Article Text


Concise report
Predictors of preterm birth in patients with mild systemic lupus erythematosus
  1. Megan E B Clowse1,
  2. Daniel J Wallace2,
  3. Michael Weisman2,
  4. Andra James3,
  5. Lisa G Criscione-Schreiber1,
  6. David S Pisetsky1
  1. 1Department of Medicine, Duke University Medical Center, Durham, North Carolina, USA
  2. 2Department of Medicine, Cedar Sinai Hospital, Los Angeles, California, USA
  3. 3Department of Obstetrics and Gynecology, University of Virginia, Charlottesville, Virginia, USA
  1. Corresponding to Dr Megan E B Clowse, Department of Medicine, Duke University Medical Center, Box 3535 Trent Drive, Durham, NC 27710, USA; megan.clowse{at}


Objective While increased disease activity is the best predictor of preterm birth in women with systemic lupus erythematosus (SLE), even women with low disease activity are at increased risk of this complication. Biomarkers that would identify at-risk pregnancies could allow interventions to prevent preterm birth.

Method Measures of SLE activity, inflammation, placental health and renal function between 20 and 28 weeks gestation (mid-gestation) were correlated to preterm birth and gestational age at delivery in a prospective cohort of pregnant women with SLE.

Result Of the 40 pregnancies in 39 women, all with mild–moderate SLE disease, 9 (23.7%) of the 38 live births were delivered preterm. Low C4 was the only marker of SLE activity associated with younger gestational age at delivery. Elevated ferritin and lower oestradiol correlated with younger gestational age at delivery. Renal function remained normal during all pregnancies at mid-gestation and did not correlate with preterm birth. Higher serum uric acid, however, correlated with younger gestational age at delivery.

Conclusions In women with SLE with mild–moderate disease activity, ferritin, oestradiol and uric acid levels at mid-gestation may predict preterm birth. These markers may prove to be clinically useful in identifying pregnancies at particularly high risk for adverse outcomes.

  • Systemic Lupus Erythematosus
  • Disease Activity
  • Autoimmune Diseases

Statistics from


Women with systemic lupus erythematosus (SLE) are at increased risk for preterm delivery, likely reflecting the immunological and vascular abnormalities associated with this disease. While SLE activity is the strongest predictor of preterm delivery, approximately a third of pregnancies in women with quiescent SLE will deliver preterm.1 ,2 Delivery even a few weeks prior to term is associated with delayed development, lung immaturity and poorer long-term outcomes for children.3

Preventing preterm delivery is a major challenge in obstetrics, in part because interventions must be initiated mid-gestation, well before overt symptoms are clinically evident. Therefore, predictors of preterm birth must be identified mid-gestation to benefit from potential therapies. Systemic inflammation and poor placental development are possible causes of preterm delivery. In the obstetrical literature, markers of these, including C reactive protein (CRP), ferritin, oestradiol and α-fetoprotein (AFP), have been identified as markers of preterm delivery. Uric acid may be a marker of preeclampsia. This study investigates whether markers of SLE activity, inflammation, placental health and renal function could predict preterm birth in women with SLE.

Patients and methods

Pregnant women with SLE according to1997 American College of Rheumatology (ACR) criteria were prospectively enrolled at two academic centres to correlate mid-gestation SLE activity, inflammation, placental and renal health with gestational age at delivery.4 Women were followed throughout pregnancy with a specific visit mid-gestation between pregnancy weeks 20 and 28. Disease activity and laboratory markers were measured at this visit and correlated with delivery data. This study was approved by the institutional review board of both hospitals; patients provided informed consent prior to enrolment.

SLE activity was measured using the SLE pregnancy disease activity index (SLEPDAI) and the physician's global assessment (PGA), both scored after ruling out pregnancy causes for possible signs and symptoms of SLE.5 Laboratory measures of SLE activity included the anti-dsDNA antibody titre (AtheNa multiplex beads), C3 and C4 (rate nephelmetry).

The erythrocyte sedimentation rate (ESR), CRP and ferritin were used to measure inflammation. Measures of placental health were AFP and oestradiol (both measured by immunoassay). Renal function was assessed using the serum creatine, urine spot protein to creatine ratio and serum uric acid.

Pregnancy outcomes

A delivery prior to 37 weeks gestation was defined as preterm. Preeclampsia was determined by the delivering obstetrician.

Statistical analysis

Pearson correlation analyses were used to examine the relationship between potential predictors with gestational age at delivery; a p value <0.05 was considered statistically significant. Correlations were only calculated using available data and do not include pregnancies with missing laboratory data. Data were analysed using STATA, V.10 (StataCorp, College Station, Texas, USA).


A total of 40 pregnancies in 39 women were included: 37 pregnancies through Duke University and three through Cedars Sinai Hospital. A total of 96 pregnancies were followed in these two centres between 2008 and 2011, but nine suffered a pregnancy loss before a mid-gestation visit, 38 did not have an office visit between 20 and 28 weeks gestation, and nine had a mid-pregnancy visit but were lost to follow-up. Two pregnancies resulted in a stillbirth at 21.4 and 24 weeks; neither patient had oestradiol, AFP or ferritin tested, but both had markers of SLE activity, renal function and inflammation measured mid-gestation. While none of the women had antiphospholipid syndrome, four had positive antiphospholipid antibodies. One of these ended with a stillbirth and the other three delivered at term.

Of the 38 live births, 9 (23.7%) delivered preterm. Preterm labour was spontaneous in three pregnancies and induced in six pregnancies for SLE activity (1), oligohydramnios (1), non-reassuring fetal heart tones (1) and preeclampsia.3 Preeclampsia was diagnosed in 7 (18.9%) pregnancies of which one also had HELLP syndrome (haemolysis, elevated liver enzymes, low platelets).

There were no differences in the maternal age or racial distribution of pregnancies that delivered at term and preterm (see table 1). The majority of patients took medication for SLE during pregnancy, most commonly hydroxychloroquine. Of these medications, only azathioprine exposure was associated with preterm delivery, likely because it was only used in women with a history of more significant SLE activity.

Table 1

Demographic information and medication use in pregnancies that resulted in term and preterm deliveries

SLE activity

The only marker of SLE activity that correlated with preterm delivery was C4 (see table 2). The level of SLE activity, however, was low to moderate in all pregnancies. The average SLEPDAI score was 3.2 and the highest noted at mid-gestation was 10 with 34.2% of patients having a SLEPDAI score of zero. The average PGA was 0.43 and the maximum was 1.5 (out of maximum of 3), denoting mild–moderate disease activity. The PGA was zero, indicating no SLE activity, in 43.2% of the cohort. C4 was low in 34.2%, C3 was low in 7.9% and dsDNA titre positive in 25.7% of pregnancies at mid-gestation.

Table 2

Markers of preterm birth at mid-gestation

Markers of inflammation

Neither the CRP nor the ESR correlated with pregnancy outcomes or SLE activity. All of the pregnancies had an elevated ESR and 36.8% had a CRP above normal at mid-gestation. Elevated ferritin, however, correlated with younger gestational age at delivery (see figure 1).

Figure 1

When measured at mid-gestation (between 20 and 28 weeks gestation), oestradiol (A), ferritin (B) and uric acid (C) may predict a preterm delivery. The vertical line is located at 37 weeks gestation and delineates pregnancies delivered at term and preterm. (A) Low oestradiol (pg/ml) at mid-gestation correlated with earlier delivery (r=0.44, p=0.01). (B) Increased ferritin (ng/ml) at mid-gestation correlates with earlier delivery (r=−0.37, p=0.05). (C) Increased uric acid (mg/dl) at mid-gestation correlates with earlier delivery (r=−0.43, p=0.02).

Markers of placental health

Because oestradiol levels increase dramatically in pregnancy due to production by the placenta, a lower than expected oestradiol level may indicate placental dysfunction. In this cohort, a lower oestradiol level correlated with younger gestational age at delivery (r=0.44, p=0.01). AFP, also a marker of placental health, was not associated with preterm delivery in this study.

Measures of renal function and hypertension

Neither serum creatine nor spot urine protein to creatine ratio correlated with preterm birth or preeclampsia; but all except one woman had normal levels of these measures at mid-gestation. Higher serum uric acid levels were associated with younger gestational age at delivery, but did not significantly correlate with preeclampsia (r=0.26, p=0.19). The uric acid level at mid-gestation correlated well with creatine (r=0.54, p<0.01).


The results of this study help define the role of biomarkers in assessing the risk of preterm delivery in women with SLE with low disease activity. In this cohort of women with mild–moderate SLE, low oestradiol, high ferritin or uric acid levels at mid-gestation correlated with gestational age at delivery. In contrast to previous studies, mid-gestational SLE activity was not a clear marker of preterm birth, likely due to low SLE activity in this cohort.1 Preterm delivery was medically induced for increased SLE activity or preeclampsia in the majority of cases of early delivery. The preterm birth rate in this study is modestly lower than other recent SLE pregnancy cohorts, but is twice the preterm birth rate in healthy US pregnancies.2 ,6

Levels of oestrogens have been used for decades as a marker of placental health and correlate in other studies with gestational age at delivery and infant birth weight.7 ,8 These hormones are produced by the placenta and increase dramatically as pregnancy progresses. Women with SLE have been shown to have a lower oestradiol level during pregnancy, on average, compared with a healthy population.9 In the current cohort, serum oestradiol varied widely at mid-pregnancy and lower levels of this hormone were associated with earlier gestational age at delivery.

Normal physiological changes of pregnancy can hamper the measurement of markers of inflammation. Overproduction of proteins in the liver during pregnancy limits the reliability of ESR as a marker of inflammation.10 CRP, however, does not typically increase as dramatically during pregnancy.11 While CRP and ESR correlated well with each other in this study, neither measure correlated with SLE activity or pregnancy outcomes in this cohort. On the other hand, studies in healthy women suggest that an elevated serum ferritin level mid-gestation is associated with an increased risk for preterm delivery.12 ,13 In our study, ferritin was associated with earlier delivery, suggesting a role of inflammation in the preterm delivery in this population.

While lupus nephritis has been associated with preterm birth and preeclampsia in women with SLE, this relationship could not be evaluated in this study because only one woman had current lupus nephritis.2 Mild renal insufficiency may be overlooked during pregnancy, however, because increased renal blood flow during pregnancy lowers the serum creatine. Preeclampsia or renal insufficiency may increase serum uric acid in pregnancy. In this cohort, an increased level of uric acid at mid-gestation correlated with preterm delivery, but was not statistically associated with preeclampsia. In addition, uric acid correlated well with serum creatine and thus may serve as a marker of subclinical renal insufficiency.

We have identified low oestradiol, elevated ferritin and elevated uric acid levels at mid-gestation (between gestational weeks 20 and 28) as potential markers of subsequent preterm delivery in a cohort of women with mild to moderate SLE activity. It is worth noting that the women all received their care in the university setting, which may somewhat limit the generalisability of this study. Further confirmation in larger studies, followed by intervention studies, is required to assess the clinical utility of measuring oestradiol, ferritin and uric acid in SLE pregnancy.


The authors would like to thank Martin Tochacek, Loren Caldwell and Lindsay Carr for their assistance in data management.


View Abstract


  • Handling editor Tore K Kvien

  • Contributors All of the authors included in this manuscript have been participants in this study and assisted in patient recruitment, study design and/or data interpretation. All authors have reviewed and made comments on the drafts involved in the development of this manuscript and have approval of the final version.

  • Funding This work was funded by the Arthritis Foundation through an Arthritis Investigator Award.

  • Competing interests None.

  • Ethics approval Institutional Review Boards for both Duke University and CedarsSinai Hospital.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement Further information about pregnancy outcomes are included in the complete database from this project. Collaborative work on these data would be considered by the authors.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.