DOI 10.1515/jpm-2013-0314 J. Perinat. Med. 2014; 42(5): 635–639
Ann-Christin Warby*, Susanne Amler, Annett M. Jacobi, Kerstin Hammer, Ute Möllmann, Maria K. Falkenberg, Mareike Möllers, Ludwig Kiesel, Walter Klockenbusch and Ralf Schmitz
Imaging of fetal thymus in pregnant women with rheumatic diseases Abstract Objective: To determine whether certain rheumatic diseases will affect the fetal thymus diameter when compared to uncomplicated singleton pregnancies. Additionally, we created a reference chart for fetal thymus size in healthy singleton pregnancies from 19 to 37 weeks of gestation. Methods: Sonographic fetal thymus size was retrospectively evaluated in 190 healthy pregnant women, and 84 pregnancies of mothers suffering from systemic lupus erythematosus, antiphospholipid syndrome (APS), or Sjögren’s syndrome between 19 and 37 weeks of gestation. These fetuses were matched one-to-one for gestational age with control fetuses. The thymic-thoracic ratio (TTratio) was defined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter. Results: Rheumatic diseases often affect pregnancy outcome, especially in case of primary APS. The TT ratio of fetuses of mothers suffering from rheumatic disease was equal to controls (P = 0.807). Conclusions: Ours is the first study to assess the correlation of fetal thymus size in high-risk pregnancies with rheumatic diseases in comparison to controls. Women with rheumatic diseases deal with pregnancy complications more frequently than controls. Our data suggest that maternal rheumatic diseases do not affect the fetal thymus size. Keywords: Antiphopholipid syndrome; lupus erythematodes, systemic; pregnancy, high risk; pregnancy outcome; Sjogren’s Syndrome; thymus gland; ultrasonography. *Corresponding author: Ann-Christin Warby, Department of Gynecology and Obstetrics, University Hospital Münster, AlbertSchweitzer-Campus 1, Gebäude: A1, 48149 Münster, Germany, Tel.: 0176/96446374, E-mail: [email protected] Ann-Christin Warby, Kerstin Hammer, Ute Möllmann, Maria K. Falkenberg, Mareike Möllers, Ludwig Kiesel, Walter Klockenbusch and Ralf Schmitz: Department of Gynecology and Obstetrics, University Hospital Münster, Münster, Germany Susanne Amler: Institute of Biostatistics and Clinical Research, University of Münster, Münster, Germany Annett M. Jacobi: Division of Rheumatology and Clinical Immunology/Department of Medicine D, University Hospital Münster, Münster, Germany
Introduction Systemic lupus erythematodes (SLE) is a rheumatic disease that often affects women at child-bearing age. In most cases, this disease is diagnosed before pregnancy. If possible, remission should be achieved at least 6 months before conceiving. It is commonly known that SLE can flare up during pregnancy. The risk of exacerbation rises directly after childbirth. Lupus nephritis worsens the prognosis for a successful pregnancy. Pregnancy complications, including preeclampsia, HELLP syndrome, intrauterine growth restriction, stillbirth, and preterm birth appear more often than in healthy pregnancies [2, 10, 11, 13]. Notching in the uterine arteries in the second trimester seems to have a predictive value for a worse pregnancy outcome [12]. Antiphospholipid syndrome (APS), being defined as habitual abortions and hypercoagulation caused by these antibodies, can either be a disease on its own or can develop from a pre-existing SLE. The pregnancy complications do not differ much from those with SLE. Low-dose acetylsalicylic acid and low-molecular weight heparin can be used as options to prevent poor outcome [13]. The clinical manifestation of patients with primary Sjögren’s syndrome (SS) includes xerostomia (dry mouth) and xerophthalmia (dry eyes), accompanied by swelling of lacrimal and salivary glands. Anti-Ro (SS-A) antibodies causing congenital heart block are often found in these patients [1] The thymus is a primary lymphatic organ that serves the development of T-lymphocytes. Recently, several studies have evaluated the importance of fetal thymus measurement in prenatal diagnosis of growth-restricted (intra-uterine growth restriction, IUGR) fetuses [5, 16], fetuses with Down syndrome [7], and preterm prelabor rupture of membranes (PPROM) [15, 18]. Additionally, the size of the thymus was smaller in fetuses of mothers with preeclampsia [14], and in fetuses with congenital heart defects in combination with microdeletion 22q11 [4]. Fetal thymus was first measured by ultrasound in 1989 [8]. Reference charts were introduced in 2002 by Zalel et al. [19], presenting a normogram for the thymus perimeter. The gender of the fetus does not have an impact
Brought to you by | Karolinska Institute Authenticated Download Date | 5/23/15 11:53 AM
636 Warby et al., Fetal thymus and pregnant women with rheumatic diseases on the thymus size [6]. In twin pregnancies, no difference could be seen in contrast to singleton pregnancies [9]. A recent study by Cromi et al. [5] on fetuses with intrauterine growth restriction has shown a decreased thymus perimeter in comparison to control fetuses. Thymic volume divided by abdominal circumference was significantly smaller in IUGR fetuses with abnormal umbilical Doppler flow velocity waveforms in contrast to healthy fetuses, small for gestational age (SGA) fetuses, and fetuses with IUGR, but without Doppler abnormalities [16]. In cases of preeclampsia and fetuses with Down syndrome, the fetal thymus size was reduced compared to controls [7, 14]. In a study by Musilova et al. [15], the transverse diameter of fetuses in patients with preterm premature rupture of membranes was below the 5th percentile, whereas in the study by Yinon et al. [18] the perimeter was decreased, especially in cases of PPROM and chorionamnionitis. Fetuses with congenital heart defects and a reduced TTratio have shown a higher rate of microdeletion 22q11 [4]. To the best of our knowledge, the fetal thymus size has not been investigated in pregnancies with maternal rheumatic diseases. The aim of our study was to compare the fetal thymus size between pregnant women with rheumatic disease and healthy pregnant women. Furthermore, reference charts containing fetal thymus size in uncomplicated pregnancies have been created.
thymus was measured retrospectively in the three-vessel view by one observer. Ultrasound screenshots of our electric database Professional Image Archiving Fetal Database (General Electric, Fairfield, CT, USA) have been re-evaluated. The thymus could be identified in the anterior mediastinum due to it homogeneous structure [6]. The intrathoracic mediastinal diameter was measured along a midline drawn through the aortic arch vessel from the anterior edge of the thoracic vertebral body to the internal edge of the sternum. The anteroposterior diameter of the thymus was defined as the parallel traced distance between the transverse aortic arch border and the posterior chest wall. The TTratio was determined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter (Figure 1). To test the intraobserver variability, the same screenshot was measured twice by one observer. A second observer carried out a single measurement of the same case. This was compared with the first measurement made by the previous operator to determine the interobserver reproducibility. Statistical analyses were performed using IBM SPSS® Statistics 21 for Windows (IBM Corporation, Somers, NY, USA). A scatter plot of TT-ratio against weeks of gestation has been created. Differences in the pregnancy outcome and sonographic measurements between the study group and the control group were evaluated by using the Mann-Whitney test and the Chi-square test. Subgroup analysis was performed for cases of maternal APS matched for gestational age with controls by using Student’s t-test. Inferential statistics are intended to be exploratory (hypotheses generating), not confirmatory, and are interpreted accordingly – i.e., P-values are interpreted in Fisher’s sense, representing the metric weight of evidence against the respective null hypothesis of no effect. Neither a global significance level nor local levels are determined. The P-values are considered noticeable in
Methods Eighty-four pregnant women with SLE, APS, and SS have been measured retrospectively in this study. All ultrasound examinations were performed in the Department of Gynecology and Obstetrics at the University Hospital in Münster, Germany during the period from January 2002 to December 2012. We measured the thymic-thoracic ratio (TT-ratio) in the plane of the three vessel view as described by Chaoui et al. [3] in accordance with the DEGUM quality requirements. Seventy women were recorded in the database with one pregnancy and ten patients with two pregnancies. Four women were examined in all of three pregnancies. Due to abortion before the 19th week of gestation in 6 pregnancies, 78 pregnancies could be finally included in the analysis. The population was subdivided into 5 groups: SLE (n = 47), primary APS (n = 11), SS (n = 8), antibodies positive (n = 6), but without established diagnosis (ABP), and SLE with secondary APS (n = 5). For the reference chart, patients were chosen randomly. We selected 10 cases for every week from the 19th to 37th week of gestation. These 190 normal subjects for the reference chart fulfilled our entry criteria: (1) pregnancies without pre-existing disease; (2) singleton pregnancies; and (3) pregnancies with anamnestic, uneventful prior pregnancies. Of these 190 pregnancies, 78 were chosen for matched pairs. We included ultrasound examinations performed between 19 and 37 weeks of gestation using a transabdominal transducer. The
Figure 1 Three vessel view of a normal fetus at 21 weeks of gestation. Intrathoracic mediastinal diameter (22.3 mm) was measured along a midline drawn through the aortic arch vessel from the anterior edge of the thoracic vertebral body to the internal edge of the sternum. Anteroposterior diameter (7.74 mm) was defined as the parallel traced distance between the transverse aortic arch border and the posterior chest wall. The thymic-thoracic ratio was determined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter.
Brought to you by | Karolinska Institute Authenticated Download Date | 5/23/15 11:53 AM
Warby et al., Fetal thymus and pregnant women with rheumatic diseases 637
case P ≤ 0.05 and highly noticeable in case P ≤ 0.01. These findings will be used to generate new hypotheses. Spearman’s test was used for notnormally distributed data to test the correlation between TT-ratio and gestational age, body mass index, birth weight, appearance, pulse, grimace, activity, respirance, after Virginia Apgar (APGAR score) after 5 min, and medication. The intraclass correlation coefficients (ICCs) to test the intra- and interobserver variability were calculated.
Results Study population The study population is described in Table 1. Statistically noticeable differences between cases with rheumatic Table 1 Characteristics and outcome of the study population. Parameter
Study cohort
Control group
Spontaneous abortion [n (%)] Late abortion [n (%)] PE [n (%)] Super-imposed PE [n (%)] HELLP [n (%)] Preterm delivery [n (%)] Cesarean section [n (%)]
6 (7%) 5 (6%) 6 (8%) 2 (3%) 3 (4%) 28 (36%) 34 (46%)
a a a a 6 (8%) 19 (24%) 23.90 39.51 3483 9.68 7.29 0.344
BMI (kg/m2)b GA at delivery (weeks)b Birth weight (g)b APGAR after 5 minb pH umbilical arteryb TT-ratiob
25.65 36.99 2878 8.88 7.27 0.345
P-value
disease and controls can be seen in gestational age (GA) at delivery, preterm delivery, cesarean section, body mass index (BMI), APGAR score after 5 min, and birth weight. The study group was divided into five groups: SLE, APS, SS, cases without established diagnosis, and SLE with secondary APS. Between these subgroups, noticeable differences can be seen as well in GA at delivery, preterm delivery, cesarean section, BMI, APGAR score after 5 min, and birth weight, but not in TT-ratio (P = 0.571) (Table 2). The TT-ratio had a mean ± standard deviation value of 0.344 ± 0.064 for the study cohort and 0.343 ± 0.058 for the matched control group (Figure 2). The subgroup analysis was performed for the cases with primary APS (n = 11) and matched pairs. A noticeable trend towards an elevated TTratio was observed (P = 0.078) (Figure 3).
Normal fetuses and reference chart The reference chart for TT-ratio was created using the data of 190 healthy fetuses between 19 and 37 weeks of gestation. The TT-ratio slightly increased throughout gestation at an average of 0.365 ± 0.064 (r2 = 0.082) (Figure 4).
a
Ann-Christin Warby*, Susanne Amler, Annett M. Jacobi, Kerstin Hammer, Ute Möllmann, Maria K. Falkenberg, Mareike Möllers, Ludwig Kiesel, Walter Klockenbusch and Ralf Schmitz
Imaging of fetal thymus in pregnant women with rheumatic diseases Abstract Objective: To determine whether certain rheumatic diseases will affect the fetal thymus diameter when compared to uncomplicated singleton pregnancies. Additionally, we created a reference chart for fetal thymus size in healthy singleton pregnancies from 19 to 37 weeks of gestation. Methods: Sonographic fetal thymus size was retrospectively evaluated in 190 healthy pregnant women, and 84 pregnancies of mothers suffering from systemic lupus erythematosus, antiphospholipid syndrome (APS), or Sjögren’s syndrome between 19 and 37 weeks of gestation. These fetuses were matched one-to-one for gestational age with control fetuses. The thymic-thoracic ratio (TTratio) was defined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter. Results: Rheumatic diseases often affect pregnancy outcome, especially in case of primary APS. The TT ratio of fetuses of mothers suffering from rheumatic disease was equal to controls (P = 0.807). Conclusions: Ours is the first study to assess the correlation of fetal thymus size in high-risk pregnancies with rheumatic diseases in comparison to controls. Women with rheumatic diseases deal with pregnancy complications more frequently than controls. Our data suggest that maternal rheumatic diseases do not affect the fetal thymus size. Keywords: Antiphopholipid syndrome; lupus erythematodes, systemic; pregnancy, high risk; pregnancy outcome; Sjogren’s Syndrome; thymus gland; ultrasonography. *Corresponding author: Ann-Christin Warby, Department of Gynecology and Obstetrics, University Hospital Münster, AlbertSchweitzer-Campus 1, Gebäude: A1, 48149 Münster, Germany, Tel.: 0176/96446374, E-mail: [email protected] Ann-Christin Warby, Kerstin Hammer, Ute Möllmann, Maria K. Falkenberg, Mareike Möllers, Ludwig Kiesel, Walter Klockenbusch and Ralf Schmitz: Department of Gynecology and Obstetrics, University Hospital Münster, Münster, Germany Susanne Amler: Institute of Biostatistics and Clinical Research, University of Münster, Münster, Germany Annett M. Jacobi: Division of Rheumatology and Clinical Immunology/Department of Medicine D, University Hospital Münster, Münster, Germany
Introduction Systemic lupus erythematodes (SLE) is a rheumatic disease that often affects women at child-bearing age. In most cases, this disease is diagnosed before pregnancy. If possible, remission should be achieved at least 6 months before conceiving. It is commonly known that SLE can flare up during pregnancy. The risk of exacerbation rises directly after childbirth. Lupus nephritis worsens the prognosis for a successful pregnancy. Pregnancy complications, including preeclampsia, HELLP syndrome, intrauterine growth restriction, stillbirth, and preterm birth appear more often than in healthy pregnancies [2, 10, 11, 13]. Notching in the uterine arteries in the second trimester seems to have a predictive value for a worse pregnancy outcome [12]. Antiphospholipid syndrome (APS), being defined as habitual abortions and hypercoagulation caused by these antibodies, can either be a disease on its own or can develop from a pre-existing SLE. The pregnancy complications do not differ much from those with SLE. Low-dose acetylsalicylic acid and low-molecular weight heparin can be used as options to prevent poor outcome [13]. The clinical manifestation of patients with primary Sjögren’s syndrome (SS) includes xerostomia (dry mouth) and xerophthalmia (dry eyes), accompanied by swelling of lacrimal and salivary glands. Anti-Ro (SS-A) antibodies causing congenital heart block are often found in these patients [1] The thymus is a primary lymphatic organ that serves the development of T-lymphocytes. Recently, several studies have evaluated the importance of fetal thymus measurement in prenatal diagnosis of growth-restricted (intra-uterine growth restriction, IUGR) fetuses [5, 16], fetuses with Down syndrome [7], and preterm prelabor rupture of membranes (PPROM) [15, 18]. Additionally, the size of the thymus was smaller in fetuses of mothers with preeclampsia [14], and in fetuses with congenital heart defects in combination with microdeletion 22q11 [4]. Fetal thymus was first measured by ultrasound in 1989 [8]. Reference charts were introduced in 2002 by Zalel et al. [19], presenting a normogram for the thymus perimeter. The gender of the fetus does not have an impact
Brought to you by | Karolinska Institute Authenticated Download Date | 5/23/15 11:53 AM
636 Warby et al., Fetal thymus and pregnant women with rheumatic diseases on the thymus size [6]. In twin pregnancies, no difference could be seen in contrast to singleton pregnancies [9]. A recent study by Cromi et al. [5] on fetuses with intrauterine growth restriction has shown a decreased thymus perimeter in comparison to control fetuses. Thymic volume divided by abdominal circumference was significantly smaller in IUGR fetuses with abnormal umbilical Doppler flow velocity waveforms in contrast to healthy fetuses, small for gestational age (SGA) fetuses, and fetuses with IUGR, but without Doppler abnormalities [16]. In cases of preeclampsia and fetuses with Down syndrome, the fetal thymus size was reduced compared to controls [7, 14]. In a study by Musilova et al. [15], the transverse diameter of fetuses in patients with preterm premature rupture of membranes was below the 5th percentile, whereas in the study by Yinon et al. [18] the perimeter was decreased, especially in cases of PPROM and chorionamnionitis. Fetuses with congenital heart defects and a reduced TTratio have shown a higher rate of microdeletion 22q11 [4]. To the best of our knowledge, the fetal thymus size has not been investigated in pregnancies with maternal rheumatic diseases. The aim of our study was to compare the fetal thymus size between pregnant women with rheumatic disease and healthy pregnant women. Furthermore, reference charts containing fetal thymus size in uncomplicated pregnancies have been created.
thymus was measured retrospectively in the three-vessel view by one observer. Ultrasound screenshots of our electric database Professional Image Archiving Fetal Database (General Electric, Fairfield, CT, USA) have been re-evaluated. The thymus could be identified in the anterior mediastinum due to it homogeneous structure [6]. The intrathoracic mediastinal diameter was measured along a midline drawn through the aortic arch vessel from the anterior edge of the thoracic vertebral body to the internal edge of the sternum. The anteroposterior diameter of the thymus was defined as the parallel traced distance between the transverse aortic arch border and the posterior chest wall. The TTratio was determined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter (Figure 1). To test the intraobserver variability, the same screenshot was measured twice by one observer. A second observer carried out a single measurement of the same case. This was compared with the first measurement made by the previous operator to determine the interobserver reproducibility. Statistical analyses were performed using IBM SPSS® Statistics 21 for Windows (IBM Corporation, Somers, NY, USA). A scatter plot of TT-ratio against weeks of gestation has been created. Differences in the pregnancy outcome and sonographic measurements between the study group and the control group were evaluated by using the Mann-Whitney test and the Chi-square test. Subgroup analysis was performed for cases of maternal APS matched for gestational age with controls by using Student’s t-test. Inferential statistics are intended to be exploratory (hypotheses generating), not confirmatory, and are interpreted accordingly – i.e., P-values are interpreted in Fisher’s sense, representing the metric weight of evidence against the respective null hypothesis of no effect. Neither a global significance level nor local levels are determined. The P-values are considered noticeable in
Methods Eighty-four pregnant women with SLE, APS, and SS have been measured retrospectively in this study. All ultrasound examinations were performed in the Department of Gynecology and Obstetrics at the University Hospital in Münster, Germany during the period from January 2002 to December 2012. We measured the thymic-thoracic ratio (TT-ratio) in the plane of the three vessel view as described by Chaoui et al. [3] in accordance with the DEGUM quality requirements. Seventy women were recorded in the database with one pregnancy and ten patients with two pregnancies. Four women were examined in all of three pregnancies. Due to abortion before the 19th week of gestation in 6 pregnancies, 78 pregnancies could be finally included in the analysis. The population was subdivided into 5 groups: SLE (n = 47), primary APS (n = 11), SS (n = 8), antibodies positive (n = 6), but without established diagnosis (ABP), and SLE with secondary APS (n = 5). For the reference chart, patients were chosen randomly. We selected 10 cases for every week from the 19th to 37th week of gestation. These 190 normal subjects for the reference chart fulfilled our entry criteria: (1) pregnancies without pre-existing disease; (2) singleton pregnancies; and (3) pregnancies with anamnestic, uneventful prior pregnancies. Of these 190 pregnancies, 78 were chosen for matched pairs. We included ultrasound examinations performed between 19 and 37 weeks of gestation using a transabdominal transducer. The
Figure 1 Three vessel view of a normal fetus at 21 weeks of gestation. Intrathoracic mediastinal diameter (22.3 mm) was measured along a midline drawn through the aortic arch vessel from the anterior edge of the thoracic vertebral body to the internal edge of the sternum. Anteroposterior diameter (7.74 mm) was defined as the parallel traced distance between the transverse aortic arch border and the posterior chest wall. The thymic-thoracic ratio was determined as the quotient of the anteroposterior thymic and the intrathoracic mediastinal diameter.
Brought to you by | Karolinska Institute Authenticated Download Date | 5/23/15 11:53 AM
Warby et al., Fetal thymus and pregnant women with rheumatic diseases 637
case P ≤ 0.05 and highly noticeable in case P ≤ 0.01. These findings will be used to generate new hypotheses. Spearman’s test was used for notnormally distributed data to test the correlation between TT-ratio and gestational age, body mass index, birth weight, appearance, pulse, grimace, activity, respirance, after Virginia Apgar (APGAR score) after 5 min, and medication. The intraclass correlation coefficients (ICCs) to test the intra- and interobserver variability were calculated.
Results Study population The study population is described in Table 1. Statistically noticeable differences between cases with rheumatic Table 1 Characteristics and outcome of the study population. Parameter
Study cohort
Control group
Spontaneous abortion [n (%)] Late abortion [n (%)] PE [n (%)] Super-imposed PE [n (%)] HELLP [n (%)] Preterm delivery [n (%)] Cesarean section [n (%)]
6 (7%) 5 (6%) 6 (8%) 2 (3%) 3 (4%) 28 (36%) 34 (46%)
a a a a 6 (8%) 19 (24%) 23.90 39.51 3483 9.68 7.29 0.344
BMI (kg/m2)b GA at delivery (weeks)b Birth weight (g)b APGAR after 5 minb pH umbilical arteryb TT-ratiob
25.65 36.99 2878 8.88 7.27 0.345
P-value
disease and controls can be seen in gestational age (GA) at delivery, preterm delivery, cesarean section, body mass index (BMI), APGAR score after 5 min, and birth weight. The study group was divided into five groups: SLE, APS, SS, cases without established diagnosis, and SLE with secondary APS. Between these subgroups, noticeable differences can be seen as well in GA at delivery, preterm delivery, cesarean section, BMI, APGAR score after 5 min, and birth weight, but not in TT-ratio (P = 0.571) (Table 2). The TT-ratio had a mean ± standard deviation value of 0.344 ± 0.064 for the study cohort and 0.343 ± 0.058 for the matched control group (Figure 2). The subgroup analysis was performed for the cases with primary APS (n = 11) and matched pairs. A noticeable trend towards an elevated TTratio was observed (P = 0.078) (Figure 3).
Normal fetuses and reference chart The reference chart for TT-ratio was created using the data of 190 healthy fetuses between 19 and 37 weeks of gestation. The TT-ratio slightly increased throughout gestation at an average of 0.365 ± 0.064 (r2 = 0.082) (Figure 4).
a
