MRI Before Pregnancy Predicts Early Postpartum MS Relapse

Erik Greb

April 19, 2021

For women with multiple sclerosis (MS), results of MRI performed during the year before they become pregnant appear to accurately predict early relapse after childbirth, new research shows.

Investigators found that women for whom there was evidence of activity on MRI during the year before pregnancy had more than a ninefold likelihood of postpartum relapse. The researchers were also able to predict which women would not experience postpartum relapse by disability and by activity on MRI before pregnancy.

"This is a new observation indicating the importance of prepartum clinical, as well as imaging, disease stability to reduce the risk of postpartum relapse," study investigator Adi Vaknin-Dembinsky, MD, PhD, professor of neuroimmunology at Hadassah Medical Center, Jerusalem, Israel, told Medscape Medical News.

"We confirmed the pattern previously observed that relapses are significantly less frequent during pregnancy but increased in the early postpartum period: a transient, pregnancy-induced remission, followed by a rebound in the postpartum period," she said.

The findings were published online March 30 in Multiple Sclerosis Journal.

A More Accurate Predictor?

During pregnancy, relapse rates can be approximately 70% lower than before pregnancy. However, relapse rates rebound during the postpartum period. The Pregnancy in MS (PRIMS) study showed that disease activity in the year before and during pregnancy, as well as higher levels of disability, were predictive of postpartum relapse.

Recent research has used MRI to detect postpartum MS activity. For the current study, the investigators examined whether MRI in the year before pregnancy was predictive of postpartum relapse.

The investigators retrospectively reviewed the files of women with MS who presented to the Hadassah Medical Center. Eligible participants completed a pregnancy between 2008 and 2018 and gave birth to live infants.

For each woman, the investigators collected data on delivery date, brain MRI during the year before and after pregnancy, follow-up visits with Expanded Disability Status Scale (EDSS) score evaluation, and data on relapses during the year before pregnancy, during pregnancy, and during the year after childbirth.

The analysis included 118 women, who completed 172 pregnancies. Women were included regardless of treatment status. For 83 women, treatment was stopped at least 6 months before pregnancy. For other patients, treatment was stopped at least 1 month before pregnancy. For no patients were medications switched.

In this study population, 81 women had one pregnancy, 22 had two, 13 had three, and two had four. The mean age at delivery was 31.8 years. The mean duration of disease at time of conception was 5.7 years, and the mean EDSS score before pregnancy was 1.3.

The mean annual relapse rate (ARR) during the year before pregnancy was 0.34. Approximately 90% of pregnancies were relapse free. The mean ARR decreased to 0.069 during the second trimester.

During follow-up, 44 women experienced 48 relapses during the first 3 months post partum. Twenty-three patients experienced 23 relapses at 3 to 6 months post partum, and 15 patients experienced 17 relapses at 6 to 9 months post partum. At the end of the first postpartum year, 90 patients remained relapse free.

In approximately 76% of pregnancies, women underwent brain MRI during the year before pregnancy and in the year after birth. Of these women, 36.6% showed activity on MRI before pregnancy. This increased to 48.9% in the postpartum period.

MRI activity during the year before pregnancy was significantly associated with MRI activity in the postpartum period (P = .011). The odds ratio (OR) of postpartum MRI activity for patients with MRI activity during the year before pregnancy was 8.331. MRI activity post partum (OR, 5.143) was more sensitive than clinical follow-up (OR, 2.912) for revealing disease activity.

Relapse during pregnancy also correlated with early postpartum relapse (OR, 3.792). Higher level of disability (as measured by EDSS) before pregnancy correlated with higher relapse rate during the first 3 postpartum months (P = .003). However, relapse during 1 or 2 years before pregnancy was not predictive of relapses during the first 3 postpartum months.

Using a multivariate model, the researchers predicted which women would not experience postpartum relapse using EDSS and disease activity before pregnancy (specificity, 96.7%; P < .001).

The researchers found that women in whom disease was active on MRI during the year before pregnancy were at significantly increased risk of having a postpartum relapse (OR, 9.375).

Unreported Relapses?

The fact that clinical relapses during pregnancy, but not relapses during 1 or 2 years before pregnancy, are associated with risk for postpartum relapse is an apparent discrepancy. Such a result may be due to the treating neurologist's not recording or being informed of minor relapses in this large cohort during the period before pregnancy, said Vaknin-Dembinsky. "This is one of the limitations of a retrospective study," she added.

In contrast to the prepregnancy period, during pregnancy and in the immediate postpartum period, women may be more aware that they should report each relapse to their physician.

The findings underscore the importance of a multidisciplinary team that includes a neurologist and neuroimaging and obstetric specialists before, during, and after pregnancy, said Vaknin-Dembinsky.

"We think it is important to understand and study the effect of clinical and imaging disease activity during the year before pregnancy, not only for the immediate postpartum time but also for MS disease outcome," she added.

The investigators plan to study the use of disease-modifying therapies (DMTs) during these periods and their influence on fetal and maternal health.

"We also have special interest in evaluating not only the effective treatment but also the exact timing in which they should be given in order to prevent postpartum relapses," she added.

New Guidance

Commenting on the findings for Medscape Medical News, Jeffrey A. Cohen, MD, director of experimental therapeutics at the Cleveland Clinic's Mellen Center for MS Treatment and Research, Cleveland, Ohio, said the study's findings may have clinical implications and may be helpful for planning disease management, specifically, when to reinstate DMT.

Clinical relapse in the year before pregnancy was not as predictive of postpartum relapse as MRI activity during the year before pregnancy. This finding probably indicates that clinical relapse is not as sensitive to disease activity as MRI is, said Cohen, who was not involved in the research.

The finding that higher EDSS scores before pregnancy predicted higher risk for postpartum relapse is somewhat counterintuitive, he said. EDSS scores were generally low in the study population, so higher EDSS scores may reflect more active disease.

"These findings provide some guidance for whom disease therapy should be reinitiated after pregnancy," said Cohen. "Specifically, women with active disease before pregnancy should consider early reinstitution of DMT. The main ramification is that breastfeeding, which has a number of benefits, may need to be abbreviated. We have very little data to guide which MS disease therapies are safe for use during breastfeeding."

He noted that the study's main weakness is its small sample size, which limited the statistical power of some of the analyses.

Also commenting on the findings, Riley Bove, MD, assistant professor of neurology at the University of California, San Francisco, Weill Institute for Neurosciences, said the high percentage of women who did not receive any treatment before pregnancy "presumably gives a better sense of the impact of pregnancy itself on MRI activity, rather than an effect of discontinuing treatments."

However, she added, it is not clear how the researchers defined or modeled breastfeeding, which could influence the findings. It is also unclear how in their models they accounted for patients who received fingolimod and natalizumab.

"These patients are usually treated as a separate category because of the risk of rebound relapses before and during and possibly even after a pregnancy," said Bove, who was not involved in the study. Likewise, the article does not include information on patients who received B-cell-depleting agents or other newer therapies.

"In some clinical settings, educating patients and clinicians about the safety of MRI, and specifically of contrast use, and breastfeeding in the postpartum period may be needed," Bove told Medscape Medical News. "If contrast is not used, a postpartum scan can still be compared to preconception scan, if this was indeed obtained for new lesions."

Future research should include studies with more patients receiving newer MS drugs and the examination of MRI scans collected prospectively at specific times to prevent sampling bias. The possible association between breastfeeding and MRI activity should also be examined. In addition, "tying the MRI activity to biomarkers will be key," said Bove.

The study had no external funding. Vaknin-Dembinsky has disclosed no relevant financial relationships. Cohen has received personal compensation for consulting for Adamas, Atara, Bristol-Myers Squibb, Convelo, MedDay, and Mylan. He also serves as an editor of Multiple Sclerosis Journal. Bove has received consulting and advisory board fees from Alexion, Biogen, EMD-Serono, Genzyme-Sanofi, Novartis, and Roche-Genentech. She has received research support from Biogen and Roche Genentech.

Mult Scler. Published online March 30, 2021. Abstract

For more Medscape Neurology news, join us on Facebook and Twitter.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.