Risk Factors And A Predictive Model For Pregnancy-related Myasthenia Gravis Worsening

Objective: The main objective was to investigate the risk factors for pregnancy-related myasthenia gravis(MG)relapse during pregnancy and delivery/abortion,to develop a Nomogram model to predict the probability of relapse,and further guide women of childbearing age with MG to plan their pregnancies rationally.Methods: This study included 113 patients with pregnancy-related MG registered from January 2015 to October 2021 at Huashan Hospital,and Chifeng Clinical Medical College,all with preconception onset.Baseline data such as demographic and clinical characteristics were extracted from the medical records,designed into a uniform questionnaire,and retrospectively analyzed.The MG patients were divided into the MG relapse group and non-relapse group according to the changes in MG-specific activities of daily living(MG-ADL)scores during one year of pregnancy and delivery/abortion.All MG patients were pre-pregnancy to reach a stable state of MG-ADL≦2.52 patients in the relapse group had worsening MG symptoms during pregnancy and delivery /abortion,and had a 1-point increase in ADL;61 patients in the non-relapse group had stable or improving MG symptoms during pregnancy and delivery/abortion,and had a 1-point decrease in ADL.The pre-pregnancy and 1-year period of pregnancy and delivery/abortion were divided into seven stages: Before pregnancy(BP),0-3 months of pregnancy(DP1),4-6 months of pregnancy(DP2)and 7-9 months of pregnancy(DP3)and 0-3 months after delivery/abortion(PP1),4-6 months after delivery/abortion(PP2)and 7-12 months after delivery/abortion(PP3+4).1.The number of attacks in the relapse group during DP1,DP2,DP3,PP1,PP2,and PP3+4 was recorded.2.The mean MG-ADL and MG QOL-15 scores of DP1,DP2,DP3,PP1,PP2,PP3+4 and BP within the relapse group were compared and analyzed for differences.3.The demographic and clinical characteristics of the two groups were compared,and risk factors that may lead to relapse of the condition during pregnancy and delivery/abortion were analyzed.Statistical analysis of data was performed by SPSS26.0 software,and study indicators with P<0.05 were included in logistic regression for multifactorial analysis and screening of risk factors affecting relapse of MG condition.A Nomogram predictive relapse probability model was drawn using R software(version 4.2.0)to visualize the predicted relapse probability during pregnancy-related MG and its associated risk factors,the level of model prediction was tested by the consistency index(C-index)and the area under the subject’s operating characteristic curve(ROC),Statistical significance was set at p<0.05.Results: A total of 113 pregnancy-related MG patients were collected in this study.Among them,52(46%)patients in the relapse group deteriorated during pregnancy and delivery/abortion,and 61(54%)patients in the non-relapse group,43(38%)of whom were stable and 18(16%)improved during pregnancy and delivery/abortion.1.In the relapse group,there were 58 episodes in each pregnancy stage,20 in DP1(34.5%),3 in DP2(5.2%),2 in DP3(3.4%),28 in PP1(48.3%),2 in PP2(3.4%),and 3 in PP3+4(5.2%).2.In the relapse group,compared with BP,the average scores of MG-ADL and MG-QOL15 in DP1,and PP1 stages were significantly increased(P<0.05).3.The age of onset of the patients in this study was 15 to 35(26.08±2.39)years,there was no statistical difference between the two groups in the age of first onset,MG type,MGFA typing,MG antibody type,thyroid abnormality,the relapse of myasthenia gravis crisis before pregnancy,glucose abnormality,blood pressure abnormality,birth outcome,and treatment during the before 6 months of pregnancy(P>0.05).Variables that were significant on the univariate analysis included: age at delivery/abortion(20-35years)(P=0.021),time to stabilization before pregnancy(P=0.001),treatment during pregnancy and delivery/abortion(P=0.001),combined thymic hyperplasia before pregnancy(P=0.001),combined thymoma(P=0.046),and thymectomy(P=0.024)Statistically between the two groups differ.Results of multivariate analysis: variables that were significant in univariate analysis were subjected to multivariate logistic analysis,and the differences between the two groups were age at delivery/abortion(20-35 years)(OR=4.746,P=0.008),duration of stable disease before pregnancy(<2 years)(OR=4.359,P=0.005),and inadequate treatment during pregnancy and delivery/abortion(OR=3.363,P=0.042),thymic hyperplasia(OR=3.258,P=0.014),and no thymectomy before pregnancy(OR=7.912,P=0.007).Conclusion:1.MG relapse is more common in women at 0-3 months of pregnancy and 0-3 months after delivery/abortion.2.Exploration of risk factors for relapse during pregnancy and delivery/abortion in patients with pregnancy-related MG:(1)age at delivery/abortion is 20-35 years,and the disease is likely to recur during pregnancy and delivery/abortion;(2)the disease is stable for less than two years before pregnancy,and the disease is likely to recur during pregnancy and delivery/abortion;(3)inadequate treatment during pregnancy and delivery/abortion,and the disease is likely to recur during pregnancy and delivery/abortion;(4)Thymic hyperplasia may increase the risk of MG relapse during pregnancy and delivery/abortion;(5)MG patients with thymic hyperplasia/thymoma and without thymectomy before pregnancy may relapse during pregnancy and delivery/abortion.3.Nomogram prediction model development:(1)In clinical application,the higher the score is used to predict the probability of relapse of pregnancy-related MG patients during pregnancy and delivery/abortion,the higher the total score,the higher the risk of relapse;(2)The prediction model has a certain predictive ability,which can provide a certain reference for the reasonable pregnancy of MG women.