Anaphylactoid reactions associated with menstruation affecting two sisters

The day before her next period started, she had a similar, but milder, episode. The following three menstrual cycles were uneventful, but with the fourth she had a further episode of severe hypotension and generalised rash, necessitating acute admission to hospital.

She was admitted electively two days before her next period was due. On day 3 of this period she developed lower abdominal pain, pruritus, a diffuse erythematous rash, oedema of the face and lips, hypotension and bradycardia. Full blood count, routine biochemical test results, erythrocyte sedimentation rate, and concentrations of C-reactive protein, C3, C4, IgE, mast-cell tryptase and urinary 5-hydroxyindolacetic acid were all normal during and after the attack, and there was no peripheral blood eosinophilia.

At laparoscopy tubal ligation clips were found to be still in situ and there was no evidence of intraperitoneal bleeding. After treatment with intravenous fluids, intramuscular adrenaline and antihistamines, she recovered. Menstrual fluid was collected on sterile sanitary towels during her admission. The towels were rinsed in sterile saline and a skinprick test was performed with the eluted fluid and with other standard skin test antigens. She was found to be atopic, with positive reactions to house dust mite and cockroach antigens, but she did not react to her own menstrual fluid.

She was treated with medroxyprogesterone 10 mg twice daily to suppress menstruation. Over the next four months she had minor itching and swelling associated with spotting at the time when her periods would be expected. Ibuprofen 400 mg three times daily was added to the treatment regimen.

She remained well for one year, then had a further admission with facial swelling, urticaria and hypotension, though without vaginal blood loss. She responded rapidly to intravenous fluids and intramuscular adrenaline. This attack had been preceded by vigorous exertion (playing netball), but previous attacks had not, to her knowledge, been associated with exercise. The ibuprofen was replaced with the selective COX-2 inhibitor celecoxib (200 mg twice daily). Over the subsequent 18 months she has had no further menstrual loss and only three episodes of very minor itching, which responded to oral antihistamines.

Concentrations of IgE, C3, C4, C1 esterase inhibitor and mast-cell tryptase in blood taken during an attack were all normal. Radioallergosorbent testing (RAST) against common antigens was negative. The patient elected to have a trial of celecoxib 200 mg daily for three months: for each of the next three cycles, she started taking the drug three days before her period was due and took it for 10 days. Over the three months, she had no recurrence of facial swelling or itching, her menstrual loss diminished and her periods became irregular (although serum gonadotropin levels were low, suggesting that she was not menopausal). She then discontinued the celecoxib and symptoms recurred. Because of the irregularity of the periods, intermittent therapy became impractical and she elected to take celecoxib daily. She has had no further recurrence of symptoms.

In the first report,¹ the patient was 38 years old at the onset of symptoms. The patient had previously suffered one attack of urticaria attributed to seafood. She was non-atopic and had negative skinprick tests to progesterone and oestrogens, but did produce a positive skin reaction to her own menstrual fluid. However, RAST testing showed no evidence of an IgE response to her menstrual fluid. Her serum IgE and complement levels were normal, but it is not clear from the report whether these were estimated during an attack.

In the second report,² the patient was a 35-year-old who had experienced previous attacks of urticaria for which no cause had been found. She too had normal complement levels during an attack and negative skinprick reactions to progestagens and oestrogens. She had a one-month trial of non-steroidal anti-inflammatory therapy with indomethacin but then elected to have a hysterectomy.

Clinically, the episodes in our Case 1 resemble anaphylaxis caused by immune mechanisms, although the lack of eosinophilia and presence of bradycardia are unusual. We found no evidence of activation of the complement system in either case, and mast-cell tryptase concentrations were not elevated, although a normal concentration does not exclude mast-cell degranulation through IgE-mediated mechanisms. The positive skin test in one of the previous reports¹ was interpreted as either an IgE-mediated response or an abnormal vasodilator response to prostaglandins in the menstrual fluid. This seemed to be an idiosyncratic response, as normal control subjects tested with the fluid did not respond. Such a reaction was not seen in our Case 1.

The clinical features of Case 2 are more reminiscent of cyclical urticaria, which has been ascribed to an autoimmune reaction to progestagens⁷ or to oestrogen effects on mast-cell function.⁸ Hypersensitivity to progesterone has been suggested as a mechanism of recurrent anaphylaxis of uncertain cause,⁹ but it is difficult to understand why these attacks should occur at the time of menstruation, when progesterone concentrations are low. Progesterone has in fact been used to treat severe premenstrual exacerbations of asthma¹⁰ and was given to our Case 1 patient without any adverse effects. If progestagens are involved in these reactions, it seems more likely that they do so indirectly by modulating mast cell function and other immune reactions.^11,12

Menstrual fluid is known to contain high levels of prostaglandins, particularly PGF_2α, which has been shown to modulate mediator release in mast-cells.¹³ It was because of this observation that one of the patients in the previous reports² had been treated with indomethacin and our Case 1 patient was treated initially with ibuprofen. However, as COX-2 is present in the endometrium, induction of COX-2 at the end of pregnancy seems to be important in the onset of labour,^14,15 and, as selective COX-2_{inhibitors are effective in treating primary dysmenorrhea,¹⁶ we elected to trial the selective COX-2 inhibitor celecoxib.}

In Case 1, symptom control with celecoxib was better than with the non-selective inhibitor ibuprofen; in Case 2, complete control of the mild symptoms was obtained with celecoxib alone. This may reflect reduction in proinflammatory prostaglandins (such as PGF_2α) by celecoxib without concurrent reduction in immunomodulatory prostaglandins (such as PGE₂), as would occur with non-selective COX inhibition.

We can not offer a complete explanation of this syndrome, but hypothesise that these severe anaphylactoid reactions are an extreme form of the better-known cyclical urticaria. The occurrence of these conditions in sisters suggests a genetic predisposition. On the basis of the response to treatment, it seems possible that altered production of prostaglandins in the uterus influences mast cell function and causes degranulation sufficient to produce severe reactions. Withdrawal of progesterone has been shown to upregulate COX-2 activity in the human endometrium.¹⁷ We have demonstrated that medical treatment of such patients, using a combination of suppression of cyclical hormonal changes and COX-2 inhibition, is possible without resorting to hysterectomy. We would also suggest that a detailed menstrual history be taken in women with apparently idiopathic anaphylaxis in case similar mechanisms are operating.

1. Basomba A, Guerro M, Campos A, et al. Grave anaphylactic-like reaction in the course of menstruation. Allergy 1987; 42: 477-479.
2. Burstein M, Rubinow A, Shalit M. Cyclic anaphylaxis associated with menstruation. Ann Allergy 1991; 66: 36-38.
3. Farah FS, Shbaklu A. Autoimmune progesterone urticaria. J Allergy Clin Immunol 1971; 48: 257-262.
4. Leech SH, Kumar P. Cyclic urticaria. Ann Allergy 1981; 46: 201-203.
5. Gibbs O, Couts II, Lock R, Finnegan OC, White RJ. Premenstrual exacerbations of asthma. Thorax 1984; 39: 833-836.
6. Eliasson O, Scherzer HH, De Graff AC. Morbidity in asthma in relation to the menstrual cycle. J Allergy Clin Immunol 1986; 77: 87-94.
7. Moghadam BK, Hersini S, Barker BF. Autoimmune progesterone dermatitis and stomatitis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1998; 85: 537-541.
8. Slater JE, Raphael G, Cutler GB, et al. Recurrent anaphylaxis in menstruating women; treatment with a luteinising hormone-releasing hormone agonist — a preliminary report. Obstet Gynecol 1987; 70: 542-546.
9. Leylek OA, Unlu S, Ozturkcan S, et al. Estrogen dermatitis. Eur J Obstet Gynecol Reprod Biol 1997; 72: 97-103.
10. Meggs WJ, Pescovitz OH, Metcalfe D, et al. Progesterone sensitivity as a cause of recurrent anaphylaxis. N Engl J Med 1984; 311: 1236-1238.
11. Wilkinson SM, Beck MH, Kingston TP. Progesterone-induced urticaria — need it be autoimmune? Br J Dermatol 1995; 133: 792-794.
12. Mittman RJ, Bernstein DI, Steineberg DR, et al. Progesterone-responsive urticaria and eosinophilia. J Allergy Clin Immunol 1989; 84: 304-310.
13. Kaliner M. Human lung tissue and anaphylaxis I. The role of cGMP as a modulator of the immunologically induced secretory process. J Allergy Clin Immunol 1977; 60: 204-211.
14. Crankshaw DJ, Dyal R. Effects of some naturally occurring prostaconoids and some cyclo-oxygenase inhibitors on the contraction of the human uterine segment in vivo. Can J Physiol Pharmacol 1994; 72: 870-874.
15. Slater D, Allport V, Bennett P. Changes in the expression of the type-2 but not the type-1 cyclo-oxygenase enzyme in chorion-decidua with the onset of labour. Br J Obstet Gynaecol 1998; 105: 745-748.
16. Morrison BW, Daniels SE, Kotey P, et al. Rofecoxib, a specific cyclo-oxygenase-2 inhibitor, in primary dysmenorrhea: a randomized controlled trial. Obstet Gynecol 1999; 94: 504-508.
17. Critchley HO, Jones RL, Lea RG, et al. Role of inflammatory mediators in human endometrium during progesterone withdrawal and early pregnancy. J Clin Endocrinol Metab 1999; 84: 240-248.

Department of Obstetrics and Gynaecology, North Queensland School of Medicine, James Cook University, Cairns, QLD.
Michael D Humphrey, FRANZCOG, FRCOG, Professor.

Reprints will not be available from the authors.
Correspondence: Associate Professor G Simpson, Department of Thoracic Medicine, Cairns Base Hospital, PO Box 902, Cairns, QLD.
marjoATiig.com.au

Make a comment