Correspondence: Mr D. Jurkovic, Early Pregnancy and Gynaecology Ultrasound Unit, Department of Obstetrics and Gynaecology, King's College Hospital, Denmark Hill, London SE5 8RX, UK.
In a number of pregnant women ovarian cysts are found incidentally during the routine first trimester scan. These cysts may pose diagnostic difficulties, and the measurement of serum CA125 levels can be used to aid management. In this study we measured maternal serum CA125 levels in 188 women with uncomplicated pregnancies between 11–14 weeks of gestation. All women had morphologically normal ovaries observed on ultrasound examination. The median serum CA125 levels were 23.4 U/mL (range 2.2–166.3 U/mL, 95% reference interval 5.28–70.15) and did not change significantly with gestation. We conclude that CA125 levels are increased at 11–14 weeks of gestation and cut off values which are used to assess the nature of ovarian cysts in nonpregnant women cannot be applied to pregnant women at this gestation.
In pregnancy most women are offered an ultrasound scan in the first trimester, either for measurement of fetal crown-rump length in order to date the pregnancy or for measurement of fetal nuchal translucency as a method of screening for chromosomal abnormalities1. The incidental detection of ovarian cysts during this routine scan can pose major difficulties in terms of diagnosis and subsequent management. In the majority of cases the cysts are physiological (corpora lutea cysts) and will resolve spontaneously. In some women, however, the cysts persist and malignancy needs to be excluded. In such cases measurement of serum CA125 is used widely, as a second line investigation, because elevated levels are observed in more than 80% of patients with ovarian cancer2. However, serum levels of CA125 are altered in pregnancy3 and therefore non-pregnant reference ranges cannot be used for the investigation of cysts observed in pregnancy. In this study we establish reference range for maternal serum CA125 at 11–14 weeks, when routine nuchal translucency screening is carried out.
At King's College Hospital women attending for routine antenatal care are offered screening for chromosomal defects by a combination of fetal nuchal translucency thickness and maternal serum free β-human chorionic gonadotrophin and pregnancy associated plasma protein-A at 11–14 weeks of gestation4. The ultrasound examinations are carried out transabdominally (Aloka SSD-2000 and 1700, Aloka Co, Tokyo, Japan). In all cases the fetal crown-rump length and nuchal translucency thickness are measured and an attempt is made to examine the ovaries. During a three month period (April-June 1999), the maternal blood sample taken from women with morphologically normal ovaries for measurement of β-human chorionic gonadotrophin and pregnancy associated plasma protein-A was also used for measurement of CA125 by radio-immunoassay (Kryptor, CIS bio international, France). The detection limit of the assay was 1 U/mL, the coefficients of intra-assay and inter-assay variance at the levels of 32.5 U/mL and 279 U/mL were 2%, 1.1% and 4.5%, 1.1%, respectively.
During the study period 291 women were examined, but 103 (35%) were excluded for the following reasons: both ovaries were not visualised (n= 92), one or both of the ovaries contained a cyst of > 2 cm in diameter (n= 4) or a blood sample was not taken for CA125 levels (n= 7). In the remaining 188 cases the median maternal age was 30 years (range 17–41 years) and the median gestation was 12 weeks (range 11–14 weeks). The median serum CA125 was 23.4 U/mL (range 2.2–166.3 U/mL, 95% reference interval 5.28–70.15). CA125 did not change significantly with gestation (Fig. 1).
This study provides a reference range for maternal serum CA125 at 11–14 weeks, which may prove to be useful in the evaluation of ovarian cysts identified incidentally during routine ultrasound examination at this gestation. The median value was 23.4 U/mL and did not change significantly within this narrow gestational range. Previous cross-sectional studies of serum CA125 in heterogeneous groups of women, including some who were pregnant, noted that the levels are higher in early pregnancy compared with nonpregnant women. The increased maternal serum levels in the first trimester are largely attributed to increased production by the decidua3. Haga et al.5 measured CA125 in 71 women at 5–41 weeks of gestation and found that the median value in the first 10 weeks was 70 U/mL, decreasing thereafter to about 20 U/mL. Similarly, Niloff et al.6 examined 81 women and reported that the levels were above 65 U/mL in 16% of those examined before 12 weeks and in none of those examined later in pregnancy.
In nonpregnant women a serum CA125 concentration > 35 U/mL is found in more than 80% cases of epithelial ovarian carcinoma compared, with only 1% of apparently healthy individuals; the corresponding values for a cut off of 65 U/mL are 75% and 0.2%1. Our data indicate that at 11–14 weeks of gestation the cut offs of 35 U/mL and 65 U/mL were exceeded by about 20% and 4%, respectively, of women with morphologically normal ovaries. These data demonstrate that in the investigation of ovarian cysts in pregnancy the nonpregnant cut offs cannot be used. If the discriminatory power of CA125 in pregnancy is similar to that in nonpregnant women, a more appropriate cut off in pregnancy may be 112 U/mL, which corresponds to the 99th centile in our study.