No increased risk of adverse pregnancy outcomes for women with
myasthenia gravis: a nationwide population-based study
J.-C. Wen
a,b, T.-C. Liu
c,d, Y.-H. Chen
e, S.-F. Chen
b, H.-C. Lin
band W.-C. Tsai
faDepartment of Internal Medicine, PoJen General Hospital;bSchool of Health Care Administration, Taipei Medical University;cCenter for
Sleeping Research, Taipei Medical University Hospital;dDepartment of Public Finance, National Taipei University;eSchool of Public Health,
Taipei Medical University; andfDepartment of Obstetrics & Gynecology, PoJen General Hospital, Taipei, Taiwan
Keywords:
adverse pregnancy out-come, low birthweight, myasthenia gravis Received 2 March 2009 Accepted 16 April 2009
Background: This study aims to examine the risk of adverse pregnancy outcomes [low birthweight (LBW), preterm birth, cesarean sections (CS) and babies born small for gestational age (SGA)] in pregnant women with myasthenia gravis (MG), using a 3-year population-based database, taking characteristics of infant and mother into consideration.
Methods: This study used two nationwide population-based datasets: the Taiwan National Health Insurance Research Dataset and the Taiwan birth certificate registry. We identified 163 pregnant women with MG during 2001–2003 as the study cohort and 815 randomly selected pregnant women as a comparison cohort. Conditional logistic regression analyses were performed.
Results: The results showed that, although these patterns did not reach a statistically significant level, mothers with MG had higher percentages of LBW (6.8%, vs. 5.6%), SGA (17.8%, vs. 14.1%) and cesarean deliveries (44.8%, vs. 37.4%), except for pre-term births (8.1%, vs. 8.1%). After adjusting for highest maternal education level, marital status, family monthly income and infant gender and parity, the odds ratios (OR) of LBW, preterm birth, SGA infants, and cesarean delivery for mothers with MG were 1.19 (95% CI = 0.60–2.38), 1.00 (95% CI = 0.54–1.87), 1.30 (95% CI = 0.83–2.04), and 1.33 (95% CI = 0.94–1.88), respectively, as compared to unaffected mothers.
Conclusions: We conclude that there were no statistically significant differences in the risk of having preterm, LBW, SGA infants and cesarean deliveries between women with and without MG.
Introduction
Myasthenia gravis (MG) is a chronic autoimmune dis-order of neuromuscular transmission characterized by varying degrees of weakness and easy fatigability of the skeletal muscles. The disease is twice as common in females; it frequently affects young women of child-bearing age and is diagnosed in an estimated 1 in 20 000 pregnancies [1–3]. Pregnancy has a variable effect on disease relapse and the course of MG during pregnancy
is unpredictable; it is equally likely that MG will remain stable, improve, or worsen during pregnancy [2,4–6].
Despite long concern about the relationship between MG and pregnancy, the existing literature on MG and pregnancy outcomes and has yielded contradictory findings. For example, some studies found that women with MG had a higher prevalence of premature births and/or low birthweight (LBW) [3,6,7] compared to the general population, whereas an equal number of studies failed to prove such associations [8–10]. Furthermore, the findings regarding whether there was an increased risk of cesarean sections among women with MG was still controversial [8–10]. Since prior studies dealing with this topic have generally relied on a small number of cases from one hospital [3,6–9] or in population sub-groups [10], inconsistent results could be due to the use of selective data, limited sample sizes and inadequate control of confounders. Furthermore, the non-repre-sentative nature of their data means these studies lack statistical rigor for attempting to detect differences from Correspondence: Herng-Ching Lin, School of Health Care
Adminis-tration, Taipei Medical University, 250 Wu-Hsing St., Taipei 110, Taiwan (tel.: 886 2 2736 1661 ext. 3613; fax: 886 2 2378 9788; e-mail: [email protected]).
This is a Continuing Medical Education article, and can be found with corresponding questions on the internet at http://www.efns.org/content.php?pid=132. Certificates for correctly answering the questions will be issued by the EFNS.
the population as a whole or for generalizing from their findings to the entire population.
This study aims to examine the risk of adverse pregnancy outcomes [LBW, preterm birth, infants born small for gestational age (SGA) and cesarean sections (CS)] among women with MG, using a 3-year nation-wide population-based database. Taiwan initiated its National Health Insurance program in 1995 to finance healthcare for all of the islandÕs 22 million citizens. The data now available from Taiwan presents a unique opportunity to clarify the relationship between MG and adverse pregnancy outcomes.
Methods
Database
This study used two nationwide population-based datasets. The first dataset was the Taiwan National Health Insurance Research Dataset (NHIRD), released by Taiwan National Health Research Institutes. The NHIRD includes all medical claims data (inpatient and ambulatory care visits) and registration files (contracted medical facilities, board-certified specialists, medical personnel and beneficiaries) under the NHI, for over 22 million enrollees, representing over 98% of the Tai-wanese population. The NHIRD data are quite accu-rate because the Bureau of the NHI audits claims regularly. Fines for fraud are 100 times the amount of the false claim charged to the NHI.
The second dataset was the Taiwan birth certificate registry. The birth certificate registry includes birth-dates for both infants and their parents, gestational week at birth, birthweight, gender, parity, place of birth, parental educational levels and maternal marital status. The government in Taiwan mandates all births and deaths be registered, so the birth certificate data are considered to be accurate.
The two datasets were linked by the motherÕs and infantÕs unique personal identification numbers, with assistance from the Bureau of the NHI. All personal identifiers were encrypted by the Bureau of NHI before release to the researchers. Confidentiality assurances were addressed by abiding by the data regulations of the Bureau of NHI. Since the NHIRD consists of de-identified secondary data released to the public for re-search purposes, this study was exempt from full review by the Internal Review Board.
Study sample
The study group was selected from a total of 473 529 pregnant women, identified as having had singleton live births in Taiwan between 1 January 2001 and 31
December 2003. If a mother had more than one sin-gleton birth during 2001–2003, we only selected the first one for inclusion in the study sample. Of these women, 163 were identified as having visited ambulatory care centers for the treatment of myasthenia gravis (ICD-9-CM code 358.0) within the 2 years prior to their index deliveries.
The comparison group was extracted from the remaining 473 361 mothers. We randomly selected 815 women (five for every woman with MG) matched with the study group in terms of age (<20, 20–24, 25–29, 30– 34 and ‡35 years), the year of delivery and whether a woman had diabetes, hypertension, anemia, coronary heart disease (CHD) or hyperlipidemia. Ultimately, 978 patients were included in our sample.
Variables of interest
The outcome variables of this study were all dichoto-mous, including low birthweight (LBW) (<2500 g), preterm birth (<37 weeks), infants born small for gestational age (SGA) (birthweight below the tenth percentile for gestational age), and cesarean delivery. The key independent variable was whether or not a mother had visited an ambulatory care center for the treatment of MG in the 2 years prior to their index deliveries.
This study also took potential confounding factors into consideration in the regression modeling. Such fac-tors included characteristics of the mother (age, the highest educational level and marital status), infant (gender and parity) and monthly family income. Mater-nal education was grouped into four levels: elementary school or lower, junior high school, senior high school, college or above. Monthly income was divided into four categories: <NT$15,000, NT15,000-NT30,000, NT30,001-NT50,000, ‡NT50,001 (US$1.00 = NT$ 33.00 in 2003).
Statistical analysis
TheSASstatistical package (SAS System for Windows,
Version 8.2) was used to perform the analyses. The chi-squared tests were carried out to examine differences in the characteristics of mother and infant, comparing mothers with MG and unaffected mothers. Conditional logistic regression analyses, which were conditioned on maternal age, the year of delivery, diabetes, hyperten-sion, anemia, CHD and hyperlipidemia, were also performed to compare the risk of LBW, preterm birth, SGA, and cesarean delivery between these two groups, after adjusting for the potential confounders. A two-sided P-value of <0.05 was considered statistically significant for this study.
Results
Table 1 presents the distribution of characteristics of mothers and infants among mothers with MG and unaffected mothers. After matching for maternal age, the year of delivery and whether the woman had dia-betes, hypertension, anemia, CHD and hyperlipidemia,
no significant differences in maternal educational level, marital status, infant gender and parity or family monthly income were observed.
Table 2 illustrates the distributions of LBW, preterm birth, SGA, and cesarean deliveries among women with MG and unaffected women. Although they did not reach a statistically significant level, Table 2 consis-tently shows that women with MG had higher per-centages of LBW (6.8% vs. 5.6%) and SGA infants (17.8% vs. 14.1%) and cesarean deliveries (44.8% vs. 37.4%), but not preterm births (8.1% vs. 8.1%). Fur-ther of those having cesarean deliveries, six (out of 73 = 8.2%) and 21 (out of 305 = 6.9%) patients with and without MG, respectively, had elective surgeries (P = 0.69; data not shown in table). Conditional lo-gistic regressions showed that the crude ORs of LBW, preterm birth, SGA infants, and cesarean deliveries for women with MG were 1.21 (95% CI = 0.61–2.39), 0.98 (95% CI = 0.53–1.83), 1.32 (95% CI = 0.84–2.61) and 1.36 (95% CI = 0.97–1.91), respectively, com-pared to unaffected women.
Table 3 shows the adjusted ORs of LBW, preterm birth, SGA infants, and cesarean delivery between these two groups by conditional logistic regression. After adjusting for maternal education level, marital status and family monthly income, infant gender and parity, the ORs of LBW, preterm births, SGA infants, and cesarean delivery for women with MG were 1.19 (95% CI = 0.60–2.38), 1.00 (95% CI = 0.54–1.87), 1.30 (95% CI = 0.83–2.04), and 1.33 (95% CI = 0.94– 1.88), respectively, as compared to unaffected women.
Discussion
As far as we know, this is the first report of pregnancy outcomes among women with MG in Asia. Our nationwide population-based study clearly demon-strated that after adjusting for potential confounders, women with MG were not at increased risk of having preterm, LBW and SGA infants. Similarly, women with MG did not have a higher risk of cesarean delivery compared to unaffected women.
Myasthenia gravis commonly affects women who are in their child-bearing years [1,3]. Therefore, it is important to evaluate the effects of pre-existing MG on delivery and outcomes for newborns. As MG is a rare disease, the bulk of scholarship in this area has been largely observational in nature, and, particularly in older studies, derives from relatively small study co-horts [3,5–9,11] producing contradictory results in the literature. Only recently has interest fueled the organi-zation of multicenter prospective cohorts of pregnant MG patients and analysis of nationwide administrative claims datasets with adequate numbers of patients for Table 1 Comparisons of mothers with myasthenia gravis and
unaf-fected mothers in relation to maternal and infant characteristics in Taiwan, 2001–2003 (n = 978) Variable Mothers with myasthenia gravis (n = 163) Mothers in comparison group (n = 815) Pvalue Total No. % Total No. % Maternal characteristics Age (years) <20 6 3.7 30 3.7 1.000 20–24 16 9.8 80 9.8 25–29 65 39.9 325 39.9 30–34 49 30.1 245 30.1 >34 27 16.6 135 16.6 Education level
Elementary school or lower 3 1.8 13 1.6 0.829 Junior high school 26 16.0 133 16.3 Senior high school 104 63.8 542 66.5 College or above 30 18.4 127 15.6 Marital status
Married 155 95.1 797 97.8 0.051
Others 8 4.9 18 2.2
Family monthly income
NT$15 000 46 28.2 243 29.8 0.856 NT$15 000–30 000 43 26.4 203 24.9 NT$30 001–50 000 44 27.0 236 29.0 >NT$50 000 30 18.4 133 16.3 Gestational diabetes Yes 9 5.5 45 5.5 1.000 No 154 94.5 770 94.5 Hypertension Yes 6 3.7 30 3.7 1.000 No 157 96.3 785 96.3 Anemia Yes 18 11.0 90 11.0 1.000 No 145 89.0 725 89.0
Coronary heart disease
Yes 4 2.5 20 2.5 1.000 No 159 97.5 795 97.5 Hyperlipidemia Yes 3 1.8 15 1.8 No 160 98.2 800 98.2 Infant characteristics Gender Male 89 54.6 446 54.7 0.977 Female 74 45.4 369 45.3 Parity 1 80 49.1 385 47.2 0.856 2 62 38.0 313 38.4 3 or more 21 12.9 117 14.4
Table 2 The distribution of LBW, preterm birth, infants small for gestational age and cesarean deliveries among mothers with myasthenia gravis and unaffected mothers, 2001–2003 (n = 978) Variable Mothers with myasthenia gravis (n = 163) Mothers in com-parison group (n = 815) Pvalue No. % No. % Low birthweight Yes 11 6.8 46 5.6 0.583 No 152 93.2 769 94.4 Crude OR, 95% CI 1.21 (0.61–2.39) 1.00 Preterm birth Yes 13 8.1 66 8.1 0.958 No 150 91.9 749 91.9 Crude OR, 95% CI 0.98 (0.53–1.83) 1.00 Small for gestational age
Yes 29 17.8 115 14.1 0.226 No 134 82.2 700 85.9 Crude OR, 95% CI 1.32 (0.84–2.61) 1.00 Cesarean delivery Yes 73 44.8 305 37.4 0.078 No 90 55.2 510 62.6 Crude OR, 95% CI 1.36 (0.97–1.91) 1.00
Crude ORs were calculated by conditional logistic regressions which were conditioned on maternal age, the year of delivery and whether a womanÕs condition was complicated by diabetes, hypertension, anemia, coronary heart disease and hyperlipidemia.
Table 3 Adjusted Odds ratios of LBW, preterm birth, infants small for gestational age and cesarean deliveries for women with myasthenia gravis and unaffected mothers, 2001–2003 (n = 978)
Variable
Low birthweight Preterm birth SGA CS
Adjusted OR 95% CI Adjusted OR 95% CI Adjusted OR 95% CI Adjusted OR 95% CI
Cohort
Myasthenia gravis 1.19 (0.60–2.38) 1.00 (0.54–1.87) 1.30 (0.83–2.04) 1.33 (0.94–1.87)
Comparison cohort 1.00 1.00 1.00 1.00
Maternal characteristics Education level
Elementary school or lower 2.62 (0.56–12.30) 0.75 (0.10–5.85) 2.08 (0.65–6.67) 1.10 (0.39–3.10) Junior high school 1.31 (0.66–2.61) 1.18 (0.65–2.14) 1.41 (0.89–2.24) 1.05 (0.73–1.51)
Senior high school 1.00 1.00 1.00 1.00
College or above 0.77 (0.33–1.80) 0.73 (0.35–1.53) 1.08 (0.64–1.80) 1.02 (0.70–1.47) Marital status Married 0.58 (0.16–2.05) 0.69 (0.20–2.43) 0.63 (0.25–1.63) 0.81 (0.36–1.82) Other 1.00 1.00 1.00 1.00 Monthly income <NT$15 000 1.00 1.00 1.00 1.00 NT$15 000–30 000 1.20 (0.56–2.56) 1.22 (0.62–2.40) 1.14 (0.70–1.84) 1.42 (0.99–2.04) NT$30 001–50 000 1.42 (0.69–2.92) 1.65 (0.88–3.10) 1.03 (0.64–1.66) 1.32 (0.93–1.88) >NT$50 000 1.49 (0.64–3.46) 1.34 (0.63–2.85) 0.99 (0.56–1.75) 2.12 (1.42–3.16) Infant characteristics Gender Male 0.64 (0.37–1.10) 1.12 (0.70–1.79) 0.76 (0.53–1.08) 0.99 (0.76–1.29) Female 1.00 1.00 1.00 1.00 Parity 1 1.00 1.00 1.00 1.00 2 0.43* (0.22–0.83) 0.97 (0.57–1.64) 0.82 (0.55–1.22) 0.99 (0.75–1.33) 3 or more 0.78 (0.36–1.70) 1.58 (0.84–3.00) 1.11 (0.66–1.87) 0.72 (0.48–1.09)
*P < 0.05. Adjusted ORs were calculated ‘y conditional logistic regressions which were conditioned on maternal age, the year of delivery and whether a motherÕs condition was complicated by diabetes, hypertension, anemia, coronary heart disease and hyperlipidemia.
reaching reliable conclusions [10]. In addition, some studies suffer from inadequate control of confounders, which could undermine the strength of their findings. For example, recent studies indicate that the most common complications during pregnancy are anemia and hypertensive disorders; anemia in particular was demonstrated to be associated with adverse pregnancy outcomes [12]. As a consequence, the extent of preg-nancies affected by MG is not well established and assessments of the MG of pregnancy complications remain inconsistent.
To fill this gap in the literature, this report, based on a large-scale population-based dataset from Taiwan, provides enough statistical power to fully evaluate the association between MG in pregnant women and ad-verse pregnancy outcomes. Furthermore, our study has taken comorbidities and other potential confounders into consideration. On balance, the our results were largely consistent with prior studies conducted in dif-ferent regions and documented no association between MG and LBW [8–10], preterm birth [8,10] and CS [8]. As previous studies investigating pregnancy outcomes for women with MG have been conducted in western countries, our findings not only add evidence that MG has no significant adverse effect on pregnancy out-comes, but also extend the literature to include the experiences of an Asian population.
The reported prevalence of MG has increased in every decade since the 1950s [13,14]. Our study likewise found that MG prevalence among pregnant women in Taiwan was much higher (3.4/10 000) than that documented by previous studies (1/20 000) [2]. The increase in the num-ber of mothers with MG can be attributed to several factors, including improved recognition of the disease, the availability of diagnostic tests with higher sensitivity and specificity, and longer life spans in affected patients due to more effective treatment. For example, under TaiwanÕs NHI program, all pregnant women are allowed to have ten free scheduled prenatal care visits, which may increase the rate at which MG is diagnosed among them. By screening for and aggressively treating pregnant wo-men with MG in Taiwan, it is possible to significantly decrease the incidence of adverse pregnancy outcomes.
Other issues also require further consideration. It has been proposed that due to the muscular fatigue, MG mothers have difficulty with vaginal delivery, especially during the second stage of labor. The use of forceps delivery and vacuum extraction was essential in these urgent situations [15]. However, we observed that a total of nine (out of 90 = 10.0%) and 52 (out of 510 = 10.2%) patients with and without MG, respec-tively, underwent forceps/vacuum vaginal delivery. This insignificant difference between groups (P = 0.955) does not indicate problems during vaginal delivery
among patients with MG. Consistent with previous findings, caesarean delivery is recommended only in cases where there is obstetric need [15].
A wide variation in cesarean section rates among countries worldwide has been reported, ranging from 0.4 to 40 percent. The median cesarean section rates were 4.0%, 16.1% and 17% among the low-, medium-, and high-income countries respectively [16]. An upward trend over the past decades has been further identified. In the United States, cesarean delivery increased from 20.7% in 1996 to 31.1% in 2006 [17]. In Taiwan, a high cesarean section rate of 32.3% of all deliveries was reported [18]. With an older population, we observed a cesarean rate of 37.4% among women without MG, compared to 44.8% among those with MG (P = 0.08). The difference in the proportion of elective cesarean sections among patients with MG was not significant, compared to unaffected women. Despite these insignificant differences, it is worth noting that a relatively high cesarean rate in the general population can mask the possibility of an even higher rate of cesarean delivery among specific risk groups, com-pared to unaffected women.
Maternal MG might also contribute to neonatal complications. A 21% incidence of transient neonatal MG in infants born to mothers with MG was reported [19]. Sucking, swallowing, and respiratory difficulties are the most commonly observed signs. Although transient, and in most cases very mild, about eighty percent of patients may require supportive management and anticholinesterase agents prior to feedings. Thy-mectomy, usually recommended and performed on all early-onset MG patients, seemed to provide a protec-tive effect against neonatal MG [9,20]. Although thy-mectomy may indicate the severity of the maternal MG, it was not associated with pregnancy outcomes in our study. Specifically, among MG patients, 6.7% had re-ceived thymectomy. No significant difference was identified in terms of the proportion of LBW (9.1% vs. 6.6%, P = 0.75), preterm births (9.1% vs. 7.9%, P= 0.89), SGA infants (18.2% vs. 17.8%, P = 0.97), and cesarean deliveries (63.6% vs. 43.4%, P = 0.19) among women who had thymectomies compared to those who did not. In addition, maternal MG might also cause a rare fetal condition, arthrogryposis multi-plex congenita [21]. While some diseases are fatal early in life, early identification and appropriate treatment can be essential in prolonging survival into adulthood. Prior studies have proposed that it is not the clinical symptoms of the MG mother, but rather, her type of antibodies, which is a critical determinant of whether her newborn will develop neonatal MG, and in a few cases, arthrogryposis multiplex congenita [22]. More studies are needed that further investigate the connec-tion between maternal MG and associated neonatal
complications in order to give more reliable genetic advice to parents, and to organize more effective and realistic management.
Despite the strengths of our study, findings must still be interpreted with caution due to the following limi-tations. First, the NHIRD database only represented patients who had sought treatment for MG. Certain factors, such as socioeconomic status, could affect healthcare utilization. For example, despite the exis-tence of universal healthcare, individuals with lower incomes and education received a lower rate of physi-cian services [23]. In addition, MG mothers were iden-tified through having visited ambulatory care centers for treatment of MG within the 2 years prior to their index deliveries. As only symptomatic MG women were recruited as cases, our results would likely be biased towards the null. Second, the NHIRD lacks detailed clinical information and therefore did not allow us to differentiate study participants according to the severity of their MG. Third, NHIRD uses discharge diagnoses provided by treating physicians, and standardized cri-teria to define cases were not imposed, which could leave some room for bias due to case misclassification. Finally, because the NHIRD does not include complete information regarding medications taken during preg-nancy, it is not possible for us to assess the confounding role of medications in the relationship between MG and pregnancy outcomes. Meanwhile, information on neo-natal complications, including neoneo-natal MG and arthrogryposis multiplex congenital, were also unavailable in our dataset.
This study using a large, unselected national dataset has demonstrated that women with MG are not at in-creased risk for having preterm, LBW and SGA babies, and for delivery by CS, compared to unaffected moth-ers. We suggest that although women with MG need not panic about adverse pregnancy outcomes, the management of MG should not be altered during pregnancy. Women with MG who choose to become pregnant should discuss their plan for pregnancy with their neurologist and their gynecologist and should get good prenatal care with doctors who are experienced in treating MG. Further large-scale studies in other re-gions or countries should also be carried out to confirm the findings of the present one.
Disclosure
No financial conflict of interest to declare.
References
1. Burke ME. Myasthenia gravis and pregnancy. J Perinat Neonatal Nurs1993; 7: 11–21.
2. Mitchell PJ, Bebbington M. Myasthenia gravis in preg-nancy. Obstet Gynecol 1992; 80: 178–181.
3. Plauche WC. Myasthenia gravis. Clin Obstet Gynecol 1983; 26: 592–604.
4. Carr SR, Gilchrist JM, Abuelo DN, Clark D. Treatment of antenatal myasthenia gravis. Obstet Gynecol 1991; 78: 485–489.
5. Eden RD, Gall SA. Myasthenia gravis and pregnancy: a re-appraisal of thymectomy. Obstet Gynecol 1983; 62: 328–333. 6. Plauche WC. Myasthenia gravis in mothers and their
newborns. Clin Obstet Gynecol 1991; 34: 82–99.
7. Giwa-Osagie OF, Newton JR, Larcher V. Obstetric per-formance of patients with my asthenia gravis. Int J Gynaecol Obstet1981; 19: 267–270.
8. Batocchi AP, Majolini L, Evoli A, Lino MM, Minisci C, Tonali P. Course and treatment of myasthenia gravis during pregnancy. Neurology 1999; 52: 447–452. 9. Djelmis J, Sostarko M, Mayer D, Ivanisevic M.
Myas-thenia gravis in pregnancy: report on 69 cases. Eur J Obstet Gynecol Reprod Biol2002; 104: 21–25.
10. Hoff JM, Daltveit AK, Gilhus NE. Myasthenia gravis: consequences for pregnancy, delivery, and the newborn. Neurology2003; 61: 1362–1366.
11. Daskalakis GJ, Papageorgiou IS, Petrogiannis ND, Ant-saklis AJ, Michalas SK. Myasthenia gravis and pregnancy. Eur J Obstet Gynecol Reprod Biol2000; 89: 201–204. 12. Bruce FC, Berg CJ, Hornbrook MC, et al. Maternal
morbidity rates in a managed care population. Obstet Gynecol2008; 111: 1089–1095.
13. Phillips LH II. The epidemiology of myasthenia gravis. Ann N Y Acad Sci2003; 998: 407–412.
14. Phillips LH II, Torner JC. Epidemiologic evidence for a changing natural history of myasthenia gravis. Neurology 1996; 47: 1233–1238.
15. Ferrero S, Pretta S, Nicoletti A, Petrera P, Ragni N. Myasthenia gravis: management issues during pregnancy. Eur J Obstet Gynecol Reprod Biol2005; 121: 129–138. 16. Althabe F, Sosa C, Belizan JM, Gibbons L, Jacquerioz F,
Bergel E. Cesarean section rates and maternal and neo-natal mortality in low-, medium-, and high-income countries: an ecological study. Birth 2006; 33: 270–277. 17. MacDorman MF, Menacker F, Declercq E. Cesarean
birth in the United States: epidemiology, trends, and outcomes. Clin Perinatol 2008; 35: 293–307, v.
18. Lin HC, Xirasagar S. Institutional factors in cesarean delivery rates: policy and research implications. Obstet Gynecol2004; 103: 128–136.
19. Papazian O. Transient neonatal myasthenia gravis. J Child Neurol1992; 7: 135–141.
20. Hoff JM, Daltveit AK, Gilhus NE. Myasthenia gravis in pregnancy and birth: identifying risk factors, optimising care. Eur J Neurol 2007; 14: 38–43.
21. Hoff JM, Daltveit AK, Gilhus NE. Artrogryposis multi-plex congenita – a rare fetal condition caused by maternal myasthenia gravis. Acta Neurol Scand Suppl 2006; 183: 26–27.
22. Vernet-der Garabedian B, Lacokova M, Eymard B, et al. Association of neonatal myasthenia gravis with antibodies against the fetal acetylcholine receptor. J Clin Invest 1994; 94: 555–559.
23. Dunlop S, Coyte PC, McIsaac W. Socio-economic status and the utilisation of physiciansÕ services: results from the Canadian National Population Health Survey. Soc Sci Med2000; 51: 123–133.
