88 1
PREGNANCY IN PATIENTS WITH A HISTORY OF JUVENILE RHEXJMATOID ARTHRITIS MONIKA @STENSEN The present investigation was undertaken to study the relationship between pregnancy and juvenile rheumatoid arthritis (JRA), with regard to possible reactivation of the disease, complications at delivery, and postpartum complications. Data on 76 pregnancies in 51 JRA patients were collected retrospectively. Comparison of pre-pregnancy disease activity with the course during gestation showed that pregnancy did not cause reactivation of the symptoms of quiescent .JRA. Patients with minor symptoms at conception and the majority of those with active inflammation experienced improvement or total remission in the second half of gestation. Four JRA patients with active anterior uveitis had active eye disease during pregnancy also. Seventyfour of the pregnancies resulted in births of infants that were healthy and of normal birth weight; 1 infant had low birth weight and 1 was stillborn. Twenty children were delivered by cesarean section, and in 15 cases this was related to sequelae of JRA. A flare 3-6 months postpartum was reported after 45 pregnancies. However, in none of the patients whose disease was quiescent before or during pregnancy did this cause permanent reactivation of the JRA. Comparison of these 51 patients with 45 age-matched female patients without children Presented in part at the XXIII Scandinavian Congress of Rheumatology, Tampere, Finland, June 1990. From the Oslo Sanitetsforening Rheumatism Hospital, University of Oslo, Oslo, Norway. Supported in part by the Grethe Harbitz Fund and the Norwegian Women’s Health Association. Monika @stensen, MD: Consultant in Rheumatology, and Senior Investigator. Address reprint requests to Monika @stensen, MD, Oslo Sanitetsforening Rheumatism Hospital, Akersbakken 27, N 0172 Oslo 1, Norway. Submitted for publication July 23, 1990; accepted in revised form December 3 1, 1990. Arthritis and Rheumatism, Vol. 34, No. 7 (July 1991)
revealed that disease severity and functional impairment were the limiting factors in the decision for or against having children.
Juvenile rheumatoid arthritis (JRA) is a rheumatic disease affecting slightly more females than males (1). By definition (2), the disease starts before the age of 16, affects 1 or more joints, and sometimes also affects extraarticular sites, such as the eye. There are subgroups of JRA, which differ with regard to extraarticular and systemic features as well as the impact on growth and development. Spontaneous remission of the disease is reported to occur in approximately 70% of the patients, allowing most to conduct a normal life during adulthood (3). The interaction between pregnancy and JRA has not been studied in detail. If the diagnostic criterion of disease onset before the age of 16 (2) is strictly applied, only 2 true cases of pregnancy in patients with JRA have been reported in the literature. The first, a patient who had had polyarticular-onset JRA since the age of 12, had a flare of systemic JRA at midgestation but improved gradually after delivery (43).The second, a patient with polyarticular JRA, experienced remission of disease symptoms during pregnancy. She had a severe flare of JRA 4 months after delivery (6). There are several case reports of adult-onset Still’s disease, which started during pregnancy in 3 cases (7-9). was active but controlled by drugs in another pregnant patient (lo), and flared postpartum after 2 pregnancies in 1 patient ( I 1). The anecdotal data cited do not provide any adequate basis for counseling of JRA patients who are planning to become or already are pregnant. Young women with a history of JRA often feel concerned
882
DSTENSEN
about having a family. T h e y are worried about the or inactive-on their impact of their disease-active offspring. Some wonder if their previous treatment with potent disease-modifying or immunosuppressive drugs could harm their children. Will the disease, if quiescent, b e reactivated by pregnancy? Can the disease be inherited by the children? A r e complications during delivery t o be expected due t o growth inhibition of the pelvis or early involvement of the hip joints? Although there have been numerous reports on the effect of pregnancy on various inflammatory rheumatic diseases (12), there is insufficient information regarding t h e relationship between pregnancy and JRA. The present study was conducted t o address some of t h e questions cited above.
PATIENTS AND METHODS The files of all patients who were hospitalized in the Department of Pediatric Rheumatology at Oslo Sanitetsforening Rheumatism Hospital with a diagnosis of JRA during the period 1958-1968 were examined. All female patients between the ages of 18 and 45 years who met the American Rheumatism Association criteria for JRA (2) were initially selected for study. Of the 228 patients who met these criteria, 191 could be traced. One hundred thirty-seven patients had not borne children, whereas 54 had. A total of 129 patients, 51 with children and 78 without, agreed to participate in the study. Three JRA patients who had children did not participate for unknown reasons. Seven patients without children were excluded because the followup disclosed diagnoses other than JRA, i.e., dermatomyositis, systemic lupus erythematosus, and scleroderma. Fifty-two patients without children failed to return the questionnaire. This low rate of participation among the group of women without children may be due to their lack of interest in a study addressing a topic not directly relevant to them, i.e., pregnancy. An analysis of their files did not reveal any particular disease pattern or drug regimen that would distinguish them from the group who did return the questionnaires. The retrospective data were obtained from the patients' hospital records, and a questionnaire (see below). Basic disease-related information, such as the year of onset of JRA, type of onset, course (polyarticular, pauciarticular, or systemic), severity, treatment, extraarticular features, presence of rheumatoid factor or antinuclear antibodies, and surgery, was derived from the in- and outpatient files for each patient. Twenty-three of the patients had continued with regular outpatient visits through adulthood, and in 20, data on the course of JRA during or after pregnancy were found in the records. Whenever possible, patient records from primary care physicians and obstetric departments were also obtained. All patients were sent a questionnaire consisting of some structured and some open-ended questions and were asked to complete it and return it in a stamped, self-
Table 1. Clinical data o n 51 juvenile rheumatoid arthritis (JRA) patients with children and 45 JRA patients without children
Mean age at time of study (years) Mean age at disease onset (years) Onset type Pauciarticular Polyarticular Systemic No. rheumatoid factor positive Course Pauciarticular Polyarticular Mean disease duration (years) No. with joint surgery No. with hip prosthesis No. in remission
With children
Without children
30.2 8
29.8 8
24 24 3 9
27 2 9
14 37 18.9 30 8 29
16
4 41
19.4 33 17 19
addressed envelope that was provided. A reminder was sent to patients who did not respond. In addition, the 23 patients who continued with regular visits were personally interviewed to verify the answers given on the questionnaires. The questions to JRA patients with children addressed information on personal data, history of JRA, functional status ( 13), mobility, household activities, activities of daily living, employment, disease activity before, during, and after pregnancy, offspring, and personal views about motherhood and JRA. The questionnaire for JRA patients without children was identical, but excluded questions on maternity and included instead questions about the reasons for not having children. General disease characteristics of the 2 JRA groups are given in Table 1. Differences between the groups were analyzed by chi-square test with Yates' correction. Determination of disease activity. For comparison of disease activity before, during, and after pregnancy, several criteria were assessed. Active articular disease was defined as 2 or more of the following: morning stiffness >30 minutes, swelling of 1 or several joints, joint pain on motion, extraarticular features, and need for nonsteroidal antiinflammatory drug (NSAID) treatment. Active systemic JRA required temperature >39"C, JRA rash, or serositis. Remission was defined as complete absence of the above-mentioned symptoms for at least 6 months. Minor symptoms included occasional episodes of joint pain and stiffness, but without joint swelling.
RESULTS T h e 5 1 JRA patients had given birth to a total of 76 children. W h e n grouped according to type of disease onset, the 24 women with polyarticular onset had borne 44 children, the 24 patients with pauciarticular onset had borne 25 children, a n d the 3 patients with systemic onset had borne 7 children. Twenty-seven patients had I child, 20 had 2 children, and 3 had 3 children.
883
PREGNANCY AND JRA Table 2. Activity of juvenile rheumatoid arthritis at conception and during 50 pregnancies
Activity at conception (no. of pregnancies) In remission (20) Minor symptoms (20) Active (10)
Activity during pregnancy Better
No change
Worse
0 10
19 8
7
1
1 2 2
Effect of pregnancy on the course of JRA. Long periods of remission are common in JRA, especially in early adulthood (14). Therefore, the attempt was made to correlate disease activity during pregnancy with disease activity prior to conception. Reliable data about pre-pregnancy disease activity could be obtained for 50 pregnancies, from in- and outpatient records or from the files of primary care physicians. As shown in Table 2, pregnancy did not reactivate the symptoms of quiescent JRA, which was present in 12 patients with pauciarticular onset, 7 with polyarticular onset, and 1 with systemic onset. An episode of acute arthritis in 1 or several joints occurred during the first trimester in 15 pregnancies. However, most patients with minor symptoms at the start of pregnancy and the majority of those with active inflammation experienced improvement or total remission in the second half of gestation. An analysis of the possible influence of onset type on disease activity during pregnancy revealed the following: Among patients with polyarticular-onset JRA, improvement was noted during 26 Pregnancies (59%), there was no change during 15 pregnancies (34%), and exacerbation occurred during 3 pregnancies (7%). JRA patients with pauciarticular onset had improvement during 9 pregnancies (36%), no change during 14 pregnancies (56%), and aggravation of disease during 2 pregnancies (8%). The latter occurred in 2 HLA-B27 positive patients who developed clinical and radiologic signs of ankylosing spondylitis (AS) during adulthood. Disease activity improved during 2 pregnancies, was unchanged during 4 pregnancies, and worsened during 1 pregnancy in patients with systemic-onset JRA. Among all 76 pregnancies taken together., improvement was noted during 35, there was no change in disease activity reported during 34, and signs and symptoms worsened during 7. However, as indicated in Table 2, 20 of the patients were known to have had inactive disease at the time of conception. None of the patients reported serious extraarticular disease related
to JRA during pregnancy. However, 4 of the 6 patients with chronic uveitis experienced active eye disease during pregnancy. Multiple pregnancies. Of the 23 women with multiple pregnancies, the course of JRA was the same during both or all pregnancies in 17, whereas 6 had diEering courses during the different pregnancies. Improvement was noted in 13, no change in 9, and disease exacerbation in 1 of the subsequent pregnancies. Pregnancies that ended in abortion. Among the JRA patients with children, 9 had abortions, either spontaneous (5 cases) or elective (4 cases), during the first trimester (week 8-16). The course of JRA before and after 7 first-trimester abortions in 5 JRA patients without children was also reported. There was no change in disease activity during 14 of these early pregnancies, and improvement was reported in 2. After the termination of pregnancy, disease exacerbation occurred in 4 cases, whereas there was no change in disease activity in 10. There was no particular pattern of disease found in the 8 patients who had spontaneous abortion. None of them had used cytotoxic drugs prior to or during the pregnancy. Drug treatment before and during pregnancy. Data on drug treatment at the onset of pregnancy was available for 20 of the pregnancies. This treatment included use of NSAIDs (19 cases), prednisolone (6 cases), hydroxychloroquine (5 cases), gold (1 case), and penicillamine (1 case). In 4 additional pregnancies, treatment with azathioprine (1 case) or methotrexate (3 cases) had been stopped shortly before conception. Immunosuppressive treatment was withdrawn during the first trimester of pregnancy in all patients. Eight patients continued taking NSAIDs during pregnancy, 4 took low doses of prednisolone, 2 received intraarticular corticosteroids, and 2 took analgesics occasionally. Outcome of pregnancy. The outcome of the 76 pregnancies is shown in Table 3. Seventy-five of the pregnancies resulted in the delivery of live, normal children of normal birth weight, and there was 1 stillbirth. Seventy babies were born at full term; 5 were born a little early (15-23 days before term) but were not considered premature. Only 1 child was considered premature, with a birth weight of 1,800 gm. This infant was delivered by cesarean section at week 33 due to very active disease in the mother and fetal distress detected by ultrasound. The mother had been receiving low-dose pulse methotrexate until 3 weeks before conception, and prednisolone treatment had been continued throughout pregnancy. Twenty children were delivered by cesarean
@STENSEN
884 Table 3. Outcome of 76 pregnancies in 51 women with juvenile rheumatoid arthritis (JRA)
Live birthslstillbirths Male childrenlfemale children Full-terdprematurelearly but not considered premature Mean birth weight (gm) Type of delivery Normal Cesarean section Other* Reason for cesarean section Contracted pelvis Hip prosthesis Non-JRA-related
* Five forceps deliveries; 2 vacuum deliveries.
7511 30146 701115 3.760 49 20
7 7 8
5 -
section, and in 15 cases, this was related to sequelae of JRA, including contracted pelvis or hip prosthesis. No complications due to anesthesia were reported. At the time of this study, the median age of the 75 live children was 5 years (range 0.6-24 years). All children, including the 1 born prematurely and the 14 whose mothers had received drug treatment during pregnancy, had normal mental and physical development. Activity of JRA after pregnancy. Investigation of the activity of JRA within 1 year after delivery revealed the occurrence of a postpartum flare of the disease after 45 of 50 pregnancies reviewed. The exacerbation or reappearance of disease symptoms was most often encountered 3-6 months after delivery. In general, symptoms eased or disappeared again within 2 years after pregnancy. In none of the patients who had quiescent disease before and during pregnancy did the postpartum flare cause permanent reactivation of the JRA. Of the 23 patients with multiple pregnancies, 9 noted a more protracted period of reactivation after the second or third pregnancy than after the first, and in 6 cases this started a permanent recrudescence of the JRA. Decision to become pregnant and severity of JRA. To evaluate whether, among patients with JRA, those with less severe disease were more likely to choose to have children, 45 age-matched female JRA patients were recruited from the 78 without children. They had answered the same questionnaire, but excluding questions about maternity. While the 2 groups were comparable with respect to age at onset of JRA, disease duration, and frequency of remission, there were differences between the 2 groups in some respects: JRA patients without children more often had polyarticular disease at onset and during followup, more of them had undergone joint surgery and, most
importantly, twice as many had had hip prosthesis (Table 1). A comparison of drug treatment given during the course of the JRA revealed that patients without children were more likely to have received immunosuppressive or cytotoxic drugs (Table 4); these included azathioprine, methotrexate, cyclophosphamide, and chlorambucil. Alkylating drugs had been given to 6 patients without children, whereas only 1 JRA patient with offspring had received cyclophosphamide previously. Functional ability also differed in the 2 groups. Forty-six patients in the group with children, versus 37 of those without children, judged their functional ability as good or fairly good, with no or only occasional need for assistance in daily activities. Six patients in the group without children, versus 2 in the group with children, were severely disabled. Moreover, 3 of the JRA patients without children had developed amyloidosis, 2 of them requiring kidney transplantation. Employment. Of the JRA patients with children, 4 were still in school, 21 had full-time or part-time employment, and 26 were homemakers. Of those without children, 31 were employed full-time or parttime, 9 were in school, and 5 were unable to work. Family planning. All patients were asked to give their opinion about the advisability of pregnancy in a woman with a history of JRA, and the number of children she should have. Of the 51 women with children, 13 recommended having 1 child, 21 recommended having 2 children, 2 recommended having more than 2 children, and 15 recommended having children but were uncertain about the number of children desirable. Thirty patients had been or were currently using oral contraceptives. Eleven of the patients without children advised against pregnancy, 7
Table 4. Drug treatment in 51 juvenile rheumatoid arthritis (JRA) patients with children and 45 JRA patients without children*
Treatment NSAIDs NSAIDs + DMARDs NSAIDs + DMARDs + steroids NSAIDs + DMARDs + steroids + cytotoxic agents
With children
Without children
7 15 22 7
7 10 13 15
* Values are the number of patients who received any of the treatments any time during the course of the disease (i.e., for patients with children, the numbers include treatment before, during, or after pregnancy). NSAIDs = nonsteroidal antiinflammatory drugs; DMARDs = disease-modifying antirheumatic drugs.
885
PREGNANCY AND JRA
Table 5. Activity ofjuvenile rheumatoid arthritis (JRA) at conception and during 1 year postpartum Activity at conception (no. of pregnancies)
Course postpartum
No change
Worse
9 10
11 10
4
6
~
In remission (20) Minor symptoms (20) Active (10)
recommended having 1 child, 23 recommended having 2 children, and 4 recommended having more than 3 children. Reasons given for not having children included functional impairment (13 patients) and fear of heredity of JRA (12 patients). Infertility was reported in 2 patients, and 2 had undergone voluntary sterilization. Thirty-seven of the 45 women without children reported having a normal menstrual cycle. Twenty-six used or had used contraceptives, mainly oral contraceptives. Heredity. Two of the children born to the 51 JRA patients had rheumatic disease. In both cases, the mothers had polyarticular-onset JRA, 1 associated with chronic uveitis and the other with psoriasis. Both were married to men with HLA-B27 positive AS. The 2 affected daughters were both HLA-B27 positive. One, like her mother, had polyarticular-onset JRA with chronic uveitis. The other had radiologically verified AS.
DISCUSSION This is the first retrospective study of a large series of JRA patients with regard to the mutual effects of pregnancy and the disease. The analysis of 76 pregnancies occurring in 51 JRA patients provided 2 major conclusions: First, pregnancy per se had no adverse effect on the signs and symptoms of JRA. Second, fetal and maternal outcome of pregnancy was good. Inactive disease did not flare during gestation, though a small proportion of patients experienced a short episode of active arthritis during the first trimester. This finding parallels observations made in other seronegative arthritides, e.g., AS and psoriatic arthritis (PA) (15,16). Active polyarticular or pauciarticular JRA was improved by pregnancy, particularly during the second half of gestation. However, the activity of chronic anterior uveitis present in 4 cases was not influenced by pregnancy. Pregnancy ending in firsttrimester abortion had no major influence on the activity of JRA. As in AS and PA (12), the disease
course during one pregnancy did not automatically predict the course during subsequent pregnancies. However, in the majority of cases, JRA activity did not worsen during subsequent pregnancies. The beneficial effect of pregnancy occurred mainly during gestation, whereas the postpartum period carried the risk of a transient aggravation or recrudescence of the disease. An arthritis flare was most often experienced during the first 6 months after delivery, similar to the findings reported in rheumatoid arthritis ( l l ) , AS (17), and PA (12). The postpartum flare was independent of factors such as pre-pregnancy disease activity (Table 5 ) , type of onset, and course of JRA. In the majority of patients, this reactivation of JRA did not result in permanent exacerbation of the disease. However, a note of caution seems appropriate: Women with multiple pregnancies noted a tendency for a more protracted reactivation of disease after later pregnancies than after the first one. This implies that multiple pregnancy could cause a more permanent recrudescence of JRA. However, it is also possible that overall, pregnancy is associated with a better prognosis in JRA, especially in patients with transient monarticular pauciarticular disease that never requires care by a specialist. Selection bias toward more severe disease cannot be ruled out in a study population derived from hospital inpatients. The present data are reassuring with regard to fetal and maternal outcome of pregnancy in patients with JRA. A history of JRA and previous treatment in the mother did not imply increased frequency of gestational disease, fetal loss, or neonatal disease. However, possible long-term effects of previous drug treatment cannot be ruled out by this study since the average followup time in the offspring was only 5 years. Other than congenital malformations, possible effects on the child resulting from immunosuppressive treatment of the mother have rarely been addressed in the literature. A followup of 7 children exposed to gold compounds in utero showed normal development in all (18). In 1 study, no major disease was found in the offspring of survivors of treated childhood cancer (19); however, the average time of observation was only 4 years. With regard to the type of delivery, JRA was associated with a higher frequency of cesarean section, with bilateral hip prosthesis as the main reason. Obstetricians were apparently concerned, perhaps unduly so, about possible parturitional stress on the prosthesis. Because hip involvement in JRA frequently results in hip prosthesis (20), surgeons should
@STENSEN
886 counsel their adolescent female patients about possible consequences and precautions. It might also be useful to inform JRA patients with involvement of the cervical spine, the temporomandibular joint, and the cricoarytenoid joint about possible hazards of general anesthesia (21,22). If cesarean section is proposed, regional anesthesia may be the preferable method (23). In any case, emergency anesthesia should be avoided if possible. The results of this study indicate that a history of JRA influences the decision for or against having a child. Willingness to have offspring seemed more frequent among women with intermittent disease activity or long-lasting remission, minor functional impairment, and absence of potential hazardous medical treatment. In contrast, the decision not to have children appeared to be related to the presence of polyarticular, nonremitting disease. Polyarticular JRA is known to be frequently associated with persistent active disease, serious complications such as amyloidosis, and the need for aggressive therapy (24). Moreover, spontaneous remission occurs less frequently in the polyarticular than the pauciarticular subtype (25). The findings in the present study were consistent with other reports of a remission rate of approximately 60% in JRA (3). Patients without offspring more frequently used immunosuppressive and alkylating agents. Factors such as fear of long-term effects of cytotoxic agents, fear of heredity of JRA, and sometimes severe functional impairment definitely influenced their decision against pregnancy. This was reflected by the frequent use of contraceptives and even sterilization. Moreover, 24% of these patients recommended not having children at all. Severe polyarticular disease with aggressive drug treatment and functional impairment thus has long-term lifestyle consequences for the female patient. With regard to the advisability of pregnancy for women with a history of JRA, the data from this retrospective study indicate the following: JRA patients who consider pregnancy can be reassured that, as a rule, pregnancy will not aggravate or reactivate their disease, nor will their disease harm the fetus. Complications at delivery are to be expected when cephalopelvic dystocia or bilateral hip involvement is present. Special anesthetic techniques may be necessary if cesarean section is required. Of note, the risk for a postpartum flare is present even in patients with inactive disease. Patients themselves advise having no more than 2 children, and indeed, multiple pregnancy
may carry the risk of more permanent reactivation of JRA. Heredity is not directly associated with JRA. However, there appears to be a risk of inherited disease in children whose mothers have seronegative JRA, and whose fathers have an HLA-linked rheumatic disorder such as AS.
REFERENCES 1. Hanson V, Kornreich HK, Bernstein B, King KK, Singsen BH: Three subtypes of juvenile rheumatoid arthritis: correlations of age at onset, sex, and serologic factors. Arthritis Rheum 20 (suppl 2):184-186, 1977 2. JRA Criteria Subcommittee of the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association Section of the Arthritis Foundation: Current proposed revision of JRA criteria. Arthritis Rheum 20 (suppl 2): 195-199, 1977 3. Hull RG: Outcome in juvenile arthritis. Br J Rheumatol 27 (suppl 1):66-71, 1988 4. Cantor B, Tyler T, Nelson RM, Stein GH: Oligohydramnios and transient neonatal anuria. J Reprod Med 24: 220-223, 1980 5. Stein GH, Cantor B, Panush RS: Adult Still’s disease associated with pregnancy. Arthritis Rheum 23:248-250, 1980 6. Hodgkinson R: Anesthetic management of a parturient with severe juvenile rheumatoid arthritis. Anesth Analg 60:611412, 1981 7. Elkon KB, Hughes GRV, Bywaters EGL, Ryan PFJ, Inman RD, Bowley NB, James MP, Eady RAJ: Adultonset Still’s disease: twenty-year followup and further studies of patients with active disease. Arthritis Rheum 255474554, 1982 8. Green J, Kanter Y, Barzilai D: Adult Still’s disease associated with pregnancy. Isr J Med Sci 18:1037-1039, 1982 9. Bango MY, Garcia Paez JM, Solovera JJ, Merino MF, Giron Gonzalez JA: Adult-onset Still’s disease: a case with onset during pregnancy (letter). Arthritis Rheum 28:957, 1985 10. Unsworth J, d’Assis-Fonseca A, Beswick DT, Blake DR: Serum salicylate levels in a breast fed infant. Ann Rheum Dis 46:638-639, 1987 11. Katz WE, Starz TW, Winkelstein A: Recurrence of adult Still’s disease after pregnancy. J Rheumatol 17: 373-374, 1990 12. @stensen M, Husby G: Pregnancy and rheumatic disease: a review of recent studies in rheumatoid arthritis and ankylosing spondylitis. Klin Wochenschr 62:891895, 1984 13. Steinbrocker 0, Traeger CH, Batterman RC: Therapeutic criteria in rheumatoid arthritis. JAMA 140:659-662, 1949
PREGNANCY AND JRA
14. Calabro JJ, Burnstein SL, Staley HL, Marchesano .JM: Prognosis in juvenile rheumatoid arthritis: a fifteen-year followup of 100 patients. Arthritis Rheum 20 (suppl 2):285, 1977 15. @stensen M, Husby G: A prospective clinical study of the effect of pregnancy on rheumatoid arthritis and ankylosing spondylitis. Arthritis Rheum 26:1155-1 159, 1983 16. @stensen M: Pregnancy in psoriatic arthritis. Scand J Rheumatol 17:67-70, 1988 17. @stensen M, Romberg @, Husby G: Ankyloising spondylitis and motherhood. Arthritis Rheum 25: 140143, 1982 18. Tarp U, Graudal H: A followup study of children exposed to gold compounds in utero (letter). Arthritis Rheum 28:235-236, 1985 19. Li FP, Jaffe N: Progeny of childhood-cancer survivors. Lancet II:707-709, 1974
887
20. Harris CM, Baum J: Involvement of the hip in juvenile rheumatoid arthritis: a longitudinal study. J Bone Joint Surg 70A:821-833, 1988 21. Smith BL: Anesthesia and Still’s disease (letter). Anesthesia 40:209, 1985 22. Person DA: Juvenile rheumatoid arthritis: anesthetic and surgical considerations. AORN J 44:439-449, 1986 23. Spiechowicz J, Pahle JA: Clinical aspects of postoperative continuous regional analgesia in the child with JRA. Adv Pain Res 15:189-200, 1990 24. Jeremy R, Schaller J, Arkless R, Wedgwood RJ, Healey LA: Juvenile rheumatoid arthritis persisting into adulthood. Am J Med 45:419-434, 1968 25. Hanson V, Kornreich H , Bernstein B, King KK, Singsen B: Prognosis of juvenile rheumatoid arthritis. Arthritis Rheum 20 (suppl2):279-284, 1977
PREGNANCY IN PATIENTS WITH A HISTORY OF JUVENILE RHEXJMATOID ARTHRITIS MONIKA @STENSEN The present investigation was undertaken to study the relationship between pregnancy and juvenile rheumatoid arthritis (JRA), with regard to possible reactivation of the disease, complications at delivery, and postpartum complications. Data on 76 pregnancies in 51 JRA patients were collected retrospectively. Comparison of pre-pregnancy disease activity with the course during gestation showed that pregnancy did not cause reactivation of the symptoms of quiescent .JRA. Patients with minor symptoms at conception and the majority of those with active inflammation experienced improvement or total remission in the second half of gestation. Four JRA patients with active anterior uveitis had active eye disease during pregnancy also. Seventyfour of the pregnancies resulted in births of infants that were healthy and of normal birth weight; 1 infant had low birth weight and 1 was stillborn. Twenty children were delivered by cesarean section, and in 15 cases this was related to sequelae of JRA. A flare 3-6 months postpartum was reported after 45 pregnancies. However, in none of the patients whose disease was quiescent before or during pregnancy did this cause permanent reactivation of the JRA. Comparison of these 51 patients with 45 age-matched female patients without children Presented in part at the XXIII Scandinavian Congress of Rheumatology, Tampere, Finland, June 1990. From the Oslo Sanitetsforening Rheumatism Hospital, University of Oslo, Oslo, Norway. Supported in part by the Grethe Harbitz Fund and the Norwegian Women’s Health Association. Monika @stensen, MD: Consultant in Rheumatology, and Senior Investigator. Address reprint requests to Monika @stensen, MD, Oslo Sanitetsforening Rheumatism Hospital, Akersbakken 27, N 0172 Oslo 1, Norway. Submitted for publication July 23, 1990; accepted in revised form December 3 1, 1990. Arthritis and Rheumatism, Vol. 34, No. 7 (July 1991)
revealed that disease severity and functional impairment were the limiting factors in the decision for or against having children.
Juvenile rheumatoid arthritis (JRA) is a rheumatic disease affecting slightly more females than males (1). By definition (2), the disease starts before the age of 16, affects 1 or more joints, and sometimes also affects extraarticular sites, such as the eye. There are subgroups of JRA, which differ with regard to extraarticular and systemic features as well as the impact on growth and development. Spontaneous remission of the disease is reported to occur in approximately 70% of the patients, allowing most to conduct a normal life during adulthood (3). The interaction between pregnancy and JRA has not been studied in detail. If the diagnostic criterion of disease onset before the age of 16 (2) is strictly applied, only 2 true cases of pregnancy in patients with JRA have been reported in the literature. The first, a patient who had had polyarticular-onset JRA since the age of 12, had a flare of systemic JRA at midgestation but improved gradually after delivery (43).The second, a patient with polyarticular JRA, experienced remission of disease symptoms during pregnancy. She had a severe flare of JRA 4 months after delivery (6). There are several case reports of adult-onset Still’s disease, which started during pregnancy in 3 cases (7-9). was active but controlled by drugs in another pregnant patient (lo), and flared postpartum after 2 pregnancies in 1 patient ( I 1). The anecdotal data cited do not provide any adequate basis for counseling of JRA patients who are planning to become or already are pregnant. Young women with a history of JRA often feel concerned
882
DSTENSEN
about having a family. T h e y are worried about the or inactive-on their impact of their disease-active offspring. Some wonder if their previous treatment with potent disease-modifying or immunosuppressive drugs could harm their children. Will the disease, if quiescent, b e reactivated by pregnancy? Can the disease be inherited by the children? A r e complications during delivery t o be expected due t o growth inhibition of the pelvis or early involvement of the hip joints? Although there have been numerous reports on the effect of pregnancy on various inflammatory rheumatic diseases (12), there is insufficient information regarding t h e relationship between pregnancy and JRA. The present study was conducted t o address some of t h e questions cited above.
PATIENTS AND METHODS The files of all patients who were hospitalized in the Department of Pediatric Rheumatology at Oslo Sanitetsforening Rheumatism Hospital with a diagnosis of JRA during the period 1958-1968 were examined. All female patients between the ages of 18 and 45 years who met the American Rheumatism Association criteria for JRA (2) were initially selected for study. Of the 228 patients who met these criteria, 191 could be traced. One hundred thirty-seven patients had not borne children, whereas 54 had. A total of 129 patients, 51 with children and 78 without, agreed to participate in the study. Three JRA patients who had children did not participate for unknown reasons. Seven patients without children were excluded because the followup disclosed diagnoses other than JRA, i.e., dermatomyositis, systemic lupus erythematosus, and scleroderma. Fifty-two patients without children failed to return the questionnaire. This low rate of participation among the group of women without children may be due to their lack of interest in a study addressing a topic not directly relevant to them, i.e., pregnancy. An analysis of their files did not reveal any particular disease pattern or drug regimen that would distinguish them from the group who did return the questionnaires. The retrospective data were obtained from the patients' hospital records, and a questionnaire (see below). Basic disease-related information, such as the year of onset of JRA, type of onset, course (polyarticular, pauciarticular, or systemic), severity, treatment, extraarticular features, presence of rheumatoid factor or antinuclear antibodies, and surgery, was derived from the in- and outpatient files for each patient. Twenty-three of the patients had continued with regular outpatient visits through adulthood, and in 20, data on the course of JRA during or after pregnancy were found in the records. Whenever possible, patient records from primary care physicians and obstetric departments were also obtained. All patients were sent a questionnaire consisting of some structured and some open-ended questions and were asked to complete it and return it in a stamped, self-
Table 1. Clinical data o n 51 juvenile rheumatoid arthritis (JRA) patients with children and 45 JRA patients without children
Mean age at time of study (years) Mean age at disease onset (years) Onset type Pauciarticular Polyarticular Systemic No. rheumatoid factor positive Course Pauciarticular Polyarticular Mean disease duration (years) No. with joint surgery No. with hip prosthesis No. in remission
With children
Without children
30.2 8
29.8 8
24 24 3 9
27 2 9
14 37 18.9 30 8 29
16
4 41
19.4 33 17 19
addressed envelope that was provided. A reminder was sent to patients who did not respond. In addition, the 23 patients who continued with regular visits were personally interviewed to verify the answers given on the questionnaires. The questions to JRA patients with children addressed information on personal data, history of JRA, functional status ( 13), mobility, household activities, activities of daily living, employment, disease activity before, during, and after pregnancy, offspring, and personal views about motherhood and JRA. The questionnaire for JRA patients without children was identical, but excluded questions on maternity and included instead questions about the reasons for not having children. General disease characteristics of the 2 JRA groups are given in Table 1. Differences between the groups were analyzed by chi-square test with Yates' correction. Determination of disease activity. For comparison of disease activity before, during, and after pregnancy, several criteria were assessed. Active articular disease was defined as 2 or more of the following: morning stiffness >30 minutes, swelling of 1 or several joints, joint pain on motion, extraarticular features, and need for nonsteroidal antiinflammatory drug (NSAID) treatment. Active systemic JRA required temperature >39"C, JRA rash, or serositis. Remission was defined as complete absence of the above-mentioned symptoms for at least 6 months. Minor symptoms included occasional episodes of joint pain and stiffness, but without joint swelling.
RESULTS T h e 5 1 JRA patients had given birth to a total of 76 children. W h e n grouped according to type of disease onset, the 24 women with polyarticular onset had borne 44 children, the 24 patients with pauciarticular onset had borne 25 children, a n d the 3 patients with systemic onset had borne 7 children. Twenty-seven patients had I child, 20 had 2 children, and 3 had 3 children.
883
PREGNANCY AND JRA Table 2. Activity of juvenile rheumatoid arthritis at conception and during 50 pregnancies
Activity at conception (no. of pregnancies) In remission (20) Minor symptoms (20) Active (10)
Activity during pregnancy Better
No change
Worse
0 10
19 8
7
1
1 2 2
Effect of pregnancy on the course of JRA. Long periods of remission are common in JRA, especially in early adulthood (14). Therefore, the attempt was made to correlate disease activity during pregnancy with disease activity prior to conception. Reliable data about pre-pregnancy disease activity could be obtained for 50 pregnancies, from in- and outpatient records or from the files of primary care physicians. As shown in Table 2, pregnancy did not reactivate the symptoms of quiescent JRA, which was present in 12 patients with pauciarticular onset, 7 with polyarticular onset, and 1 with systemic onset. An episode of acute arthritis in 1 or several joints occurred during the first trimester in 15 pregnancies. However, most patients with minor symptoms at the start of pregnancy and the majority of those with active inflammation experienced improvement or total remission in the second half of gestation. An analysis of the possible influence of onset type on disease activity during pregnancy revealed the following: Among patients with polyarticular-onset JRA, improvement was noted during 26 Pregnancies (59%), there was no change during 15 pregnancies (34%), and exacerbation occurred during 3 pregnancies (7%). JRA patients with pauciarticular onset had improvement during 9 pregnancies (36%), no change during 14 pregnancies (56%), and aggravation of disease during 2 pregnancies (8%). The latter occurred in 2 HLA-B27 positive patients who developed clinical and radiologic signs of ankylosing spondylitis (AS) during adulthood. Disease activity improved during 2 pregnancies, was unchanged during 4 pregnancies, and worsened during 1 pregnancy in patients with systemic-onset JRA. Among all 76 pregnancies taken together., improvement was noted during 35, there was no change in disease activity reported during 34, and signs and symptoms worsened during 7. However, as indicated in Table 2, 20 of the patients were known to have had inactive disease at the time of conception. None of the patients reported serious extraarticular disease related
to JRA during pregnancy. However, 4 of the 6 patients with chronic uveitis experienced active eye disease during pregnancy. Multiple pregnancies. Of the 23 women with multiple pregnancies, the course of JRA was the same during both or all pregnancies in 17, whereas 6 had diEering courses during the different pregnancies. Improvement was noted in 13, no change in 9, and disease exacerbation in 1 of the subsequent pregnancies. Pregnancies that ended in abortion. Among the JRA patients with children, 9 had abortions, either spontaneous (5 cases) or elective (4 cases), during the first trimester (week 8-16). The course of JRA before and after 7 first-trimester abortions in 5 JRA patients without children was also reported. There was no change in disease activity during 14 of these early pregnancies, and improvement was reported in 2. After the termination of pregnancy, disease exacerbation occurred in 4 cases, whereas there was no change in disease activity in 10. There was no particular pattern of disease found in the 8 patients who had spontaneous abortion. None of them had used cytotoxic drugs prior to or during the pregnancy. Drug treatment before and during pregnancy. Data on drug treatment at the onset of pregnancy was available for 20 of the pregnancies. This treatment included use of NSAIDs (19 cases), prednisolone (6 cases), hydroxychloroquine (5 cases), gold (1 case), and penicillamine (1 case). In 4 additional pregnancies, treatment with azathioprine (1 case) or methotrexate (3 cases) had been stopped shortly before conception. Immunosuppressive treatment was withdrawn during the first trimester of pregnancy in all patients. Eight patients continued taking NSAIDs during pregnancy, 4 took low doses of prednisolone, 2 received intraarticular corticosteroids, and 2 took analgesics occasionally. Outcome of pregnancy. The outcome of the 76 pregnancies is shown in Table 3. Seventy-five of the pregnancies resulted in the delivery of live, normal children of normal birth weight, and there was 1 stillbirth. Seventy babies were born at full term; 5 were born a little early (15-23 days before term) but were not considered premature. Only 1 child was considered premature, with a birth weight of 1,800 gm. This infant was delivered by cesarean section at week 33 due to very active disease in the mother and fetal distress detected by ultrasound. The mother had been receiving low-dose pulse methotrexate until 3 weeks before conception, and prednisolone treatment had been continued throughout pregnancy. Twenty children were delivered by cesarean
@STENSEN
884 Table 3. Outcome of 76 pregnancies in 51 women with juvenile rheumatoid arthritis (JRA)
Live birthslstillbirths Male childrenlfemale children Full-terdprematurelearly but not considered premature Mean birth weight (gm) Type of delivery Normal Cesarean section Other* Reason for cesarean section Contracted pelvis Hip prosthesis Non-JRA-related
* Five forceps deliveries; 2 vacuum deliveries.
7511 30146 701115 3.760 49 20
7 7 8
5 -
section, and in 15 cases, this was related to sequelae of JRA, including contracted pelvis or hip prosthesis. No complications due to anesthesia were reported. At the time of this study, the median age of the 75 live children was 5 years (range 0.6-24 years). All children, including the 1 born prematurely and the 14 whose mothers had received drug treatment during pregnancy, had normal mental and physical development. Activity of JRA after pregnancy. Investigation of the activity of JRA within 1 year after delivery revealed the occurrence of a postpartum flare of the disease after 45 of 50 pregnancies reviewed. The exacerbation or reappearance of disease symptoms was most often encountered 3-6 months after delivery. In general, symptoms eased or disappeared again within 2 years after pregnancy. In none of the patients who had quiescent disease before and during pregnancy did the postpartum flare cause permanent reactivation of the JRA. Of the 23 patients with multiple pregnancies, 9 noted a more protracted period of reactivation after the second or third pregnancy than after the first, and in 6 cases this started a permanent recrudescence of the JRA. Decision to become pregnant and severity of JRA. To evaluate whether, among patients with JRA, those with less severe disease were more likely to choose to have children, 45 age-matched female JRA patients were recruited from the 78 without children. They had answered the same questionnaire, but excluding questions about maternity. While the 2 groups were comparable with respect to age at onset of JRA, disease duration, and frequency of remission, there were differences between the 2 groups in some respects: JRA patients without children more often had polyarticular disease at onset and during followup, more of them had undergone joint surgery and, most
importantly, twice as many had had hip prosthesis (Table 1). A comparison of drug treatment given during the course of the JRA revealed that patients without children were more likely to have received immunosuppressive or cytotoxic drugs (Table 4); these included azathioprine, methotrexate, cyclophosphamide, and chlorambucil. Alkylating drugs had been given to 6 patients without children, whereas only 1 JRA patient with offspring had received cyclophosphamide previously. Functional ability also differed in the 2 groups. Forty-six patients in the group with children, versus 37 of those without children, judged their functional ability as good or fairly good, with no or only occasional need for assistance in daily activities. Six patients in the group without children, versus 2 in the group with children, were severely disabled. Moreover, 3 of the JRA patients without children had developed amyloidosis, 2 of them requiring kidney transplantation. Employment. Of the JRA patients with children, 4 were still in school, 21 had full-time or part-time employment, and 26 were homemakers. Of those without children, 31 were employed full-time or parttime, 9 were in school, and 5 were unable to work. Family planning. All patients were asked to give their opinion about the advisability of pregnancy in a woman with a history of JRA, and the number of children she should have. Of the 51 women with children, 13 recommended having 1 child, 21 recommended having 2 children, 2 recommended having more than 2 children, and 15 recommended having children but were uncertain about the number of children desirable. Thirty patients had been or were currently using oral contraceptives. Eleven of the patients without children advised against pregnancy, 7
Table 4. Drug treatment in 51 juvenile rheumatoid arthritis (JRA) patients with children and 45 JRA patients without children*
Treatment NSAIDs NSAIDs + DMARDs NSAIDs + DMARDs + steroids NSAIDs + DMARDs + steroids + cytotoxic agents
With children
Without children
7 15 22 7
7 10 13 15
* Values are the number of patients who received any of the treatments any time during the course of the disease (i.e., for patients with children, the numbers include treatment before, during, or after pregnancy). NSAIDs = nonsteroidal antiinflammatory drugs; DMARDs = disease-modifying antirheumatic drugs.
885
PREGNANCY AND JRA
Table 5. Activity ofjuvenile rheumatoid arthritis (JRA) at conception and during 1 year postpartum Activity at conception (no. of pregnancies)
Course postpartum
No change
Worse
9 10
11 10
4
6
~
In remission (20) Minor symptoms (20) Active (10)
recommended having 1 child, 23 recommended having 2 children, and 4 recommended having more than 3 children. Reasons given for not having children included functional impairment (13 patients) and fear of heredity of JRA (12 patients). Infertility was reported in 2 patients, and 2 had undergone voluntary sterilization. Thirty-seven of the 45 women without children reported having a normal menstrual cycle. Twenty-six used or had used contraceptives, mainly oral contraceptives. Heredity. Two of the children born to the 51 JRA patients had rheumatic disease. In both cases, the mothers had polyarticular-onset JRA, 1 associated with chronic uveitis and the other with psoriasis. Both were married to men with HLA-B27 positive AS. The 2 affected daughters were both HLA-B27 positive. One, like her mother, had polyarticular-onset JRA with chronic uveitis. The other had radiologically verified AS.
DISCUSSION This is the first retrospective study of a large series of JRA patients with regard to the mutual effects of pregnancy and the disease. The analysis of 76 pregnancies occurring in 51 JRA patients provided 2 major conclusions: First, pregnancy per se had no adverse effect on the signs and symptoms of JRA. Second, fetal and maternal outcome of pregnancy was good. Inactive disease did not flare during gestation, though a small proportion of patients experienced a short episode of active arthritis during the first trimester. This finding parallels observations made in other seronegative arthritides, e.g., AS and psoriatic arthritis (PA) (15,16). Active polyarticular or pauciarticular JRA was improved by pregnancy, particularly during the second half of gestation. However, the activity of chronic anterior uveitis present in 4 cases was not influenced by pregnancy. Pregnancy ending in firsttrimester abortion had no major influence on the activity of JRA. As in AS and PA (12), the disease
course during one pregnancy did not automatically predict the course during subsequent pregnancies. However, in the majority of cases, JRA activity did not worsen during subsequent pregnancies. The beneficial effect of pregnancy occurred mainly during gestation, whereas the postpartum period carried the risk of a transient aggravation or recrudescence of the disease. An arthritis flare was most often experienced during the first 6 months after delivery, similar to the findings reported in rheumatoid arthritis ( l l ) , AS (17), and PA (12). The postpartum flare was independent of factors such as pre-pregnancy disease activity (Table 5 ) , type of onset, and course of JRA. In the majority of patients, this reactivation of JRA did not result in permanent exacerbation of the disease. However, a note of caution seems appropriate: Women with multiple pregnancies noted a tendency for a more protracted reactivation of disease after later pregnancies than after the first one. This implies that multiple pregnancy could cause a more permanent recrudescence of JRA. However, it is also possible that overall, pregnancy is associated with a better prognosis in JRA, especially in patients with transient monarticular pauciarticular disease that never requires care by a specialist. Selection bias toward more severe disease cannot be ruled out in a study population derived from hospital inpatients. The present data are reassuring with regard to fetal and maternal outcome of pregnancy in patients with JRA. A history of JRA and previous treatment in the mother did not imply increased frequency of gestational disease, fetal loss, or neonatal disease. However, possible long-term effects of previous drug treatment cannot be ruled out by this study since the average followup time in the offspring was only 5 years. Other than congenital malformations, possible effects on the child resulting from immunosuppressive treatment of the mother have rarely been addressed in the literature. A followup of 7 children exposed to gold compounds in utero showed normal development in all (18). In 1 study, no major disease was found in the offspring of survivors of treated childhood cancer (19); however, the average time of observation was only 4 years. With regard to the type of delivery, JRA was associated with a higher frequency of cesarean section, with bilateral hip prosthesis as the main reason. Obstetricians were apparently concerned, perhaps unduly so, about possible parturitional stress on the prosthesis. Because hip involvement in JRA frequently results in hip prosthesis (20), surgeons should
@STENSEN
886 counsel their adolescent female patients about possible consequences and precautions. It might also be useful to inform JRA patients with involvement of the cervical spine, the temporomandibular joint, and the cricoarytenoid joint about possible hazards of general anesthesia (21,22). If cesarean section is proposed, regional anesthesia may be the preferable method (23). In any case, emergency anesthesia should be avoided if possible. The results of this study indicate that a history of JRA influences the decision for or against having a child. Willingness to have offspring seemed more frequent among women with intermittent disease activity or long-lasting remission, minor functional impairment, and absence of potential hazardous medical treatment. In contrast, the decision not to have children appeared to be related to the presence of polyarticular, nonremitting disease. Polyarticular JRA is known to be frequently associated with persistent active disease, serious complications such as amyloidosis, and the need for aggressive therapy (24). Moreover, spontaneous remission occurs less frequently in the polyarticular than the pauciarticular subtype (25). The findings in the present study were consistent with other reports of a remission rate of approximately 60% in JRA (3). Patients without offspring more frequently used immunosuppressive and alkylating agents. Factors such as fear of long-term effects of cytotoxic agents, fear of heredity of JRA, and sometimes severe functional impairment definitely influenced their decision against pregnancy. This was reflected by the frequent use of contraceptives and even sterilization. Moreover, 24% of these patients recommended not having children at all. Severe polyarticular disease with aggressive drug treatment and functional impairment thus has long-term lifestyle consequences for the female patient. With regard to the advisability of pregnancy for women with a history of JRA, the data from this retrospective study indicate the following: JRA patients who consider pregnancy can be reassured that, as a rule, pregnancy will not aggravate or reactivate their disease, nor will their disease harm the fetus. Complications at delivery are to be expected when cephalopelvic dystocia or bilateral hip involvement is present. Special anesthetic techniques may be necessary if cesarean section is required. Of note, the risk for a postpartum flare is present even in patients with inactive disease. Patients themselves advise having no more than 2 children, and indeed, multiple pregnancy
may carry the risk of more permanent reactivation of JRA. Heredity is not directly associated with JRA. However, there appears to be a risk of inherited disease in children whose mothers have seronegative JRA, and whose fathers have an HLA-linked rheumatic disorder such as AS.
REFERENCES 1. Hanson V, Kornreich HK, Bernstein B, King KK, Singsen BH: Three subtypes of juvenile rheumatoid arthritis: correlations of age at onset, sex, and serologic factors. Arthritis Rheum 20 (suppl 2):184-186, 1977 2. JRA Criteria Subcommittee of the Diagnostic and Therapeutic Criteria Committee of the American Rheumatism Association Section of the Arthritis Foundation: Current proposed revision of JRA criteria. Arthritis Rheum 20 (suppl 2): 195-199, 1977 3. Hull RG: Outcome in juvenile arthritis. Br J Rheumatol 27 (suppl 1):66-71, 1988 4. Cantor B, Tyler T, Nelson RM, Stein GH: Oligohydramnios and transient neonatal anuria. J Reprod Med 24: 220-223, 1980 5. Stein GH, Cantor B, Panush RS: Adult Still’s disease associated with pregnancy. Arthritis Rheum 23:248-250, 1980 6. Hodgkinson R: Anesthetic management of a parturient with severe juvenile rheumatoid arthritis. Anesth Analg 60:611412, 1981 7. Elkon KB, Hughes GRV, Bywaters EGL, Ryan PFJ, Inman RD, Bowley NB, James MP, Eady RAJ: Adultonset Still’s disease: twenty-year followup and further studies of patients with active disease. Arthritis Rheum 255474554, 1982 8. Green J, Kanter Y, Barzilai D: Adult Still’s disease associated with pregnancy. Isr J Med Sci 18:1037-1039, 1982 9. Bango MY, Garcia Paez JM, Solovera JJ, Merino MF, Giron Gonzalez JA: Adult-onset Still’s disease: a case with onset during pregnancy (letter). Arthritis Rheum 28:957, 1985 10. Unsworth J, d’Assis-Fonseca A, Beswick DT, Blake DR: Serum salicylate levels in a breast fed infant. Ann Rheum Dis 46:638-639, 1987 11. Katz WE, Starz TW, Winkelstein A: Recurrence of adult Still’s disease after pregnancy. J Rheumatol 17: 373-374, 1990 12. @stensen M, Husby G: Pregnancy and rheumatic disease: a review of recent studies in rheumatoid arthritis and ankylosing spondylitis. Klin Wochenschr 62:891895, 1984 13. Steinbrocker 0, Traeger CH, Batterman RC: Therapeutic criteria in rheumatoid arthritis. JAMA 140:659-662, 1949
PREGNANCY AND JRA
14. Calabro JJ, Burnstein SL, Staley HL, Marchesano .JM: Prognosis in juvenile rheumatoid arthritis: a fifteen-year followup of 100 patients. Arthritis Rheum 20 (suppl 2):285, 1977 15. @stensen M, Husby G: A prospective clinical study of the effect of pregnancy on rheumatoid arthritis and ankylosing spondylitis. Arthritis Rheum 26:1155-1 159, 1983 16. @stensen M: Pregnancy in psoriatic arthritis. Scand J Rheumatol 17:67-70, 1988 17. @stensen M, Romberg @, Husby G: Ankyloising spondylitis and motherhood. Arthritis Rheum 25: 140143, 1982 18. Tarp U, Graudal H: A followup study of children exposed to gold compounds in utero (letter). Arthritis Rheum 28:235-236, 1985 19. Li FP, Jaffe N: Progeny of childhood-cancer survivors. Lancet II:707-709, 1974
887
20. Harris CM, Baum J: Involvement of the hip in juvenile rheumatoid arthritis: a longitudinal study. J Bone Joint Surg 70A:821-833, 1988 21. Smith BL: Anesthesia and Still’s disease (letter). Anesthesia 40:209, 1985 22. Person DA: Juvenile rheumatoid arthritis: anesthetic and surgical considerations. AORN J 44:439-449, 1986 23. Spiechowicz J, Pahle JA: Clinical aspects of postoperative continuous regional analgesia in the child with JRA. Adv Pain Res 15:189-200, 1990 24. Jeremy R, Schaller J, Arkless R, Wedgwood RJ, Healey LA: Juvenile rheumatoid arthritis persisting into adulthood. Am J Med 45:419-434, 1968 25. Hanson V, Kornreich H , Bernstein B, King KK, Singsen B: Prognosis of juvenile rheumatoid arthritis. Arthritis Rheum 20 (suppl2):279-284, 1977
