Some Misconceptions and Pitfalls in Ultrasonography

Ultrasound Quarterly
Vol. 17, No. 3, pp. 129–155

© 2001 Lippincott Williams & Wilkins, Inc., Philadelphia
Some Misconceptions and Pitfalls in Ultrasonography

Hsu-Chong Yeh, M.D.
Professor of Radiology, Department of Radiology, Mount Sinai School of Medicine and Mount Sinai New York University
Medical Center, New York, New York, U.S.A.
Summary: There are many misconceptions in the field of ultrasonography. Some are
because of poor understanding of anatomy and/or embryological or developmental
features of certain organs, which have prevailed over the course of the centuries since
the discovery of these entities (e.g., column of Bertin and hypertrophic column of
Bertin). Some misconceptions derived from misinterpretation (e.g., double decidual
sac sign) or inadequate observation of ultrasonographic findings. (e.g., hyperechoic
stroma in polycystic ovarian syndrome). Key Words: Hypertrophic column of
Bertin—Milk of calcium–Renal cyst—Double decidual sac sign—Hashimoto thyroid-
itis—Polycystic ovarian syndrome.
Objectives: After reading this article and completing the embryologic development of the kidney. The adult kidney
posttest, the learner should be able to: consists of approximately 14 lobes. During early fetal life,
● Summarize the anatomic detail and some developmental such lobes are like small individual kidneys called reniculi
features of certain common misconceptions of entities in (or renunculi) (Fig. 2). Each reniculus is like a small kidney
the kidney, early pregnancy, thyroid, ovary, diaphragm, that consists of its own cortical and medullary tissues and its
and abdominal aorta. own blood supply and collecting system. The central med-
● Describe ultrasonographic findings based on a better un- ullary tissue (pyramid) is enveloped on all sides by cortical
derstanding of the aforementioned anatomy, therefore tissue, except where the papilla (i.e., the tip of the pyramid)
avoiding misdiagnosis and achieving accurate diagnosis emerges. There are two parts to the cortical tissue: 1) cen-
of these entities. trilobe cortex (cortical arch)3 that covers the base of the
pyramid; and 2) mural cortex that wraps around the sides of
There are a few misconceptions pertaining to the column of
the pyramid. When the reniculi fuse to form a single kidney,
Bertin and renal lobes. The term “column of Bertin” is a
the centrilobar cortex of adjacent reniculi fuse to form the
misnomer. When French anatomist Dr. Exupère Joseph
renal cortex, whereas the mural cortices of adjacent reniculi
Bertin1 first described the cortical tissue between pyramids
fuse to form the cloisonné (septum) of Bertin. Because mu-
in 1744, he called it cloisonné, which means partition or
ral cortices surround each pyramid, when multiple reniculi
septum. However, the term “cloisonné” was mistranslated
fuse together, mural cortices naturally form honeycomb-like
as column in English, and column of Bertin remains popu-
partitions.
lar. Hodson’s1 attempt to revert it back to being called the
There are numerous lobules within the renal cortex and
cloisonné of Bertin was in vain. Only a few people referred
septum of Bertin, approximately 20,000 lobules in the entire
to it as the septum of Bertin.2 It is evident that the column
kidney.3 The lobules radiate from the surface of the pyramid
is not the real feature of this structure if one look at a
outward in each reniculus; hence, in an adult kidney, the
tangential section of the kidney cortex (Fig. 1). The cortical
lobules are perpendicular to the renal surface within the
tissue between the pyramids form a honeycomb-like parti-
renal cortex but parallel to the renal surface within the septa
tion, and the pyramids are embedded within this honey-
of Bertin. The lobules contain renal tubules (cortical rays or
comb. This can also be easily understood if one studies the
medullary rays of Ferrin).3 Within the interlobular spaces
are straight interlobular arteries and veins. Along the surface
of the pyramid are arcuate arteries that give rise to inter-
The author has disclosed that he has no significant relationship with or
financial interest in any commercial companies pertaining to this educa- lobular arteries.
tional activity.
Address correspondence and reprint requests to: Dr. Hsu-Chong Yeh, HYPERTROPHIC COLUMN OF BERTIN
Department of Radiology, Box 1234, Mount Sinai—NYU Medical Center,
One Gustave L. Levy Place, New York, NY 10029. E-mail: hsu- A 2- to 3-cm mass-like structure is not uncommonly seen
chong.yeh@mountsinai.org in the kidney, usually at the junction between the upper and
129
130 H. C. YEH
TABLE 1. Previous terms used to describe

junctional parenchyma
Origin or nature
of the tissue Terms References
Hypertrophic or Hypertrophic column of 4–6
abnormally large Bertin
tissue Focal cortical hyperplasia 11
Large cloison 12
Malposition or displaced Lobar dysmorphism 7,8
tissue Malposition of renal lobe 7
Infolding cortical mass 13
Cortical invagination 10
with prominent
FIG. 1. Septa of Bertin, not column of Bertin. Tangential scan through column of Bertin
the lateral aspect of the renal cortex of a transplanted kidney showing Kidney within a kidney 12
septa of Bertin with a honeycomb appearance, but not the round cross-
section of a “column.” Multiple, round, hypoechoic pyramids are embed- Mass or pseudomass Renal pseudotumor 9,10
ded in each space of the honeycomb; therefore, “column of Bertin” is a Glomerular zone 14
misnomer. pseudotumor
Renal cortical nodule 4
Primary cortical nodule 15
middle thirds of the kidney. This has been observed in in- Intermediate cortical 3
mass
travenous urography and ultrasonography. Many terms (at
least 18) have been coined to name this structure.4–18 They Embryologic anomaly Aberrant lobule of renal 16
tissue
are listed in Table 1. The most commonly used terms are Benign cortical rest 17
hypertrophic column of Bertin,4–6 lobar dysmorphism,7,8 Cortical island of kidney 17
and renal pseudotumor.9,10 Most of the previously described Developmental Abortive attempt at 8
ultrasonographic features for diagnosing this entity are non- abnormality duplication of
renal parenchyma
specific2,5,6,19 They include: 1) lesion indenting on the renal
sinus laterally; 2) renal sinus engulfing the lesion in a claw- Supernumerary lobe Accessory renal lobe 18
like fashion (split sinus sign);2 3) clearly defined from the
renal sinus; 4) maximum dimension less than 3 cm; 5) con- will have the same echogenicity as the renal cortex. This
tinuous with the renal cortex; 6) echogenicity close to that presumption, based on common sense, may appear correct,
of the renal cortex; 7) engulfs papilla;2 and 8) companion but it is another misconception. In fact, the normal septa of
cortical projection.6 The first to fourth features are nonspe- Bertin is usually much more echogenic than renal cortex
cific and can be seen in any mass or mass-like lesion. The because of the anisotropic effect. This will be discussed in
fifth and sixth features are based on the presumption that it more detail later. The seventh feature is also called impris-
is a hypertrophic septum of Bertin. Because the septum of oned papilla and has been seen in angiography. It was de-
Bertin is a cortical tissue, as is the renal cortex, it naturally scribed as the central lucency sign8 in the angiographic
FIG. 2. Formation of renal lobe, cortex, and

septum of Bertin from the reniculus. A. The
reniculus (renunculus) consists of a central
core of medulla enveloped by cortical tissue.
B. The cortical tissue is divided into a centri-
lobar and a mural cortex. C. A group of ren-
iculi (two are shown here) fuse to form a kid-
ney, and each reniculus becomes a renal lobe.
Two layers of adjacent mural cortex fuse to
form the column (septum) of Bertin, whereas
the centrilobar cortex fuses to form the renal
cortex. The interlobular arteries run perpen-
dicular to the renal surface in the renal cortex,
but parallel to the renal surface in the column
of Bertin. The renal lobules are also in the
same orientation as the interlobular arteries, as
shown in A. a ⳱ artery.
Ultrasound Quarterly, Vol. 17, No. 3, 2001

MISCONCEPTIONS IN ULTRASONOGRAPHY 131
literature. Because a pyramid (the tip of which is the papilla) from the upper pole to the lower pole, and from one side to
surrounded by cortical tissue forms a renal lobe, the term the other on sagittal or coronal scanning, we have found
“lobar dysmorphisor” was coined based on this finding in- many kidneys that appear to form by fusing of the superior
dicating that this is a large and malpositioned renal lobe. and inferior kidneys.20 We called these two kidneys supe-
The eighth feature is an ancillary sign that is seen in some rior and inferior subkidneys. The upper pole of the inferior
cases when the junctional parenchyma extends to the medial subkidney usually overlaps the lower pole of the superior
side of the kidney. subkidney. The former is located anterior to the latter at the
The so-called hypertrophic column of Bertin actually rep- area of overlap. The so-called hypertrophic column of
resents unresorbed polar parenchyma of either the superior Bertin represents unresolved renal parenchyma at the junc-
or inferior subkidney20 (Figs. 3–5), and not a single large tion of two subkidneys; therefore, we call it junctional pa-
septum of Bertin. After studying 15- to 30-mm thick em- renchyma.20 Within the junctional parenchyma, there are
bryo resin cases of the kidney, Fine and Keson21 found that normal renal cortex, pyramids, and septa of Bertin. Between
most kidneys develop by means of fusion of two kidneys the two subkidneys, an echogenic line may be seen. This is
(upper and lower ones), which correspond to the upper and called the junctional parenchymal line.20 A groove on the
lower groups of calyces. Although calyces are not usually surface of the kidney formed by this line appears as a small,
clearly delineated in ultrasonography, the echogenic renal triangular defect that is called the junctional parenchymal
sinus fat in which the calyceal system is situated can be defect. When all elements and features of junctional paren-
clearly seen by ultrasonography in a normal kidney. A kid- chyma (e.g., renal cortex, pyramids, septa of Bertin, junc-
ney is composed of echogenic sinus fat enveloped by renal tional parenchymal line, and junctional parenchymal defect)
parenchyma. The renal parenchyma consists of renal cortex, are identified and two fused subkidneys are recognized on
pyramids, and septa of Bertin, which are all less echogenic ultrasonography, the diagnosis of junctional parenchyma
than the renal sinus fat. By carefully scanning the kidney can be certain, and a tumor can be excluded. Even if two
FIG. 3. A pseudotumor caused by

junctional parenchyma. A. A hy-
poechoic, mass-like lesion (m) is
seen in the kidney. B. Slightly differ-
ent scanning plane shows a promi-
nent pyramid (arrowhead) in the le-
sion. Therefore, the lesion most
likely represents junction paren-
chyma (JP). C. Further scanning
shows a junctional parenchymal de-
fect (arrowhead) and junctional pa-
renchyma line (arrow), confirming
that the lesion is indeed junctional
parenchyma (JP). The renal sinus fat
of the upper (U) and lower (L) sub-
kidneys appeared separate. Junc-
tional parenchyma (JP) represents the
lower parenchyma of the superior
subkidney; therefore, this is a supe-
rior subkidney type of junctional pa-
renchyma. D. Transverse scan shows
junctional parenchyma (JP) to be lo-
cated in the lateral aspect of the kid-
ney between the upper (U) and lower
(L) subkidney sinus fat. Arrow ⳱
junctional parenchymal line; black
arrowhead ⳱ junctional parenchy-
mal defect. E. Diagrammatic drawing
corresponding to C.

132 H. C. YEH
FIG. 4. Example of a gross specimen. The drawing from a kidney speci-

men with a double collecting system shows inferior subkidney type of FIG. 5. Junctional parenchyma simulating a mass. A. A hyperechoic
junctional parenchyma. The double collecting system is known to fre- mass (arrowhead) is seen partly within the renal sinus. B. Scanning slightly
quently be associated with junctional parenchyma. Note that a pyramid more laterally shows a pyramid (P) in the mass and a defect (arrow) on the
(arrow) appears to be cutoff or deformed by the junctional parenchyma renal surface, associated with a demarcating line (arrowheads). This rep-
(arrowheads). From: Yeh HC, et al. Junctional parenchyma: revised defi- resents junctional parenchyma of inferior subkidney type.
nition of hypertrophic column of Bertin. Radiology 1992;185: 725–32.
the renal collecting system is dilated. However, it is quite

subkidneys are not identified and a pyramid or pyramids are possible that a bifid renal pelvis may be present if there is no
seen within it, junctional parenchyma can be highly sug- duplication of the pyelocalyceal system. In general, there is
gested. Because all elements of the tissue are normal, there higher incidence of double or bifid renal collecting systems
is nothing hypertrophic or tumor-like about it. It is not a in the kidneys with junctional parenchyma.
displaced, malpositioned, or aberrant renal lobe or an em- When two subkidneys fuse, either the upper polar paren-
bryonic rest. It is not an infolding or invagination of cortical chyma of the inferior subkidney or the lower polar paren-
tissue as previously described. In fact, all those previously chyma of the superior subkidney may be resorbed. The
created terms pertaining to this entity as listed in Table 1 are remaining unresorbed polar parenchyma, which represents
based on incorrect concepts. junctional parenchyma, may be the lower polar parenchyma
Because junctional parenchyma is located between two of the superior subkidney (type 1, the superior subkidney
overlapping poles of two subkidneys, on transverse scan, it type) or the upper polar parenchyma of inferior subkidney
can be seen located between two echogenic fatty renal si- (type 2, the inferior subkidney type), or it may be a mixed
nuses in the lateral aspect of the kidney (Fig. 3). The echo- type (type 3) in which two junctional parenchymas are seen
genic renal sinus fat of the two subkidneys are usually fused because of a large amount of overlapping between two sub-
at the renal hilum. Occasionally, the fat of the two renal kidneys and partial (i.e., anterior half or posterior half) re-
sinuses are completely separate, and transverse scan through sorption of these overlapping parts (Fig. 6).
this plane shows only renal parenchymal tissue, and a sag- The pyramid adjacent to the junctional parenchyma may
ittal or coronal scan will show a band-like area of junctional appear partly cut-off by the junctional parenchyma, and
parenchyma completely transecting the kidney. This has sometimes it even appears displaced or distorted by the
been called the intermediate cortical mass, and it is most junctional parenchyma20 (Fig. 4). This gives the appearance
likely that the kidney having this feature will have a double of “mass effect” in the junctional parenchyma. This is an-
pyelocalyceal system. When the connection of the two other reason why the junctional parenchyma may be mis-
echogenic renal sinuses is small, an intermediate cortical taken for a tumor.
mass may also be seen on certain sagittal or coronal scans. It is possible that all or most adult kidneys are formed by
Whether there is a double pyelocalyceal system in such fusion of two subkidneys. Because of various degrees of
kidneys is difficult to determine on ultrasonography unless resorption of junctional parenchyma, the amount of over-

fact, it is a line between two subkidneys and represents

the junctional parenchymal line.
● A part of the junctional parenchyma is not resorbed, and
various types of junctional parenchymas will be seen as
previously described (Fig. 6).
When two subkidneys are not in straight alignment, this will

result in a variety of appearances, some of which are shown
in Figure 8. This occurs mostly in the left kidney, and the
most common variety is a triangular kidney with dromedary
hump, i.e., a protrusion on lateral border of the kidney (Fig.
9). This protrusion represents the lateral aspect of the junc-
tional parenchyma.20 The cause of offset in alignment of
two subkidneys is not clear, but is probably caused by a
small renal fossa during early fetal life from a relatively
large spleen or some other causes. In cross-ectopic kidneys,
if the two fused kidneys are aligned in a craniocaudal di-
rection, their features may be quite similar to a kidney with
a junctional parenchyma, except that the cross-ectopic kid-
ney may be longer than normal.
FIG. 6. Classification of junctional parenchyma. From: Yeh HC, et al.

Junctional parenchyma: revised definition of hypertrophic column of TRUE HYPERTROPHIC SEPTUM OF BERTIN
Bertin. Radiology 1992;185:725–32.
In most individuals, the size of the septa of Bertin is quite
lapping of the two subkidneys, along with a difference in variable within the same kidney and in different individuals.
alignment, allows the kidneys to have a variety of appear- They range from small to larger than a pyramid. In the polar
ances. regions, the pyramids are close together, and only small or
incomplete septa of Bertin may be seen. In the middle por-
When two subkidneys are in straight alignment, the results
tion of the kidney, the pyramids are farther separated and
may be:
the septa of Bertin are larger. Because septa of Bertin rep-
● The junctional parenchyma is completely resorbed and a resents cortical tissue that surrounds a pyramid to form a
usual, normal, reniform kidney will be seen. renal lobe, it has the same histologic structure as renal cor-
● The junctional parenchyma is resorbed, but the junctional tex. It is therefore generally believed that a normal and a
parenchymal line and the junctional parenchyma defect hypertrophic septum of Bertin should have the same
remain (Fig. 7). An echogenic line will be seen at the echopattern as renal cortex. Contrary to this belief, the septa
junction between upper and middle thirds of the kidney. of Bertin are usually much more echogenic than renal cor-
This echogenic line has been called the interrenicular tex.20 This is because of the anisotropic effect20,23 (Fig. 10).
septum,22 which means a line between two renal lobes. In As previously mentioned, the renal cortex and septa of
Bertin are composed of numerous lobules. The orientation
of the lobules, the renal tubules that are contained in the
lobules (cortical rays or medullary rays of Ferrein),3 the
interlobular connective tissue, the interlobular arteries and
veins, and the like are all perpendicular to the renal surface
in the renal cortex; therefore, during ultrasonographic scan-
ning on sagittal or coronal plane, the sound beam will be
parallel to these structures, and only a few echoes will be
seen. In the septa of Bertin, these structures are parallel to
the renal surface and perpendicular to the sound beam.
Therefore, septa of Bertin will be much more echogenic
than renal cortex (Fig. 10). If one scans the kidney in the
FIG. 7. Junctional parenchymal line and defect. An echogenic line (ar- craniocaudal direction or caudocranial direction, the echo-
row) is seen in this kidney with a small or no junctional parenchyma genicity of the renal cortex and septa of Bertin will reverse,
because of near-complete resorption of polar parenchyma during fusion of
two subkidneys. The junctional parenchymal defect (arrowhead) can also i.e., the renal cortex will become more echogenic than septa
be seen. of Bertin. In the practical clinical setting, however, this is

134 H. C. YEH
FIG. 8. Three variations in offset align-

ment of two subkidneys. A, B, and C show
progressive increase in the degrees of an-
gulations of the long axis of two subkid-
neys. D, E, and F are corresponding draw-
ings. The angles between the two subkid-
neys in each kidney are shown in the right
lower corner of each drawing.
usually difficult to perform unless the kidney is horizontally ally round. Color Doppler may show no detectable flow or
oriented, and ultrasound scanning may not show good detail symmetrical vascular supply from adjacent arcuate vessels
of the parenchymal tissue in the middle portion of the kid- without any irregular tumor vessels.
ney. A typical normal septum of Bertin is pear-shape, with
a small part pointing toward the renal surface20 (Fig. 10). A
large or hypertrophic septa of Bertin retains a pear-shape FETAL LOBATION, RENAL SCARS, CORTICAL
and is highly echogenic20 (Fig. 11). It is always located TUMOR, AND HYPERTROPHIC RENAL LOBE
between two pyramids and usually partly protrudes into the
renal sinus. It may sometimes indent on the adjacent pyra- It is well known that so-called fetal lobulation is a normal
mids on both sides (Fig. 12). A single or double hypertro- variation of the kidney in which the lobular contour of the
phic septa of Bertin may be seen. It is usually located in the fetal kidney remains in the adult kidney. In fact, as previ-
anterior parenchyma in the middle portion of the kidney. ously mentioned, a kidney consists of approximately 20,000
This is different from junctional parenchyma, which is usu- lobules and 14 lobes. There are numerous lobules in each
ally located at the junction between the upper and middle lobe. The renal lobules are too small to be delineated by
third of the kidney and between two renal sinus systems on ultrasonography. What we see in fetal kidneys are lobes,
transverse scanning and along the lateral aspect of the kid- and the term “fetal lobulation” should, instead, be “fetal
ney. The characteristic pear-shape, homogeneous high-level lobation.” Although the term “lobulation” has been used to
echoes, and location between two adjacent pyramids usually describe a bumpy or undulated contour of an object, e.g., a
allow the diagnosis of a true hypertrophic septum of Bertin lobulated tumor, and is not necessarily related to true lob-
and differentiate it from an echogenic tumor, which is usu- ules. Hence, fetal lobulation may mean that kidney has an

FIG. 9. Dromedary hump caused

by offset alignment of subkidneys. A.
The left kidney appears small, 7.8
cm, because only the inferior subkid-
ney was delineated. The superior
subkidney was not well visualized
because it is in a slightly different
plane. B. Scanning slightly more me-
dially, both subkidneys are clearly vi-
sualized, and a correct measurement
(10.4 cm) was made. A triangular
kidney with a dromedary hump (ar-
rowhead) is now evident. C. Drawing
shows offset alignment of the two
subkidneys from side view (c) and
frontal view (d). Scanning along X
plane, only inferior subkidney is seen
(a). Scanning along Y plane, both
subkidneys are seen to be in offset
alignment (b), which is the cause for
kidney being triangular.
undulated contour similar to a fetal kidney, whereas fetal MILK OF CALCIUM IN A RENAL CYST
lobation means that a kidney retains a multiple lobar con- Milk of calcium in a renal cyst or calyceal diverticulum
tour similar to a fetal kidney; certainly, this is a much more had been considered to be exceedingly rare25 and, therefore,
precise term and was coined by Hodson.1,24 only a few case reports26–30 had been published because of
The diagnosis of fetal lobation can be made by finding its rarity until 1992, which is when we found that this is
that each bulge of the undulated renal contour corresponds actually common.31 Because milk of calcium contains cal-
to each renal lobe. Because each renal lobe is formed by a cium, naturally, layering of highly echogenic material with
pyramid surrounded by cortical tissue, each bulge on the an acoustic shadow was thought to be the ultrasonographic
renal contour should correspond to a renal lobe, i.e., each feature of this entity. In fact, in most instances, milk of
bulge should overlie one pyramid and each indentation calcium does not cast an acoustic shadow because it is usu-
should be between two lobes (Fig. 13). This is different ally seen in small renal cysts, smaller than 1cm and only
from renal scars, which may also result in indentations (or occasionally larger than 1.6 cm. Each cyst usually contains
cortical defects) and undulation on the renal contour. The only a small amount of milk of calcium, which produces
indentation (or indentations) from the scar is more likely to reverberation echoes (Fig. 17). This phenomenon can also
be located in the middle of a bulge (a renal lobe) on the occur with other small, highly reflecting particles, such as
renal contour or in a random fashion, rather than in between cholesterol crystals and ground sand. The reverberation ech-
two lobes (Fig. 14). A small tumor may also cause a bump oes are caused by the ultrasound beam bouncing back and
on the renal contour (Fig. 15). In diagnosing fetal lobation, forth within the aggregations of small, highly reflecting par-
one should carefully examine each bulge on the renal con- ticles. When there is a larger amount of milk of calcium,
tour to be certain that each bulge corresponds to each renal part of the sound beam will be absorbed or scattered away
lobe so that a renal mass associated with fetal lobation may from the transducer, and partial shadowing may occur.
not be missed. When cysts are larger than 1.2 cm, more apparent acoustic
A hypertrophic renal lobe may simulate a renal cortical shadowing may be seen, and cysts smaller than 0.6 cm
tumor (Fig. 16). Careful scanning to delineate the pyramid usually have no acoustic shadowing. In a larger cyst, the
within it and the symmetrical echogenic septa of Bertin on acoustic shadow will become more obvious as the layer of
both sides will allow a diagnosis of a hypertrophic renal milk of calcium increases in thickness. If one partly aspi-
lobe. rates the cystic fluid, the cyst becomes small, and the layer

136 H. C. YEH
thicker layer of milk of calcium, and acoustic shadowing

may become stronger.
The diagnosis of milk of calcium within a renal cyst is
usually easy and quite definite. When layering of highly
echogenic material is seen within the dependent portion of
a cyst with reverberation echoes and with or without only a
weak acoustic shadow, milk of calcium within a cyst is
highly suggestive. To bring out this feature, it is best to scan
the echogenic lesion with the transducer perpendicular to
the ground. If the echogenic material remains in the depen-
dent portion of the cyst when the patient turns in different
positions, the diagnosis is certain. If the cyst is small and
filled with milk of calcium, layering of milk of calcium will
be difficult to demonstrate. In such cases, the diagnosis may
be difficult because a small stone may also show no or only
weak acoustic shadowing. This, however, rarely occurs, and
most milk of calcium can be diagnosed by ultrasonography.
The diagnosis of milk of calcium in a cyst is important
because this is usually a benign condition that requires no
treatment in most cases. Misdiagnoses may result in unnec-
essary anxiety for patients, along with unnecessary follow-
up examinations and unnecessary treatments, such as litho-
tripsy, which is both ineffective32 and potentially harmful.
Milk of calcium can be misdiagnosed as other conditions,
such as:
1. Renal stone (Fig. 18). As already stated, a tangential
view of milk of calcium produces a round, echogenic
structure with an acoustic shadow, and no fluid-filled
cystic part is visualized. This can easily be mistaken for
a stone and usually occurs on a supine coronal scan of a
kidney, scanning from the flank. If reverberation echoes
are seen and the acoustic shadow is not as strong as
expected for a stone of this size, milk of calcium should
be suspected. Scanning carefully with the transducer per-
FIG. 10. Anisotropic effect on septa of Bertin. A. Diagram shows a pendicular to the ground and with the patient in different
kidney with its arterial system. The kidney is scanned (dashed lines) by a positions will clarify the diagnosis.
curve-array transducer. Note that the interlobular arteries within the renal
cortex are perpendicular to the renal surface, but those within the septa of
2. Angiomyolipoma (Fig. 19). Tangential views of milk of
Bertin are parallel to the renal surface. As shown in Figure 1, renal lobules calcium show a round, highly echogenic lesion. If little
(and cortical rays) are also in the same orientation as interlobular arteries. reverberation echoes and no acoustic shadow are seen,
B. Diagram shows that in the area where the ultrasound beam is perpen-
dicular to the renal arteries (and also renal lobules and cortical rays), the
the lesion may be mistaken for an angiomyolipoma. An
echoes are increased. These echogenic areas include the septa of Bertin aggregation of tiny cysts with milk of calcium may also
(arrows) in the middle portion of the kidney and the renal cortex in both appear as a single echogenic lesion, similar to an angio-
poles (arrowheads). Arrowheads ⳱ pyramids. C. Example of a normal
kidney showing the anisotropic effect. The septa of Bertin (black arrows)
myolipoma. Again, even small amounts of reverberation
are more echogenic than the renal cortex. The polar cortex (black arrow- echoes or acoustic shadowing should cause suspicion
heads) are also hyperechoic. White arrowheads ⳱ pyramids. that this is not an angiomyolipoma, and careful scanning
should be performed to clarify the diagnosis.
of milk of calcium becomes thicker; therefore, acoustic 3. A cyst with calcified wall (Fig. 20). Calcification in a
shadowing will appear or become more apparent. The pres- cyst wall may be caused by chronic inflammation or
ence and degree of acoustic shadowing also depends on the carcinoma; therefore, it is indicated that the cyst is not a
scanning technique. Using a higher frequency transducer or simple cyst. Two conditions with milk of calcium in a
smaller or focused ultrasound beam will result in a more cyst can be mistaken for calcification in a cyst wall: 1)
obvious acoustic shadow. By scanning tangentially through when milk of calcium is scanned with the transducer at
the milk of calcium, the sound beam may pass through a an oblique angle, the milk of calcium may appear as

FIG. 11. True hypertrophic septum

of Bertin in two patients. A. Coronal
scan of right kidney shows a large sep-
tum of Bertin (arrow) flanked by pyra-
mids (arrowheads) and protruding into
the renal sinus fat. B. Transverse scan
shows the large septum of Bertin (ar-
row) located anteriorly and flanked by
pyramids (arrowheads). C. Coronal
scan of right kidney on different pa-
tients shows double hypertrophic
septa of Bertin (arrows). They are
typically pear-shape. Arrowheads ⳱
pyramids. D. Transverse scan shows
that the large septum of Bertin (arrow)
is again located anteriorly.
calcification in a cyst wall. A proper scan as previously 1. No lesion detected.

mentioned will clarify the diagnosis; 2) when a small This happens if the cyst is small and close-spaced thin-
cyst with milk of calcium is adjacent to a larger simple sections are not performed.
cyst, the small cyst with milk of calcium may be mis- 2. Mistaken for a calcification or stone.
taken for calcification in the wall of a larger cyst. A This occurs if the fluid-containing portion of the cyst is
careful scanning to separate the two cysts will resolve the not delineated.
problem.
It is usually quite difficult to diagnose milk of calcium in
a cyst on computed tomography (CT) scan unless the cyst is
relatively large. This is because of the difficulty in delin-
eating the layering of milk of calcium. Even for a 1-cm cyst,
one may have to repeat close-space thin-section scanning to
see the layering of the milk of calcium. Depending on the
size of the cyst and amount of milk of calcium, CT scanning
may show milk of calcium in the cyst as:
FIG. 12. True hypertropic septa of Bertin compressing on pyramids. FIG. 13. Fetal lobation. A. Right kidney of a 35-week-old fetus. Note
Sagittal scan of the kidney shows two markedly enlarged septa of Bertin undulated renal contour caused by renal lobes (arrows). B. Similar lobation
(arrowheads) that indent on adjacent pyramids (P). The septa of Bertin are in an adult kidney. Each lobe contains a pyramid. Note the hyperechoic
highly echogenic. tissue between the two pyramids, representing septa of Bertin.

138 H. C. YEH
FIG. 14. Fetal lobation and scars. Sagittal scan of a kidney shows fetal
lobation (white and black arrows). An indentation (black arrow) in the
middle of the margin of a renal lobe is caused by scar. Extensive irregular
cortical defects (arrowheads) seen in the upper pole is caused by a scar or
scars. In one renal lobe, the pyramid (p) is only slightly decreased in size,
but the cortex is absent. The other pyramid (p’) is small.
3. Seen as a simple cyst.

This happens if a small amount of milk of calcium is not
delineated.
4. Mistaken for a solid lesion and a cancer cannot be ex-
cluded (Fig. 21).
FIG. 16. Hypertrophic renal lobe. A. Hypoechoic mass-like lesion (ar-
This is caused by the partial volume effect on CT scan- rowhead) is seen in the lower pole of the kidney. B. More careful scanning
ning. Because milk of calcium may be associated with in- shows a large pyramid in the lesion, indicating that this is a large renal
lobe. The pyramid is flanked by a slightly hyperechoic septa of Bertin.
flammation of the cyst, and because of the difficulty in
scanning a small lesion and precisely evaluate the density

on CT scan, the lesion may even appear slightly enhanced
on CT scan after injection of contrast material, which fur-
ther raises the possibility of cancer. Such a lesion is fre-
quently referred to ultrasonography for clarification. If one
is not aware of this problem on CT and looks for a solid
mass on ultrasonography, no lesion will be found, resulting
in a nondiagnostic study. Finding a cyst with milk of cal-
cium in the same location as found on CT (best to compare
with the transverse scan) makes the diagnosis clear.
Multiple or even numerous cysts with milk of calcium in
a kidney is not uncommonly seen (Fig. 22). Milk of calcium
may also be seen in congenital or acquired polycystic kid-
neys and in ectopic or transplanted kidneys. Because milk of
calcium in a cyst is so common, it is frequently detected in
patients with renal colic or hematuria. However, milk of
calcium is usually not the cause of pain or hematuria. Oc-
casionally, milk of calcium in a cyst may disappear spon-
taneously on a follow-up examination or appear on a pre-
viously normal kidney.
Plain radiographs may show no calcification for a small
cyst with milk of calcium, or they may show a small, cal-
cified density or mottled densities when multiple cysts with
FIG. 15. Fetal lobation with a small mass. A. Fetal lobation. An apparent milk of calcium are present. It is usually difficult to dem-
small defect (arrowhead) or indentation is seen in a renal lobe. This simu-
lates a scar. B. Scanning in a slightly different plane shows a small mass onstrate layering of a small amount of milk of calcium on an
(arrow) and no scar. upright or cross-table radiograph. Most cases of milk of
FIG. 17. Milk of calcium in a renal cyst. A 33-year-old woman with hematuria. Radiograph, ultrasonogram, and CT scan were obtained elsewhere (not
shown), and a right renal stone was diagnosed. Our ultrasonograms revealed a right renal cyst with milk of calcium rather than a stone. Computed
tomography scans and radiographs were repeated thereafter. B and D. Longitudinal scans of the right kidney with the patient in the supine and prone
positions, respectively, show a cyst (arrowheads) with milk of calcium (arrows) layering in the dependent portion of the cyst. A. Unenhanced CT scan shows
a calcified lesion (arrow) similar to a stone. C. Repeat scanning at 0.5-cm intervals shows the stone is in fact a small cyst (arrowhead) with milk of calcium
(arrow) layering in the dependent portion Thin-slice CT scan was performed twice to obtain the image showing layering of milk of calcium. From: Yeh
HC, et al. Milk of calcium in renal cysts: new sonographic features. J Ultrasound Med 1992;11:195–203.
calcium cannot be diagnosed or detected on plain radio- uterine cavity and not within the endometrium. The early
graphs because of their small size. gestational sac appears empty on ultrasonography. The first
Fine-needle aspiration may show clear, yellow fluid or structure to appear within the gestational sac is usually a
dark brown, turbid fluid. It is usually difficult to aspirate the small bleb that has been considered to represent a yolk sac.
calcified material to prove milk of calcium unless the cyst If the gestational sac grows to a certain size and still shows
contains a large amount of milk of calcium and unless a no evidence of the yolk sac, it is considered to be a nonvi-
prolonged, forceful suction is applied during aspiration. able gestation or a blighted ovum.34 The critical size of
discriminatory value of the gestational sac for diagnosing a
YOLK SAC AND FETAL POLE nonviable pregnancy has been decreased from 2 cm34 on
transabdominal scanning to 0.8 cm on current endovaginal
During early pregnancy, the earliest sign of intrauterine scanners.35 In fact, in the early gestational sac, there is an
pregnancy during ultrasonography is a small, cyst-like inner cell mass that is usually too small to be seen by current
structure “within” the anterior or posterior layer of endo- ultrasound equipment. The inner cell mass subsequently
metrium. This is called the intradecidual sign,33 and the forms an early amniotic sac, the roof of which is a germ disc
cyst-like structure represents an early gestational sac. This is (bilaminate disc), i.e., the early embryo. Therefore, the first
different from a pseudogestational sac in which the cyst-like bleb that appears in a gestational sac is an amniotic sac-
structure or fluid collection represents fluid or blood in the embryo complex36 and not the yolk sac. The yolk sac at this
140 H. C. YEH
FIG. 18. Milk of calcium in a renal cyst

simulating a stone in a patient with hema-
turia. A. Coronal view of the right kidney
shows a round, highly echogenic lesion
(arrowhead) in the upper pole of the right
kidney, casting an acoustic shadow. Pre-
liminary impression of the ultrasonogra-
pher was a renal stone. Note the acoustic
shadow is of mild to moderate degree,
which is unusual for a stone of this size.
There are reverberation echoes. B. Because
of suspicion for milk of calcium in A, scan-
ning with perpendicular ultrasound beam,
with the patient supine, from costal margin
shows layering (arrowhead) of echogenic
material. Perpendicular beam scanning was
also performed with the patient in the de-
cubitus (C) and prone positions ( D), again
showing layering of echogenic material in-
dicating milk of calcium.
early stage is called the primary yolk sac, which represents

the inner lining layer of the early gestational sac, and it is
not clearly separated from gestational sac wall. Therefore,
ultrasonography usually does not delineate the early pri-
mary yolk sac. As the gestational sac grows, the primary
yolk sac shrinks and starts to separate from the gestational
sac. The primary yolk sac may be seen at this stage if a
careful scan is performed (Fig. 23).
The term “fetal pole” has been used extensively in the
past. The term may have different meanings to different
people, and I do not think it is a good term to use. It origi-
nally meant amniotic sac-embryo yolk-sac complex.37 In
general, fetal pole refers to any structure that may be related
FIG. 19. Milk of calcium simulating an angiomyolipoma. A. Highly FIG. 20. Milk of calcium versus calcified cystic wall in two patients. A.
echogenic lesion (arrowhead) in the upper pole of the kidney. There is no Cyst (arrowhead) with small amount of milk of calcium (arrow) may
acoustic shadow, and the lesion is somewhat similar to an angiomyoli- simulate calcification in the cystic wall. Careful scanning may be necessary
poma. B. Because some reverberation was seen (A), milk of calcium was to show layering, reverberation echoes, and movement with gravity (not
suspected, and scanning with a perpendicular ultrasound beam with the shown) to confirm milk of calcium. B. Renal cyst (arrowhead) in another
patient in the prone position was performed, demonstrating layering of patient shows calcification (arrow) in the cystic wall casting only a weak
echogenic material, indicating milk of calcium. shadow.

FIG. 21. Cyst with milk of calcium simulat-

ing a tumor on CT. A. Scanning without in-
jection of contrast medium, a small calcifica-
tion (arrowhead) was seen in the medial aspect
of the left kidney, but no obvious cyst was
seen. B. After injection of contrast medium, a
lesion (arrow) was seen. The attenuation value
of the lesion increased from 22.6 HU to 58.8
HU after the injection of contrast. This is
highly suspicious for carcinoma. C. Ultrasono-
graphic scanning with patient in the decubitus
position shows a 1-cm cyst with milk of cal-
cium. A ⳱ aorta. D. Scanning with the patient
in the prone position again shows the cyst with
milk of calcium, which has moved anteriorly
to the dependent portion. S ⳱ spine.
to the fetus during early gestation. However, because we are lineate as two layers on ultrasonography (Fig. 24). Even if
now able to see the normal embryo or fetus, yolk sac, and one carefully delineates the double decidual sac at this
amniotic sac most of the time, we can use more precise stage, a well-formed fetus is always found in a normal preg-
terms, such as fetus, amniotic sac, yolk sac, or double bleb nancy, and the double decidual sac sign is not necessary to
instead of fetal pole, which is poorly defined. diagnose an intrauterine pregnancy. During earlier intrauter-
ine pregnancy, the gestational sac is located in a thickened
DOUBLE DECIDUAL SAC SIGN layer of decidua (the opposite layer of decidua is usually
thinner), and the thin line of the echogenic uterine cavity
It is believed that after the intradecidual sign seen in early between the two layers of decidua is usually straight or only
pregnancy, in the next stage of pregnancy, one may see a slightly displaced by the gestational sac. Therefore, two
double sac38 or a double decidual sac sign39 before the yolk layers of decidua really do not wrap around the gestational
sac or embryo can be seen. In fact, the intradecidual sign sac to form the double decidual sac sign (Figs. 25,26). Evi-
can be seen up to seventh week of pregnancy and not just dence that the double echogenic ring during early pregnancy
during the early weeks of pregnancy; although, the clinical does not represent double decidual sac include: 1) to form a
usefulness is during the early gestational stage when the double decidual sac, one should find a thin, echogenic, uter-
gestational sac appears empty. The double decidual sac sign ine cavity ring between two echogenic decidual rings to
is also useful in this empty-sac stage. A true double- indicate that this is indeed the double decidual rings divided
decidual sac sign, however, is usually not present before by the uterine cavity, i.e., one should find a third, thin,
seventh week.33,40 Two echogenic rings, one inside the echogenic ring between the two thick, echogenic rings;
other, during early pregnancy has been called the double however, this has never been documented; 2) the hy-
decidual sac sign. This is based on the assumption that each poechoic ring between the two echogenic rings should be
echogenic ring represents one layer of decidua. In fact, to determined. One speculation is that it represents fluid within
have double layers of decidua surrounding a gestational sac, the uterine cavity. Fluid in the uterine cavity during early
the gestational sac has to be large enough to markedly dis- gestation is unusual and may be abnormal. Fluid should be
place the decidua so that two layers of decidua will wrap flanked by two thin, echogenic lines (or rings) of decidual
around the gestational sac to form a double decidual sac. surface; 3) the outer echogenic ring represents the basal
This usually occurs only after approximately 8 weeks of layers of the decidual, and the inner echogenic ring repre-
gestation. By this time, the layers of decidua are markedly sents choriotrophoblastic tissue of the gestational sac. This
stretched and close to each other, and it is difficult to de- is usually evident on longitudinal scan of the uterus,

142 H. C. YEH
FIG. 22. Multiple cysts with milk of calcium in a kidney. A. A cyst

(arrowhead) with milk of calcium (arrow) is seen near the upper pole of the
kidney. B. Two more cysts (arrowheads) with milk of calcium (arrows) are
seen in the lower pole of the same kidney.
and double echogenic rings usually appear on oblique scan

(Fig. 26). The decidua on the same side of the gestational
sac is usually markedly thickened and poorly echogenic, FIG. 23. Single bleb stage: amniotic sac. A. Endovaginal scanning shows
a small gestational sac, with average diameter of 0.7 cm, containing a
except for the basal layer, which is highly echogenic; 4) if single bleb (a) that represents an amniotic sac. The primary yolk sac (y) is
both echogenic rings represent two layers of decidua, then also visible as it is separating from the chorionic sac. B, C, D, and E.
where is the choriotrophoblastic ring, i.e., the gestational Drawings showing formation of single bleb and double bleb. B. Drawing
shows an inner cell mass (icm) within blastocyst at the time of implanta-
sac? We should see triple echogenic rings because during tion. C. Drawing shows formation of amniotic sac (a). This is the single
early pregnancy, the gestational sac is usually clearly sepa- bleb. The roof of the amniotic sac is the early embryo, the bilaminate germ
rated from the echogenic basal layer of the decidua. disc (g). The primary yolk sac (py) is the lining membrane of the gesta-
tional sac. D. Drawing shows shrinking of primary yolk sac (py), which is
Features of the double echogenic ring may also be seen in separating from chorionic sac. g ⳱ germ disc; a ⳱ amniotic sac. E.
cases of a pseudogestational sac.41 Therefore, it is not a Drawing shows that the yolk sac (y) has shrunken to become the same size
specific sign of an intrauterine pregnancy. as amniotic sac (a). Germ disc is located between the amniotic and yolk
sacs, and the yolk sac–germ disc amniotic sac complex forms double bleb.
In sum, the ultrasonographic signs for early intrauterine
pregnancy are: 1) intradecidual sign; 2) single bleb sign
(i.e., embryo-amniotic sac complex; 3) double bleb sign gland did not contain a tumor, ultrasonography is consid-
(i.e., yolk sac-embryo amniotic-sac complex). The double ered by some authors42 to be of little value in diagnosing
decidual sac sign is not a reliable sign of an intrauterine Hashimoto thyroiditis. Other authors43–45 described the ul-
pregnancy, because it is virtually nonexistant in ultrasonog- trasonographic features of Hashimoto thyroiditis as a dif-
raphy.40 fusely hypoechoic and inhomogeneous gland. These find-
ings, however, are nonspecific. In a reported series of 360
MICRONODULATION IN THE THYROID GLAND patients44 with a diffuse hypoechoic thyroid gland, only
34.2% proved to have Hashimoto thyroiditis. Because the
Hashimoto thyroiditis is the most common form of thy- patients frequently are asymptomatic, clinical assessment
roiditis. It is one of the most common causes of hypothy- may be difficult. Sailer et al.44 performed a histologic ex-
roidism. Therefore, it is frequently encountered in daily amination of surgical thyroid specimens and found Hashi-
clinical practice. Ultrasonographic diagnosis of Hashimoto moto thyroiditis in 81 patients. The preoperative diagnosis
thyroiditis has been relatively difficult in the past. Other for the disease had only been made in three patients.
than being able to confirm that a diffusely enlarged thyroid Ultrasonography is important in diagnosing Hashimoto

FIG. 24. True double decidual sac in patient in eighth week of

gestation. A. Sagittal endovaginal scan shows an intrauterine
gestational sac with a fetus and a yolk sac. The gestational sac
is surrounded by two layers of decidua (white arrowhead).
Small segment of uterine cavity line (UC) is seen between two
layers of decidua. B. Drawing showing true double decidual sac
(DDS) caused by marked displacement of uterine cavity (UC)
by a large gestational sac. C. Sagittal scan in a slightly different
plane shows more clearly the uterine cavity line (UC) and,
hence, the double layers of decidua (arrowhead).
thyroiditis because of the recent discovery of a new sign, geneously hypoechoic, target-shape, large hyperechoic, or a
micronodulation.46 The sign is highly characteristic and complex mass with multiple small cystic areas. Therefore,
highly specific, with a positive predictive value of 94.7%. unlike the characteristic hypoechoic micronodules, the ul-
We have been using this sign to make initial diagnoses trasonographic features of masses caused by Hashimoto
before clinical awareness of the disease in 41 of 57 patients. thyroiditis may be entirely nonspecific, and needle biopsy
Micronodulation indicates numerous small, hypoechoic 0.2- may be necessary for a definite diagnosis.
to 0.3-cm nodules, which can occasionally be as large as 0.6 An increased risk of lymphoma in association with
cm, scattered throughout the entire thyroid gland (Fig. 27). Hashimoto thyroiditis has been reported in the literature.43
The micronodules are frequently surrounded by a thin, hy- Lymphoma tends to be hypoechoic, and three such hy-
perechoic rim that makes them highly recognizable. The poechoic masses, which I have performed biopsy on, in
hyperechoic rims are caused by fibrous strands, and the patients with this disease were caused by Hashimoto thy-
hypoechoic micronodules are caused by massive infiltration roiditis.
by an exudate of lymphocytes and some plasma cells.46 In There is also an increased risk of papillary carcinoma, of
pathology, the micronodules correspond to accentuated lob- which I have seen a few cases, and the smallest mass was
ules.47 The micronodules are usually more apparent on ob- 0.8 cm (Fig. 28). They all contained stippled calcifications
servation during actual real-time scanning than on a static caused by psammoma bodies. One should carefully evaluate
hard-copy image. During real-time scanning, as one slowly each nodule in Hashimoto thyroiditis, looking for stippled
sweeps the transducer across the thyroid gland, numerous calcifications, and if they are present, a fine-needle aspira-
micronodules appearing on consecutive images and are tion biopsy should be performed.
much more impressive than when seen on a static image.
The background echogenicity of thyroid gland in Hashi- THYROID CYST
moto thyroiditis is usually normal (50% of cases) or slightly
decreased (26% of cases), and in only 24% of the cases, it Unlike renal cysts, which are common, simple thyroid
was moderately hypoechoic. Therefore, previous descrip- cysts with an epithelial lining are rare,48 and they are not
tions of a diffusely hypoechoic thyroid gland apply only to mentioned in some popular pathology textbooks. Most cyst-
a few cases. like lesions in the thyroid gland actually have a thick wall
Although micronodules may sometimes increase in size, and represent benign colloid nodules or adenomatous nod-
they rarely grow beyond 0.6 cm. Occasionally, masses of ules that have undergone central cystic degeneration or
various sizes may be seen in a thyroid gland with Hashi- hemorrhage. The thick wall is usually isoechoic or only
moto thyroiditis, and most masses are caused by this dis- slightly hyperechoic and is frequently poorly visualized
ease, as proven by fine-needle aspiration biopsy. The (Figs. 29,30). Therefore, the thick walls are frequently not
masses may have a variety of echotextures, such as homo- delineated or recognized on ultrasonography, and the cystic

144 H. C. YEH
FIG. 26. Drawing depicting the formation of the double echogenic ring,
which is the so-called double decidual sac sign in a normal intrauterine
pregnancy. Left. Drawing of a sagittal scan shows a gestational sac (GS)
within the thickened posterior decidua. UC ⳱ uterine cavity line; BL ⳱
echogenic basal layer of decidua. Top right. Drawing of a transverse scan
again shows a gestational sac (GS) within the posterior decidua. A dashed
line indicates the scanning plane of the drawing in top right. Bottom
right. Drawing of oblique scan shows a double echogenic ring appearance
of the so-called double decidual sac sign. The inner echogenic ring repre-
sents the gestational sac (GS) (or choriotrophoblastic ring), and the outer
echogenic ring represents the basal layer (BL) of the decidua. The uterine
cavity line (UC) passes anterior to the gestational sac.
thick wall. Small cystic lesions smaller than 0.3 cm usually

represent dilated follicles and not true cysts (Fig. 31).
RECANALIZATION OF UMBILICAL VEIN
In adults, the umbilical vein is usually completely oc-

FIG. 25. So-called double decidual sac sign in a patient at 6 weeks’ 2 cluded, at least in its distal part, and becomes a fibrotic
days’ gestation. A. Endovaginal oblique scan of the uterus shows a double cord-like structure, i.e., a round ligament. A remnant (or
echogenic ring, which is the so-called double decidual sac sign. Note the
uterine cavity line (white arrowheads) passes the anterior border of the vestige) of the umbilical vein of up to 10 cm in length may
inner echogenic ring and does not run in a circle between the two echo- be seen in 50% of healthy individuals.50 This is called
genic rings. B. Longitudinal scan of the same patient shows the inner Baumgarten’s recess. Therefore, visualization of a short
echogenic ring actually represents the gestational sac (G), which is located
posterior to the uterine cavity line (white arrowheads). The uterine cavity segment of a patent umbilical vein within the round liga-
line is straight and not curved around the gestational sac (G). The outer ment does not indicate portal hypertension. Baumgarten’s
echogenic ring (B) actually represents the basal layer of decidua. The recess is usually small but may be more prominent in cir-
decidua is thick, including the basal echogenic line (or ring) and the less
echogenic area, and not just the echogenic ring. The yolk sac is seen within rhosis, and it is always smaller than 3 mm in diameter.51
the gestational sac. The embryo is not shown in this section. From: Yeh The inferior part of the round ligament is always occluded,
HC. Efficacy of the intradecidual sign and fallacy of the double decidual even in cirrhotic patients.50 Usually, there are one or two
sac sign in the diagnosis of early intrauterine pregnancy (letter). Radiology
1999;210:579–81. paraumbilical veins connected to Baumgarten’s recess. This
is called Burrow’s vein, which is one of the three groups of
paraumbilical veins. The commonly described paraumbili-
lesions are frequently diagnosed as simply a “cyst;” there- cal vein is called Sappey’s inferior vein.52 The veins are
fore, only the cystic areas are measured for their size. Care- located in the extraperitoneal space of the anterior abdomi-
ful scanning from various angles and with appropriate set- nal wall and communicate with inferior epigastric veins and
tings with moderate contrast are important to bring out the cutaneous veins at the umbilical region. As they ascend, the
full detail of the thick wall and diagnose the nodule as a veins traverse the inferior part of the falciform ligament,
mass with central cystic degeneration or hemorrhage rather running alongside the round ligament within the falciform
than as a simple cystic lesion. Some articles stress the un- ligament (Fig. 32), and enter the liver to communicate with
common occurrence of malignant tumor in thyroid the portal system in a variety of pathways. There can be one
cysts.48,49 However, researchers also observe that cysts in or a few tiny collapsed paraumbilical veins in healthy indi-
patients with cancer appeared to originate from necrosis of viduals.50 They are usually not visualized by color Doppler
the tumor;49 therefore, those cysts are more likely to have a ultrasound. In portal hypertension, the paraumbilical veins

FIG. 27. Micronodules: variation in size in four different patients with Hashimoto thyroiditis. A. Tiny micronodules, smaller than 0.2 cm. B. Small
micronodules (0.2 cm). C. Moderate micronodules (0.25–0.35 cm). D. Large micronodules (0.3–0.4-cm). From: Yeh HC, et al. Micronodulation:
Ultrasonographic sign of Hashimoto thyroiditis. J Ultrasound Med 1996;15:813–9.
engorge to allow collateral circulation. In 24 patients with vein connecting via the superficial epigastric vein in the
portal hypertension and paraumbilical collateral circulation, subcutaneous space.
Lafortune et al.50 found on umbilicoportography that most
patients (91.7%) had a single, dilated, paraumbilical vein, HYPERECHOIC STROMA IN POLYCYSTIC
and only two patients showed two veins. In all of our 27 OVARIAN SYNDROME
patients, ultrasonography showed only a single vein.
The extrahepatic systemic connections of paraumbilical Polycystic ovarian syndrome (PCOS) is one of the most
veins have rarely been investigated in the past, and some common ovarian abnormalities in women of reproductive
authors suggest in their illustrations that the vein connects age. It is one of the most important causes of irregular
directly to the inferior vena cava at the umbilical region.53 menstruation, amenorrhea, and infertility in women. Ultra-
Based on our research of 27 patients, there are at least four sonography has an important role in the diagnosis of PCOS.
types of pathways, none through the inferior vena cava54 Hyperechoic stroma in the ovaries has been considered one
(Fig. 33). The most common pathway is through the left, of the most important ultrasonographic signs for the diag-
right, or both external iliac veins (Fig. 34) via the inferior nosis of PCOS by many researchers.55–58 In fact, I have
epigastric veins within the extraperitoneal fat space (Fig. rarely seen a truly hyperechoic stroma in patients with
35). In the second common pathway, the vein makes a sharp PCOS. With PCOS, there are always multiple, moderate-
turn upward to connect to the innominate vein in the chest size, developing follicles, usually 0.5 to 0.8 cm, and there is
via the superior epigastric vein. In the third type, the vein through transmission of ultrasound beams through the fol-
divides after exiting from the liver to connect to both the licles. This results in a hyperechoic area in the ovarian
external iliac vein inferiorly and the innominate vein supe- stroma behind the follicles. For areas not behind the fol-
riorly. The least common pathway is the greater saphenous licles, the echotexture of the stroma is similar to that of the

146 H. C. YEH
FIG. 29. Cystic lesion of the thyroid gland. A. Sagittal scan of right lobe
shows a lesion possibly representing a simple cyst. B. Scanned more care-
fully, an isoechoic thick wall (between [+]) is delineated. C. After aspira-
tion, a solid mass (arrowheads) with a collapsed slit-like central cystic area
is seen. Aspiration at this time gained more tissue than did aspiration of the
cystic fluid.
FIG. 28. Hashimoto thyroiditis with a papillary carcinoma. A. Sagittal

scan of right lobe of thyroid shows numerous hypoechoic micronodules in not resorb. This results in more developing follicles in both
the thyroid gland. A hypoechoic mass (arrowheads) is seen on the anterior ovaries than in normal individuals.
surface of the lower pole of the right lobe of thyroid gland. Multiple
stippled calcifications are seen in the mass, which is highly suspicious for
papillary carcinoma. B. Transverse scan shows more clearly the multiple
stippled calcifications within the mass (arrowheads). Biopsy was per-
DIAPHRAGMATIC SLIP SIMULATES
formed with a 25-gauge needle on the mass, which proved to be a papillary LIVER MASS
carcinoma.
A diaphragmatic slip is a fold or strip of muscle protrud-
ing from the inferior surface of the diaphragm. In ultraso-
uterus and is not hyperechoic (Fig. 36). This effect can be nography, a diaphragmatic slip frequently simulates a liver
observed in the illustrated figures of the articles in which the mass60–62 (Figs. 37–39). The mass becomes more promi-
authors claim to have observed hyperechoic ovarian stroma nent during deep inspiration because the diaphragmatic slip
in PCOS.55,58 Therefore, the so-called hyperechoic stroma will enlarge because of muscular contraction, forming a
as the most important sign for PCOS actually is caused by deeper fold (Fig. 37). The mass may appear irregular in
through-transmission artifact. The most important diagnos- shape, with heterogeneous echoes, or it may have a smooth
tic sign for PCOS is multiple (more than five, based on our contour. An indentation may be seen on the superior surface
research)59 developing follicles in each ovary. The size of of the diaphragm at the diaphragmatic slip, especially dur-
the ovary may not be important because in more than 25% ing inspiration, and a small piece of gas-containing lung
of patients, the ovaries are normal in size.59 In healthy in- may be seen entering the indentation. It is more commonly
dividuals, one ovary may sometimes contains more than seen in the right anterior aspect of the dome of the dia-
five moderate-size, developing follicles; however, the other phragm. Multiple, equally spaced diaphragmatic slips also
ovary usually contains less developing follicles. After ovu- may be seen (Fig. 39). In the presence of subphrenic ascites,
lation, the follicles resorb and mostly disappear. But, new a diaphragmatic slip may simulate a peritoneal tumor im-
follicles may appear during the next menstrual cycle. In plant on the inferior surface of the diaphragm60 (Fig. 38).
PCOS, patients may not have menstruation, or they may Ultrasonographic diagnosis can be easily performed by
have anovulatory cycles and the developing follicles may slowly rotating the transducer during scanning (Figs.

FIG. 30. Complex thyroid mass. A. Lesion in the thyroid gland was
initially considered a cyst (arrowheads) with internal echoes. B. On careful
scanning, a thick, solid, isoechoic wall (arrowheads) is seen, indicating that FIG. 32. Paraumbilical vein. A. Transverse scan of the epigastrium
this is a mass with central cystic degeneration. shows the falciform ligament (f) just below the liver. The round ligament
(r) is at its posterior margin, and a slightly prominent paraumbilical vein (v)
is adjacent to the round ligament. B. Drawing of the view from within the
anterior abdominal wall. f ⳱ falciform ligament; r ⳱ round ligament; v ⳱
37,38). The diaphragmatic slip will become elongated and paraumbilical vein; D ⳱ diaphragm.
can be seen obliquely across the diaphragm. The elongated
diaphragmatic slip contains multiple linear echoes, usually ABDOMINAL AORTIC ANEURSYM
up to four, caused by two layers of diaphragm (Figs. 37,38).
This feature is diagnostic for a diaphragmatic slip, and when There are some controversies and pitfalls pertaining to
this is seen, a tumor can be excluded. When multiple dia- abdominal aortic aneurysm. An abdominal aortic aneurysm
phragmatic slips are seen, one may observe scalloping of the has been frequently defined as a focal dilatation of the ab-
diaphragm during deep inspiration (Fig. 39). dominal aorta to larger than 2.966,67 or 3 cm63–65 in diam-
eter. The size of normal aorta is highly variable and can be
as small as 1.2 cm in diameter in adults (Fig. 40). It is
usually larger superiorly, especially proximal to the renal
arteries. The normal aorta gradually tapers off distally.68
Focal dilatation of 3 cm compared with an aorta of 1.2 cm
is a 150% increase in diameter, but compared with an aorta
of 2.5 cm, there is only a 20% increase in diameter. There-
fore, the size of an aneurysm may have different clinical
significance in different patients. Other authors69–71 define
the diameter of the abdominal aorta as a ratio of more than
1.5, i.e., focal dilatation of the aorta with at least a 50%
increase in diameter compared with the expected normal
diameter of the aorta. In this regard, normal arterial diam-
eters determined from selected data in the literature should
be considered.69 Given the assumption that arterial (or aor-
FIG. 31. Small cyst in the thyroid gland. Transverse scan shows a 0.2-cm
cyst (arrowhead) in the thyroid gland. This most likely represents an en- tic) diameter proximal to a dilatation is normal, increase in
larged follicle. diameter greater than 50% of the artery proximal to the

148 H. C. YEH
FIG. 33. Drawings show four major types of

paraumbilical systemic collateral venous cir-
culation. pu ⳱ dilated paraumbilical vein; ie
⳱ inferior epigastric vein; sfe ⳱ superficial
epigastric vein; se ⳱ superior epigastric vein;
it ⳱ internal thoracic vein (internal mammary
vein). From: Yeh HC, et al. Paraumbilical ve-
nous collateral circulations: color Doppler ul-
trasound features. J Clin Ultrasound 1996;24:
359–66.
dilatation may also be considered an aneurysm.69 However, than 1.5 should be followed-up to prevent spontaneous rup-
because all aneurysms grow from small size, all small fo- ture of aneurysms. Therefore, defining an abdominal aortic
cal dilatations of the aorta should be reported. Because pa- aneurysm to be more than 3 cm in diameter or as having more
tients are usually asymptomatic before the rupture of an than 1.5 in diameter ratio is indeed not appropriate. A smaller
aneurysm, and because the mortality rate of spontaneous focal dilatation should also be included with the term “aneu-
rupture of an aortic aneurysm is high (>70%),72 patients rysm,” but at the same time, one should include diameter ratio
who have a small aneurysm with a diameter ratio smaller
FIG. 35. In a patient with type 1A paraumbilical systemic collateral venous

circulation, a sagittal scan of the anterior abdominal wall just above the
FIG. 34. Sagittal scan just superior to the left inguinal region shows the umbilicus shows a dilated inferior epigastric vein in the extraperitoneal fat
left inferior epigastric vein (IEGV) joining the external iliac vein (EIV). space, i.e., posterior to the rectus muscle (M) and anterior to the peritoneum
There are mixed color Doppler signals at the junction of the two veins (P). There is a small amount of ascites (a). Note that the distended vein
caused by turbulence. From: Yeh HC, et al. Paraumbilical venous collateral displaces the peritoneum posteriorly. From: Yeh HC, et al. Paraumbilical
circulations: color Doppler ultrasound features. J Clin Ultrasound 1996; venous collateral circulations: color Doppler ultrasound features. J Clin Ul-
24:359–66. trasound 1996;24:359–66.

or percentage of dilatation. In my experience, focal dilata-

tion of 2.5 cm or less or less than 1.5 in diameter ratio is
quite common. We should diagnose these small aneurysms
and not completely ignore them. Cook and Gallard73 re-
ported a mean growth rate of aneurysm in the 2.5- to 3.9-cm
group to be 2.2 mm during the first year, 2.8 mm during the
second year, and 1.8 mm during the third year. Correspond-
ing growth rates in the 4- to 4.9-cm group were 2.7 mm, 4.2
mm, and 2.2 mm for the first, second, and third years,
respectively. They suggest a policy of annual screening for
aneurysms measuring 2.5 to 2.9 cm and six monthly screen-
ings for those ⱖ4 cm. The growth rate for aneurysms
smaller than 2.5 cm is not known, but they probably should
be screened annually or at least biannually. The risk factors
associated with rapid expansion should also be considered
when determining the interval for follow-up examination.
According to Chang et al.,74 the risk factors for rapid ex-
pansion are advanced age, severe cardiac disease, previous
stroke, and history of cigarette smoking.
FIG. 36. Normal echogenicity of ovarian stroma in polycystic ovarian The measurement of an abdominal aortic aneurysm has
syndrome. A. Transverse endovaginal scan shows numerous, prominent, been considered to be more accurate on CT than on ultra-
developing follicles in the right ovary. There is a highly echogenic area sonography,72 and ultrasonographic measurements are fre-
(arrowhead) medially in the ovary caused by through transmission artifact
distal to a prominent follicle. In the area (open arrow) where no near-field quently performed on transverse images. Correct measure-
follicle is seen, the ovarian stroma has the same echogenicity as the uterus ment on ultrasonography is frequently more accurate than
(U) does. B. Transverse scan of left ovary shows a similar appearance as on CT without high-resolution, three-dimensional recon-
in right ovary, i.e., echogenic area (arrowhead) is distal to follicles, and in the
area not behind the follicle, the echogenicity is the same as in the uterus (U). struction, and the measurement on sagittal or coronal im-
FIG. 37. Diaphragmatic slip

simulates a liver mass. A. Sagittal
scan of the right lobe of the liver
shows a mass-like lesion (white
arrowheads) at the dome of the
liver. B. Scanning during deep in-
spiration, the lesion (white arrow-
heads) becomes more prominent.
It is associated with an indenta-
tion (black arrowhead) of the dia-
phragm superior to the lesion.
The indentation contains gas that
may represent lung. C. Rotating
the transducer while scanning,
the lesion (white arrowheads) be-
comes elongated and finally ap-
pears to run obliquely across the
diaphragm. It contains linear ech-
oes. This confirms the lesion to
be a diaphragmatic slip.

FIG. 38. Diaphragmatic slip
simulates peritoneal masses. A.
Sagittal scan of right hemidia-
phragm shows large amount of as-
cites in subphrenic region. Two
mass-like lesions, a and b, are at-
tached to the inferior surface of the
diaphragm. Note the slight indenta-
tion (arrowheads) of the diaphragm
at the mass-like lesion. B. As one
rotates the transducer to the oblique
position, the lesion (a) stretch out to
become a four-line structure. This
represents two layers of diaphragm
that fold to form a band-like struc-
ture. The central two lines represent
space between two layers of dia-
phragm. Another diaphragmatic slip
(b) is also seen, but not as clearly.
C. Scanning in at a slightly different
angle, the inferior slip (b) is now
clearly visualized as a four-line
structure. D. Computed tomography
scan from a different patient shows
a diaphragmatic slip (arrowhead)
with a two-layered structure. Com-
puted tomography scan from: Yeh
HC, et al. Anatomic variations and
abnormalities in the diaphragm seen
with US. Radiographics 1990;10:
1019–30.
FIG. 39. Diaphragmatic slips associated

with scalloping of the diaphragm. A. A mass-
like lesion in the liver and some smaller pro-
trusions (small arrowheads) from the dia-
phragm represent diaphragmatic slips. B. On
deep inspiration, the diaphragmatic slips (ar-
rowheads) become more prominent because of
muscular contraction. This is associated with
indentations (arrows) of the diaphragm. Be-
tween the indentations are superior bulgings
(between arrows) of the diaphragm, i.e., scal-
loping of the diaphragm. C. Computed tomo-
gram from a different patient shows similar
findings, scalloping of the diaphragm, because
of contraction of the diaphragmatic slips (ar-
rowheads). Note that the indentations (arrows)
of the diaphragm are seen. Computed tomog-
raphy scan from: Yeh HC, et al. Anatomic
variations and abnormalities in the diaphragm
with US. RadioGraphics 1990;10:1019–30.
tilt the transducer slightly caudad to obtain an appropriate

scanning plane for measuring the aneurysm. However, pre-
cisely tilting the transducer to the correct degree can be
difficult, especially if the aorta is tortuous. One may think
that an appropriate transverse scanning plane can be
determined by performing sagittal scanning, then turning
the transducer 90° to the selected point. Frequently, this is
not quite accurate, and it is much easier and more precise to
measure the maximum diameter on longitudinal scan-
ning because real-time ultrasound scanning is good for fol-
lowing the long axis of a tortuous aorta or aneurysm. One
FIG. 40. Small aortic aneurysm. Sagittal scan of the distal abdominal should measure the anterior-posterior diameter on sagittal
aorta shows a 1.6-cm aneurysm (arrow). The normal aorta superior to the scanning and the transverse diameter on coronal scan-
aneurysm measures only 1.2 cm in diameter (arrowhead). Therefore, the
aneurysm is 33.3% larger than the normal aorta. ning, the latter can be performed in the decubitus or, some-
times, the supine position. By measuring on the longitu-
dinal scan, an appropriate plane can be selected and the
ages is more accurate than on transverse images. To accu- maximum diameter is easier to determine. Furthermore, be-
rately measure diameter, one should measure the maximum cause lateral resolution usually is not as good as depth
diameter at the plane that is perpendicular to the long axis of resolution, and because refractory shadowing may also
the aneurysm. In the supine position, the lumbar spine obscure the lateral wall on ultrasonography, measuring
curves forward distally, as does the aorta. Therefore, to the transverse diameter on transverse section is usually less
accurately measure diameter on transverse scan, one has to desirable.
FIG. 41. Iliac artery aneurysm

simulates an abdominal aortic aneu-
rysm on CT. A. Computed tomogra-
phy scan shows a calcified, normal-
size aorta deviated to the right. B. An
abdominal aortic aneurysm of 5.9 cm
in transverse diameter was diagnosed
on this image. Note the multiple sur-
gical clips to the right of the aneu-
rysm. C. A 3-cm aneurysm is seen in
the left iliac artery. These were three
consecutive CT images. D. Ultra-
sound scan shows normal-size aorta
down to the bifurcation (arrow-
heads). E. Oblique scan shows a 2.5-
cm aneurysm (arrow) in the left com-
mon iliac artery. F. Drawing showing
how the left iliac aneurysm can be
mistaken for a large aortic aneurysm.
a, b, and c represent the section
planes of the CT scans on the top,
middle, and bottom left images, re-
spectively. d and e represent ultra-
sound-section planes on the top and
middle right images, respectively.
Note that in the b section plane, the
transverse diameter, including nor-
mal distal aorta and left iliac aneu-
rysm, simulates a large aortic aneu-
rysm.

152 H. C. YEH
Because most CT scans are performed in the axial plane delineated, unless heavy calcification is present. One should
only, an aneurysm of a tortuous aorta can be difficult to measure from outer wall to outer wall of the aneurysm.
accurately measure. An iliac artery aneurysm may even be When the wall of the aneurysm is thick, it may result in a
mistook for a large aortic aneurysm (Fig. 41). Unless a significant difference in the result of measurement if the
high-resolution, three-dimensional reconstruction is per- measurements are performed from the outer wall to inner
formed, an accurate measurement of an aortic aneurysm on wall or the inner wall to inner wall. The measurement of
a tortuous aorta with a complicated curve can be difficult on outer wall to outer wall will correlate much better with the
CT scanning. In a tortuous aorta with a complicated curve, surgical findings, and if one consistently measures in the
one part of the aorta may be in one oblique direction and same manner each time, the progressive change of an an-
another part in another oblique direction. When measuring eurysm during follow-up examinations will be more pre-
these two aneurysms, one should scan along the central axis cise, even if performed by a different ultrasonographer. Ac-
of each aneurysm and measure each aneurysm in two dif- curate measurement of the aneurysm is important. One of
ferent images. This can be performed with ultrasonography the criteria for surgical indication for abdominal aortic an-
(Fig. 42).75 Double aneurysms adjacent to each other may eurysm is an increase in diameter of 1 cm over the course of
sometimes be difficult to recognized on the axial CT scan 1 year. If follow-up examination is performed within 6
(Fig. 42). A small aneurysm is also easily missed on CT months or less, an increase in diameter of 0.5 cm will be
scan (Fig. 43). clinically highly significant. If intraobserver or interob-
How should the aneurysm be measured? Some people server error is larger than 0.3 cm, then, when an increase in
measure from the outer wall to inner wall, from the inner diameter of 0.6 cm is encountered on follow-up examina-
wall to inner wall,75 or with no criteria at all. Because tion, the clinical significance of this finding may be uncer-
gray-scale ultrasound has been greatly improved, the actual tain. This is because this may actually mean a difference of
thickness of the wall of an aortic aneurysm can be clearly only 0.3 cm, or maybe up to 0.9 cm, which is highly sig-
FIG. 42. Double abdominal aortic aneurysms in a

markedly tortuous aorta causing difficult interpreta-
tion on an axial CT scan. A. Ultrasound scan shows
two aneurysms in the distal abdominal aorta. The aorta
is markedly tortuous, curved to the right, then buckled
forward toward the left, then turns to the right again.
Dashed line indicates the long axis of the aorta. Maxi-
mum external diameter (between arrowheads of the
superior aneurysm) is measured. The wall of the distal
aneurysm is not clearly delineated in this scan and,
therefore, the distal aneurysm cannot be measured ac-
curately on this scan. B. Scanning along the long axis
of distal aneurysm, now the wall of the distal aneu-
rysm (arrows), is better seen and the aneurysm can be
measured. The portion of aorta between the two an-
eurysms is also better visualized. The superior aneu-
rysm, however, is not seen along its central axis and
appears smaller. C. Left drawing shows the three-
dimensional view of the double aneurysms. Right
drawing shows three consecutive scanning planes for
CT scans in the top, middle, and bottom right images.
Dashed line ⳱ central axis of aorta and two aneu-
rysms. Dotted line ⳱ correct diameters of two aneu-
rysms measured with ultrasound. Open dotted line ⳱
maximum diameter measured on CT in bottom left
image. D, E, and F. Three consecutive CT scans from
the superior aneurysm down to the distal aneurysm.
This was read as a single aneurysm, with the maxi-
mum dimension measured at the distal aneurysm (be-
tween arrowheads). The distal aneurysm is actually
smaller than the superior one, but because of the near-
horizontal position, it was markedly overmeasured
and was thought to represent the maximum diameter
of the single aneurysm. Ultrasound measurements: su-
perior aneurysm ⳱ 5 cm, distal aneurysm ⳱ 3.7 cm.
Computed tomography measurement: only one aneu-
rysm, 6.8 cm in maximum diameter.

FIG. 43. Small abdominal aortic

aneurysm. A. Sagittal scan shows a
small aneurysm in the distal abdomi-
nal aorta 2.16 cm in external diam-
eter. Normal aorta superior to the an-
eurysm measures 1.71 cm, i.e., the an-
eurysm is 20.8% larger than normal.
B. Transverse scan shows an oval
(taller than it is wide) 2.42-cm aneu-
rysm. It was measured along the plane
of the solid line in A rather than the
dashed line. The dashed line is the
correct plane for measurement be-
cause it is perpendicular to the long
axis of the aorta. Note the indistinct
lateral wall of the aneurysm, which
makes it difficult to obtain an accurate
transverse diameter. C, D, and E.
Consecutive CT scans from the nor-
mal aorta to the aneurysm. Because of
the small difference in aortic caliber,
the aneurysm in E was not detected.
Note the oval shape of the aneurysm
similar to that in B.
nificant. In my experience, when the aneurysmal wall is called “fetal lobation.” To differentiate this entity from scars
clearly delineated on both sides and the measurement is and tumors, one should carefully examine each renal lobe
performed on proper sagittal or coronal plane, the intraob- that consists of a pyramid overlaid by a “bump” of cortex on
server error is usually less than 0.1 cm. the renal contour.
Milk of calcium in a renal cyst is common, contrary to the
CONCLUSION belief that it is rare. Ultrasonography is much better than CT
for diagnosing milk of calcium, and even a tiny 0.2-cm
The term “column of Bertin” is a misnomer, and “septum lesion can be detected. On CT, milk of calcium can be
of Bertin” (or “cloisonné of Bertin”) is the proper term. mistaken for a stone or tumor. On ultrasonograph, however,
Although septum of Bertin has the same histologic structure milk of calcium sometimes may mimic a stone, angiomyo-
as renal cortex, it is usually much more echogenic than renal lipoma, or cyst with calcified wall. A small bleb that first
cortex because of the difference in orientation of the lobules appears in the gestational sac (single-bleb sign) represents
and vessels. This is caused by the anisotropic effect. an embryo–amniotic sac complex and not a yolk sac.
The terms “hypertrophic column of Bertin” and “lobar The double decidual sac sign has been considered one of
dysmorphism” should be called “junctional parenchyma,” the most important signs of an intrauterine pregnancy before
which represents the unresorbed renal parenchyma at the the amniotic sac, yolk sac, or embryo can be seen. In fact,
junction of two subkidneys that fuse to form a normal kid- the double decidual sac sign is formed only after approxi-
ney. The renal cortex, pyramid, and septa of Bertin within mately the eighth week of pregnancy and is difficult to
the junctional parenchyma are all normal, and there is noth- delineate. Even if it is delineated, a live fetus should be
ing hypertrophic or displaced. When the elements of the seen, and so there is no need for the double decidual sign to
junctional parenchyma are delineated by ultrasonography, diagnose an intrauterine pregnancy. The double echogenic
the diagnosis is certain and tumor can be excluded. rings during early pregnancy do not represent the double
A true hypertrophic septum (or septa) of Bertin is usually decidual sign.
hyperechoic, pear-shape, and symmetrically flanked by the Micronodulation is a highly specific and useful sign for
pyramids. It is usually located in the anterior aspect of the Hashimoto thyroiditis. Because of the increased incidence
midportion of the kidney, whereas a junctional parenchyma of papillary carcinoma in Hashimoto thyroiditis, one should
is usually located in the lateral aspect of the kidney at the carefully look for stippled calcifications in the nodules.
junction between the upper and middle thirds of the kidney. Recanalization of the umbilical vein rarely occurs. The
The term “fetal lobulation” should be more properly collateral vein seen is usually a dilated paraumbilical vein

154 H. C. YEH
that may drain through the inferior epigastric–external iliac 18. Palma LD, Bazzocchi M, Cressa C, et al. Radiological anatomy of the
kidney revisited. Br J Radiol 1990;63:680–90.
veins or the superior epigastric–internal thoracic veins. 19. Maklad NF, Chuang VP, Doust BD, et al. Ultrasonic characterization
Hyperechoic stroma is not a common or important sign in of solid renal lesions: echographic, angiographic, and pathologic cor-
polycystic ovarian syndrome. The hyperechogenicity of the relation. Radiology 1977;123:733–9.
stroma is usually caused by increased through transmission 20. Yeh HC, Halton KP, Shapiro RS, et al. Junctional parenchyma: revised
definition of hypertrophic column of Bertin. Radiology 1992;185:
caused by the presence of multiple developing follicles. 725–32.
Bilateral, multiple, developing follicles are the most impor- 21. Fine H, Keen EN. The arteries of the human kidney. J Anat 1966;100:
tant diagnostic feature for polycystic ovarian syndrome. 881–94.
22. Hoffer FA, Hanabergh AM, Teele RL. Interrenicular junction: a mimic
Diaphragmatic slips may simulate liver masses. Scanning of renal scarring on normal pediatric sonograms. AJR Am J Roentgenol
along the long axis of the slip shows a band-like structure 1985;145:1075–8.
with multiple linear echoes. This is the diagnostic feature. 23. Rubin JM, Carson PL, Meyer CR. Anisotropic ultrasonic backscatter
from the renal cortex. Ultrasound Med Biol 1988;14:507–11.
Defining an abdominal aortic aneurysm as having a di- 24. Patriquin H, Lefaivre JF, Lafortune M, et al. Fetal lobation, an
ameter greater than 3 cm or with a diameter ratio of greater anatomo-ultrasonographic correlation. J Ultrasound Med 1990;9:
than 1.5 will prevent many patients who have smaller an- 191–7.
25. Widder DJ, Newhouse JH. The sonographic appearance of milk of
eurysms from needing follow-up examinations. Measure- calcium in renal calyceal diverticuli. J Clin Ultrasound 1982;10:488.
ment of an aneurysm should be performed on a longitudinal 26. Walker WH, Pearson RE, Johnson NR. Milk of calcium renal stone:
scan along the long axis of the aneurysm. When the aorta is Case report. J Urol 1960;84:517.
markedly tortuous with multiple aneurysms, each aneurysm 27. Becker JA, Staiano S. Milk of calcium in a renal cyst: A sonolucent
mass effect. J Clin Ultrasound 1975;3:135.
should be measured along its own long axis. 28. Diakoumakis E, Vieux U, Seife B, et al. Case report: Ultrasonographic
features of “milk of calcium” in kidney. Mt Sinai J Med 1985;52:656.
Acknowledgment: The author thanks Ms. Gwen Hanley for her 29. Tey PH, Ahmed A, Chan KF. Case report: milk of calcium in kidney
excellent secretarial work. after cyst puncture. Mt Sinai J Med 1985;52:659.
30. Dyer R, Montgomery J, Thiele A, et al. Right upper quadrant renal
milk of calcium mimicking limy bile. J Clin Ultrasound 1987;15:140.
REFERENCES 31. Yeh HC, Mitty HA, Halton K, et al. Milk of calcium in renal cysts:
new sonographic features. J Ultrasound Med 1992;11:195–203.
1. Hodson J. The lobar structure of the kidney. Br J Urol 1972;44: 32. Vandeursen H, Baert L. An intrarenal opacity resisting extracorporeal
246–61. shock wave lithotripsy. J Urol 1990;144:961.
2. Mahony BS, Jeffrey RB, Laing F. Septa of Bertin: a sonographic 33. Yeh HC, Goodman JD, Carr L, et al. Intradecidual sign: an US crite-
pseudotumor. J Clin Ultrasound 1983;11:317–9. rion of early intrauterine pregnancy. Radiology 1986;161:463–7.
3. Netter F. Kidneys, ureters and urinary bladder. In: Shapter RK, Yon- 34. Bernard KG, Cooperberg PL. Sonographic differentiation between
kman FF, eds. The Ciba Collection of Medical Illustration. Summit, blighted ovum and early viable pregnancy. AJR Am J Roentgenol
NJ: Ciba Pharmaceutical, 1979;6:5. 1985;144:597.
4. Wolfman NT, Bechtold RE, Watson NE. Hypertrophic column of 35. Levi CS, Lyons EA, Lindsay DJ. Early diagnosis of nonviable preg-
Bertin. In: Resnick ML, Rifkin MD, eds. Ultrasonography of the nancy with endovaginal ultrasound. Radiology 1988;167:383–5.
Urinary Tract, 3rd ed. Baltimore: Williams & Wilkins, 1991:109–51. 36. Yeh HC, Rabinowitz JG. Endovaginal ultrasonographic measurement
5. Lafortune M, Constantin A, Breton G, et al. Sonography of the hy- of embryonic size (Letter). J Ultrasound Med 1995;14:97–100.
pertrophied column of Bertin. AJR Am J Roentgenol 1986;146:53–6. 37. Horrow SR. Yolk sac sign: sonographic appearance of the fetal yolk
6. Leekman RN, Martizinger MA, Brunelle M, et al. The sonography of sac in missed abortion. J Ultrasound Med 1986;5:435–8.
renal columnar hypertrophy. J Clin Ultrasound 1983;11:491–4. 38. Bradley WG, Fishe CE, Filly RA. The double sac sign of early intra-
7. Charghi A, Dessureault P, Drouin G, et al. Malposition of a renal lobe uterine pregnancy: use in exclusion of ectopic pregnancy. Radiology
(lobar dysmorphism): a condition simulating renal tumor. J Urol 1971; 1982;143:223–6.
105:326–9.
39. Nyberg DA, Laing FC, Filly RA, et al. Ultrasonographic differentia-
8. Dacie JE. The “central lucency” sign of lobar dysmorphism (pseudo-
tion of the gestational sac of early intrauterine pregnancy from pseu-
tumor of the kidney). Br J Radiol 1976;49:39–42.
dogestational sac of ectopic pregnancy. Radiology 1983;146:755–9.
9. Felson B, Moskowitz M. Renal pseudotumors: the segmented nodule
and other lumps, bumps and dromedary humps. AJR Am J Roentgenol 40. Yeh HC. Efficacy of the intradecidual sign and fallacy of the double
1969;107:720–9. decidual sac sign in the diagnosis of early intrauterine pregnancy.
10. Lopez FA. Renal pseudotumors. AJR Am J Roentgenol 1972;129: Radiology 1999;210:579–82.
172–84. 41. Fates MC, Laing FC. Sonographic evaluation of ectopic pregnancy: an
11. Popky GL, Bogash M, Pollack H, et al. Focal cortical hyperplasia. J update. AJR Am J Roentgenol 1995;165:251–9.
Urol 1969;102:657–60. 42. Williamson MR, Rosenberg RD. Computed tomography and ultra-
12. Hodson CJ, Mariani S. Large cloisons. AJR Am J Roentgenol 1982; sound of the thyroid and parathyroid glands. In: Eisenberg B, ed.
139:327–32. Imaging of the Thyroid and Parathyroid Glands: A Practical Guide.
13. King MC, Friedenberg RM, Tena LB. Normal renal parenchyma simu- New York: Churchill Livingstone, 1991:111.
lating tumor. Radiology 1968;91:217–22. 43. Gooding GAW. Sonography of the thyroid and parathyroid. Radiol
14. Hartman GW, Hodson CJ. The duplex kidney and related abnormali- Clin North Am 1993;31:967.
ties. Clin Radiol 1969;20:387–400. 44. Nordmeye JP, Shafeh A, Heckmann C. Thyroid sonograph in autoim-
15. Thornbury JR, McCormick TL, Silver TM. Anatomic/radiologic clas- mune thyroiditis: A prospective study on 123 patients. Acta Endocri-
sification of renal cortical nodule. AJR Am J Roentgenol 1980;134: nol 1990;122:391.
1–7. 45. James EM, Charboneau JW, Hay ID. The thyroid. In: Rumack CM,
16. Meaney TF. Errors in angiographic diagnosis of renal masses. Radi- Wilson SR, Charboneau JW, eds. Diagnostic Ultrasound. St. Louis:
ology 1969;93:361–6. Mosby-Year Book, 1991:507.
17. Flynn VJ, Gittes RF. Benign cortical rest: “a pseudotumor” of the 46. Yeh HC, Futterweit W, Gilbert P. Micronodulation: Ultrasonographic
kidney. J Urol 1972;108:54–7. sign of Hashimoto thyroiditis. J Ultrasound Med 1996;15:813–9.

47. Li Volsi. Surgical Pathology of the Thyroid. Philadelphia: WB Saun- a sonographic sign of diaphragmatic disease. J Clin Ultrasound 1982;
ders, 1990:68. 10:42–5.
48. de los Santos E, Keyhani-Rofagha S, Cunningham JJ, et al. Cystic 63. Pedersen OM, Vik-MO H, Trippestad A. Diagnosis of small infrarenal
thyroid nodules. Arch Intern Med 1990;150:1422–7. aortic aneurysms by ultrasonography. Int Angiol 1996;15(4):328–34.
49. Hammer M, Wortsman J, Folse R. Cancer in cystic lesions of the 64. Kanagasabay R, Garaj H, Pointon L, et al. Co-morbility in patients
thyroid. Arch Surg 1982;117:1020–3. with abdominal aortic aneurysms. J Med Screen 1996;3(4):208–10.
50. Lafortune M, Constantin A, Breton G, et al. The recanalized vein in 65. Gooding GAW. The abdominal great vessels. In: Rumack CM, Wilson
portal hypertension: a myth. AJR Am J Roentgenol 1985;144:549–53. SR, Charboneau JW, eds. DiagnosticUltrasound. Chicago: Mosby-
51. Saddekni S, Hutchinson DE, Cooperberg PL. The sonographic patent Year Book. 1991:335–51.
umbilical vein in portal hypertension. Radiology 1982;145:441–3. 66. Lindholt JS, Fasting H, Henneberg EW, et al. Preliminary results of
52. Martin BF, Tudor RG. The umbilical and paraumbilical veins of man. screening for abdominal aortic aneurysm in the county of Vibort.
J Anat 1980;130:305–22. Ugeskr Laeger 1997;159(13):1920–3.
53. Subramanyam BR, Balthazar EJ, Madamba MR, et al. Sonography of 67. Simoni G, Gianotti A, Andia A, et al. Screening study of abdominal
portosystemic venous collaterals in portal hypertension. Radiology aortic aneurysm in a general population: lipid parameters. Cardiovasc
1983;146:161–6. Surg 1996;4(4):445–8.
54. Yeh HC, Stancato-Pasik A, Ramos R, et al. Paraumbilical venous 68. Steinberg CR, Archer M, Steinberg I. Measurement of the abdominal
collateral circulations: color Doppler ultrasound features. J Clin Ul- aorta after intravenous aortography in health and arteriosclerotic pe-
trasound 1996;24:359–66. ripheral vascular disease. AJR Am J Roentgenol 1965;95:703–8.
55. Battaglia C, Regnani G, Antin PG, et al. Polycystic ovary syndrome: 69. Johnston KW, Rutherford RB, Tilson MD, et al. Suggested standards
a new ultrasonographic and color Doppler pattern. Gynecol Endocrinol for reporting on arterial aneurysms. J Vasc Surg 191;13:452–8.
2000;14:417–24. 70. Hollier LH, Wisselink W. Abdominal aortic aneurysm. In: Haimovici
56. Pache TD, Wladimiroff JW, Hop WCJ. How to discriminate between H, Ascer E, Standness DE, et al., eds. Haimovici’s Vascular Surgery.
normal and polycystic ovaries: transvaginal US study. Radiology Cambridge, Massachusetts: Blackwell Science, 1998:797–827.
1992;183:421–3. 71. Goldstone J. Aneurysm of the aorta and iliac artery. In: Moore WS, ed.
57. Saxton DW, Farguhar CM, Beard TR, et al. Accuracy of ultrasound Vascular Surgery, 5th ed. Philadelphia: WB Saunders, 1998:435–55.
measurements of female pelvic organ. Br J Obstet Gynecol 1990;97: 72. Pearce WH, Salyapongse AN, Fitzgerald SW, et al. Computerized
695–9. tomographic scanning and vascular disorders. In: Callow AD, Ernst
58. Adams J, Polson DW, Abdulwahid N, et al. Multifollicular ovaries: CB, eds. Vascular Surgery: Theory and Practice. Stamford, CT:
clinical and endocrine features and response to pulsatile gonadotropin Appleton & Lange, 1995:394.
releasing hormone. Lancet 1985;2:1375–8. 73. Cook TA, Gallard RB. A prospective study to define the optimum
59. Yeh HC, Futterweit W, Thornton JC. Polycystic ovarian disease: US rescreening interval for small abdominal aortic aneurysm. Cardiovasc
features in 104 patients. Radiology 1987;163:111–16. Surg 1996;4(4):441–4.
60. Yeh HC, Halton KP, Gray CE. Anatomic variations and abnormalities 74. Chang JB, Stein TA, Liu JP, et al. Risk factors associated with rapid
in the diaphragm seen with US. Radiographics 1990;10:1019–30. growth of small abdominal aortic aneurysms. Surgery 1997;121(2):
61. Auh YH, Rubenstein WA, Zirinsky K, et al. Accessory tissues of the 117–22.
liver: CT and sonographic appearance. AJR Am J Roentgenol 1984; 75. Lanne T, Sandgren T, Mangell P, et al. Improved reliability of ultra-
143:565–72. sonic surveillance of abdominal aortic aneurysm. Eur J Vasc Endovasc
62. Worthen JH, Worthen WF II. Disruption of the diaphragmatic echoes: Surg 1997;13(2):149–53.

Some Misconceptions and Pitfalls in Ultrasonography

Uploaded by

Copyright:

Available Formats

Some Misconceptions and Pitfalls in Ultrasonography

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Some Misconceptions and Pitfalls in Ultrasonography

Uploaded by

Copyright:

Available Formats

Ultrasound Quarterly

Vol. 17, No. 3, pp. 129–155

Some Misconceptions and Pitfalls in Ultrasonography

TABLE 1. Previous terms used to describe

FIG. 2. Formation of renal lobe, cortex, and

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 3. A pseudotumor caused by

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 4. Example of a gross specimen. The drawing from a kidney speci-

the renal collecting system is dilated. However, it is quite

Ultrasound Quarterly, Vol. 17, No. 3, 2001

fact, it is a line between two subkidneys and represents

When two subkidneys are not in straight alignment, this will

FIG. 6. Classification of junctional parenchyma. From: Yeh HC, et al.

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 8. Three variations in offset align-

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 9. Dromedary hump caused

Ultrasound Quarterly, Vol. 17, No. 3, 2001

thicker layer of milk of calcium, and acoustic shadowing

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 11. True hypertrophic septum

calcification in a cyst wall. A proper scan as previously 1. No lesion detected.

Ultrasound Quarterly, Vol. 17, No. 3, 2001

3. Seen as a simple cyst.

scanning a small lesion and precisely evaluate the density

FIG. 18. Milk of calcium in a renal cyst

early stage is called the primary yolk sac, which represents

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 21. Cyst with milk of calcium simulat-

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 22. Multiple cysts with milk of calcium in a kidney. A. A cyst

and double echogenic rings usually appear on oblique scan

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 24. True double decidual sac in patient in eighth week of

Ultrasound Quarterly, Vol. 17, No. 3, 2001

thick wall. Small cystic lesions smaller than 0.3 cm usually

RECANALIZATION OF UMBILICAL VEIN

In adults, the umbilical vein is usually completely oc-

Ultrasound Quarterly, Vol. 17, No. 3, 2001

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 28. Hashimoto thyroiditis with a papillary carcinoma. A. Sagittal

Ultrasound Quarterly, Vol. 17, No. 3, 2001

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 33. Drawings show four major types of

FIG. 35. In a patient with type 1A paraumbilical systemic collateral venous

Ultrasound Quarterly, Vol. 17, No. 3, 2001

or percentage of dilatation. In my experience, focal dilata-

FIG. 37. Diaphragmatic slip

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 39. Diaphragmatic slips associated

tilt the transducer slightly caudad to obtain an appropriate

FIG. 41. Iliac artery aneurysm

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 42. Double abdominal aortic aneurysms in a

Ultrasound Quarterly, Vol. 17, No. 3, 2001

FIG. 43. Small abdominal aortic

Ultrasound Quarterly, Vol. 17, No. 3, 2001

Ultrasound Quarterly, Vol. 17, No. 3, 2001

Ultrasound Quarterly, Vol. 17, No. 3, 2001