remained reasonably constant (not shown). At 26 weeks 15 patients had light, 2 had moderate, and 1 had severe pain. During the first week prednisolone had a pronounced effect on patients having moderate or severe pain, but not on patients with light pain. An effect was thus seen in patients with constant or constant plus neuralgic pain since these two types of pain tended to be moderate or severe. Discussion Patients aged over 60 years were chosen for this trial since this age-group is particularly susceptible to the development of post-herpetic neuralgia.1 3 Admission to the trial was restricted to patients whose pain had been present for less than 4 days; it would hardly have been possible to initiate corticosteroid treatment earlier than that. On admission most patients had either constant, burning pain (39%) or such pain combined with shooting neuralgic pain (34%). The rest had pure neuralgic pain. It is interesting that the type of pain and its intensity remained fairly constant after the acute phase. At the 6 month follow-up 18 (23%) patients had pain. Approximately one-third had constant and one-third had constant plus neuralgic pain, while the remaining third consisted of equal proportions of patients with pure neuralgic or triggered pain. 3 of the 18 had moderate or severe pain. Thus, only 4% of the patients entering the study had pronounced neuralgia at 6 months. The proportion of patients developing post-herpetic neuralgia was higher than expected (see introduction) at 3 months (46%) and 6 months (23%). Although the proportion of patients with trigeminal zoster who had pain at 6 months (29%) was nearly as expected, the proportion (22%) of patients with non-trigeminal zoster was four times as high as expected. We have no explanation for this finding. We have previously shown that acyclovir in a closely similar group of patients reduces the median period of acute pain from 13 to 5 days. This trial shows that prednisolone treatment gives additional pain relief during the first 3 days. However, in view of the established side-effects of corti- costeroids, we would not recommend using its analgesic potential in acute herpes zoster. In many patients the pain returned 1-2 weeks after admission and this seems to be the period required for the development of post-herpetic neuralgic pain. Triggered pain also took this time to develop. Once established pain may last for any number of weeks, but on average it recedes slowly over the whole observation period. Prednisolone had no effect on the development or the rate of disappearance of this post-herpetic pain. Acyclovir (’Zovirax’) was kindly supplied by Wellcome Research Labora- tories, England. Coded tablets containing either placebo or prenisolone were provided by the dispensary, Aarhus Kommunehospital. Mr. C. A. Burke, Wellcome, gave helpful advice on the design of the study, and J. Eriksen, Pain Clinic, Bispebjerg Hospital, on analgesic administration. Correspondence should be addressed to V. E. REFERENCES 1 Wildenhoff KE, Ipsen J, Esmann V, Ingemann-Jensen J, Hjelm Poulsen J Treatment of herpes zoster with idoxundine ointment, including a multivariate analysis of symptoms and signs. Scand J Infect Dis 1979, 11: 1-9. 2 Wildenhoff KE, Esmann V, Ipsen J, Harving H, Peterslund NA, Schønheyder H. Treatment of trigeminal and thoracic zoster with idoxuridine. Scand J Infect Dis 1981; 13: 57-62. 3. Peterslund NA, Ipsen J, Schønheyder H, Seyer-Hansen K, Esmann V, Juhl H. Acyclovir m herpes zoster. Lancet 1981; ii. 8127-30. 4. Appelman DH. Treatment of herpes zoster with ACTH. N Engl J Med 1955, 253: 693-95 5. Gelfand ML. Treatment of herpes zoster with cortisone. JAMA 1954; 154: 911-12. 6. Duke-Elder S. The clinical value of cortisone and ACTH in ocular disease. Br J Ophthalmol 1951; 35: 637-71 (p 655). 7. Olson JA, Steffensen EH, Smith RV, Margulis RR, Whitney EL. Use of adrenocorticotropic hormone and cortisone in ocular disease Arch Ophthalmol 1951; 45: 274-300 (p 288) 8 Elliott FA. Treatment of herpes zoster with high doses of prednisone. Lancet 1964; ii: 610-11 9. Carter AB, Royds JE. Treatment of ophthalmic zoster with prednisone. Br Med J 1957, ii: 746-48. 10 Scheie HG, McLellan TG. Treatment of herpes zoster ophthalmicus with corticotropin and corticosteroids. Arch Ophthalmol 1959; 62: 579-87. 11. Elliott FA Postherpetic neuralgia. Br Med J 1973, i: 679. 12. Epstein E. Treatment of zoster and postzoster neuralgia by the intralesional injection of triamcinolone: A computer analysis of 199 cases Int J Dermatol 1976; 15: 762-69. 13 Eaglstein WH, Katz R, Brown JA. The effects of early corticosteroid therapy on the skin eruption and pain of herpes zoster. JAMA 1970; 211: 16581-83. 14. Keczkes K, Basheer AM. Do corticosteroids prevent post-herpetic neuralgia? Br J Dermatol 1980; 102: 551-55 15 Clemmensen OJ, Andersen KE. ACTH versus prednisone and placebo in herpes zoster treatment. Clin Exp Dermatol 1984; 9: 557-63. 16. Bean B, Braun C, Balfour HH Jr Acyclovir therapy for acute herpes zoster Lancet 1982; ii: 1118-21. 17. Peterslund NA, Esmann V, Ipsen J, Dencker Christensen K, Munck Petersen C. Oral and intravenous acyclovir are equally effective in herpes zoster. J Antimicrob Chemother 1984; 14: 185-89. 18. McKendrick MW, McGill JI, White JE, Wood MJ Oral acyclovir in acute herpes zoster. Br Med J 1986; 293: 1529-32. 19. Esmann V, Ipsen J, Peterslund NA, Seyer-Hansen K, Schønheyder H, Juhl H. Therapy of acute herpes zoster with acyclovir in the nonimmunocompromised host. Acyclovir Symposium Am J Med 1982; 73, 1A: 320-25. FETAL BREATHING MOVEMENTS AS PREDICTOR OF FAVOURABLE PREGNANCY OUTCOME AFTER OLIGOHYDRAMNIOS DUE TO MEMBRANE RUPTURE IN SECOND TRIMESTER M. BLOTT K. H. NICOLAIDES D. GIBB A. GREENOUGH G. MOSCOSO S. CAMPBELL Departments of Child Health, Pathology, and Obstetrics, King’s College Hospital, London Summary In 11 pregnancies complicated by oligohydramnios due to spontaneous rupture of the membranes in the second trimester of pregnancy fetal breathing movements were assessed regularly by ultrasonographic examinations. In the 6 cases in which fetal breathing movements were detected the babies were liveborn and there was no evidence of pulmonary hypoplasia or the other non-renal features of Potter’s syndrome. In the other 5 cases there were no fetal breathing movements. 1 pregnancy was terminated electively, and 1 ended in an intrauterine death; the remaining 3 infants died in the neonatal period. All 5 cases showed necropsy evidence of pulmonary hypoplasia. These findings indicate that premature and prolonged rupture of membranes in the second trimester of pregnancy does not uniformly result in a poor prognosis. They suggest that fetal breathing movements could be used as a predictor of favourable neonatal outcome. Introduction OLIGOHYDRAMNIOS in the second trimester due to premature and prolonged rupture of membranes has been associated with a poor prognosis. Infants may be stillborn, while in those born alive extreme respiratory difficulty, possibly fatal within a few hours despite assisted ventilation, may result from pulmonary hypoplasia with pneumothorax