oocyte development have been found to correlate with a poor outcome.2"122 Once again an explanation is lacking, although the association may be secondary to the polycystic ovary syndrome, which is present in 23% of all women and in some 80% of patients with recurrent miscarriage (diagnosed with ultrasonography).2' Alternatively, it may be explained by the effect of luteinising hormone on the production of ovarian steroids or on the maturation of oocytes.24 Whichever it is, knowing that a raised concentration of serum luteinising hormone increases the risk of miscarriage provides us with a predictive test for women before pregnancy. An abnormal result may identify a subgroup of patients with a potentially remediable cause for their pregnancy losses: hypersecretion of luteinising hormone, which may be suppressed by gonadotrophic releasing agonists ,2526 ovarian diathermy,27 and somatostatin. Clinicians must remember that women who have miscarried recurrently from one cause are not protected from a further miscarriage from another cause. Most importantly, when assessing the efficacy of treatment for patients with recurrent miscarriage we must remember that even without treatment about 60% of these women will be successful in their next pregnancy. LESLEY REGAN Senior Lecturer in Obstetrics and Gynaecology, St Mary's Hospital, London W2 1PG I World Health Organisation. Recommended definitions, terminology and format for statistical tables related to the perinatal period. Acia Obsiei GynecoiScand 1977;56:247-53. 2 Warbtirton D, Fraser FC. Spontaneous abortion risks in man: data from reproductive histories collected in a medical genetics unit. Human Genet 1964;16:1-25. 3 HLisjes HJ. Spontaneous abortion. Edinburgh: Churchill Livingstone, 1984. 4 Alberman E. The epidemiology of repwated abortion. In: Beard RW, Sharp F, eds. Early pregnancy loss: mechanisms and treatment. London: Royal College of Obstetricians and Gynaecologists, 1988:9-17.
S Regan L, Braude IPR, 'I'rembath P1L. Influence of past reproductive performance on risk of spontaneous abortion. BMj 1989;299:541-5. 6 Warburton 1), Strobino B. Rccurrcnt sp)ntaneous abortion. In: Bennett MJ, Edmonds 1DK, cds. Sponianeous and recurreni abortion. Oxford: Blackwcll Scicntific, 1987:193-213. 7 Bennctt MJ. Congenital abnormalities of the fundus. In: Bennett MJ, Edmonds D)K, eds. Spontaneous and recurrent abortion. Oxford: Blackwell Scicntific, 1987:109-29. 8 Medical Research Council and Royal College of Obstetricians and Gynaccologists Working Parts on Cervical Cerclage. Interim report. Br] Obsiet fiynaecol 1988;95:437-45. 9 Watts D)H, Eschenbach DA. Reproductive tract infections as a cause of abortion and prcterm birth. Seminars in Reproductive Endocrinology 1988;6:203-15. 10 Lubbe WF, Liggins (iC. Role of' lupus anticoagulant and pregnancy. Seminars in Reproductive Endocnnology 1988;6:181-90. 11 D)erue GJ, Englert HJ, Harris EN, ei al. Fetal loss in systemic lupus erythematosus: ass(oiation with anti-cardiolipin antibodies. Am] O)bsie Gynecol 1985;5:207-9. 12 Fault WI', Temple A. Distribution of beta-2 microglobulin and Hl A in chorionic villi of human placentac. Nature 1976;262:799-802. 13 McIntyre JA, Faulk WP. Trophoblast modulation of maternal allogeneic recognition. I-roc Nail AcadSci USA 1979;76:4029-32. 14 McIntyre JA. In search of trophoblast-lymphocyte cross-reactive ([LX) antigens. American Jfournal of Reproducitve Immunology and M icrobiology 1988;17: 101)-10. 15 Johnson I'M, Chia KV, Hart CA, Griffith HB, Francis WJA. I'rophoblast membrane infusion for unexplained recurrent miscarriage. Br] Obstet Gynaecol 1988;95:342-7. 16 Mowbray JF, Gibbings C, Liddell H, Reginald PW, Underwood jL, Beard RW. Controlled trial of treatment of recurrent miscarriage with paternal cells. Lancet 1984;i:941-3. 17 King A, Wellings V, Gardner I, Loke YW. Immunocytochemical characterisation of the unusual large granular lymphocytes in human endometrium throughout the menstrual cycle. Hum Immunol 1989;24:195-205. 18 Kirng A, Birkby C, Loke YW. Early human decidual cells exhibit NK activity against the K562 cell line but not against first trimester trophoblast. Cell Immunol 1989;121:166-73. 19 Reijnders F, Thomas CMG, Doesburg WH, Rolland J, Eskes TK. Endocrine effects of 17 alpha hydroxyprogesterone caproate during early pregnancy: a double blind clinical trial. Br] Obstei
Gvnaecol 1988;95:262-8. 20 Stanger JD, Yovich IL. Reduced in-vitro fertilisation of human oocytes from patients with raised basal LH levels during the follicular phase. Br] Obstei G(vnaecol 1985;92:385-93. 21 Punnonen R, Heinonen PK, Ashorn R, Kujansuu E, Vil ja I', I uohimaa P. Spontaneous luteinising hormone surge and cleavage of in vitro fertilised embryos. Fertil Steril 1988;49:479-82. 22 Regan L, Owen EJ, Jacobs HS. Hypersecretion of luteinising hormone, infertility and miscarriage. Lancei 1990;336:1141-4. 23 Sagle M, Bishop K, Ridlev N, et al. Recurrent early miscarriage and polycystic ovaries. BM] 1988;297: 1027-8. 24 Jacobs HS, Homburg RR. The endocrinology of conception. Ballteres Clin Endocrinol Metab 1990;4: 195-205. 25 Johnson P, Pearce JM. Recurrent spontaneous abortion and polycystic ovary disease: comparison of two regimens to induce ovulation. BM] 1990;300:154-6. 26 Homburg R, Eshel A, Kilborn J, Adams J, Jacobs HS. Combined luteinising hormone releasing hormone analogue and exogenous gonadotrophins for the treatment of infertility associated with polycystic ovaries. Hum Reprod 1990;5:32-5. 27 Armar NA, MacGarrigle HHG, Honour JW, Holownia I', Jacobs HS, Lachelin GCL. Laparoscopic ovarian diathermy in the management of anovulatory infertility in women with polycystic ovaries: endocrine changes and clinical outcome. Fertil Steril 1990;53:45-9. 28 Prelevic GM, Wurzburger MI, Balint-Peric L, Nesic JS. Inhibitory effect of sandostatin on sccretion of luteinising hormone and ovarian steroids in polycystic ovarv syndrome. Lancet 1990;336:900-3.
Oral lichen planus Not rare -and not easily treated The false idea that lichen planus is rare goes back to 1869, when Erasmus Wilson presented the condition to the medical world'-just a few years before he presented Cleopatra's needle to London. The notion has stuck for two main reasons. Firstly, many people with lichen planus, particularly when it is confined to the mouth, have no idea that there is anything wrong with them. Secondly, some doctors find it hard to recognise lichen planus-in one study of primary care physicians fewer than one in five could do so.' The truth emerges only when large populations are examined by experts. Even so, some dermatologists were astonished to read that lichen planus had been found in 14% of 20 749 Americans aged under 75, thought to be representative of the general population.' Admittedly only one in 100 of these people had lichen planus badly enough to need medical attention, but the numbers seemed unexpectedly large. The prevalence of oral lichen planus is now known for several countries and runs at about 1% or 2%.4 In the cases found in such studies there are seldom lesions on the skin as well as in the mouth. These patients may show the classical lacework of fine white lines and papules inside their cheeks, but their symptoms are few and they need no treatment. In contrast are the patients who attend dental clinics with pain and burning arising from atrophic areas, erosions, and 544
ulcers. For them the future seems bleak, and their symptoms are likely to persist: in one series of 611 patients only 17% of cases cleared completely over a mean observation period of seven and a half years.' These patients have an increased risk of developing oral squamous cell carcinoma6 and perhaps psychiatric disturbances.7 What can be done to help them? Firstly, bear in mind the possibility of a lichenoid drug reaction: stopping non-steroidal anti-inflammatory agents may be particularly helpful.' Next, consider lichenoid reactions to dental materials, especially if the lesions lie near to old amalgam fillings. If patch testing shows mercury allergy changing to another type of filling may be worth while.9'" Rough fillings should be smoothed and oral hygiene improved,'2 but giving up smoking makes little difference." Beyond this the treatments most in vogue now are corticosteroids, retinoids, and cyclosporin, used topically rather than systemically if possible. Triamcinolone can be injected into lesions, applied in Orabase (which sticks for a short while to the mucosa), or bonded to the molars in slow release preparations.'4 Mouths affected by lichen planus carry no greater risk of excessive absorption of topical corticosteroids than normal ones.'9 Isotretinoin is effective used as a gel, but the condition tends to relapse quickly when treatment is stopped.'6 Most recently, in a double blind trial a topical BMJ VOLUME 302
9 MARCH 1991
cyclosporin rinse (swish and spit) proved helpful without causing systemic side effects and offered the prospect of longer remissions. 7 Finally, in recalcitrant cases treatment with oral psoralens and ultraviolet A (PUVA) may be considered, though it may confer that most absurd of all status symbols, a genuine tan inside the mouth." Erasmus Wilson would have been glad to hear of these developments, but sorry that the cause of his disease is still not known. J A SAVIN
Consultant Dermatologist, Royal Infirmary, Edinburgh EH3 9YW Samman P1). Lichen planus: a (lctcrmatological cntrenary. BIri7 )ermatol 1969;81:306-7. 2 Ramsay 1)1,, ox A1. 'Ihe ahility ol primary carc physicians to rcognisc the common dcrmatoses. A rc-h )ermatuol 198 ; 117:620-2. 3 Johnson M-LI', RoertsJ. Skin condtlisons andrelatednted/l(r medIcal carea mong persons 1-74 years. Hyatsvillc: US D)epartment ol Health Education, 1978. ()HEW lPublication Ntt (PHS)
79-1660).) 4 Salonen L, Axell 'F, Helldcn L. Occurrence ol oral mucosal lcsions, the influence ol tobacco habits and an estimatc of treatmcnt timc in an adult Swcdish p)pulation. 7 Oral lathl Med
1990;19:17)-6. 5 'I'horn JJ, Holmstrup P, Rindtim J, IPindborg JJ. Course of various clinical lorms ot lichcn planus: a prospectivc lollow-up. 7 (Oral P'aiho 1988; 17:213-8.
6 Holmstrup 1' 'I'horn JJ. Malignant development of lichenr planus--alTCccid oral mucosa. 7 Oral l'athol 19X8;17:219-25. 7 Hamp' l3(, Malmstron MJ, Aalhcrg VA, Hannula JA,. Vikkula J. Psychiatric distuLrbance in paticnts with oral lichcn planus. Oral Surg Oral Med Oral l'Pathl 1987;63:429-32. 8 lPotts (J, Hamhcrgcr J, Scully C. 'I'hc medication of' paticnis with oral lichcn planits attd the association ot non-stcroiidal anti-inflammatory drugs with crosive lcsions. Oral Surg Oral Med Oral Plathol 1987;64:541-3. 9 Lund P0, Huirlcn 13, Lyberg ', Aas F. Amalgam rclatcd oral lichcnoidi reaction. Scand7 I)ent Res 1986;94:448-5 1. Ill James J, lerguison MM. Oral lichcnoiid reactiions rclated to mercury sensitivity. lir 7 Oral MaxilloJac Surg 1987;25:474-86. II (iarioch J, 'I'oddl'J ewis Mli;orsyth A Radcmakcr M. 'I'hc significancc ol'a )siltivc patLh test to mercury in oral discasc. Brr7 )ermnaiol 199();123 (suppl 37):25-6. 12 Holmstrup 1P Schiotz AW, Wcstcrgaard J. Ef7ccts of dcntal plaquc control in gingival lichen planus. Oral Surg Oral Med Oral P'aihol 1990;63:585 -90. 13 Silvcrman S, (iorsky M, LIozada-Nur 1i. A prospectivc I`61ow-up oi 5711 paticnts with oral lichcn planus. lcrsistcnc, remission and malignant assucatiumn. Oral Surg Oral Med Oral l'Paihol 1985;60: 3-4. 14 l)easy PlB, Collins AEM, lurke FM, Shanlcy )B. In vitrit and in visvo esaluatinoiif'a hitndablc tompact o`6r the prolontind dclivery oi triamcinolonc acettonidcloe the oral cas'ity in patictts with lichcn planus. Ptharn Acia flelv 1989;64:276-9. 15 'lcmitns JM, Rccs ''1), Zacharin NY. Ahsorptiion if a topical stcroidi and evaluation ol adrcnal suppression in patients with crosivc lichcn planus. Oral Surg Oral Med Oral I'aihol 199();69: 688-93. 16 (Giustina IA, Stcwart JCB, Ellis CN, ci al. 'I'opical applicatioit ol isotrctinoin gcl improtves oral lichcn planus. Arch Dernautol 1986;122:534-6. 17 Eiscn 1), E'llis CN, I)ucll EA, (irifiths CEM, Voorliccs JJ. Ef'ect if' topical cycliosporin rinsc on oral lichen planus. N l nglJ Med 1990;323:29)-4. 18 jatiscii C[,1'chtincn R, Haprp)ncn R', ILchtincn A, Soderlund K. Mouth l'UVA: a ncw treatmcnt fur rccalcitrant oral lichcn planus. /thitodermaiol 1987;4: 165-6.
Infantile colitis Allergy to food is the main non-infective cause and is easily treatable Infancy is one of the two periods of childhood in which colitis may present, the second being in late childhood and early adolescence.' The obvious symptom in affected infants is chronic bloody diarrhoea in the first few weeks or months of life. Our ability to diagnose the cause of infantile colitis has recently improved considerably with the arrival of small calibre fibreoptic colonoscopes. These allow the infant's colon to be visualised and multiple histological specimens to be obtained. The most common cause is infection-at least 2500 cases were reported nationally in 1989 (J Cowden, Communicable Disease Surveillance Centre, personal communication). In Britain the pathogens seen most often are Campylobacter jejuni,' Salmonella spp, Shigella spp, enterotoxigenic Escherichia coli, which produces a Shiga-like toxin, verotoxin,' and Clostridium difficile. The course of colonic infections in infancy may be more severe and prolonged than in adults because the immune system is still developing, and certain organisms that do not usually cause a colitis may do so in small babies. For example, Yersinia enterocolitica,4 Cyptosporidium,9 Aeromonas hydrophilia," and Clostridium difficile may be associated with colitis even without antecedent treatment with antibiotics.7 When no infectious agent is detectable the colitis may have a more serious prognosis. Classic inflammatory bowel disease may present at any age, and Crohn's disease is more common than ulcerative colitis in children in Britain.' It now seems that the peak in incidence of colitis in infancy is not due to classic inflammatory bowel disease but is attributable to colitis due to food allergy. "' " In one study of non-infectious colitis in infancy all eight patients had an allergic colitis"'; in another two of five children under 1 year of age were similarly afflicted, the three others having indeterminate colitidesinflammation with no diagnostic features.'" The less common causes include colitis with coeliac disease' " and two newly described conditions, autoimmune enteropathy with colitis'4 and a familial necrotising enterocolitis (J A Walker-Smith, personal communication).` BMJ
Vo(1UME 302
9 MARCII 1991
Allergic colitis is a disorder related to atopy which occurs in association with the ingestion of specific dietary antigens, usually in cows' milk or soya based formula feeds, or, if the infant is breast fed, with antigens ingested by the mother which have passed into her milk. Although it first presents in early infancy, symptoms may recur when new antigens are encountered on weaning: wheat, soya, beef, pork, fish, and goats' milk have all been implicated.'6 The macroscopic features seen at colonoscopy in infants with colitis are often distinctive. Allergic colitis has a characteristic patchy appearance with areas ofinflammation (without ulceration) separated by intervening normal mucosa. Crohn's disease and familial necrotising enterocolitis may also appear patchy, but usually there is ulceration-aphthous ulcers and snail track ulcers in Crohn's disease and deep "punched out" ulcers in familial necrotising enterocolitis.s' Inflammation is diffuse in ulcerative colitis and autoimmune colitis (usually affecting the colon with similar severity throughout'4). These clear cut findings and the ability provided by colonoscopy for multiple biopsies throughout the colon have made precise diagnosis possible in most infants with colitis. Nevertheless, while the histological features of classic inflammatory bowel disease are well known,'7 those of allergic colitis are not. The key feature is the infiltrate of the lamina propria, which consists largely of eosinophils (which clump together and invade the epithelium) and plasma cells."' In infancy-as in later life-the non-infective infantile colitides run a chronic fluctuating course which may result in both mortality and severe morbidity if not controlled. The treatment of classic inflammatory bowel disease, autoimmune disease, and familial necrotising enterocolitis is-as in older children and adults-with anti-inflammatory and immunosuppressive drugs and surgery. In colitis due to food allergy, removal of offending foods from the diet (usually from two to five foods in an individual patient) will result in the relief of symptoms, and biopsies will show a fall in the numbers of eosinophils and plasma cells bearing IgE. "' Treatment need be only temporary as patients