Вы находитесь на странице: 1из 5

British Journal of Rheumatology 1997;36:234–238

BEHC¸ ET’S DISEASE AND PREGNANCY RELATIONSHIP STUDY

S. MARSAL,* C. FALGA , C. P. SIMEON, M. VILARDELL and J. A. BOSCH

Internal Medicine Service and *Rheumatology Unit, Hospital General i Universitari Vall dHebron, Barcelona, Spain

SUMMARY The effects of pregnancy on the course of Behc¸et’s disease (BD), and vice versa, are unknown and little has been reported. We have studied three groups of women: (1) group A included 61 pregnancies in 23 women with BD, 25 pregnancies took place in 10 patients already diagnosed (group 1A) and 36 pregnancies occurred in 13 patients before disease diagnosis (group 2A); (2) group B included 30 females with 83 pregnancies affected by recurrent oral ulcers (ROU); (3) group C included 20 healthy women with 61 pregnancies. We investigated the effects of BD on pregnancy and fetal outcome, and the influence of gestation on the course of BD. A questionnaire was used in which specific information about each pregnancy, labour and puerperium was collected. We looked for medical confirmation in all cases where any pathology had been identified. No significant differences were found in the incidence of pregnancy complications between groups. The incidence of perinatal death was also similar and neither congenital abnormalities nor neonatal BD were observed. Only two patients observed a flare of the disease and in two cases the diagnosis of BD was made during the pregnancy. In our series, the outcome of pregnancy was generally good in BD patients, disease manifestations were not consistently worsened and fetal outcome was excellent. The first case of Budd–Chiari syndrome during the puerperium in a BD patient is reported.

K : Behc¸et’s disease, Pregnancy, Fetal outcome, Budd–Chiari syndrome.

B¸disease (BD) is a multisystemic disorder of unknown aetiology characterized by recurrent oral and genital ulcers, and eye inflammation. Other less frequent manifestations include skin lesions, arthritis, and neurological, gastrointestinal and vascular abnor- malities. Although BD is mainly diagnosed during the fertile years [1], little is known about its influence on the outcome of gestation [2–7]. In addition, the effect of pregnancy and the puerperium on BD activity is also poorly understood. So far, only a few cases analysing this relationship have been reported in the literature. The aim of this study is to analyse the relationship between gestation and BD.

SUBJECTS AND METHODS We studied 61 pregnancies in 23 women with BD attending our BD clinic. All these patients were diagnosed according to the 1990 revised criteria of the International Study Group for Behc¸et’s Disease [8]. Twenty-five pregnancies took place in 10 patients who had already been diagnosed (group 1A); the other 36 pregnancies occurred in 13 patients before disease diagnosis (group 2A). In two patients from the latter group, the diagnosis of BD was made during the pregnancy (see below). In the first part of this study, we investigated the effect of BD on pregnancy. The patients were interviewed using a questionnaire (Table I) in which specific information about each pregnancy, labour and puerperium (the period between delivery and

Submitted 23 April 1996; revised version accepted 3 September

1996.

Correspondence to: J. A. Bosch, 3rd Floor, Internal Medicine Service, Hospital General i Universitari Vall d’Hebron, Passeig Vall d’Hebron 119–129, 08035 Barcelona, Spain.

234

first menstruation) was collected. We investigated the number and dates of pregnancies and the following conditions: (1) threatened abortion; (2) abortions (pregnancy loss before 28 weeks); (3) placenta previa; (4) toxaemia; (5) premature labour (pregnancy loss between 28 and 37 weeks); (6) jaundice during the third trimester; (7) Caesarean section; (8) infections; (9) other abnormal events. Since the adverse events investigated are so important to the outcome of the pregnancy, we felt they would be easily recalled by all the women. This part of the questionnaire was also applied to 30 females with 83 pregnancies (group B) affected by recurrent oral ulcers (ROU) without BD and to 20 healthy women with 61 pregnancies (group C). The former group of patients was randomly selected from the BD clinic where we regularly see a large number of patients having only ROU. The healthy individuals were randomly selected from the Emergency Service at our hospital where they had been seen for an acute minor illness such as influenza. The second part of this study examined the influence of BD on fetal outcome. The information about newborns was also obtained by questionning the three groups of mothers. We focused on: (1) the incidence of perinatal death; (2) congenital abnormalities; (3) neonatal BD (only for the BD group). In the third part of this study, the effects of gestation on the course of BD were analysed by a third group of questions recording the pregnancy and puerperal incidence of (1) oral and genital ulcer outbreaks, (2) anterior and posterior segment eye inflammation, (3) arthritis, (4) skin lesions, and (5) neurological, (6) gastrointestinal and (7) vascular manifestations. Questions about the presence or absence of symptoms at the time of onset of pregnancy were finally eliminated from the questionnaire since many

1997 British Society for Rheumatology

MARSAL ET AL.: BEHC¸ ET’S DISEASE AND PREGNANCY OUTCOME

235

women failed to recall these non-obstetric data. As the natural history of BD is one of exacerbations and remissions, we understood that it was difficult to fit in one of these periods the onset of pregnancy, especially if symptoms were minor. On the other hand, no clinical score or laboratory measures have been described to assess disease activity. The necessity of treatment at onset of one pregnancy may be used as an indicator of disease activity, but these data were also excluded from the questionnaire for the same reason and because several pregnancies occurred years ago when some drugs were not available. In all groups, the questionnaire was administered by the same physician. After analysing the questionnaires, we looked for medical confirmation in all cases where any pathology had been identified. The agreement between the patients’ answers and the medical records was complete in all these cases even when the interval between the first pregnancy and the questionnaire was maximum (21 years). The confirmatory evidence from the notes emphasized that recall of abnormal events during pregnancy has a high degree of reliability. Differences in the prevalence of conditions between groups were analysed using the 2 test, with Yates’

TABLE I

Questionnaire

Name:

Date of birth:

Number of pregnancies (including abortions):

Age at the beginning of each pregnancy:

1

2

3

4

5

6

Behc¸et disease patient Recurrent oral ulcers Healthy control

disease patient Recurrent oral ulcers Healthy control Part I (write yes/no, number and dates) 1.—threatened

Part I (write yes/no, number and dates)

1.—threatened abortion

2.—abortions

3.—placenta praevia

4.—toxaemia

5.—premature labour 6.—jaundice during 3rd trimester 7.—Caesarean section

8.—infections

9.—other abnormal events

Part II (write yes/no, number and dates)

1.—perinatal death 2.—congenital abnormalities (type) 3.—neonatal Behc¸et disease

Part III (write yes/no, type, number and dates)

1.—oral and genital ulcer outbreaks 2.—eye inflammation

3.—arthritis

4.—skin lesions 5.—neurological manifestations 6.—gastrointestinal manifestations 7.—vascular manifestations

TABLE II Complications of pregnancy, labour and the puerperium in patients with Behc¸et’s disease (group A), recurrent oral ulcers (group B) and healthy controls (group C)

Group

 

1A

2A

B

C

No. of patients/ pregnancies

10/25

13/36

30/83

20/61

Age at pregnancy* 21 6.3

19 6.4

20 6.5

18 6.8

Threatened abortions Abortions Placenta previa Toxaemia Premature labour Jaundice in 3rd trimester Caesarean sections Episiotomy infections Budd–Chiari syndrome

1 (4%)

0 (0%)

0 (0%)

0 (0%)

1 (4%)

4 (11%)

10 (12%)

5 (8%)

0 (0%)

0 (0%)

0 (0%)

2 (3%)

0 (0%)

1 (3%)

0 (0%)

2 (3%)

1 (4%)

0 (0%)

0 (0%)

2 (3%)

0 (0%)

1 (3%)

0 (0%)

0 (0%)

0 (0%)

1 (3%)

3 (4%)

3 (5%)

0 (0%)

3 (8%)

0 (0%)

0 (0%)

1 (4%)

0 (0%)

0 (0%)

0 (0%)

Perinatal death

0 (0%)

1 (3%)

3 (4%)

3 (5%)

Congenital

abnormalities

0 (0%)

0 (0%)

0 (0%)

0 (0%)

Neonatal BD

0 (0%)

0 (0%)

0 (0%)

0 (0%)

Complicated

pregnancies

4 (16%)

11 (30%)

16 (19%)

17 (28%)

Uncomplicated

pregnancies

21 (84%)

25 (70%)

67 (81%)

44 (72%)

No statistically significant differences between all three groups; *mean .. Group 1A, pregnancies in BD patients already diagnosed; group 2A, pregnancies prior to BD diagnosis.

correction when appropriate. P values of 0.05 were considered significant.

RESULTS The mean age at pregnancy was similar in the three groups (Table II). No significant differences in mean age at pregnancy in both groups of women were observed. There were no significant differences between the incidence of complications during pregnancy, labour and the puerperium and the period in which the disease was diagnosed. Only 4/15 observed com- plications (abortion, threatened abortion, premature labour and Budd–Chiari syndrome) occurred in patients previously diagnosed with BD. The remaining 11 occurred in women in whom the disease was diagnosed months or years later. No significant differences were found in the incidence of pregnancy complications between groups A, B and C (Table II). The higher number of abortions in group B does not reach statistical significance. As shown in Table II, five abortions were observed in group A, 10 in group B and five in group C. Therefore, 56 (group A), 73 (group B) and 56 (group C) newborns were analysed in the second part of this study. Only one perinatal death was observed in group A. No significant differences were seen in the incidence of perinatal death between the three groups. Neither

236

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 2

TABLE III Complications of Behc¸et’s disease during the pregnancy in patients diagnosed before (group 1A) or after (group 2A) pregnancy

 

Group 1A

Group 2A

No.

of

patients

(n =10)

(n =13)

No.

of

pregnancies

(n =25)

(n =36)

Oral/genital ulcer outbreaks Eye inflammation Arthritis Skin lesions Neurological manifestations Gastrointestinal manifestations Vascular manifestations

Recurrent episiotomy infections Oral ulcers as a first symptom

1

0

0

1†

0

1†‡

0

1†‡

0

0

0

0

1*

0

0

3*

0

1*

*Placed in puerperium. †BD diagnosis made during pregnancy. ‡Same patient.

congenital abnormalities nor neonatal BD were observed (Table II). In the third part of the analysis, a small number of abnormalities were found (Table III). From all 23 BD

patients interviewed, two patients from group 1A gave

a positive answer. One patient showed an evident

increase in the frequency and intensity of outbreaks of oral ulcers during all periods of pregnancy, and the other patient developed an inferior vena caval and suprahepatic venous thrombosis (Budd–Chiari syn- drome) in the puerperium. As mentioned previously, in two cases from group 2A the diagnosis of BD was made during the pregnancy. Eye involvement in the first trimester was the defining feature in one patient. Anterior uveitis was the diagnosis made by an ophthalmologist and recorded in the notes. The other case was a woman with ROU who in the first trimester developed oligoarthritis, erythema nodosum and positive cutaneous pathergy. Also in group 2A, one patient started the first symptoms (oral ulcers) of BD in the puerperium, although she did not have sufficient criteria to make a definite diagnosis until 16 yr later. Another woman, who developed episiotomy infections in all three consecutive labours, presented 17 yr later with genital and oral ulcers, uveitis and positive cutaneous pathergy.

DISCUSSION We have studied the relationship between pregnancy and BD by analysing retrospectively the evolution of pregnancy, fetal outcome, and disease activity during pregnancy, comparing BD patients, women having only ROU and healthy controls. Specific abnormalities did occur in some patients, but on the basis of our results we felt that, in general, pregnancy in BD does not represent a major risk for the mother or the fetus.

A degree of variability seems to exist in the influence

of pregnancy on BD, although the limited number of

cases reported and the absence of prospective studies makes it difficult to reach a consensus. We have run a

large BD clinic for several years but, despite this, would not have sufficient patients to undertake prospective studies evaluating the relationship between BD and gestation, except over an impracticably long time period. The necessity of making available the maximum amount of information encouraged us to perform this retrospective study. This same motive led us to include the pregnancies previous to disease diagnosis in order to examine for the first time what happened in this subgroup of patients. A further reason for including pregnancies before diagnosis was that, in BD, the period between the first symptom and the definitive diagnosis may be very long, as we could see

in two patients from group 2A, since we lack a specific

diagnostic test. We are aware that many more patients

may be required to confirm our findings, but in relation

to BD they are important given the low prevalence rate

and irregular global distribution of this disease. In an extended literature review, individual case reports describe pregnancy-associated remissions or flares of disease. Madkour and Kudwah [2] reported mucocutaneous and articular exacerbations during

pregnancy in four women. Farrag et al. [9] described

a woman who developed severe genital ulceration in

the third trimester of pregnancy in whom prednisolone

was necessary after delivery. Hurt et al. [10] described

a recurrence of severe oral and genital mucosal

ulceration followed by iridocyclitis during the late second trimester. Conversely, several authors report remissions during pregnancy in BD patients. Chajek and Fainaro [4] describe a woman with persistent BD who after 20 yr follow-up was only asymptomatic during her two pregnancies. Larsson and Baum [6] and Ferraro et al. [5] published two similar cases of complete clinical remission during gestation with a flare after the delivery. Hamza et al. [7] studied retrospectively 21 pregnancies in eight women with BD and observed remission unrelated to therapy in 12 pregnancies, while exacerbations of disease occurred in nine pregnancies, mainly as a painful genital ulceration during the last period of pregnancy. In our series, gestation had a negative influence on the disease activity in only two patients with BD. One patient showed a clear increase in the frequency and intensity of outbreaks of oral ulcers during all her pregnancy and was treated with topical corticosteroids. The second patient was a 26-yr-old woman with known BD, who developed Budd–Chiari syndrome in the puerperium. In this case, treatment with heparin was instituted and then replaced by acenocumarol and colchicine. This represented the only serious effect on BD activity observed in our series. Although a few cases of Budd–Chiari syndrome in BD patients have been reported [11], an extended literature review shows that none of them occurred in the puerperium. In fact, there is little information about an increase in thrombotic events during the last months of pregnancy and puerperium in BD patients. One report of cerebral venous thrombosis at 31 weeks of gestation in a BD patient has also been published [12]. A reduced fibrinolytic activity in addition to a vasculitic process

MARSAL ET AL.: BEHC¸ ET’S DISEASE AND PREGNANCY OUTCOME

237

has been suggested as an explanation for venous thrombosis in BD, but like other alterations of coagu- lation they are neither constant nor specific and they have a variable correlation with clinical manifestations [13–15]. In healthy women, reduced fibrinolytic activity may also be frequently demonstrated during these periods [16], therefore the co-existence of two hypercoagulability conditions might have been ex- pected to increase the risk of thrombotic events. However, in our series, the incidence of thrombosis was really low. A controversial point is whether asymp- tomatic BD patients with altered coagulation tests should be treated. Further research is required to establish a therapeutic consensus. In group 2A, the appearance of new symptoms during the pregnancy that allowed the fulfilment of BD diagnostic criteria was observed in only two cases. Both patients received topical and systemic cortico- steroids, showing a favourable evolution. Certainly, the proportion is very small and insufficient to make any correlation, although it is important to note that both women already had symptoms consistent with BD. Also in this group, two patients developed the first expression of the disease during the pregnancy. The presence of recurrent episiotomy infections in one of these women led us to speculate that these infections could represent the same phenomenon as cutaneous pathergy. We believe that it is remarkable that the unique infectious event observed in the study was recurrent episiotomy infections in a woman who developed definitive BD several years later. Whether they are an expression of the influence of pregnancy on BD or are independent coincidences is unresolved. We may only suggest that in BD patients, and in those who will eventually be diagnosed as having BD, gestation did not appear to result in a specific situation which reliably worsened the disease or favoured its diagnosis. It is important to mention that in those patients whom we certainly knew were in treatment at the onset of pregnancy, it was withdrawn without incident. Systemic corticosteroids (three patients), colchicine (one patient) and cyclosporin (one patient) were discontinued when the women become aware of the pregnancy. No exacerbations of the disease were observed and in no case was it necessary to reintroduce the treatment. These observations also suggest that pregnancy is not a negative factor on BD activity. The second aspect of our study was to examine the effect of BD on the fetus. Clausen and Bierring [17] reported aortic lesions in a 16-week-old fetus from a BD patient and other authors have observed transient mucocutaneous lesions in newborns [18–20]. It has been postulated that the infant’s transient illness is caused by transplacental passage of maternal anti- bodies [18–20], although there is no evidence for this. We think that this is exceptional since in our series of 55 newborns we did not observe any pathology. Our data suggest that the fetal outcome is generally excellent since the frequency of spontaneous abortions, congenital abnormalities and perinatal death in babies born to BD patients included in our series was not

significantly different from that in those born to healthy women and ROU patients. In relation to perinatal death, it might be helpful to give the perinatal mortality rate in our district (1.1%) [21] which is very similar to the rate calculated for our BD group

(1.6%).

The influence of BD on the outcome of pregnancy is even less investigated. Hamza et al. [7] just give a brief reference since they did not find abortions, prematurity or perinatal death in any of the 21 pregnancies that they studied in eight women with BD. Our findings show that BD does not have an unfavourable influence on pregnancy, labour or the puerperium since their course was similar in BD patients and in healthy or ROU women. No cases of placenta previa were observed in BD patients although, as discussed above, episiotomy infection and Budd–Chiari syndrome were only found in this group. As we said previously, we are conscious that the group sizes are too small to exclude adverse effects completely; however, we feel that they are sufficient to detect gross effects of the disease on pregnancy and fetal outcome. Although it was not an objective of this study, the impression that the number of pregnancies per woman was lower in BD was not confirmed by statistical methods (binomial distribution probability). In any case, the apparent lower fertility ratio in BD vs controls would magnify any risk differences between groups for pregnancy and fetal outcome. In summary, in our series, the outcome of pregnancy is generally good in BD, disease manifestations are not consistently worsened, and fetal outcome is excellent. The first case of Budd–Chiari syndrome during the puerperium in a BD patient is reported.

A We would like to thank Dr G. H. Kingsley for her comments.

R

1. O’Duffy JD. Behc¸et’s syndrome [Editorial]. N Engl J Med 1990;322:326–8.

2. Madkour M, Kudwah A. Behc¸et’s disease. Br Med J

1978;2:1786.

3. Plouvier B, Devulder B. Behc¸et’s disease. Br Med J

1979;1:690.

4. Chajek T, Fainaro M. Behc¸et’s disease. Report of 41 cases and review of the literature. Medicine 1975;54:

179–96.

5. Ferraro G, Lo Meo C, Moscarelli G, Assennato E. A case of pregnancy in a patient suffering from the Behc¸et’s syndrome: immunological aspects. Acta Eur Fertil

1984;15:67–72.

6. Larsson L-G, Baum J. Behc¸et’s syndrome in pregnancy and after delivery. J Rheumatol 1987;14:183.

7. Hamza M, Eleuch M, Zribi A. Behc¸et’s disease and pregnancy. Ann Rheum Dis 1988;47:350–2.

8. International Study Group for Behc¸et Disease. Criteria for Behc¸et’s disease. Lancet 1990;335:1078–80.

9. Farrag OA, Al-Suleiman SA, Bella H, Al-Omari H. Behcet disease in pregnancy. Aust NZ J Obstet Gynaecol

1987;27:161–3.

238

BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 2

10. Hurt WG, Cooke LL, Jordan WP, Bullock JP, Rodrı´guez GE. Behc¸et’s syndrome associated with pregnancy. Obstet Gynecol 1979;53:31S–3S.

11. Bismuth E, Hadengue A, Hammel P, Benhamou J-P. Hepatic vein thrombosis in Behc¸et’s disease. Hepatology

1990;11:969–74.

12. Wechsler B, Genereau T, Biousse V et al. Pregnancy complicated by cerebral venous thrombosis in Behc¸et’s disease. Am J Obstet Gynecol 1995;173:1627–9.

13. Conard J, Horellou MH, Wechsler B et al. La maladie de Behc¸et est-elle associe´e a des anomalies car- acte´ristiques de la coagulation et de la fibrinolyse? A propos de 70 observations. J Mal Vasc (Paris)

1988;13:257–61.

14. Schmitz-Huebner U, Knop J. Evidence for endothelial cell dysfunction in association with Behc¸et’s disease. Thromb Res 1984;34:227–85.

15. Aitchison R, Chu P, Cater DR, Harris RJ, Powell RJ. Defective fibrinolysis in Behc¸et’s syndrome: sig-

nificance and possible mechanisms. Ann Rheum Dis

1989;48:590–3.

16. Stirling Y, Woolf L, North WRS, Seghatchian MJ, Meade TW. Haemostasis in normal pregnancy. Thromb Haemostasis 1984;52:176–82.

17. Clausen J, Bierring F. Fetal arterial involvement in Behc¸et’s disease: an electron microscope study. Acta Pathol Microbiol Immunol Scand 1983;91:133–6.

18. Thivolet J, Cambazard F, Genvo M-F. Grande aphtose ne´o-natale de transmission maternelle. Ann Dermatol Venerol (Paris) 1982;109:815–6.

19. Lewis MA, Priestley BL. Transient neonatal Behc¸et’s disease. Arch Dis Child 1986;61:805–6.

20. Fam AG, Siminovitch KA, Carette S, From L. Neonatal Behc¸et’s syndrome in an infant of a mother with the disease. Ann Rheum Dis 1981;40:509–12.

21. L’enquesta confidencial de mortalitat perinatal a la ciutat de Barcelona. In: La salut a Barcelona. Ajuntament de Barcelona: Area de Salut Pu´ blica, 1992.