Australian and New Zealand Journal of Obstetrics and Gynaecology 2006; 46: 413 418

Original Article

Blackwell Publishing Asia

Yolk sac in early pregnancy loss

The quality and size of yolk sac in early pregnancy loss

Fu-Nan CHO,1 San-Nung CHEN,1 Ming-Hong TAI2 and Tsung-Lung YANG3
1

Department of Obstetrics and Gynecology, 2Department of Medical Education and Research and 3Department of Radiology,
Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan

Abstract
Background: Accurate differentiation between normal pregnancy and pregnancy loss in early gestation remains a
clinical challenge.
Aims: To determine whether ultrasound findings of yolk sac size and morphology are valuable in relation to
pregnancy loss at six to ten weeks gestation.
Methods: Transvaginal ultrasonography was performed in 111 normal singleton pregnancies, 25 anembryonic
gestations, and 18 missed abortions. Mean diameters of gestational sac and yolk sac were measured. The relationship
between yolk sacs and gestational sacs in normal pregnancies was depicted. The yolk sacs ultrasound findings in
cases of pregnancy loss were recorded.
Results: In normal pregnancies with embryonic heartbeats, a deformed or an absent yolk sac was never detected.
Sequential appearance of yolk sac, embryonic heartbeats and amniotic membrane was essential for normal pregnancy.
The largest yolk sac in viable pregnancies was 8.1 mm. Findings in anembryonic gestations included an absent yolk
sac, an irregular-shaped yolk sac and a relatively large yolk sac (> 95% upper confidence limits, in 11 cases). In
cases of missed abortion with prior existing embryonic heartbeats, abnormal findings included a relatively large, a
progressively regressing, a relatively small, and a deformed yolk sac (an irregular-shaped yolk sac, an echogenic spot,
or a band).
Conclusion: A very large yolk sac may exist in normal pregnancy. When embryonic heartbeats exist, the poor quality
and early regression of a yolk sac are more specific than the large size of a yolk sac in predicting pregnancy loss. When
an embryo is undetectable, a relatively large yolk sac, even of normal shape, may be an indicator of miscarriage.
Key words: gestational sac, miscarriage, pregnancy loss, transvaginal ultrasonography, yolk sac.

Introduction

Materials and methods

Accurate differentiation between normal pregnancy and

pregnancy loss in early gestation remains a clinical challenge.
Previous studies have described the association between
embryonic well-beings and the characteristics of gestational
sac,1,2 yolk sac,38 amniotic cavity,9,10 and embryonic
heartbeats.11,12 Despite the advent of three-dimensional
ultrasound,1315 two-dimensional transvaginal ultrasound
remains a simple and convenient way to assess pregnancy status.
There are discriminatory criteria in predicting spontaneous
pregnancy losses, including 8-mm mean diameter of
gestational sac (MD-GS) without a visible yolk sac4 and
1620-mm MD-GS without embryonic heartbeats;16,17
however, various exceptions to these criteria exist.1,18 In
addition, pregnancies with a very large yolk sac are almost
always associated with poor outcomes.7 This study was
designed to determine the value of yolk sac findings in
predicting early pregnancy outcomes.

One hundred and fifty-four women with a singleton pregnancy

between six and ten weeks gestation were enrolled in this
study from June 1998 to May 2004. Gestational age was based
on one or more of the following criteria: (i) last menstrual
period if the woman had a regular menstrual cycle; (ii) basal
body temperature charts; (iii) sperm insemination date;
(iv) or the crownrump length (CRL) from early ultrasound
if available. The institutional human ethics committee in
our hospital had approved this research.

Correspondence: Dr Fu-Nan Cho, Department of Obstetrics

and Gynecology, Kaohsiung Veterans General Hospital, 386
Ta-Chung 1st Road, Kaohsiung, Taiwan 813.
Email: fncho@isca.vghks.gov.tw
DOI: 10.1111/j.1479-828X.2006.00627.x
Received 16 February 2006; accepted 06 June 2006.

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists

413

F.-N. Cho et al.

All patients received transvaginal ultrasonography (ATL,

HDI-3000, Washington, USA; 7-MHz transducer) by the
same senior staff (Fu-Nan Cho) member to eliminate interobserver variance. Patients presenting with obvious bleeding
(including incomplete miscarriage), incomplete data (including three cases lost to follow up), and examination by other
staff members were excluded from this study. One hundred
and eleven normal cases were followed until 36 weeks
gestation or birth. Forty-three cases with poor outcome were
identified as either miscarriage without a visible embryo
(anembryonic gestation, at adjusted gestation > 6+4 weeks
or at least 18 days after positive urine pregnancy test) or
miscarriage with a visible embryo lacking heartbeats (missed
abortion, with a CRL > 5 mm or adjusted gestation > 6+4
weeks). The mean values of the three inner diameters of
gestational sac and yolk sac were measured. The CRL and
embryonic heart rate, if detectable, were measured at the same
time. Serum beta-human chorionic gonadotropin (-HCG)
(3rd IS WHO 75/537, BRAHMS, Germany) and follow-up
ultrasound examinations (in three to ten days) were performed
in cases with uncertain diagnosis and unknown dating to
confirm the poor outcomes. The relationship between yolk
sac and gestational sac in normal cases was determined, with
the mean, upper, and lower 95% confidence limits being
calculated and depicted. The findings in cases with miscarriage
were presented and analysed.

Figure 1 The relationship between the mean diameter of the

yolk sac (MD-YS) and the gestational sac (MD-GS) in 111
normal cases with a living embryo. The mean, upper, and lower
95% confidence limits are indicated by three curves. YS diameter
= 2.73 + 0.0623(GS diameter) 0.0028(GS diameter2) + 0.000047
(GS diameter3); r = 0.549, n = 110, P < 0.0001. Of note, with
normal embryonic heartbeats, six yolk sacs are located above the
curve of the upper 95% confidence limits.

Results
Findings in normal pregnancies
In 111 normal pregnancies with embryonic heartbeats, an
absent or a deformed yolk sac was never detected. The yolk
sacs were nearly spherical in shape. The mean diameter of
yolk sac (MD-YS) increased progressively with advancing
gestation (Fig. 1). Six of 111 (5.8%) cases were above the upper
95% confidence limits. The largest MD-YS was 8.1 mm with
a CRL of 28.7 mm. The smallest MD-YS was 2.3 mm, with
a living embryo detected. The smallest MD-GS with a living
embryo was 9.3 mm. The smallest CRL with positive heartbeats
was 3 mm. The embryonic heart rates increased with
advancing gestation. The slowest embryonic heart rate was
100 b.p.m. with a CRL of 3 mm. The fastest embryonic
heart rate was 192 b.p.m. with a CRL of 35 mm.

Findings in 25 cases of miscarriage without a

visible embryo (anembryonic gestation)
1 Twelve cases without a visible yolk sac: The diagnosis was
made only when the fertilisation date or serum-HCG value
was available. Of these cases, the smallest MD-GS was 11 mm
when pregnancy loss was confirmed (-HCG: 6210 miu/mL).
2 Thirteen cases with a visible yolk sac: One case showed
an irregular-shaped yolk sac that turned into an echogenic
band eventually. Another case showed a normal yolk sac
(2.8 mm in MD). The remaining 11 cases had a normal
shape, but relatively large yolk sacs (4.2, 4.7, 4.9, 5.3, 5.4,
414

Figure 2 The relationship between yolk sac and gestational sac

in 25 cases of miscarriage without a visible embryo. Three
curves are derived from Figure 1. Of note, in the absence of a
visible living embryo, a relatively large yolk sac may be an
indicator of miscarriage.

5.9, 6.2, 6.3, 7.8, 8.6, and 8.7 mm in MD, respectively)

in relation to the gestational sac (beyond upper 95%
confidence limits) (Figs 2, 3). Regression of the yolk sac
(to 6.8 mm) was found four days later in the last case.
The size of yolk sac in most normal pregnancies (105/
111) was within upper and lower 95% confidence limits. The
yolk sacs in 24 anembryonic gestations were noted beyond
upper or lower 95% confidence limits, including a relatively
large yolk sac, a deformed yolk sac, and an absent yolk sac.
A very small gestational sac (relative to the gestational sacs

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists; 46: 413 418

Yolk sac in early pregnancy loss

Figure 3 Ultrasound findings in a case of anembryonic gestation.

Without a visible embryo, a very large yolk sac (arrow; 8.7 mm
in mean diameter) may render the diagnosis of miscarriage.

of normal pregnancies at the same gestation) was shown in

three cases of anembryonic gestation without a visible yolk sac.

Findings in 18 cases of miscarriage with a

visible embryo lacking heartbeat
The data are summarised in Table 1. In case one, a yolk sac
was undetectable at first. Three days later, an echogenic
spot of yolk sac and embryonic heartbeat were detected.

Unfortunately, miscarriage occurred over the subsequent seven

days. In case two, a yolk sac was invisible initially. After one week,
a yolk sac was visible, but without embryonic heartbeat.
Regression of the yolk sac occurred over the next seven days.
In case three, the embryo was undetectable with MD-YS of
3.3 mm. Miscarriage was diagnosed ten days later. In case four,
a small gestational sac and a very large yolk sac (6.6 mm in
MD) were found with normal embryonic heartbeat. The yolk
sac turned into an echogenic band with normal embryonic
heartbeat six days later (Fig. 4). In case five, a relatively large
yolk sac was noted with positive heartbeat. After one week,
miscarriage was noted. Regression of the yolk sac was noted
over the next week. In cases six to 12, a deformed yolk sac and
an absent yolk sac were noted. In case 13, a relatively large
amniotic cavity (11.1 mm in mean diameter) with a very small
embryo was found. In case 14, a relatively large amniotic cavity
(14 mm in mean diameter) with a small CRL was noted.
In cases 15 and 16, the amniotic membrane was visible,
but neither yolk sac nor embryonic heartbeat was detected.
Interestingly, the amniotic cavity continued to grow even without
embryonic heartbeat. The yolk sac appeared as an echogenic
spot six days later in case 15 (Fig. 5). In case 17, slow embryonic
heartbeat and a small yolk sac were noted. Miscarriage was
found ten days later. In case 18, pregnancy loss was unexpectedly noted. These findings showed that, even with positive
embryonic heartbeats, miscarriage ensued when associated

Table 1 The detailed information in 18 cases of miscarriage with an embryo

Patients
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma
Ma

1
1.1
1.2
2
2.1
2.2
3
3.1
4
4.1
5
5.1
5.2
6
7
8
9
10
11
12
13
14
15
15.1
16
17
18

CRL

FHB

YS

5
7.6

100

Spot
Spot

5.7

4.4
4.3
8.9
5.1
6.8
6.8
11.8
16.5
8.5
9.7
19.6
7.6
3.9
3
6.7
13.2
14
5.5
5.7
8.8

120
150
100

170

4.6
3.4
3.3
3.4
6.6
Band
5.3
3.6
2.6
2.7
3.8
2.2

1.7
2.2
Spot
85
160

1.9
3.7

GS
10.0
17.3
20.6
19.3
24.7
25.3
12.7
19.3
9.0
10.0
18.0
19.3
20.3
27.4
36.3
17.7
25.0
28.3
20.7
21.6
17.6
28.0
36.0
32.7
9.8
15.3
26.7

Remarks
Invisible YS and embryo
Echogenic YS
Echogenic YS, EHB ()
-HCG = 65000 miu/mL
Large YS
Regressive YS, no embryo
Large YS, no embryo
Absolutely large YS
Echogenic band of YS
Large YS
Regression of YS
Regression of YS
Deformed YS
Deformed YS. No EHB later
Deformed YS
Invisible YS
Invisible YS
Invisible YS
Invisible YS
Large AC
Large AC
Visible AC (13.3), invisible YS
Growing AC (18.3), echogenic YS
Visible AC, invisible YS
Slow EHB, small YS
No EHB later

Ma1, first case of miscarriage; Ma1.1, the first follow up; Ma 1.2, the second follow up. EHB, embryonic heartbeats; CRL, crownrump length;
YS, yolk sac; GS, gestational sac; and AC, amniotic cavity were stated as mm in mean diameter.

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists; 46: 413 418

415

F.-N. Cho et al.

Figure 5 A case of miscarriage with a yolk sac shown as an

echogenic spot (large arrow). The amniotic cavity (small arrows)
kept growing despite negative embryonic heartbeats (the
embryo was not visible in this view).

Figure 6 Diagram of the relationship between yolk sac (YS)

and gestational sac in 18 cases of miscarriage with a visible
embryo lacking heartbeats. Six cases with serial sonography are
depicted by arrows. Yolk sacs are typical of abnormal quality,
especially as an echogenic spot, an echogenic band, or a
progressively regressing YS. For cases with YS located in the
normal range, they were associated (with exception of one case)
with other abnormal findings, such as a deformed YS, a relatively
large amniotic cavity, and slower embryonic heartbeats.
Figure 4 Ultrasound findings of a case with miscarriage. (A) A
large yolk sac (6.6 mm in mean diameter; arrowheads) in a small
gestational sac; an embryo (+) with heartbeats (120 b.p.m.).
(B) The yolk sac turned into an echogenic band (arrow) 6 days
later with an embryo (F). (C) Pregnancy loss ensued despite
normal heartbeats (150 b.p.m.; large arrows).

with a relatively large, a progressively regressing, a relatively small

(with slow heartbeats), and a deformed yolk sac (an irregularshaped yolk sac, an echogenic spot, or a band) (Fig. 6).

Discussion
Sequential appearance of a yolk sac, embryonic heartbeat,
and an amniotic cavity is essential for normal pregnancy. In
416

our cases with a live birth, a deformed yolk sac never occurred.
In addition, from observation of normal pregnancies in our and
another reports,4 totalling 382 cases, embryonic heartbeats
have always been detectable when a yolk sac reaches 3.3 mm
in mean diameter. We believe that, if a living embryo is not
detected by transvaginal ultrasound, the presence of a
relatively large yolk sac (beyond upper 95% confidence limits,
4.2 mm in mean diameter at least), even with normal shape, can
still lead to the diagnosis of miscarriage, as in this study there
were no cases of successful pregnancy following a ultrasound
scan reporting a yolk sac greater than the 95th centile unless
a viable embryo was also seen during the assessment.
Of note, a relatively large yolk sac, with normal shape and
normal embryonic heartbeat, should not be hastily identified

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists; 46: 413 418

Yolk sac in early pregnancy loss

Table 2 Brief summary of published reports regarding the characteristics of yolk sac in pregnancy loss using two-dimensional transvaginal
ultrasound
Authors

Gestation

Cases number

Kurtz et al.3

< 12 weeks

Lindsay et al.4

< 10 weeks

Stampone et al.5

< 12 weeks

Rowling et al.1

< 13 weeks

163 normal,
49 miscarriage
327 normal,
159 miscarriage
101 normal,
16 miscarriage
NA

Kucuk et al.7

< 12 weeks

Mara and Foster6

710 weeks

Chama et al.8

< 12 weeks

Our study

6 10 weeks

Study, sonologist

219 normal,
31 miscarriage
1 case report

Prospective,
multiple
Prospective,
multiple
Prospective,
multiple
Retrospective,
multiple
Prospective,
multiple

70 normal,
35 miscarriage
111 normal,
43 miscarriage

Prospective,
multiple
Prospective,
one staff member

Remarks
The presence or absence of YS is
not a predicting sign
Miscarriage occur with (i) YS > 5.6 mm,
(ii) absent YS in a 8-mm GS
Predict miscarriage by > or < 2 SD
of YS diameter
Live birth may occur with absent YS
in a 8-mm GS
Predict miscarriage by > or < 2 SD
of YS diameter
Embryonic death when associated
with YS regression
Predict miscarriage by > or < 2 SD
of YS diameter
An echogenic spot or a band,
A large YS in anembryonic gestation,
Detailed follow-up findings available

GS, gestational sac; YS, yolk sac; SD, standard deviation.

NA, not available because of mixed results from transabdominal or vaginal scan.

as an abnormal pregnancy. In the normal cases of our study,

the maximal MD-YS was 8.1 mm, which is higher than the
value of 5.6 mm as reported in a previous study.4 To our
knowledge, this is the largest yolk sac associated with a live
birth.20 In this case, the placenta was implanted into the septum
of a bicornuate uterus. The unusual growth of the yolk sac
may be influenced by environmental factors.20
To precisely diagnose pregnancy failure in an empty
gestational sac (without an embryo and a yolk sac), the
discriminatory gestational date should be confirmed.17 A
significant difference between the size of the gestational sac
in normal and abnormal pregnancies should appear after
six weeks gestation (8.2 vs 4.5 mm) due in part to the different
vascularisation.2 In one of our normal cases, neither a living
embryo nor a yolk sac was detectable initially with a MD-GS
of 10 mm. Accordingly, the diagnosis of anembryonic
gestation without a yolk sac should not be made if a MD-GS
is 10 mm or less, which is larger than the value of 8 mm
mentioned in previous studies,1,4 unless the adjusted gestation
reaches 6+4 weeks. This is possibly attributed to the fact
that some normal cases may have a relatively fast-growing
gestational sac.
The findings of an invisible yolk sac and a deformed yolk
sac have been reported as typical signs of pregnancy loss.46
Furthermore, as a deformed yolk sac has never occurred in
normal cases, it may be used to predict pregnancy loss
before embryonic heartbeats appear.19 The abnormal yolk
sac findings in viable pregnancies (with normal embryonic
heartbeat) that went on miscarriage include a large yolk sac,
an echogenic spot, an echogenic band, an irregular-shape
yolk sac, and an early regressing yolk sac during the followup examination. Besides, the existence of a visible amniotic
cavity, but without a living embryo, has been reported as a

sign of early miscarriage.911 We observed that an absent yolk

sac or a deformed yolk sac has been associated with this sign.
Interestingly, the growth of amniotic cavity can still proceed
despite embryo demise. To predict early pregnancy loss,
use of the integration of abnormal findings of yolk sac,
gestational sac and amniotic cavity may be more helpful
and convincible. These concepts may be exploited to clinical
practice in embryo reduction. Our study is limited by a small
number of patients with pregnancy loss. Further prospectively
large-scale studies are warranted to define the new criteria
fit for new high-resolution ultrasound equipment. A brief
summary of previous reports is showed in Table 2.

Conclusion
The sequential appearance of yolk sac, embryonic heartbeat
and amniotic membrane is essential for normal pregnancy.
In normal pregnancy, a very large yolk sac may exist. When
embryonic heartbeats exist, the poor quality and early
regression of the yolk sac may be more specific than a large
yolk sac in predicting early pregnancy loss. When an embryo
is undetectable, a relatively large yolk sac, even of normal
shape, may be an indicator of miscarriage.

References
1 Rowling SE, Coleman BG, Langer JE, Arger PH, Nisenbaum HL,
Horii SC. First-trimester US parameters of failed pregnancy.
Radiology 1997; 203: 211217.
2 OH JS, Wright G, Coulam B. Gestational sac in very early
pregnancy as a predictor of fetal outcome. Ultrasound Obstet
Gynecol 2002; 20: 267269.

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists; 46: 413 418

417

F.-N. Cho et al.

3 Kurtz AB, Needleman L, Pennell RG, Baltarowich O, Vilaro M,
Goldberg BB. Can detection of the yolk sac in the first trimester
be used to predict the outcome of pregnancy? AJR Am J
Roentgenol 1992; 158: 843847.
4 Lindsay DJ, Lovett IS, Lyons EA et al. Yolk sac diameter and
shape at endovaginal US: Predictors of pregnancy outcome in
the first trimester. Radiology 1992; 183: 115118.
5 Stampone C, Nicotra M, Muttinelli C, Cosmi EV. Transvaginal
sonography of the yolk sac in normal and abnormal pregnancy.
J Clin Ultrasound 1996; 24: 39.
6 Mara E, Foster GS. Spontaneous regression of a yolk sac
associated with embryonic death. J Ultrasound Med 2000; 19:
655656.
7 Kucuk T, Duru NK, Yenen MC, Dede M, Ergun A, Baser I.
Yolk sac size and shape as predictors of poor pregnancy
outcome. J Perinat Med 1999; 27: 316 320.
8 Chama CM, Marupa JY, Obed JY. The value of the secondary
yolk sac in predicting pregnancy outcome. J Obstet Gynaecol
2005; 25: 245247.
9 Horrow MM. Enlarged amniotic cavity. A new sonographic
sign of early embryonic death. AJR Am J Roentgenol 1992;
158: 359362.
10 McKenna KM, Feldstein VA, Goldstein RB, Filly RA. The
empty amnion: A sign of early pregnancy failure. J Ultrasound
Med 1995; 14: 117121.
11 Benson CB, Doubliet PM. Embryonic heart rate in early first
trimester: indicators of poor pregnancy outcome. Radiology
1994; 192: 343344.

418

12 Martinez JM, Comas C, Ojuel J, Borrell A, Puerto B,

Firtuny A. Fetal heart rate patterns in pregnancies with
chromosomal disorders or subsequent fetal loss. Obs Gyn 1996;
87: 118 121.
13 Kupesic S, Kurjak A, Ivani-Koauta M. Volume and vascularity
of the yolk sac studied by three-dimensional ultrasound and
color Doppler. J Perinat Med 1999; 27: 9196.
14 Shih JC, Jaffe R, Hsieh FJ. Three-dimensional ultrasonography
in early pregnancy. Seminar Perinat 2001; 25: 3 10.
15 Figueras F, Torrents M, Munoz A et al. Three-dimensional
yolk and gestational sac volume. A prospective study of
prognostic value. J Reprod Med 2003; 48: 252256.
16 Schouwink MH, Fong BF, Mol BW, van der Veen F.
Ultrasonographic criteria for non-viability of first trimester
intra-uterine pregnancy. Early Pregnancy [Electronic Resource]
2000; 4: 203203.
17 Nyberg DA, Filly RA. Predicting pregnancy failure in empty
gestational sacs. Ultrasound Obstet Gynecol 2003; 21: 9 12.
18 Falco P, Zagonari S, Gabrielli S, Bevini M, Pilu G, Bovicelli
L. Sonography of pregnancies with first-trimester bleeding
and a small intrauterine gestational sac without a demonstrable
embryo. Ultrasound Obstet Gynecol 2003; 21: 62 65.
19 Howe RS, Isaacson KJ, Albert JL, Coutifaris CB. Embryonic
heart rate in human pregnancy. J Ultrasound Med 1991; 10:
367371.
20 Cho FN, Kan YY, Chen SN, Yang TL, Hsu PH. Very large
yolk sac and bicornuate uterus in a live birth. J Chin Med Assoc
2005; 68: 535 537.

2006 The Authors

Journal compilation 2006 The Royal Australian and New Zealand College of Obstetricians and Gynaecologists; 46: 413 418