Lupus Low Disease Activity State Achievement Is Important for Reducing Adverse Outcomes in Pregnant Patients With Systemic Lupus Erythematosus

DISCUSSION

Over the past 40 years, improvements in disease care and regular prenatal monitoring have decreased adverse pregnancy outcomes¹⁸. Most pregnancies end in successful livebirths; nonetheless, women with SLE have more than twice the cumulative rate of pregnancy complications than the general population^19,20,21. In this study, there was a noticeable difference in total pregnancy-related complications between patients with SLE and controls (55.2% vs 36.1%). Our findings were considerably higher than the rates reported in most previous studies (22.0–38.7%)^22,23. One possible reason for the higher number of unsuccessful pregnancies in our data is the length of the study, in that some participants may not have received specialized close monitoring. However, the livebirth rate of patients with SLE was 72.4%, similar to rates in other studies in Korea²⁴. Several Korean studies have reported livebirth rates ranging between 75% and 82%, which are comparable to those reported in Western countries^24,25,26. Differences in the incidence of comorbidities and rates of cesarean section were not observed between women with and without SLE. These results are probably due to the need for referral of mothers without SLE having medical comorbidities to a tertiary medical center to reduce postpartum complications and mortality.

The results of our study confirm that pregnancy-related complications are significant problems in patients with SLE compared to controls. Similar to previous studies, pregnancies in SLE show high rates of miscarriage, stillbirth, IUGR, NICU admission, preeclampsia, emergency cesarean section, and preterm delivery^11,27. In a subgroup analysis, we observed 2.9% cases of preeclampsia, which is similar to that of previous studies involving Japan and other cohorts in Korea^26,28. However, we noticed significantly lower rates of preeclampsia in our cohorts compared to Western countries. Several studies conducted in the US and Europe reported that the preeclampsia rates range from 12% to 35%^29,30. In terms of fetal complications, we found a high rate of stillbirths (7.4% vs 0.5%), preterm deliveries (16.4% vs 7.4%), IUGR (9.8% vs 4.0%), low birthweight infants (16.0% vs 6.4%), and NICU admission (17.2% vs 9.7%) in patients with SLE; this was consistent with the results of previous studies³¹. However, there was no statistical difference in the incidence of miscarriages among women with SLE and controls (20.2% vs 15.2%, P = 0.13). These rates were similar considering that the miscarriage rate of pregnant Korean women is 14.0–23.0%³².

We further analyzed the increased rates of adverse pregnancy outcomes after adjusting for important confounding factors. The OR for all complications proved to be higher in patients with SLE than in controls. The higher OR in our analysis was stillbirth (adjusted OR 13.2, P < 0.001) and preeclampsia (adjusted OR 4.3, P = 0.025). A previous metaanalysis by Bundhun, et al supported tendencies similar to the present study³³. However, there have been no published studies comparing the frequency of pregnancy complications between SLE and the general obstetric population in Korea.

As a result of investigating pregnancy-related complications, factors affecting complications were mostly consistent with other previous analyses conducted over the years^34,35. One important difference is that a positive test for aPL was not an independent risk factor for adverse pregnancy outcomes³⁶. This might be explained by the widespread consensus on the benefit of antithrombotic prophylaxis³⁷, as we adequately treated with ASA or low molecular weight heparin for the presence of aPL. We also found that some variables were only associated with maternal complications; these included a high cumulative CS dose during pregnancy and a history of CYC treatment. Among the various immunosuppressants, the only medication that demonstrated a meaningful relationship with adverse maternal outcomes was CYC. Despite possible teratogenic agents being discontinued in preparation for conception, ovarian failure and sustained amenorrhea might persist owing to a cumulative effect of CYC. As previously reported, the incidence of infertility by CYC was approximately 14.9–37.3%³⁸; thus, it is recommended to limit the use of CYC in fertile women. For patients planning to continue CYC therapy, hormonal co-therapy during treatment is suggested to reduce the risk of ovarian insufficiency³⁹.

The reduction of adverse pregnancy outcomes by achieving LLDAS has not been studied previously. Even after validation in a multivariate model, the achievement of LLDAS was the only factor that affected both maternal and fetal complications. An agreement on the principles of the definition of remission in SLE has been proposed by several research groups over the years, which includes SLEDAI, and mostly coincided with SLEDAI ≤ 4^40,41. In our analysis, to identify the SLEDAI that leads to better outcomes, we obtained similar results described in previous studies. However, many experts argue that the drawback of SLEDAI is that it does not reflect important systems such as hemolytic anemia or gastrointestinal activity, and treatment domains¹⁴. Indeed, currently there is a growing opinion that LLDAS has excellent validity and represents a much more attainable clinical target state⁴². In this regard, it is noteworthy that our study is the first study, to our knowledge, to analyze the relationship between LLDAS and pregnancy outcomes. Moreover, we investigated factors affecting LLDAS, and the results were largely consistent with factors affecting pregnancy outcomes (Supplementary Table 1, available with the online version of this article). When we compared the pregnancy-related complications between patients with SLE who achieved LLDAS and controls, there was no significant difference in the incidence of pregnancy-related complications other than stillbirth (Supplementary Table 2, available with the online version of this article). These results suggest that achieving LLDAS is important to preventing pregnancy-related complications in patients with SLE.

The areas with high disagreement on the validity of LLDAS are concerned with CS dose. The opinion that CS-related side effects generally occur at daily doses of ≥ 7.5 mg prednisone-equivalent is predominant, and this dosage is also used in the LLDAS criteria⁴³. Contrary to the prevailing view, several studies reported that the CS dose accommodates daily doses of 10 mg prednisone-equivalent because this dosing has the benefit of rapid onset of action⁴⁴. Further, CS are safe even during pregnancy; thus, this is the first-line therapy commonly used for acute flares throughout pregnancy. However, it is clear that CS are not innocuous because of the increased risk of congenital malformation and pregnancy outcomes⁴⁵. The controversy over the safe maintenance CS dose of SLE pregnancies remains; thus, our available data analyzed the adequate CS doses and timing for pregnancy. Our study revealed that a mean CS dose during pregnancy of < 6 mg/day prednisone-equivalent was a reasonable value for reducing maternal and fetal risk. Cumulative CS doses exceeding 3.5 g prednisone-equivalent for women with SLE have contributed to the higher rate of adverse maternal outcomes, whereas CS doses taken 3 months before conception were not influential. Even in obstetric medicine, it is crucial to not exceed such doses while maintaining quiescent SLE. There is a general consensus that safe immunosuppressants such as AZA and tacrolimus can be administered by reducing CS to manage SLE flares in pregnant women⁴⁶.

The present study is relevant for several reasons. One of the strengths of this analysis is that it is the first study conducted in Korea to compare obstetric complications in patients with SLE with the general population, to our knowledge. Second, it is the first study to reveal that achieving LLDAS plays an important role in minimizing adverse pregnancy outcomes. Third, it is meaningful that our study has tried to determine the optimal CS dose, considering that we are uncertain what CS dose is best for mothers and babies.

We encountered some limitations in our attempt. First, this is a retrospective study in a single center, with selection and information bias. Additionally, as the general population was recruited from a tertiary care center, we may have recruited relatively high-risk pregnancies. Further, our study encompasses a vast period from 1999 to 2019, and during this period many advances in diagnostic criteria and medication use have changed the outcomes of the disease.

In conclusion, this study demonstrated adverse pregnancy outcomes are still higher in pregnant women with SLE than in the general population. A comprehensive analysis of risk factors and predictors of maternal and fetal complications was provided. Minimizing the occurrence of complications during pregnancy was strongly associated with achieving LLDAS and keeping daily steroid doses of prednisone < 6 mg/day. Targeting inactive disease and preventing a disease flare is important to allow for successful outcomes. Therefore, pregnancies in SLE should receive preconception counseling with a multidisciplinary team in a specialized center and choose an optimal time for pregnancy.