- Email: [email protected]
Serum progesterone and human chorionic gonadotropin in very early pregnancy: implications for clinical management
European Journal of Obstetrics & Gynecology and Reproductive Bioloa, Elsevier
EUROBS
32 (1989) 71-77
71
00782
Serum progesterone and human chorionic gonadotropin in very early pregnancy: implications for clinical management Paul A. Riss,
Karl
Radivojevic
and
Christian
Bieglmayer
.?nd Department of Obstetrics and Gynecology, Universrty Hospital, Vienna, Austria Accepted
for publication
15 November
1988
Summary We studied beta-human chorionic gonadotropin (beta-HCG) and progesterone (P) in maternal serum in 71 patients with a positive urine pregnancy test with a threshold of 50 U HCG/liter urine and an ultrasound examination without demonstrable fetal heart beat. Patients were divided into three groups: ongoing intra-uterine pregnancy of more than 12 weeks duration (n = 23) ectopic pregnancy (n = 14), and abnormal intra-uterine pregnancy (blighted ovum, early spontaneous abortion) resulting in dilatation and curettage (n = 34). We found that both beta-HCG and progesterone levels were significantly lower in abnormal pregnancies but that there was less overlap between the progesterone values in normal, ectopic and abnormal pregnancy as compared to beta-HCG levels. There was a positive correlation between beta-HCG levels < 5000 mu/ml and serum progesterone in ectopic pregnancies but not in normal or abnormal intrauterine pregnancies. The size of the gestational sac correlated with serum beta-HCG levels but not with serum progesterone levels in normal intra-uterine pregnancies. We conclude that serum progesterone levels (cut-off point 15 ng/ml) in very early pregnancy can be helpful to differentiate between normal intra-uterine pregnancy and abnormal intra-uterine or ectopic pregnancy. Serum progesterone;
Human
chorionic
gonadotropin;
Ectopic
pregnancy;
Early pregnanq
Introduction Immediately secretes human
Correspondence:
after implantation of the blastocyst chorionic gonadotropin (HCG) into
P.A. Riss, 2nd Univ. Frauenklinik,
0028-2243/89/$03.50
0 1989 Elsevier Science
Spitalgasse
Publishers
the developing trophoblast the maternal circulation [2].
23. A-1090
B.V. (Biomedical
Vienna.
Division)
Austria.
12
This hormone can easily be assayed by sensitive, commercially available, urinary tests which provide positive readings at the time of the expected menstruation. HCG maintains and stimulates the corpus luteum and as a consequence the secretion of progesterone (P). Ectopic and abnormal intrauterine pregnancies are associated with low HCG and low progesterone (P) levels [3,6,15]. It would be highly desirable in clinical practice to be able to differentiate between normal intra-uterine and pathological pregnancy before fetal heart beats can be seen on ultrasound. We therefore tested whether maternal serum levels of progesterone can be used for diagnostic purposes in very early pregnancy. In addition we compared progesterone levels in very early pregnancies to HCG levels and to the size of the gestational sac in intra-uterine pregnancies.
Patients and methods
71 patients were included in our study (Table I). The patients presented consecutively at the out-patient clinic of our department during a 5 month period. The criteria for inclusion in the study were a positive urine pregnancy test (Tandem Icon HCG, Threshold 50 U HCG/liter urine) as well as an ultrasound examination without demonstrable fetal heart beat. Patients who had an induced abortion of a viable pregnancy were excluded from the study. Patients were followed at our department until one of three possible outcomes was confirmed: ongoing intrauterine pregnancy of more than 12 weeks duration (n = 23), ectopic pregnancy (n = 14), and abnormal intra-uterine pregnancy (blighted ovum, early spontaneous abortion) resulting in dilatation and curettage (n = 34). At the time of the first visit a venous blood sample was collected and beta-HCG and progesterone levels were measured by radio-immunoassay (commercial kits from Serono, FRG, and Coat a Count, U.S.A., respectively). On the same day, an ultrasound examination was done and the following parameters were noted: presence of a gestational sac, presence of an embryo within the sac, demonstrable fetal heart beat, adnexal pathology, presence of fluid in the pouch of Douglas. The mean diameter of the gestational sac was calculated from three ultrasound measurements. Correlation coefficients were calculated by a spreadsheet on personal computer [4], thresholds for significance were taken from the tables in Wissenschaftliche Tabellen Geigy [ 141.
TABLE
I
Patients
admitted
into the study (mean,
range)
Diagnosis
n
Age
Week of gestation
Normal pregnancy Ectopic pregnancy Abnormal pregnancy
23 14 34
23.9 (18-34) 27.1 (15-37) 26.9 (18-47)
6.2 (5-8) 7.5 (5-12) 9.2 (6-12)
Total
71
26.0 (15-47)
7.9 (5-12)
73
Results Fig. 1 shows the median values of beta-HCG and progesterone in the three groups. beta-HCG scattered over an extremely wide range in all groups with considerable overlap. The variations of serum progesterone values were less pronounced. It has been suggested that a progesterone value of 15 ng/ml could be used as a cut-off point between normal and ectopic pregnancy [6]. In our study 3/23 patients with normal pregnancies had progesterone values of < 15 ng/ml, and l/14 patients with ectopic pregnancy had a progesterone value of > 15 ng/ml. Progesterone values for abnormal intra-uterine pregnancies ranged from 0.4 to 27.9 ng/ml. A preliminary calculation of the overall correlation coefficient between beta-HCG and progesterone values gave r = 0.280 (p < 0.05 for 69 degree of freedom (df)). We then looked separately at the three groups and at the relationship between beta-HCG and progesterone at different levels of chorionic gonadotropin, choosing as the cut-off points for beta-HCG the median values from our laboratory for pregnancy week 5, 6 and 7 (Table II). It is interesting to note that the only significant correlation between beta-HCG and progesterone was found in ectopic pregnancies with very low levels of beta-HCG. In intra-uterine pregnancies the size of the gestational sac correlated with the serum levels of beta-HCG but not P. A detailed analysis revealed that in normal intra-uterine pregnancies the correlation coefficient between size of the gestational sac and beta-HCG was r = 0.526 (p < 0.05 for 19 df) and in abnormal intra-uterine pregnancies r = 0.391 (p < 0.01 for 20 df). Low progesterone values were a very sensitive indicator of ectopic pregnancies (93%), but less so for abnormal intra-uterine pregnancy (79%) (Table III). It is evident that low progesterone values alone are not sufficient to differentiate between
TABLE
II
Correlation between serum beta-HCG and progesterone in (0) normal intra-uterine abnormal intra-uterine pregnancy resulting in curettage, and (E) ectopic pregnancy. coefficient; x-axis, serum beta-HCG; y-axis, serum progesterone. r
Type of pregnancy
n
c3000
0 C E
6 13 10
0.749 0.053 0.827
(5000
0 C E
10 18 12
- 0.332 0.436 0.642
<16000
0 C E
17 27 13
0.120 0.360 0.140
HCG (mu/ml)
NS. not significant.
Regression
line
pregnancy, (C) r, correlation
P
y = 0.005000x + 24.0 y = 0.000351x + 5.1 y = 0.015160x + 0.5
-c 0.1 NS i 0.01
y = -0.005000x + 27.8 y = 0.001685x + 3.8 y = 0.002418x + 3.8
NS < 0.1 < 0.05
y = 0.005000x + 24.0 y = 0.000351x+ 5.1 y = 0.015160x+0.5
NS i 0.01 NS
74
Serum
Chorionic
Gonadotropin
(N-71)
20 19 -
tt+t+
+
tt++t+
18 -
+
17 18 -
+
15 -
+
14 -
+
+
13 12 -
+
+
11 -
+ +
+
10 9-
-+-
8-
+ +
7-
*
65-
+ +
43-
: +
21-
0
+
+
A
a I
-1
+ +
+ I
1
1
I
2
1 =normol
3
2=ectopic
Serum
50 -
; +
+
3=obnormol
Progesterone
t
40 -
+ t
30 -
* --
20 -
+
; +
*
t !
+
: +
+ $ 10 0
B
1
+
1
1 1 1 =normal
Fig. 1. Median
values of (A) serum beta-HCG,
-*-
+1
l
--
,
1
2 2=cctopic
,
3 3=obnormol
and (B) serum progesterone
in early pregnancy.
TABLE
III
Sensitivity,
specificity,
Serum progesterone
and predictive (P) and outcome
value (pred. val.) of serum progesterone of pregnancy
P i 15 ng/ml + ectopic pregnancy sensitivity specificity pred. val. pos. pred. val. neg. P i 15 ng/ml --f abnormal intra-uterine pregnancy sensitivity specificity pred. val. pos. pred. val. neg. P > 15 ng/ml --) normal intra-uterine pregnancy sensitivity specificity pred. val. pos. pred. val. neg.
and abnormal intra-uterine diagnoses was only around 50%. very sensitive to and specific for The possible clinical usefulness were 93% for a low progesterone pregnancy, and 71% for a high pregnancy (Table III). It should been influenced by the possible viable pregnancies.
ectopic
in very early pregnancy
93% 47% 30% 96% 79% 57% 63% 75% 87% 83% 71% 93%
pregnancy, since specificity with regard to both On the other hand high progesterone values were a normal intra-uterine pregnancy (87% and 83%). can be derived from the predictive values, which value and a diagnosis of ectopic or abnormal progesterone value and the presence of a normal be added that the predictive values would have inclusion of patients with induced abortions of
Discussion While it has recently been shown that tiny amounts of human chorionic gonadotropin are secreted in all normal adults in a pulsatile fashion, probably by the pituitary, the demonstration of beta-HCG in larger amounts is generally taken to indicate the presence of trophoblastic tissue somewhere in the patient [9]. Progesterone, on the other hand, is thought to provide some quantitative information on the ‘vitality’ of the pregnancy: withdrawal of progester0r.e support of the early pregnancy by antagonization results in termination of the pregnancy and has been used as a contraceptive method [8,13]. The production of progesterone by the corpus luteum depends on the maintenance of the corpus luteum by chorionic gonadotropin [12]. Eventually, at about the 8th week of gestation, the so-called luteo-placental shift occurs and the placenta takes over production of progesterone until delivery of the fetus [lO,ll]. Data from artificially induced pregnancies in the absence of ovaries indicate that placental takeover may occur later [5,7].
Our study of the very early pregnancy focuses on the time period of about 2 weeks between the detection of urinary beta-HCG levels and the demonstration of a fetal heart beat by ultrasound. Since our patients were taken consecutively from the gynecologic out-patient clinic of our department, we were able to calculate predictive values reflecting the actual situation in clinical practice. When we looked at the relationship between beta-HCG and progesterone, we found a positive correlation only in ectopic pregnancies with low levels of beta-HCG. Obviously, in normal intrauterine pregnancies, beta-HCG rises so rapidly that the progesterone levels do not follow, maybe indicating a supramaximal stimulation of the corpus luteum by the trophoblast in early intra-uterine pregnancies. In ectopic pregnancies, on the other hand, it is possibly not the corpus luteum which responds poorly to stimulation by beta-HCG, but insufficient synthesis of beta-HCG by the trophoblast. Our findings confirm and extend the observation of other authors, who found that P levels in ectopic pregnancies are lower than would be expected in a normal pregnancy with similar beta-HCG levels’[3,12]. The observation that the size of the gestational sac correlates with maternal beta-HCG levels but not with maternal progesterone levels is further evidence of the lack of direct relationship between chorionic gonadotropin and progesterone. Ectopic pregnancy can certainly not be diagnosed on the basis of progesterone measurements alone. Indeed, low progesterone values are non-specific in regard to a diagnosis of ectopic pregnancy (Table III). However, we wanted to know whether a normal intra-uterine pregnancy can be diagnosed by endocrine methods before the pregnancy can be visualized by ultrasound. Though the sensitivity and specificity of high progesterone values with regard to normal intra-uterine pregnancy were over 808, the predictive value of the positive test in our patient sample was only 71%. One explanation for the seemingly low predictive value is the fact that the mean duration of pregnancy varied between the three study groups (Table I), which in turn was due to our excluding all intra-uterine pregnancies with a demonstrable fetal heart beat. Recent advances in ultrasound technology have improved the possibilities of diagnostic imaging in very early pregnancy [l]. In particular the use of a vaginal ultrasound probe often permits the differential diagnosis between a gestational sac indicative of intra-uterine pregnancy - and a decidual sac, which is highly suspicious of ectopic pregnancy. Furthermore, a second ultrasound examination after 1 week will confirm the presumptive diagnosis in most cases. To summarize, we found less overlap between the progesterone values in normal, ectopic and abnormal pregnancies compared to beta-HCG levels. Low beta-HCG levels (less than 5000 mu/ml) and serum progesterone levels showed a positive correlation in ectopic pregnancies but did not correlate in normal or abnormal intra-uterine pregnancies. The size of the gestational sac correlated with beta-HCG levels but not with serum progesterone levels in normal intra-uterine pregnancies. Though the differential diagnosis between ectopic and abnormal intra-uterine pregnancies cannot be made on the basis of biochemical methods, serum progesterone levels in very early pregnancy can be used as a screening test because of the predictive value of high progesterone for normal intra-uterine pregnancy.
References
9 10 11 12 13 14 15
Bernaschek G, Rudelstorfer R, Csaicsich P. Vaginal sonography versus serum human chorionic gonadotropin in early detection of pregnancy. Am J Obstet Gynecol 1988;158:608-612. Deutinger J, Neumark J, Reinthaller A, Riss P, Miiller-Tyl E, Fischl F, Bieglmayer Ch, Janisch H. Pregnancy-specific parameters in early pregnancies after in vitro fertilization: prediction of the course of pregnancy. Fertil Steril 1986;46:77-80. Hubinot CJ, Thomas C, Schwers JF. Luteal function in ectopic pregnancy. Am J Obstet Gynecol 1987;156:669-674. Jeanty P. Performing linear regressions with l-2-3. PC Magazine Ott 16, 1984;243-248. Lutjen P, Trounson A, Leeton J, Findlay J, Wood C, Renou P. The establishment and maintenance of pregnancy using in vitro fertilization and embryo donation in a patient with primary ovarian failure. Nature 1984;307:174-175. Matthews CP, Coulson PB, Wild RA. Serum progesterone levels as an aid in the diagnosis of ectopic pregnancy. Obstet Gynecol 1986;68:390-394. Navot D, Laufer N, Kopolovic J, Rabinowitz R, Birkenfeld A, Lewin A, Granat M, Margalioth EJ. Schenker JG. Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N Engl J Med 1986;314:806-811. Nieman LK, Choate TM, Chrousos GP, Healy DL, Morin M, Renquist D, Merriam GR, Spitz IM, Bardin CW, Baulieu EE, Leriaux DL. Progesterone antagonist RU-486: a potential new contraceptive agent. N Engl J Med 1987;316:187-191. Ode11 WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688-1691. Runnebaum B, Rabe T. Zur Regulation der plazentaren Progesteronsynthese beim Menschen. Geburtsh Frauenheilk 1983;43:87-92 (Sonderheft). Ryan KJ. Placental synthesis of steroid hormones. In: Tulchinsky D, Ryan KJ, eds. Maternal-fetal endocrinology. Philadelphia: WB Saunders, 1980:3-16. Sauer MV, Gorill MJ, Rodi IA, Yeko TR, Buster JE. Corpus luteum activity in tubal pregnancy. Obstet Gynecol 1988;71:667-670. Vervest HAM, Haspels AA. Preliminary results with the antiprogestational compound RU-486 (mifepristone) for interruption of early pregnancy. Fertil Steril 1985;44:627-632. Wissenschaftliche Tabellen Geigy. Bd. 3, Statistik. Basel: Ciba-Geigy, 1980. Yeko TR, Gorill MJ, Hughes LH, Rodi IA, Buster JE, Sauer MV. Timely diagnosis of early ectopir pregnancy using a single blood progesterone measurement. Fertil Steril 1987;48:1048-1050.
EUROBS
32 (1989) 71-77
71
00782
Serum progesterone and human chorionic gonadotropin in very early pregnancy: implications for clinical management Paul A. Riss,
Karl
Radivojevic
and
Christian
Bieglmayer
.?nd Department of Obstetrics and Gynecology, Universrty Hospital, Vienna, Austria Accepted
for publication
15 November
1988
Summary We studied beta-human chorionic gonadotropin (beta-HCG) and progesterone (P) in maternal serum in 71 patients with a positive urine pregnancy test with a threshold of 50 U HCG/liter urine and an ultrasound examination without demonstrable fetal heart beat. Patients were divided into three groups: ongoing intra-uterine pregnancy of more than 12 weeks duration (n = 23) ectopic pregnancy (n = 14), and abnormal intra-uterine pregnancy (blighted ovum, early spontaneous abortion) resulting in dilatation and curettage (n = 34). We found that both beta-HCG and progesterone levels were significantly lower in abnormal pregnancies but that there was less overlap between the progesterone values in normal, ectopic and abnormal pregnancy as compared to beta-HCG levels. There was a positive correlation between beta-HCG levels < 5000 mu/ml and serum progesterone in ectopic pregnancies but not in normal or abnormal intrauterine pregnancies. The size of the gestational sac correlated with serum beta-HCG levels but not with serum progesterone levels in normal intra-uterine pregnancies. We conclude that serum progesterone levels (cut-off point 15 ng/ml) in very early pregnancy can be helpful to differentiate between normal intra-uterine pregnancy and abnormal intra-uterine or ectopic pregnancy. Serum progesterone;
Human
chorionic
gonadotropin;
Ectopic
pregnancy;
Early pregnanq
Introduction Immediately secretes human
Correspondence:
after implantation of the blastocyst chorionic gonadotropin (HCG) into
P.A. Riss, 2nd Univ. Frauenklinik,
0028-2243/89/$03.50
0 1989 Elsevier Science
Spitalgasse
Publishers
the developing trophoblast the maternal circulation [2].
23. A-1090
B.V. (Biomedical
Vienna.
Division)
Austria.
12
This hormone can easily be assayed by sensitive, commercially available, urinary tests which provide positive readings at the time of the expected menstruation. HCG maintains and stimulates the corpus luteum and as a consequence the secretion of progesterone (P). Ectopic and abnormal intrauterine pregnancies are associated with low HCG and low progesterone (P) levels [3,6,15]. It would be highly desirable in clinical practice to be able to differentiate between normal intra-uterine and pathological pregnancy before fetal heart beats can be seen on ultrasound. We therefore tested whether maternal serum levels of progesterone can be used for diagnostic purposes in very early pregnancy. In addition we compared progesterone levels in very early pregnancies to HCG levels and to the size of the gestational sac in intra-uterine pregnancies.
Patients and methods
71 patients were included in our study (Table I). The patients presented consecutively at the out-patient clinic of our department during a 5 month period. The criteria for inclusion in the study were a positive urine pregnancy test (Tandem Icon HCG, Threshold 50 U HCG/liter urine) as well as an ultrasound examination without demonstrable fetal heart beat. Patients who had an induced abortion of a viable pregnancy were excluded from the study. Patients were followed at our department until one of three possible outcomes was confirmed: ongoing intrauterine pregnancy of more than 12 weeks duration (n = 23), ectopic pregnancy (n = 14), and abnormal intra-uterine pregnancy (blighted ovum, early spontaneous abortion) resulting in dilatation and curettage (n = 34). At the time of the first visit a venous blood sample was collected and beta-HCG and progesterone levels were measured by radio-immunoassay (commercial kits from Serono, FRG, and Coat a Count, U.S.A., respectively). On the same day, an ultrasound examination was done and the following parameters were noted: presence of a gestational sac, presence of an embryo within the sac, demonstrable fetal heart beat, adnexal pathology, presence of fluid in the pouch of Douglas. The mean diameter of the gestational sac was calculated from three ultrasound measurements. Correlation coefficients were calculated by a spreadsheet on personal computer [4], thresholds for significance were taken from the tables in Wissenschaftliche Tabellen Geigy [ 141.
TABLE
I
Patients
admitted
into the study (mean,
range)
Diagnosis
n
Age
Week of gestation
Normal pregnancy Ectopic pregnancy Abnormal pregnancy
23 14 34
23.9 (18-34) 27.1 (15-37) 26.9 (18-47)
6.2 (5-8) 7.5 (5-12) 9.2 (6-12)
Total
71
26.0 (15-47)
7.9 (5-12)
73
Results Fig. 1 shows the median values of beta-HCG and progesterone in the three groups. beta-HCG scattered over an extremely wide range in all groups with considerable overlap. The variations of serum progesterone values were less pronounced. It has been suggested that a progesterone value of 15 ng/ml could be used as a cut-off point between normal and ectopic pregnancy [6]. In our study 3/23 patients with normal pregnancies had progesterone values of < 15 ng/ml, and l/14 patients with ectopic pregnancy had a progesterone value of > 15 ng/ml. Progesterone values for abnormal intra-uterine pregnancies ranged from 0.4 to 27.9 ng/ml. A preliminary calculation of the overall correlation coefficient between beta-HCG and progesterone values gave r = 0.280 (p < 0.05 for 69 degree of freedom (df)). We then looked separately at the three groups and at the relationship between beta-HCG and progesterone at different levels of chorionic gonadotropin, choosing as the cut-off points for beta-HCG the median values from our laboratory for pregnancy week 5, 6 and 7 (Table II). It is interesting to note that the only significant correlation between beta-HCG and progesterone was found in ectopic pregnancies with very low levels of beta-HCG. In intra-uterine pregnancies the size of the gestational sac correlated with the serum levels of beta-HCG but not P. A detailed analysis revealed that in normal intra-uterine pregnancies the correlation coefficient between size of the gestational sac and beta-HCG was r = 0.526 (p < 0.05 for 19 df) and in abnormal intra-uterine pregnancies r = 0.391 (p < 0.01 for 20 df). Low progesterone values were a very sensitive indicator of ectopic pregnancies (93%), but less so for abnormal intra-uterine pregnancy (79%) (Table III). It is evident that low progesterone values alone are not sufficient to differentiate between
TABLE
II
Correlation between serum beta-HCG and progesterone in (0) normal intra-uterine abnormal intra-uterine pregnancy resulting in curettage, and (E) ectopic pregnancy. coefficient; x-axis, serum beta-HCG; y-axis, serum progesterone. r
Type of pregnancy
n
c3000
0 C E
6 13 10
0.749 0.053 0.827
(5000
0 C E
10 18 12
- 0.332 0.436 0.642
<16000
0 C E
17 27 13
0.120 0.360 0.140
HCG (mu/ml)
NS. not significant.
Regression
line
pregnancy, (C) r, correlation
P
y = 0.005000x + 24.0 y = 0.000351x + 5.1 y = 0.015160x + 0.5
-c 0.1 NS i 0.01
y = -0.005000x + 27.8 y = 0.001685x + 3.8 y = 0.002418x + 3.8
NS < 0.1 < 0.05
y = 0.005000x + 24.0 y = 0.000351x+ 5.1 y = 0.015160x+0.5
NS i 0.01 NS
74
Serum
Chorionic
Gonadotropin
(N-71)
20 19 -
tt+t+
+
tt++t+
18 -
+
17 18 -
+
15 -
+
14 -
+
+
13 12 -
+
+
11 -
+ +
+
10 9-
-+-
8-
+ +
7-
*
65-
+ +
43-
: +
21-
0
+
+
A
a I
-1
+ +
+ I
1
1
I
2
1 =normol
3
2=ectopic
Serum
50 -
; +
+
3=obnormol
Progesterone
t
40 -
+ t
30 -
* --
20 -
+
; +
*
t !
+
: +
+ $ 10 0
B
1
+
1
1 1 1 =normal
Fig. 1. Median
values of (A) serum beta-HCG,
-*-
+1
l
--
,
1
2 2=cctopic
,
3 3=obnormol
and (B) serum progesterone
in early pregnancy.
TABLE
III
Sensitivity,
specificity,
Serum progesterone
and predictive (P) and outcome
value (pred. val.) of serum progesterone of pregnancy
P i 15 ng/ml + ectopic pregnancy sensitivity specificity pred. val. pos. pred. val. neg. P i 15 ng/ml --f abnormal intra-uterine pregnancy sensitivity specificity pred. val. pos. pred. val. neg. P > 15 ng/ml --) normal intra-uterine pregnancy sensitivity specificity pred. val. pos. pred. val. neg.
and abnormal intra-uterine diagnoses was only around 50%. very sensitive to and specific for The possible clinical usefulness were 93% for a low progesterone pregnancy, and 71% for a high pregnancy (Table III). It should been influenced by the possible viable pregnancies.
ectopic
in very early pregnancy
93% 47% 30% 96% 79% 57% 63% 75% 87% 83% 71% 93%
pregnancy, since specificity with regard to both On the other hand high progesterone values were a normal intra-uterine pregnancy (87% and 83%). can be derived from the predictive values, which value and a diagnosis of ectopic or abnormal progesterone value and the presence of a normal be added that the predictive values would have inclusion of patients with induced abortions of
Discussion While it has recently been shown that tiny amounts of human chorionic gonadotropin are secreted in all normal adults in a pulsatile fashion, probably by the pituitary, the demonstration of beta-HCG in larger amounts is generally taken to indicate the presence of trophoblastic tissue somewhere in the patient [9]. Progesterone, on the other hand, is thought to provide some quantitative information on the ‘vitality’ of the pregnancy: withdrawal of progester0r.e support of the early pregnancy by antagonization results in termination of the pregnancy and has been used as a contraceptive method [8,13]. The production of progesterone by the corpus luteum depends on the maintenance of the corpus luteum by chorionic gonadotropin [12]. Eventually, at about the 8th week of gestation, the so-called luteo-placental shift occurs and the placenta takes over production of progesterone until delivery of the fetus [lO,ll]. Data from artificially induced pregnancies in the absence of ovaries indicate that placental takeover may occur later [5,7].
Our study of the very early pregnancy focuses on the time period of about 2 weeks between the detection of urinary beta-HCG levels and the demonstration of a fetal heart beat by ultrasound. Since our patients were taken consecutively from the gynecologic out-patient clinic of our department, we were able to calculate predictive values reflecting the actual situation in clinical practice. When we looked at the relationship between beta-HCG and progesterone, we found a positive correlation only in ectopic pregnancies with low levels of beta-HCG. Obviously, in normal intrauterine pregnancies, beta-HCG rises so rapidly that the progesterone levels do not follow, maybe indicating a supramaximal stimulation of the corpus luteum by the trophoblast in early intra-uterine pregnancies. In ectopic pregnancies, on the other hand, it is possibly not the corpus luteum which responds poorly to stimulation by beta-HCG, but insufficient synthesis of beta-HCG by the trophoblast. Our findings confirm and extend the observation of other authors, who found that P levels in ectopic pregnancies are lower than would be expected in a normal pregnancy with similar beta-HCG levels’[3,12]. The observation that the size of the gestational sac correlates with maternal beta-HCG levels but not with maternal progesterone levels is further evidence of the lack of direct relationship between chorionic gonadotropin and progesterone. Ectopic pregnancy can certainly not be diagnosed on the basis of progesterone measurements alone. Indeed, low progesterone values are non-specific in regard to a diagnosis of ectopic pregnancy (Table III). However, we wanted to know whether a normal intra-uterine pregnancy can be diagnosed by endocrine methods before the pregnancy can be visualized by ultrasound. Though the sensitivity and specificity of high progesterone values with regard to normal intra-uterine pregnancy were over 808, the predictive value of the positive test in our patient sample was only 71%. One explanation for the seemingly low predictive value is the fact that the mean duration of pregnancy varied between the three study groups (Table I), which in turn was due to our excluding all intra-uterine pregnancies with a demonstrable fetal heart beat. Recent advances in ultrasound technology have improved the possibilities of diagnostic imaging in very early pregnancy [l]. In particular the use of a vaginal ultrasound probe often permits the differential diagnosis between a gestational sac indicative of intra-uterine pregnancy - and a decidual sac, which is highly suspicious of ectopic pregnancy. Furthermore, a second ultrasound examination after 1 week will confirm the presumptive diagnosis in most cases. To summarize, we found less overlap between the progesterone values in normal, ectopic and abnormal pregnancies compared to beta-HCG levels. Low beta-HCG levels (less than 5000 mu/ml) and serum progesterone levels showed a positive correlation in ectopic pregnancies but did not correlate in normal or abnormal intra-uterine pregnancies. The size of the gestational sac correlated with beta-HCG levels but not with serum progesterone levels in normal intra-uterine pregnancies. Though the differential diagnosis between ectopic and abnormal intra-uterine pregnancies cannot be made on the basis of biochemical methods, serum progesterone levels in very early pregnancy can be used as a screening test because of the predictive value of high progesterone for normal intra-uterine pregnancy.
References
9 10 11 12 13 14 15
Bernaschek G, Rudelstorfer R, Csaicsich P. Vaginal sonography versus serum human chorionic gonadotropin in early detection of pregnancy. Am J Obstet Gynecol 1988;158:608-612. Deutinger J, Neumark J, Reinthaller A, Riss P, Miiller-Tyl E, Fischl F, Bieglmayer Ch, Janisch H. Pregnancy-specific parameters in early pregnancies after in vitro fertilization: prediction of the course of pregnancy. Fertil Steril 1986;46:77-80. Hubinot CJ, Thomas C, Schwers JF. Luteal function in ectopic pregnancy. Am J Obstet Gynecol 1987;156:669-674. Jeanty P. Performing linear regressions with l-2-3. PC Magazine Ott 16, 1984;243-248. Lutjen P, Trounson A, Leeton J, Findlay J, Wood C, Renou P. The establishment and maintenance of pregnancy using in vitro fertilization and embryo donation in a patient with primary ovarian failure. Nature 1984;307:174-175. Matthews CP, Coulson PB, Wild RA. Serum progesterone levels as an aid in the diagnosis of ectopic pregnancy. Obstet Gynecol 1986;68:390-394. Navot D, Laufer N, Kopolovic J, Rabinowitz R, Birkenfeld A, Lewin A, Granat M, Margalioth EJ. Schenker JG. Artificially induced endometrial cycles and establishment of pregnancies in the absence of ovaries. N Engl J Med 1986;314:806-811. Nieman LK, Choate TM, Chrousos GP, Healy DL, Morin M, Renquist D, Merriam GR, Spitz IM, Bardin CW, Baulieu EE, Leriaux DL. Progesterone antagonist RU-486: a potential new contraceptive agent. N Engl J Med 1987;316:187-191. Ode11 WD, Griffin J. Pulsatile secretion of human chorionic gonadotropin in normal adults. N Engl J Med 1987;317:1688-1691. Runnebaum B, Rabe T. Zur Regulation der plazentaren Progesteronsynthese beim Menschen. Geburtsh Frauenheilk 1983;43:87-92 (Sonderheft). Ryan KJ. Placental synthesis of steroid hormones. In: Tulchinsky D, Ryan KJ, eds. Maternal-fetal endocrinology. Philadelphia: WB Saunders, 1980:3-16. Sauer MV, Gorill MJ, Rodi IA, Yeko TR, Buster JE. Corpus luteum activity in tubal pregnancy. Obstet Gynecol 1988;71:667-670. Vervest HAM, Haspels AA. Preliminary results with the antiprogestational compound RU-486 (mifepristone) for interruption of early pregnancy. Fertil Steril 1985;44:627-632. Wissenschaftliche Tabellen Geigy. Bd. 3, Statistik. Basel: Ciba-Geigy, 1980. Yeko TR, Gorill MJ, Hughes LH, Rodi IA, Buster JE, Sauer MV. Timely diagnosis of early ectopir pregnancy using a single blood progesterone measurement. Fertil Steril 1987;48:1048-1050.