Abstract

Objective. MMF is teratogenic and needs to be replaced before pregnancy. This change may lead to flares. Our aim was to determine the risk of renal flares in women with LN who switched treatment from MMF to AZA before conception and to evaluate the outcome of their pregnancies.

Methods. Medical records of women with LN counselled for pregnancy wish were reviewed. Women receiving treatment with either MMF or AZA (control group), with inactive lupus (SLEDAI ≤ 4) and quiescent LN (serum creatinine <1.5 mg/dl, inactive sediment and proteinuria <1 g/24 h for the preceding 6 months) were eligible for this study.

Results. We identified 54 women [23 treated with MMF (group 1) and 31 treated with AZA (group 2)]. MMF dosage was tapered and subsequently transferred to AZA, which was maintained throughout pregnancy. Three (13%) patients (group 1) vs none (group 2) developed a renal flare 3–6 months after transitioning from MMF to AZA (P = 0.14) before pregnancy ensued. The only parameter with a significant difference in those with flare compared with those without was younger age (median 27 vs 30 years; P = 0.03). Risk for adverse outcome within 48 pregnancies (pre-eclampsia 9%, preterm delivery 20.5%) increased with every milligramme of prednisone dosage [odds ratio (OR) 2.03] and every single unit of SLEDAI score (OR 3.92). Renal flares occurred post-partum in two women. No patient developed worsening of renal function.

Conclusion. Replacing MMF with AZA in patients with quiescent LN for pregnancy planning rarely leads to renal flares. Pregnancy outcome was favourable.

Introduction

LN is one of the most serious manifestations of SLE and an important factor in mortality and morbidity. Advances in immunosuppressive treatment options resulted in significant improvement in both preserved renal function and patient survival. Once successfully treated, long-lasting remissions of LN are common. Consequently a growing number of women with LN wish to fulfil their desire to have a child.

Pregnancy may increase SLE activity, potentially leading to impaired renal function up to end-stage renal disease. Guidelines for optimal counselling and clinical management of patients with LN and pregnancy wish are limited. A number of studies have shown that both maternal and foetal outcomes of lupus pregnancies are better if conception is delayed until the disease has been inactive for at least 6 months [1]. A common concern is the optimal choice of immunosuppressive treatment during the preconception period in the case of planned pregnancy. In this period, medications with teratogenic properties must be stopped, while stable low disease activity needs to be preserved. Currently the two most common choices for maintenance therapy in LN are MMF and AZA [2]. There is still controversy on the superiority of MMF over AZA as maintenance therapy. Findings of the ALMS showed significant superiority of MMF to AZA after successful induction therapy (lower flares in MMF vs AZA, 13% vs 23%) [3]. In contrast, the MAINTAIN nephritis trial demonstrated equivalence (similar flare rates in MMF vs AZA, 19% vs 25%) [4]. Up to now there are no data to guide the duration of treatment beyond 3 years, so continuing treatment for longer time periods must be individualized [2]. As AZA can be used safely during pregnancy [5], the current recommendation is transition from MMF to AZA at least 6 weeks before conception [2]. However, there are few data supporting this recommendation. In particular, any change in medication can induce flares in stable disease. Moreover, the above-mentioned superiority of MMF would be expected to further increase the risk of flare if exchanged for AZA. It is also not clear that such flares may not evolve during pregnancy.

Accordingly, our purpose was twofold: to evaluate the risk of renal relapse in patients with stable LN who switched from MMF to AZA for planned pregnancy and to study the frequency and outcome of subsequent pregnancies. The results were compared with patients with stable LN who continued AZA treatment during the preconception period.

Patients and methods

At the Department of Rheumatology, Heinrich-Heine-University, patients with inflammatory rheumatic diseases and a pregnancy wish are counselled prior to conception and followed during pregnancy and the post-partum period according to a standard protocol [6]. The medical records of women who asked for pregnancy counselling between January 2000 and December 2010 were reviewed. Inclusion criteria were (i) diagnosis of SLE based on the 1997 ACR criteria [7], (ii) renal involvement assessed by a renal biopsy, (iii) maintenance treatment with either MMF or AZA, (iv) stable disease defined as SLEDAI ≤ 4 [8] and (v) renal remission defined as stable serum creatinine <1.5 mg/dl, proteinuria <1 g/24 h and absence of active sediment (<5 red blood cells/hpf, ≤5 white blood cells/hpf, no cellular casts) for at least 6 months. The study was approved by the local ethics committee at Heinrich-Heine-University Duesseldorf, and all subjects consented to the use of their records for research.

At the pre-pregnancy counselling visit (baseline), data on medical history were recorded, including renal and extra-renal symptoms of SLE, treatment at the onset of disease and maintenance therapy. Histological features of LN were classified according to the 1995 World Health Organization classification [9]. Clinical and laboratory assessment included ANA, anti-dsDNA antibodies, aCL antibodies, lupus anticoagulant, immunoreactive C3 and C4, serum creatinine, 24-h proteinuria and urinary sediment microscopy. During the preconception period, disease activity, laboratory data and medication record were collected every 3–6 months. From the time the pregnancy was detected, the patients were evaluated every 4–8 weeks throughout their pregnancy until 6 months after delivery.

Disease activity before conception was assessed by SLEDAI [8]. Disease activity during pregnancy was determined by the SLE-Pregnancy Disease Activity Index (SLEPDAI), which takes into account the physiological changes of pregnancy [10].

Patients receiving MMF were informed about the possible teratogenic risk and transitioning to AZA prior to conception was recommended. The MMF dose was tapered in steps of 500 mg/day every 4 weeks to 500 mg/day and transferred to AZA (2 mg/kg) if there were no signs of renal or extra-renal flare at this time point. Women were advised to conceive at the earliest after another 3 months to ensure stabilization. Treatment with AZA was maintained throughout the preconception period and pregnancy.

The use of angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin receptor II blockers (ARBs) is probably associated with an increase in congenital malformations [11]. Therefore patients who were treated with ACEIs/ARBs for blood pressure control were switched to alternative antihypertensives regarded as safe in pregnancy. In patients with mild proteinuria controlled by ACEIs/ARBs, these medications were stopped while patients tried to conceive and blood pressure was monitored closely.

APS was defined according to the Sapporo criteria [12], because the study period had started before more recent guidelines were published [13]. Women with obstetric APS received low-dose aspirin from preconception and additional low-molecular-weight heparin (LMWH) from a positive pregnancy test. Women with thrombotic APS treated with low-dose aspirin before pregnancy continued this medication throughout pregnancy and post-partum. In addition, these patients started with LMWH as soon as they had a positive pregnancy test. Women on warfarin before conception were switched to adjusted doses of LMWH (plus aspirin 100 mg daily) as soon as pregnancy was confirmed.

In addition, regardless of their aPL status, all women who asked for pregnancy counselling after 2007 received low-dose aspirin during pregnancy, as at this time point a meta-analysis showed significant reductions in the frequency of pre-eclampsia among women at risk [14].

Definitions of the terms used in this study

Renal flare: worsening of proteinuria (increase by >1000 mg/24 h, during pregnancy in the absence of preeclampsia) associated with urinary sediment turning active from inactive or an increase in serum creatinine of >30%.

Hypertension: systolic blood pressure >140 mm Hg and/or diastolic blood pressure >90 mm Hg in a sitting position in three consecutive measurements, or use of anti-hypertensive drugs.

Pre-eclampsia: in women without baseline hypertension and proteinuria <300 mg/day, pre-eclampsia is defined as new-onset hypertension and proteinuria >300 mg/day after 20 weeks of gestation. In women without hypertension and proteinuria >300 mg/day at baseline, superimposed pre-eclampsia is defined as new-onset hypertension and doubling 24-h urinary proteins or urine/creatinine ratio after 20 weeks of gestation. In women with hypertension and proteinuria >300 mg/day at baseline, diagnosis of superimposed pre-eclampsia requires both worsening hypertension (increase of systolic or diastolic blood pressure of 30 mm Hg or greater and 15 mm Hg or greater, respectively, above baseline values) and doubling proteinuria. Diagnosis of pre-eclampsia was taken into consideration only in the presence of inactive urinary sediment (<5 red blood cells/hpf).

Other obstetric definitions were early miscarriage (spontaneous termination of pregnancy prior to 10 weeks gestation), foetal death (death of a foetus demonstrated to be alive at or beyond 10 weeks gestation), premature birth (termination of pregnancy with a live birth of < 37 weeks’ gestation), full-term birth (termination of pregnancy with a live birth between 38 and 40 weeks) and small for gestational age (birth weight below the 10th centile for the stated gestation).

Statistical analysis

All tests were performed by the statistical software R version 2.12 (R Foundation for Statistical Computing). Descriptive statistics for the patients were obtained and are reported as median, mean (s.d.), range, frequency or proportion as appropriate. Differences among study groups were analysed by tests of equal proportions, Student’s t-tests and Pearson’s χ2 test of independence. The difference in renal flares between the two groups was predefined as the main statistical outcome. Adverse pregnancy outcomes were considered secondary. Odds ratios (ORs) were calculated after performing logistic regression. Results were considered statistically significant when two-sided P-values were <0.05.

Results

Of 58 patients with a SLEDAI ≤4 at counselling, 4 patients had to be excluded because of record irregularities or absent follow-up. Thus 54 patients were analysed, of whom 23 received MMF (group 1) and 31 received AZA (group 2) at the time of counselling (Fig. 1). The median duration of MMF treatment was 28 (range 16–96) months. Nine (39%) patients received MMF as primary induction treatment and 14 (61%) as maintenance regimen following treatment with CYC (mean cumulative dose 10.4 ± 3.1 g). Patients’ baseline characteristics are summarized in Table 1.

Fig. 1

Treatment regimen.

Fig. 1

Treatment regimen.

Table 1

Characteristics of all SLE patients at baseline

 All patients Group 1 (previous MMF treatment) Group 2 (continuous AZA treatment) P value 
No. of women 54 23 31  
Age (years), median (range) 30 (22–37) 28 (24–37) 31 (22–35) 0.55 
Duration of SLE (years), median (range) 8 (3–15) 8 (3–15) 8 (3–12) 0.11 
Duration of renal disease (years), median (range) 5.5 (2.5–15) 6 (3–15) 5 (2.5–12) 0.05 
SLEDAI, mean (s.d.2.3 (1.2) 2.6 (1.1) 2.1 (1.3) 0.16 
Hypertension, n (%) 13 (24) 6 (26) 7 (23) 1.00 
Serum creatinine (mg/dl), mean (s.d.0.7 (0.1) 0.8 (0.2) 0.7 (0.1) 0.13 
24-hour urinary protein (mg/day), mean (s.d.212 (324) 262 (360) 174 (295) 0.34 
C3 (mg/dl) (90–180), mean (s.d.95 (21) 95 (19) 95 (22) 0.97 
Hypocomplementaemia, n (%) 19 (35.2) 8 (34.7) 11 (35.4) 0.81 
anti-dsDNA positive, n (%) 47 (87) 21 (91.3) 26 (83.8) 0.69 
WHO class nephritis, n (%)     
    Class III 9 (17) 3 (13) 6 (19)  
    Class IV 36 (67) 17 (74) 19 (62) 1.00 
    Class V 9 (17) 3 (13) 6 (19)  
Prednisone dosage (mg/day), median (range) 5 (0–10) 5 (0–7.5) 3 (0–10) 0.73 
HCQ, n (%) 31 (58.5) 13 (56.5) 18 (60) 1.00 
ACEI or ARB, n (%) 20 (37.5) 11 (47.8) 9 (29.0) 0.26 
 All patients Group 1 (previous MMF treatment) Group 2 (continuous AZA treatment) P value 
No. of women 54 23 31  
Age (years), median (range) 30 (22–37) 28 (24–37) 31 (22–35) 0.55 
Duration of SLE (years), median (range) 8 (3–15) 8 (3–15) 8 (3–12) 0.11 
Duration of renal disease (years), median (range) 5.5 (2.5–15) 6 (3–15) 5 (2.5–12) 0.05 
SLEDAI, mean (s.d.2.3 (1.2) 2.6 (1.1) 2.1 (1.3) 0.16 
Hypertension, n (%) 13 (24) 6 (26) 7 (23) 1.00 
Serum creatinine (mg/dl), mean (s.d.0.7 (0.1) 0.8 (0.2) 0.7 (0.1) 0.13 
24-hour urinary protein (mg/day), mean (s.d.212 (324) 262 (360) 174 (295) 0.34 
C3 (mg/dl) (90–180), mean (s.d.95 (21) 95 (19) 95 (22) 0.97 
Hypocomplementaemia, n (%) 19 (35.2) 8 (34.7) 11 (35.4) 0.81 
anti-dsDNA positive, n (%) 47 (87) 21 (91.3) 26 (83.8) 0.69 
WHO class nephritis, n (%)     
    Class III 9 (17) 3 (13) 6 (19)  
    Class IV 36 (67) 17 (74) 19 (62) 1.00 
    Class V 9 (17) 3 (13) 6 (19)  
Prednisone dosage (mg/day), median (range) 5 (0–10) 5 (0–7.5) 3 (0–10) 0.73 
HCQ, n (%) 31 (58.5) 13 (56.5) 18 (60) 1.00 
ACEI or ARB, n (%) 20 (37.5) 11 (47.8) 9 (29.0) 0.26 

WHO: World Health Organization.

Follow-up after pregnancy counselling visit and before conception

Two women in whom MMF was substituted by AZA developed mild leukopenia; no other adverse side effects of AZA were observed. Three additional women (13%) in group 1 experienced a renal flare after a median time of 4 (range 3–6) months after switching treatment and prior to conception. The only parameter with a significant difference in those with flare compared with those without was younger age [median 27 (renal flare) vs 30 years (no renal flare); P = 0.03). There was a trend towards shorter duration of treatment with MMF in patients with renal flare (29 ± 6 months vs 38 ± 23 months; P = 0.17). Only one of the three women with renal flare had stopped ACEI treatment 3 months before flare occurrence, while the other two patients had not received ACEIs or ARBs when presenting for pregnancy counselling. We were unable to detect additional predictors concerning the occurrence of flares, probably due to the low number of renal flares. All patients were treated again with MMF, leading to complete remission in two cases. The third patient developed increasing proteinuria and was treated successfully with pulse CYC, followed by MMF, during further follow-up. All three patients have remained on MMF since then (for 6, 8 and 14 months, respectively). No renal flare was documented prior to conception within group 2, but one 34-year-old woman developed an extra-renal flare with arthritis and severe pleuritis after three in vitro fertilizations. She received prednisone (initial dosage 25 mg daily) and HCQ. She has not yet conceived. Since one could challenge our definition of a renal flare, and in view of possible pure membranous (Class V) LN, in particular, we have also looked at proteinuria alone. In fact, none of the patients not having a renal flare as defined above reached a proteinuria of ≥1 g/day.

Outcome of pregnancies

Forty-eight (89%) women conceived during follow-up, 18 of the 20 women successfully switched from MMF to AZA (group 1) and 30 of the 31 women on stable AZA (group 2). The baseline characteristics of patients with pregnancies are summarized in Table 2.

Table 2

Characteristics of patients with pregnancies at baseline

 All patients Group 1 (previous MMF treatment) Group 2 (continuous AZA treatment) P value 
No. of women 48 18 30  
Age (years), median (range) 30 (22–35) 28.5 (24–32) 30.5 (22–35) 0.37 
Duration of SLE (years), median (range) 8 (3–15) 8.5 (3–15) 8 (3–12) 0.04 
Duration of renal disease (years), median (range) 6 (2.5–15) 8 (3–15) 5 (2.5–12) 0.03 
Nulliparity, n (%) 34 (71) 17 (94) 17 (57) 0.24 
APS, n (%)     
    Obstetric, late pregnancy loss 3 (6.3) 1 (5.5) 2 (6.6) 1.00 
    Thrombotic 2 (4.1) 1 (5.5) 1 (3.3) 0.81 
Hypertension, n (%) 11 (22.9) 4 (22.2) 7 (23.3) 1.00 
Autoantibodies, n (%)     
    Anti-Ro/SSA 19 (39.6) 7 (38.9) 12 (40) 1.00 
    aPL 6 (12.5) 3 (16.7) 3 (10) 0.82 
Serum creatinine (mg/dl), mean (s.d.0.7 (0.1) 0.7 (0.0) 0.7 (0.1) 0.67 
24-hour urinary protein (mg/day), mean (s.d.224 (338) 301 (389) 176 (300) 0.25 
Hypocomplementaemia, n (%) 19 (39.5) 8 (44) 11 (37) 0.81 
anti-dsDNA positive, n (%) 42 (87.5) 17 (94.4) 25 (83.3) 0.10 
SLEDAI, mean (s.d.2.4 (1.3) 2.7 (1.0) 2.1 (1.4) 0.07 
Prednisone use, n (%) 31 (65) 13 (72) 18 (60) 0.58 
    Dosage in users (mg/dl), median (range) 5 (3–10) 5 (0–7.5) 3 (0–10) 0.30 
Medication during pregnancy, n (%)     
    Aspirin 39 (81) 15 (83) 24 (80) 1.00 
    LMWH 4 (8.3) 2 (11.1) 2 (6.7) 1.00 
    HCQ 30 (62.5) 12 (67) 18 (60) 0.87 
 All patients Group 1 (previous MMF treatment) Group 2 (continuous AZA treatment) P value 
No. of women 48 18 30  
Age (years), median (range) 30 (22–35) 28.5 (24–32) 30.5 (22–35) 0.37 
Duration of SLE (years), median (range) 8 (3–15) 8.5 (3–15) 8 (3–12) 0.04 
Duration of renal disease (years), median (range) 6 (2.5–15) 8 (3–15) 5 (2.5–12) 0.03 
Nulliparity, n (%) 34 (71) 17 (94) 17 (57) 0.24 
APS, n (%)     
    Obstetric, late pregnancy loss 3 (6.3) 1 (5.5) 2 (6.6) 1.00 
    Thrombotic 2 (4.1) 1 (5.5) 1 (3.3) 0.81 
Hypertension, n (%) 11 (22.9) 4 (22.2) 7 (23.3) 1.00 
Autoantibodies, n (%)     
    Anti-Ro/SSA 19 (39.6) 7 (38.9) 12 (40) 1.00 
    aPL 6 (12.5) 3 (16.7) 3 (10) 0.82 
Serum creatinine (mg/dl), mean (s.d.0.7 (0.1) 0.7 (0.0) 0.7 (0.1) 0.67 
24-hour urinary protein (mg/day), mean (s.d.224 (338) 301 (389) 176 (300) 0.25 
Hypocomplementaemia, n (%) 19 (39.5) 8 (44) 11 (37) 0.81 
anti-dsDNA positive, n (%) 42 (87.5) 17 (94.4) 25 (83.3) 0.10 
SLEDAI, mean (s.d.2.4 (1.3) 2.7 (1.0) 2.1 (1.4) 0.07 
Prednisone use, n (%) 31 (65) 13 (72) 18 (60) 0.58 
    Dosage in users (mg/dl), median (range) 5 (3–10) 5 (0–7.5) 3 (0–10) 0.30 
Medication during pregnancy, n (%)     
    Aspirin 39 (81) 15 (83) 24 (80) 1.00 
    LMWH 4 (8.3) 2 (11.1) 2 (6.7) 1.00 
    HCQ 30 (62.5) 12 (67) 18 (60) 0.87 

Forty-four pregnancies (91.6%) resulted in live birth (Table 3). The median duration of the successful pregnancies was 38 (range 28–41) weeks. Using logistic regression analysis, independent explanatory variables associated with combined adverse pregnancy outcome (miscarriage, pre-eclampsia and/or preterm delivery) included higher disease activity as reflected by a higher SLEDAI score, higher prednisone dosage and older age at preconception visit. More precisely, the risk of adverse pregnancy outcome increased with every milligramme of prednisone dosage at baseline [OR 2.02 (95% CI 1.19, 3.44)] and every single unit of SLEDAI score [OR 3.92 (95% CI 1.18, 13)]. In addition, the risk of adverse pregnancy outcome increased with every year of age [OR 1.31 (95% CI 1.02, 1.68)].

Table 3

Outcome of pregnancies

 All patients Group 1 Group 2 P value 
No. of women 48 18 30  
SLEPDAI (first trimester), mean (s.d.), range 0.6 (1.0), 0–4 0.2 (0.4), 0–1 0.8 (1.1), 0–4 0.05 
SLEPDAI (second trimester), mean (s.d.), range 0.3 (0.6), 0–2 0.2 (0.4), 0–1 0.4 (0.7), 0–2 0.17 
SLEPDAI (third trimester), mean (s.d.), range 0.4 (0.8), 0–4 0.1 (0.3), 0–1 0.5 (0.8), 0–4 0.05 
Increase of steroid dosage during pregnancy, n (%) 13 (27) 2 (11) 11 (36) 0.11 
Live birth, n (%) 44 (92) 17 (94) 27 (90) 1.00 
First trimester miscarriage, n (%) 4 (8.3) 1 (5.5) 3 (10) 0.06 
Preterm delivery, n (%) 9 (20.5) 3 (17.6) 6 (22.2) 1.00 
Pre-eclampsia, n (%) 4 (9.1) 4 (14.8) 0.26 
Caesarean section, n (%) 22 (50) 9 (52.9) 13 (48.1) 1.00 
Birth weight (mg), median (range) 3195 (1200–4700) 3100 (1200–4590) 3310 (1950–4700) 0.14 
Small for gestation age, n (%) 6 (13.6) 3 (17.6) 3 (11.1) 0.87 
 All patients Group 1 Group 2 P value 
No. of women 48 18 30  
SLEPDAI (first trimester), mean (s.d.), range 0.6 (1.0), 0–4 0.2 (0.4), 0–1 0.8 (1.1), 0–4 0.05 
SLEPDAI (second trimester), mean (s.d.), range 0.3 (0.6), 0–2 0.2 (0.4), 0–1 0.4 (0.7), 0–2 0.17 
SLEPDAI (third trimester), mean (s.d.), range 0.4 (0.8), 0–4 0.1 (0.3), 0–1 0.5 (0.8), 0–4 0.05 
Increase of steroid dosage during pregnancy, n (%) 13 (27) 2 (11) 11 (36) 0.11 
Live birth, n (%) 44 (92) 17 (94) 27 (90) 1.00 
First trimester miscarriage, n (%) 4 (8.3) 1 (5.5) 3 (10) 0.06 
Preterm delivery, n (%) 9 (20.5) 3 (17.6) 6 (22.2) 1.00 
Pre-eclampsia, n (%) 4 (9.1) 4 (14.8) 0.26 
Caesarean section, n (%) 22 (50) 9 (52.9) 13 (48.1) 1.00 
Birth weight (mg), median (range) 3195 (1200–4700) 3100 (1200–4590) 3310 (1950–4700) 0.14 
Small for gestation age, n (%) 6 (13.6) 3 (17.6) 3 (11.1) 0.87 

Flares/disease outcome

Overall, SLE activity during pregnancy remained low (Table 3). Mild clinical worsening consisting of cutaneous or joint symptoms was observed in 13 (27%) pregnancies [11% (group 1) vs 36% (group 2), P = 0.11]. While no renal flares were diagnosed during pregnancy, one woman in each group [5.5% (group 1) vs 3.3% (group 2)] experienced renal flare (2 and 4 months after delivery, respectively). Both responded to an increase in steroid dose, and renal symptoms recovered completely. No patient developed worsening of renal function during a mean follow-up period of 31 (± 27) months post-partum. Since no cases of pre-eclampsia developed in patients switched from MMF to AZA (Fig. 1), the combination of flare and pre-eclampsia could only be suspected for the long-term AZA group. None of the four women thus affected had any additional indication of concomitant renal flares, and all four showed markedly reduced proteinuria within 6 weeks after giving birth.

Discussion

Our purpose was to study the risk of renal flares and adverse pregnancy outcomes in women with LN in remission who switched from MMF to AZA and to compare the results to women with LN without pre-conceptional change of immunosuppressive treatment. While the switch from MMF to AZA is necessary in preparation for pregnancy given the teratogenic potential of MMF, there have been concerns that such a change may lead to SLE flares. In addition, MMF probably is somewhat more efficacious compared with AZA, potentially increasing the risk. However, until today no study has evaluated the risk for renal flares in women with LN who switch from MMF to AZA for pregnancy planning.

Within our study population, transition to AZA was tolerated remarkably well. Only three women experienced an increase in renal activity prior to pregnancy and had to be switched back to MMF. These flares all occurred within 6 months. None of the 18 patients who later conceived developed renal relapse during pregnancy. Thus the risk of renal flare seems to be highest during the first months after switching. However, although the difference in renal flares between both groups was smaller than expected, and not statistically significant, we only observed flares in women who switched from MMF to AZA. While the number of patients clearly limits firm conclusions, we would expect this difference to become statistically significant in a larger sample, reflective of some risk involved. Probably the low risk in our cohort can be attributed to the combined fact that all patients had to be in renal remission and that the majority of our patients received MMF longer than 2 years. Within a retrospective analysis determining the safety of tapering of MMF in 44 patients with stable LN, 56% of patients reducing MMF developed a renal flare compared with 23% of patients with no drug tapering. Patients in whom MMF was reduced before 18 months were more prone to relapsing, while a reduction after 18 months was as safe as the continuation of full-dose treatment [15]. We saw a trend towards shorter duration of treatment with MMF in patients with renal flare (29 ± 6 months vs 38 ± 23 months; P = 0.17). One could also argue for stopping MMF instead of changing to AZA in a population of SLE patients beyond (or even close to) 3 years of MMF treatment. However, despite a lack of objective arguments against such an approach, we have always maintained active immunosuppression when patients prepared for pregnancy in order to prevent possible flares.

No women developed a renal flare during pregnancy and only two (4.5%) experienced renal relapse in the postpartum period. The available evidence shows that quiescence of renal disease is the best predictor of favourable maternal and foetal outcome [16]. Our renal flare rate in the context of pregnancy is lower than most rates reported in women with stable LN (10–15%) and obviously much lower than rates reported in pregnancies in women with active LN (up to 48%) [16–20]. Within a meta-analysis of 37 studies, including 1842 women with LN with a wide range of baseline renal function, 16% of pregnancies were complicated by renal disease activation [16]. On the other hand, a flare rate similar to ours was described within a retrospective analysis of 51 pregnancies, in which renal flares occurred in 5% of women with stable renal disease [21]. Our results can be explained by the fact that (i) all our patients had no signs of renal activity at conception, (ii) all our patients continued maintenance immunosuppressive treatment with AZA compared with <50% in most other studies including women with stable LN and (iii) a high percentage of 60% of our patients received antimalarials during pregnancy (compared with 20–50% in earlier studies) [19–23]. For example, within a multicentre study, patients with pre-existing LN who were not in complete remission and without any specific therapy showed a high rate of flare (66%) during and after pregnancy [19].

Pregnancy of patients with LN can be complicated by a number of obstetric and neonatal problems. Besides an increased rate of pregnancy losses (up to 40%), the most frequent complications observed are premature birth (13–53%) and pre-eclampsia (9–35%) [16, 17, 19, 23, 24]. Higher rates of foetal loss are associated with impaired renal function, whereas pre-eclampsia is mostly reported in women diagnosed with unplanned pregnancies or with active disease at conception [18, 24, 25]. In contrast, a number of investigations observed no significant differences in pregnancy outcomes between women with quiescent LN at conception and those with SLE but without renal involvement [17, 18, 20, 26]. Our study adds to the literature regarding pregnancy outcome with predominantly quiescent LN and preserved renal function. Our live birth rate (91.6%) is consistent with other studies including women with inactive LN [18, 19, 21]. However, the rates of pre-eclampsia (9%) and preterm deliveries (20.5%) observed in our study cohort are even lower than the rates from recent research. For example, a retrospective analysis of 43 pregnancies in women with previous LN described a high number of successful pregnancies (live birth rate 100%) [21]. Despite this, pregnancies were complicated with pre-eclampsia (28%) and prematurity (30%). Factors associated with these complications were older maternal age (>35 years), impaired renal function, higher values of proteinuria and high blood pressure.

Overall, women with a combination of clinical and serologic SLE activity have been shown to be at highest risk for pregnancy loss and preterm delivery [27]. High disease activity (reflected by a SLEDAI score of ≥4) 6 months before conception predicts adverse maternal outcomes (mainly pre-eclampsia) [28]. Interestingly, although we only included patients with stable disease (SLEDAI ≤ 4), we recognized an association between higher disease activity at baseline and adverse pregnancy outcome. In our study cohort, higher SLEDAI scores were mainly due to positive anti-dsDNA and low complement and not due to clinically active disease. Hypocomplementaemia itself is a significant predictor of unsuccessful pregnancies in SLE patients with and without LN [19, 27]. Accordingly, the optimal basis for successful pregnancy outcome would be a SLEDAI score <4 with inclusion of normal complement in the preceding months before conception. However, not all patients will achieve normal complement levels, and patients with C4 deficiency will evidently not normalize their C4 levels.

Finally, our study cohort included only 17% of patients with persistent aPL positivity, which is associated with prematurity in patients with LN [16]. Nevertheless, >80% received low-dose aspirin during pregnancy. We had adopted aspirin use because of a reduced frequency of pre-eclampsia (and associated prematurity) shown for women at high risk [14]. In addition, low-dose aspirin has been specifically associated with favourable pregnancy outcome in women with LN [20]. Accordingly, the recently published combined European League Against Rheumatism and European Renal Association/European Dialysis and Transplant Association recommendations include the use of aspirin for all women with LN considering pregnancy [2].

The strength of our study is the prospective follow-up after pregnancy counselling, which enabled us to use disease activity indices. Limitations of our research include the relatively small number of patients, which prevented identification of subtle associations with pregnancy complications, and the fact that all the women are Caucasian. Larger studies involving women from a variety of ethnic and racial groups need to be conducted to further validate these findings. With state-of-the-art care, even SLE patients considered at high risk for pregnancy complications, such as patients with renal involvement, can have successful pregnancies without deteriorating the underlying disease.

graphic

Acknowledgements

We thank Martina Kümmel (Rheumazentrum Rhein-Ruhr) for revising the manuscript.

Disclosure statement: The authors have declared no conflicts of interest.

References

1
Ruiz-Irastorza
G
Khamashta
MA
Lupus and pregnancy: integrating clues from the bench and bedside
Eur J Clin Invest
 , 
2011
, vol. 
41
 (pg. 
672
-
8
)
2
Bertsias
GK
Tektonidou
M
Amoura
Z
, et al.  . 
Joint European League Against Rheumatism and European Renal Association-European Dialysis and Transplant Association (EULAR/ERA-EDTA) recommendations for the management of adult and paediatric lupus nephritis
Ann Rheum Dis
 , 
2012
, vol. 
71
 (pg. 
1171
-
82
)
3
Dooley
MA
Jayne
D
Ginzler
EM
, et al.  . 
Mycophenolate versus azathioprine as maintenance therapy for lupus nephritis
N Engl J Med
 , 
2011
, vol. 
365
 (pg. 
1886
-
95
)
4
Houssiau
F
D’Cruz
D
Sangle
S
, et al.  . 
Azathioprine versus mycophenolate mofetil for long-term immunosuppression in lupus nephritis: results from the MAINTAIN nephritis trial
Ann Rheum Dis
 , 
2010
, vol. 
69
 (pg. 
2083
-
9
)
5
Østensen
M
Khamashta
M
Lockshin
M
, et al.  . 
Anti-inflammatory and immunosuppressive drugs and reproduction
Arthritis Res Ther
 , 
2006
, vol. 
8
 (pg. 
209
-
28
)
6
Fischer-Betz
R
Specker
C
Brinks
R
, et al.  . 
Pregnancy outcome in patients with antiphospholipid syndrome after cerebral ischaemic events: an observational study
Lupus
 , 
2012
, vol. 
21
 (pg. 
1183
-
9
)
7
Hochberg
MC
Updating the American College of Rheumatology revised criteria for the classification of systemic lupus erythematosus
Arhritis Rheum
 , 
1997
, vol. 
40
 pg. 
1725
 
8
Bombardier
C
Gladman
DD
Urowitz
MB
, et al.  . 
Derivation of the SLEDAI. A disease activity index for lupus patients. The Committee on Prognosis Studies in SLE
Arthritis Rheum
 , 
1992
, vol. 
35
 (pg. 
630
-
40
)
9
Churg
J
Bernstein
J
Glassock
RJ
Renal disease
1995
Tokyo
Ikagu-Shoin
(pg. 
151
-
79
)
10
Buyon
JP
Kalunian
KC
Ramsey-Goldman
R
, et al.  . 
Assessing disease activity in SLE patients during pregnancy
Lupus
 , 
1999
, vol. 
8
 (pg. 
677
-
84
)
11
Cooper
WO
Hernandez-Diaz
S
Arbogast
PG
, et al.  . 
Major congenital malformations after first-trimester exposure to ACE inhibitors
N Engl J Med
 , 
2006
, vol. 
354
 (pg. 
2443
-
51
)
12
Wilson
WA
Gharavi
AE
Koike
T
, et al.  . 
International consensus statement on preliminary classification criteria for definite antiphospholipid syndrome: report of an international workshop
Arthritis Rheum
 , 
1999
, vol. 
42
 (pg. 
1309
-
11
)
13
Miyakis
S
Lockshin
MD
Atsumi
T
, et al.  . 
International consensus statement on an update of the classification criteria for definite antiphospholipid syndrome (APS)
J Thromb Haemost
 , 
2006
, vol. 
4
 (pg. 
295
-
306
)
14
Askie
LM
Duley
L
Henderson-Smart
DJ
, et al.  . 
Antiplatelet agents for prevention of preeclampsia: a meta-analysis of individual patient data
Lancet
 , 
2007
, vol. 
369
 (pg. 
1791
-
8
)
15
Laskari
K
Tzioufas
AG
Antoniou
A
, et al.  . 
Longterm followup after tapering mycophenolate mofetil during maintenance treatment for proliferative lupus nephritis
J Rheumatol
 , 
2011
, vol. 
38
 (pg. 
1304
-
8
)
16
Smyth
A
Oliveira
GH
Lahr
BD
, et al.  . 
A systematic review and meta-analysis of pregnancy outcomes in patients with systemic lupus erythematosus and lupus nephritis
Clin J Am Soc Nephrol
 , 
2010
, vol. 
5
 (pg. 
2060
-
8
)
17
Rahman
FZ
Rahman
J
Al-Suleiman
SA
, et al.  . 
Pregnancy outcome in lupus nephropathy
Arch Gynecol Obstet
 , 
2005
, vol. 
271
 (pg. 
222
-
6
)
18
Wagner
SJ
Craici
I
Reed
D
, et al.  . 
Maternal and foetal outcomes in pregnant patients with active lupus nephritis
Lupus
 , 
2009
, vol. 
18
 (pg. 
342
-
7
)
19
Imbasciati
E
Tincani
A
Gregorini
G
, et al.  . 
Pregnancy in women with pre-existing lupus nephritis: predictors of foetal and maternal outcome
Nephrol Dial Transplant
 , 
2009
, vol. 
24
 (pg. 
519
-
25
)
20
Gladman
DD
Tandon
A
Ibañez
D
, et al.  . 
The effect of lupus nephritis on pregnancy outcome and foetal and maternal complications
J Rheumatol
 , 
2010
, vol. 
37
 (pg. 
754
-
8
)
21
Bramham
K
Hunt
BJ
Bewley
S
, et al.  . 
Pregnancy outcomes in systemic lupus erythematosus with and without previous nephritis
J Rheumatol
 , 
2011
, vol. 
38
 (pg. 
1906
-
13
)
22
Moroni
G
Quaglini
S
Banfi
G
, et al.  . 
Pregnancy in lupus nephritis
Am J Kidney Dis
 , 
2002
, vol. 
40
 (pg. 
713
-
20
)
23
Carmona
F
Font
J
Moga
I
, et al.  . 
Class III–IV proliferative lupus nephritis and pregnancy: a study of 42 cases
Am J Reprod Immunol
 , 
2005
, vol. 
53
 (pg. 
182
-
8
)
24
Moroni
G
Ponticelli
C
Pregnancy after lupus nephritis
Lupus
 , 
2005
, vol. 
14
 (pg. 
89
-
94
)
25
Huong
DL
Wechsler
B
Vauthier-Brouzes
D
, et al.  . 
Pregnancy in past or present lupus nephritis: a study of 32 pregnancies from a single centre
Ann Rheum Dis
 , 
2001
, vol. 
60
 (pg. 
599
-
604
)
26
Julkunen
H
Kaaja
R
Palosuo
T
, et al.  . 
Pregnancy in lupus nephropathy
Acta Obstet Gynecol Scand
 , 
1993
, vol. 
72
 (pg. 
258
-
63
)
27
Clowse
ME
Magder
LS
Petri
M
The clinical utility of measuring complement and anti-dsDNA antibodies during pregnancy in patients with systemic lupus erythematosus
J Rheumatol
 , 
2011
, vol. 
38
 (pg. 
1012
-
6
)
28
Kwok
LW
Tam
LS
Zhu
T
, et al.  . 
Predictors of maternal and foetal outcomes in pregnancies of patients with systemic lupus erythematosus
Lupus
 , 
2011
, vol. 
20
 (pg. 
829
-
36
)

Comments

0 Comments