Fetal medicine
DOI: 10.1111/j.1471-0528.2011.03253.x www.bjog.org
First trimester maternal serum free b-human chorionic gonadotropin and pregnancy-associated plasma protein A in pregnancies complicated by diabetes mellitus MD Savvidou,a,b A Syngelaki,b M Muhaisen,b E Emelyanenko,b KH Nicolaidesb,c a
Academic Department of Obstetrics and Gynaecology, Imperial College School of Medicine, Chelsea and Westminster Hospital, London, UK Harris Birthright Research Centre for Fetal Medicine, King’s College Hospital, London, UK c Department of Fetal Medicine, University College Hospital, London, UK Correspondence: Dr MD Savvidou, Department of Maternal Fetal Medicine, Imperial College School of Medicine, Chelsea and Westminster Hospital, 369 Fulham Road, London, SW10 9NH, UK. Email
[email protected] b
Accepted 14 November 2011.
Objective To investigate whether markers of first trimester
screening for aneuploidies, including fetal nuchal translucency (NT), maternal serum free b-human chorionic gonadotropin (b-hCG) and pregnancy-associated plasma protein A (PAPP-A), are altered in women with pre-existing type-1 and type-2 diabetes mellitus, and in women that subsequently develop gestational diabetes mellitus (GDM). Design Retrospective analysis of prospective combined screening for aneuploidies in singleton pregnancies at 11+0–13+6 weeks of gestation. Setting Antenatal clinic. +0
+6
Population Singleton pregnancies at 11 –13
weeks of gestation resulting in the delivery of phenotypically normal neonates. The study included 194 women with type-1 diabetes, 122 women with type-2 diabetes, 779 women who developed GDM and 41 007 non-diabetic controls.
and fetal NT was expressed as a difference from the expected median (D). Main outcome measures Comparison of median MoM maternal
free b-hCG and PAPP-A, and fetal NT, in the four outcome groups. Results There were no significant differences between the groups in median DNT and maternal free b-hCG MoM. Maternal median PAPP-A in type-2 diabetes, compared with the non-diabetic group, was reduced (0.75 MoM, IQR 0.50–1.09 MoM versus 1.00 MoM, IQR 0.68–1.42 MoM; P < 0.001), which resulted in doubling in the false-positive rate in the combined screening in this population. There were no significant differences in maternal PAPP-A between the other groups. Conclusions In women with type-2 diabetes, the estimation of
accurate patient-specific risk in the first trimester combined screening for aneuploidies necessitates an adjustment of maternal serum PAPP-A.
Methods Maternal free b-hCG and PAPP-A levels were
Keywords Diabetes mellitus, free b-hCG, nuchal translucency,
expressed as multiples of the respective normal median (MoM),
PAPP-A, screening for aneuploidy.
Please cite this paper as: Savvidou M, Syngelaki A, Muhaisen M, Emelyanenko E, Nicolaides K. First trimester maternal serum free b-human chorionic gonadotropin and pregnancy-associated plasma protein A in pregnancies complicated by diabetes mellitus. BJOG 2012;119:410–416.
Introduction Effective screening for aneuploidies in the first trimester of pregnancy is provided by a combination of maternal age, fetal crown–rump length (CRL), nuchal translucency (NT) thickness, maternal serum free b-human chorionic gonadotropin (free b-hCG) and pregnancy-associated plasma protein-A (PAPP-A).1 The measurements of free b-hCG and PAPP-A are affected by maternal and pregnancy character-
410
istics, including maternal racial origin, weight, smoking status and method of conception, and these are taken into account in the calculation of ‘multiple of the median’ (MoM) values.1 Some studies have reported that consideration should also be given to pre-existing diabetes mellitus. Four studies examining 35, 79, 489 and 178 women with pre-existing diabetes mellitus reported that the median free b-hCG was 0.74, 0.87, 1.0 and 1.0 MoM, respectively, and that the median PAPP-A MoM was 0.79, 1.00, 0.85 and
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
0.78 (95% CI 0.59–0.98)* 0.86 0.90 (IQR 0.60–1.60) 0.86 Values are expressed as multiples of the median (MoM), and median MoM (IQR; 95% CI) or median (SD) are given when available. *P < 0.05.
1.01 (95% CI 0.98–1.04) 0.99 1.00 (IQR 0.70–1.50) 1.00 0.85 (95% CI 0.69–1.00)* 0.98 0.70 (IQR 0.50–1.20)* 0.85 1.05 (95% CI 1.03–1.07) 1.01 1.20 (IQR 0.80–1.60) 1.05 49 27 228 304 Gestational: Gestational: Gestational: Gestational: 4297 1109 228 5634 Ong et al.2 Tul et al.21 Beneventi et al.22 Overall
– (95% CI 0.00–2.04) (95% CI 0.75–1.16) (log10 SD 0.27) (SD 0.29) – 1.01 (95% CI 0.98–1.04) 1.00 1.01 (log10 SD 0.26) 0.99 (SD 0.24) 1.00 Lower by 0.20 (95% CI 0.03–0.38)* 0.84 (95% CI 0.67_1.00)* 1.02 (95% CI 0.83–1.05) 0.88 (log10 SD 0.25)* 0.93 (SD 0.28) 0.88 – 1.05 (95% CI 1.03–1.07) 1.01 1.03 (Log10 SD 0.23) 0.98 (SD 0.27) 1.02 79 35 79 489 178 860 Pre-existing: Pre-existing: Pre-existing: Pre-existing: Pre-existing: Pre-existing: 93 4297 16 366 83 972 186 104 914
Controls Diabetics Controls
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
Pedersen et al.6 Ong et al.2 Spencer et al.3 Spencer et al.4 Kuc et al.5 Overall
Diabetics
Controls
0.75 0.87 1.01 0.99 0.93
Free b-hCG (MoM) PAPP-A (MoM) Number
This study was part of an ongoing prospective screening study to identify first trimester potential biomarkers of pregnancy complications in women attending for their first routine hospital visit in pregnancy at King’s College Hospital, London, UK. At this visit, which is held at 11+0– 13+6 weeks of gestation, all women have an ultrasound scan to: confirm gestational age from the measurement of the fetal crown–rump length (CRL); diagnose any major fetal abnormalities; and measure the fetal NT thickness as part of screening for chromosomal abnormalities. In addition, the maternal serum free b-hCG and PAPP-A are determined, and the results are combined with maternal age and fetal NT to calculate the patient-specific risk for trisomy 21 and trisomy 13/18.1,9 All blood samples were processed immediately, and an automated machine that provides
Study
Methods
Table 1. First trimester studies that assessed the levels of maternal serum PAPP-A and free b-hCG in women with pre-existing or gestational diabetes mellitus (GDM)
0.94 MoM, respectively (Table 1).2–5 A further study examining 79 women with pre-existing diabetes and assessing only maternal PAPP-A showed that the median PAPP-A level was 0.80 MoM (Table 1).6 But, in all four studies the data from type-1 and type-2 diabetes mellitus were presented together. One recent study examined 331 women with type-1 diabetes mellitus only, and reported that the median maternal serum free b-hCG and PAPP-A were 1.01 and 0.85 MoM, respectively.7 Although type-1 and type-2 diabetes mellitus are considered and managed similarly in pregnancy, they are distinct entities with differences in epidemiology and pathophysiology. Type-1 diabetes mellitus is a pure insulin-deficiency disease that affects mainly children and young people. In pregnancy, recurrent episodes of hypoglycaemia are more likely in type-1 compared with type-2 diabetic women. Conversely, type-2 diabetes mellitus is mainly caused by insulin resistance and relative insulin deficiency, as is gestational diabetes mellitus (GDM), and is more commonly associated with obesity, age, multiparity, non-white racial origin and social deprivation.8 The aim of the current study was to investigate whether markers used in first trimester screening for aneuploidies, including fetal NT, maternal serum free b-hCG and PAPP-A, in women with pre-existing type-1 and type-2 diabetes mellitus are altered. It is established that women who subsequently develop GDM have a metabolic profile similar to the women with type-2 diabetes.8 Certainly the diagnosis of GDM is not known at the time of the first trimester screening, but we sought to investigate whether the above markers used in the first trimester screening for aneuploides are also different in women who subsequently develop GDM, and if this were the case whether these analytes could be used for future prediction of the disease.
Diabetics
First trimester serum free b-hCG and PAPP-A in diabetes
411
Savvidou et al.
reproducible results within 30 minutes was used to measure free b-hCG and PAPP-A (Delfia Express System, Perkin Elmer, Waltham, MA, USA). Written informed consent was obtained from the women agreeing to participate in the study, which was approved by the King’s College Research Ethics Committee. The inclusion criteria for the present study were singleton pregnancy, delivering a phenotypically normal neonate at or after 30 weeks of gestation. We excluded pregnancies with missing outcome data, those ending in miscarriage or termination, those with fetal defects and women who delivered before 30 weeks of gestation because they may not have had screening and diagnosis of GDM.
Maternal history and characteristics Women were asked to complete a questionnaire on maternal age, racial origin (white European, African, South Asian, East Asian and mixed), cigarette smoking during pregnancy (yes or no), method of conception (spontaneous or assisted conception requiring the use of ovulation drugs) and medical history, including pre-pregnancy diabetes mellitus, with a distinction made between types 1 and 2. The questionnaire was then reviewed by a doctor together with the patient and, for the purpose of this study, women were classified as parous or nulliparous if they had no previous pregnancies that reached 24 weeks of gestation or later. Maternal weight and height were measured, and body mass index (BMI) was calculated in kg/m2.
Screening and diagnosis of gestational diabetes mellitus Screening for GDM in our hospital is based on a two-step approach. For all women, plasma glucose was measured randomly at 24–28 weeks of gestation, and if the concentration was at least 6.7 mmol/l an oral glucose tolerance test (OGTT) was carried out within the subsequent 2 weeks. A diagnosis of GDM is made if the fasting plasma glucose level is at least 6 mmol/l, and/or the plasma glucose level 2 h after the oral administration of 75 g of glucose is 7.8 mmol/l or more (WHO). In women with normal random blood sugar, an OGTT is performed if they have persistent glucosuria, develop polyhydramnios or the fetus becomes macrosomic. Women with the diagnosis of GDM are given dietary and exercise advice, and are encouraged to test capillary blood glucose before and 1 h after each meal. If during a period of 1–2 weeks the pre-meal or 1 h post-meal blood glucose level is higher than 5.5 and 7 mmol/l, respectively, the women are treated with insulin. Details of maternal characteristics and the findings of the 11–13 weeks of gestation assessment were recorded in our database, and pregnancy outcomes were added as soon as they became available.
412
Statistical analysis In the cases and controls the measured serum PAPP-A and free b-hCG were converted to multiples of the expected normal median (MoM), corrected for fetal CRL, maternal weight, smoking, parity, racial origin and method of conception, as previously described.1 The measured NT was expressed as a difference from the expected normal median for gestation (4 value).10 Birthweight percentiles were calculated as previously described.11 Comparison between continuous and categorical variables was performed using the Mann–Whitney U-test and a chi-square test or Fisher’s exact test, both with Bonferroni correction, respectively. The risk for trisomy 21 was estimated by combined screening for each pregnancy, and the false-positive rate for each outcome group was estimated at risk cut-offs of 1 in 50, 100 and 150. The statistical software package spss 16.0 (SPSS Inc., Chicago, IL, USA) and prism 5 (GraphPad Software Inc., La Jolla, CA, USA) were used for data analyses.
Results Study population During the study period (from March 2006 to August 2010), first trimester combined screening for aneuploidies was carried out in 45 497 singleton pregnancies. We excluded 3395 cases because they had missing outcome data (n = 1690), the pregnancies resulted in miscarriage, termination or the birth of babies with major defects (n = 1428) or they delivered before 30 weeks of gestation (n = 277). In the remaining 42 102 cases, there were 194 women (0.5%) with type-1 diabetes, 122 women (0.3%) with type-2 diabetes and 779 (1.9%) that subsequently developed GDM, of which 454 (58.3%) were treated by diet alone and 325 (41.7%) were treated by diet in combination with pharmacological intervention. The maternal characteristics of the study population are given in Table 2. In the type-1 diabetes mellitus group, compared with the unaffected group, women had higher BMIs and were more likely to be of white European origin, to have conceived with assisted conception techniques and to have delivered neonates with higher birthweight percentiles. In the type-2 diabetes and GDM groups, compared with the unaffected group, the maternal ages, BMIs and neonatal birthweight percentiles were higher, and more women were parous and of non-white racial origin.
Markers of aneuploidy There was no significant difference in DNT and maternal serum free b-hCG MoM between the outcome groups (Table 2). Similarly, there was no significant difference in the median PAPP-A MoM between the unaffected group and type-1 or GDM groups, even when the latter group
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
First trimester serum free b-hCG and PAPP-A in diabetes
Table 2. Maternal and pregnancy characteristics in the outcome groups Maternal characteristics
Maternal age in years, median (IQR) Maternal body mass index in kg/m2, median (IQR) Crown–rump length in mm, median (IQR) Days of gestation at sampling, median (IQR) Racial origin White European, n (%) African, n (%) South Asian, n (%) East Asian, n (%) Mixed (%) Parity Nulliparous, n (%) Parous, n (%) Cigarette smoker, n (%) Conception Spontaneous, n (%) Assisted, n (%) Nuchal translucency (mm) D nuchal translucency Free b-human chorionic gonadotropin MoM, median (IQR) Pregnancy-associated plasma protein MoM, median (IQR) Birthweight in percentile, median (IQR)
Unaffected pregnancies (n = 41 007)
Type-1 diabetes mellitus (n = 194)
Type-2 diabetes mellitus (n = 122)
Gestational diabetes mellitus (n = 779)
32.1 (27.6–35.8) 24.2 (21.9–27.6)
32.2 (28.2–35.7) 25.8 (23.8–30.2)*
34.5 (29.4–38.6)* 31.2 (26.3–37.2)*
33.7 (29.7–37.5)* 28.3 (24.5–33.5)*
63.8 (58.9–69.2)
63.8 (59.2–70.0)
64.0 (58.6–69.9)
63.3 (58.6–69.0)
89 (87–92)
89 (87–92)
89 (87–92)
89 (87–92)
29 372 7890 1734 867 1144
(71.6) (19.2) (4.2) (2.1) (2.8)
169 16 3 2 4
(87.1)* (8.2)* (1.5) (1.0) (2.1)
44 63 14 0 1
(36.1)* (51.6)* (11.5)* (0.8)
(58.8)* (27.7)* (7.3)* (3.5) (2.7)
19 918 (48.6) 21 089 (51.4) 3313 (8.1)
102 (52.6) 92 (47.4) 19 (9.8)
39 542 (96.4) 1465 (3.6) 1.8 (1.6–2.1) 0.00 ()0.21 to 0.22) 1.00 (0.68–1.52)
179 (92.3) 15 (7.7)* 1.8 (1.6–2.1) )0.02 ()0.20 to 0.22) 1.00 (0.74–1.66)
116 (95.1) 6 (4.9) 1.8 (1.6–2.0) )0.04 ()0.22 to 0.10) 0.86 (0.61–1.42)
746 (95.8) 33 (4.2) 1.9 (1.6–2.1) 0.03 ()0.19 to 0.24) 0.95 (0.64–1.51)
0.91 (0.60–1.59)
0.75 (0.50–1.09)*
0.94 (0.65–1.39)
1.00 (0.68–1.42) 50.8 (27.1–73.3)
81.2 (53.3–95.0)*
31 (25.4) 91 (74.6)* 8 (6.6)
458 216 57 27 21
66.2 (29.1–84.7)*
327 (42.0) 452 (58.0)* 51 (6.5)
62.5 (36.0–87.3)*
Comparison between categorical and continuous variables by chi-square or Fisher’s exact test and Mann–Whitney U-test, both with Bonferroni correction. *Adjusted significance value P < 0.0167.
was split between those treated by diet alone and those requiring pharmacological intervention, which is considered to be a more severe form of glucose intolerance (0.95, IQR 0.68–1.45 MoM versus 0.93, IQR 0.62–1.30 MoM; P = 0.211). However, women with type-2 diabetes demonstrated a significant 25% reduction in the levels of PAPP-A MoM compared with the unaffected group (Table 2), which was apparent in all the different maternal racial origin and weight groups (Table 3). If women with pre-existing type-1 and type-2 diabetes were considered together, they would have had a 16% significant reduction in maternal serum PAPP-A, compared with the unaffected group (0.84, IQR 0.58–1.29 MoM versus 1.00, IQR 0.68–1.42 MoM; P < 0.001).
Performance of screening In the screened population of 45 497 pregnancies, there were 221 cases of trisomy 21, and the estimated risk for trisomy 21 from combined screening was 1 in 50 or more
in 200 cases (90.5%), 1 in 100 or more in 210 cases (95.0%) and 1 in 150 or more in 213 cases (96.3%). The respective values at these risk cut-offs for the euploid pregnancies according to status of diabetes mellitus are shown in Table 4. The false-positive rates for pregnancies with type-1 diabetes mellitus, GDM and no diabetes were similar, but for type-2 diabetes the rates were increased (Table 4).
Discussion The study has demonstrated that in women with preexisting type-2 diabetes mellitus maternal serum PAPP-A at 11–13 weeks of gestation is decreased by 25%, with a consequent doubling in the false-positive rate of first trimester combined screening for trisomy 21. In type-1 diabetes there was a non-significant decrease in serum PAPP-A by 9%. When we combined type-1 and type-2 diabetes mellitus in our study the overall reduction in serum PAPP-A was
ª 2012 The Authors BJOG An International Journal of Obstetrics and Gynaecology ª 2012 RCOG
413
Savvidou et al.
Table 3. The maternal PAPP-A values according to maternal racial origin (three main groups) and weight in the unaffected and type-2 diabetic pregnancies Maternal characteristic
Racial origin White Black South Asian Maternal weight
