DOI:10.2214/AJR.07.2489
AJR 2007; 189:W177-W183
© American Roentgen Ray Society
Radiologic Anatomy of the Inguinofemoral Region: Insights from MDCT
P. T. Cherian1 and
A. P. Parnell2
1 Liver Surgery Secretaries, Queen Elizabeth University Hospital, Nuffield
House, 3rd Fl., Birmingham, United Kingdom, B15 2TH.
2 Department of Radiology, Good Hope Hospital, Sutton Coldfield, United
Kingdom.
Received November 14, 2006;
accepted after revision May 19, 2007.
Address correspondence to P. T. Cherian.
WEB
This is a Web exclusive article.
Abstract
OBJECTIVE. We set out to reexamine the radiologic anatomy of the
inguinofemoral region using volume data sets obtained with an MDCT
scanner.
MATERIALS AND METHODS. We conducted a systematic prospective review
of CT scans of 20 consecutively enrolled patients, 10 men and 10 women chosen
retrospectively from our CT database. An experienced radiologist and a senior
trainee surgeon conducted an image review to maximize recognition of relevant
anatomic detail.
RESULTS. The inferior epigastric artery and femoral canal were
identified in all planes in all patients. On axial views a spur on the pubic
bone was visible in 17 (85%) of the patients, but the inguinal ligament was
not reliably identified in any. The round ligament or spermatic cord was
visible in only 15 (75%) of 20 patients. In contrast, on coronal and sagittal
views, the inguinal ligament, which is vital to reliable identification and
accurate classification of groin hernias, was visible in 19 (95%) of the 20
patients. Scans in the sagittal plane best depicted the gutter-like aspect of
the ligament, the canal and contents being clearly visible in 95% of the
patients. On sagittal views, the internal ring was identifiable in 90% and the
round ligament or spermatic cord in 95% of the patients. On coronal images,
the internal ring was identified in all and the conjoint tendon in 95% of the
patients. The round ligament or spermatic cord was not seen in 10% of the
patients.
CONCLUSION. MDCT produces images of the inguinal region in detail
not possible with previous generations of scanners. In our small series, 100%
identification of key anatomic structures was achieved when information from
all three views was combined. We found subtle differences between imaging
findings and standard anatomic teaching.
Keywords: CT • femoral hernia • inguinal hernia
Introduction
Symptomatic femoral hernia can be clinically occult in 20–25% of
cases [1]. Surgical guidelines
call for urgent operation on asymptomatic femoral hernias, whereas direct
inguinal hernias can be managed conservatively
[2]. Accurate classification of
groin hernias is important in both the presence and the absence of symptoms.
Until the advent of MDCT scanners capable of high-speed acquisition of
thin-slice data, high-resolution CT images were limited to the axial plane
[3–6].
This limitation made consistent interpretation of crucial landmarks demanding
and led to at least perceived difficulty in interpreting scans of the inguinal
region. We believed that the volume data set obtained from the helical path of
these scanners would allow us to visualize detail in both the coronal and the
sagittal views to a degree that would enable accurate and consistent
identification of structures in the inguinal region. To our knowledge, no
article has described these details, which are important because pathologic
changes in the groin are neither rare nor insignificant.
A brief overview of anatomic features relevant to CT interpretation is
essential. Detail about the anatomic course of structures that run through the
internal inguinal ring aids in identification of these structures on CT
[7]. The bone landmarks that
form the basis of interpretation are the anterior superior iliac spine, the
pubic tubercle (a forward projecting prominence in the lateral aspect of the
pubic crest), and the superior pubic ramus. The upper, sharp ridge of this
ramus is called the pectineal line, and the thickened periosteum overlying
this line forms the pectineal ligament
(Fig. 1).
The ligament that joins the iliac spine to the pubic tubercle is the
inguinal ligament (IL) of Poupart, which in essence is the folded-up lower
border of the external oblique muscle, the outermost of the three muscles of
the anterior abdominal wall. The lower aponeurotic part of the external
oblique muscle forms the anterior wall of the inguinal canal. Immediately
superior and lateral to the pubic tubercle a V-shaped gap in the external
oblique aponeurosis forms the external inguinal ring. The folded lower border
of the external oblique muscle forms a gutter that constitutes the floor of
the inguinal canal. The inner two muscles, the internal oblique and the
transversus abdominis, both of which arise in part from the lateral half of
the IL, join in the medial aspect to form the conjoint tendon, which has a
free edge that arches over the spermatic cord or round ligament to attach
itself to the pectineal line. Therefore, the conjoint tendon forms both the
roof and the posterior wall of the inguinal canal and is the layer onto which
mesh is laid when during a Liechtenstein-type inguinal hernia repair. The
conjoint tendon is in essence one of the layers the weakness of which is
strengthened by the mesh.
Deep in relation to the arch of the conjoint tendon is the weak
transversalis fascia, through which the round ligament or spermatic cord exits
the abdomen to form the internal ring. The inferior epigastric artery (IEA),
which lies on the surface of the conjoint tendon immediately medial to the
cord, defines the medial border of the internal ring. In men, the spermatic
cord is made up of three coverings derived from the layers of the abdominal
wall and contains among structures the vas deferens, testicular arteries and
veins, and the processus vaginalis (the obliterated tunica vaginalis of the
testis, which when patent forms the sac of an indirect hernia). The vas
deferens is a continuation of the testicular epididymis, runs through the
canal, and enters the abdomen at the internal ring. It then passes along the
side wall, crosses over the iliac vessels and the ureter lying on the
obturator fascia, curves medially to the floor of the pelvis, and reaches the
back of the bladder. Throughout its course, the vas deferens is in constant
contact with the peritoneum. Gonadal arteries are direct branches from the
aorta and run down over the psoas muscle. The testicular arteries pass along
the pelvic brim to exit the abdomen through the internal ring. The left
gonadal vein drains into the left renal vein and the right into the inferior
vena cava. In women, the round ligament extends from the uterus through the
broad ligament, runs along the pelvic sidewall, leaves the abdomen through the
internal ring, runs in the inguinal canal, and inserts itself into the vulval
labia. By definition, an indirect inguinal hernia is one in which the neck of
the sac is lateral in relation to the IEA, through the internal ring. A direct
inguinal hernia is one in which the sac is medial in relation to the IEA,
through the inguinal triangle of Hesselbach formed by the IEA in the lateral
aspect, the lateral border of the rectus muscle in the medial aspect, and the
IL in the inferior aspect.
The deep circumflex iliac artery arises from the external iliac artery
close to the IEA and passes toward the anterior superior iliac spine deep in
relation to the lateral aspect of the IL
[8]. From the medial end of the
IL, the lacunar ligament of Gimbernat extends backward to the pectineal line
to form the medial border of the femoral ring. The femoral ring is the widest,
most proximal part of the femoral canal, which itself is the space medial in
relation to the femoral vein within the covering femoral sheath (an extension
of the transversalis fascia). The femoral ring has three other borders, the IL
in the anterior aspect, the pectineal ligament in the posterior aspect, and
the femoral vein in the lateral aspect
(Fig. 2). A femoral hernia
enters the femoral canal through the femoral ring and lies within the femoral
triangle. The femoral triangle lies in the plane immediately below the deep
fascia of the thigh (fascia lata) and is bordered anatomically by the IL, the
medial border of the ribbonlike sartorius muscle, and the medial border of the
adductor longus muscle.
Materials and Methods
From the database of an MDCT scanner, a specialist radiologist with 18
years of practice in gastrointestinal radiology and a senior trainee surgeon
with 8 years of experience in surgical gastroenterology selected the images of
20 patients, 10 of each sex, who had undergone CT that included the entire
inguinofemoral region but was performed for symptoms not related to a groin
disorder. A prospective decision was made to select consecutively registered
patients to fill four two-by-two cohorts, one half of each group being younger
than 60 years and the other half older than 60 years. All scans were obtained
with a 16-MDCT scanner (Brilliance, Philips Medical Systems). The acquisition
collimation was 1.5 mm, and workstation reconstructions were routinely 2 mm
thick in all three planes for the purposes of the study. All scans were
obtained with IV contrast enhancement. Contrast enhancement was routinely
performed with 100 mL of ioversol 300 (Optiray 300, Tyco Healthcare). A
70-second delay between the injection and acquisition of images was used in
all cases. All scans were obtained in the portal venous phase with no scan
delay at the iliac crest. All patients received oral contrast medium in the
form of a 3% solution of meglumine diatrizoate (Gastrografin, Schering).
The exclusion criteria were the presence of pelvic or inguinal disease that
would distort the normal anatomic configuration (extremes of age were avoided
for the same reason) and the presence of metallic implants in the vicinity
that would compromise scan quality. Recruitment into the study was based on
fulfillment of the required number of patients in a fixed time frame. At
selection we ensured that one half of each cohort would be younger than 60
years to minimize bias that might have arisen from age-related muscular
atrophy. We considered this criterion especially important in evaluation of
the femoral canal and space. All scan requests routinely filed in the X-ray
film packets at our hospital were scrutinized to ensure absence of groin
symptoms and signs. We accept, however, that it is theoretically possible that
the referring clinician might have omitted the mention of groin symptoms.
Each patient's scans were reviewed on a computer workstation with regard to
specific anatomic landmarks. A checklist specific to each plane of view was
completed for the axial, coronal, and sagittal views. The IL, IEA, round
ligament or spermatic cord, internal ring, and femoral space with its contents
were sought in all three planes. In addition, the pubic spur was sought on
axial views, the inguinal canal on sagittal views, and the conjoint tendon on
coronal views. A structure was deemed visible only if the appearances and
route conformed to standard anatomic texts
[7,
8] and therefore confirmed the
identity of the structure beyond reasonable doubt. In the coronal view,
adjustments in angle of view were made to compensate for protuberance of the
anterior abdominal wall if the patient was obese, so the final image was
viewed in parallel to the abdominal wall. Adjustment was needed for most of
the patients in the study. Because an observational review of retrospective
scans with no patient-identifying details was being used, ethics board
approval was not required.
Results
A total of 20 patients fit into four groups: five women older than 60 years
(mean age, 76 years; range, 70–82 years), five women younger than 60
years (mean age, 48 years; range, 31–54 years), five men older than 60
years (mean age, 71 years; range, 62–79 years), and five men younger
than 60 years (mean age, 47 years; range, 34–58 years)
(Table 1). In isolation it
became obvious that for certain structures, some views were better than others
(Table 2). Conversely, it was
possible to use other structures as reference points in the comparisons of
views. For example, a lymph node in the femoral space and the IEA were seen in
more than one view, facilitating further identification through correlation of
adjacent structures.
On review of the axial images, the IEA and the femoral space (i.e., the
femoral canal) were positively identified in all patients. A previously
described bone spur (Fig. 3)
was visible on the pubic bone on scans of 17 (85%) of the 20 patients. We then
looked for the IL and found that despite the advances in technology, the
ligament was not reliably visible on axial images of any of the 20 patients.
The round ligament or spermatic cord was visible on the axial scans of only 15
(75%) of the 20 patients. An interesting finding was that the round ligament
(Fig. 4) was visible in only
six of the women, whereas the spermatic cord was visible in nine of the men.
In both groups of women, two patients had undergone hysterectomy, but this
factor did not seem to affect identification in our series. Visibility of the
round ligament or spermatic cord on axial images was important because we
often (although not always) relied on these structures for further
confirmation of the site of both the internal and, especially, the external
inguinal ring, leading to only 80% (16 of 20 patients) identification of the
internal ring on axial images.
View larger version (129K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4 —74-year-old woman. Axial CT scan shows round ligament
(large arrow) entering internal inguinal ring and inferior epigastric
vessels (small arrow) defining medial aspect of internal inguinal
ring. Fibroid, dilated small bowel and small amount of free peritoneal fluid
are evident.
|
|
On coronal slices, the IEA, femoral canal, and internal ring were
positively identified in all patients. The conjoint tendon
(Fig. 5) was identifiable on
the scans of 19 (95%) of the 20 patients, making the coronal view the best by
far for identification of this structure. The round ligament or spermatic cord
was not seen in two (10%) of the 20 patients, both of whom were women. In some
of the men, vascular enhancement of the testicular vessels helped in
visualization of the spermatic cord (Fig.
6). Unlike the situation with the axial images, however, the IL
(Fig. 7), which is vital to
reliable interpretation, was visible on the coronal scans of 19 patients. We
often found that identification of the IL was helped by the easy visibility of
the contrast-enhanced deep circumflex iliac artery
(Fig. 8), which arises from the
lateral aspect of the external iliac artery nearly opposite the IEA and
ascends obliquely and laterally toward the anterior superior iliac spine in a
course almost identical to that of the IL.
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6 —64-year-old man. Coronal reconstruction CT scan shows
inguinal canal. Arrowheads indicate inguinal ligaments. Thin arrows indicate
inferior epigastric vessels. Thick arrow indicates right testicular vessel
passing along right inguinal canal.
|
|
View larger version (103K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8 —82-year-old woman. Coronal reconstruction CT scan shows
inguinal ligaments (arrows) on both sides. Arrowhead indicates deep
circumflex iliac artery, which crosses immediately below inguinal
ligament.
|
|
On sagittal views, the IEA and the femoral canal were positively identified
in all patients. Again in contrast to the situation with axial images, the IL
was visible on the sagittal scans of 19 (95%) of the 20 patients. The sagittal
view best depicted the gutter formed by the IL. On the scans of 19 of the 20
patients, even the canal and its contents were clearly visible. The internal
ring (Fig. 9) was identifiable
on the scans of 18 (90%) of the 20 patients. The detail allowed us to
visualize the continuity of the external spermatic fascia with the external
oblique fascia from which it is derived. The round ligament or spermatic cord
was seen on the scans of 19 of the 20 patients.
View larger version (100K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9 —37-year-old man. Sagittal reconstruction CT scan shows
inguinal ligament and canal. Large arrowhead indicates testicular vessels
passing from psoas through transversalis fascia (small arrowheads)
and into internal inguinal ring. Floor of ring is inguinal ligament
(arrow).
|
|
When information derived from all three views was combined
(Table 3), the IEA, the
internal ring, the IL and canal with the round ligament or spermatic cord, and
the femoral canal were positively identified in all patients. In other words,
in every patient every structure vital to complete interpretation of CT scans
of the groin was reliably visible in one view or another.
Discussion
In a study [9] in which the
rates of detection of inguinal hernia with physical examination, sonography,
and MRI were compared with those for laparoscopy as the reference standard,
the sensitivity and specificity were 74.5% and 96.3% for physical examination,
92.7% and 81.5% for sonography, and 94.5% and 96.3% for MRI. Even when
symptoms were severe enough to warrant surgical intervention, 20–25% of
femoral hernias were not diagnosed preoperatively by surgical staff
[1]. It is clear that the
complex anatomic features of the groin occasionally necessitate adjuvant
investigations.
CT has inherent flaws that make it a less than ideal tool in hernia
detection in outpatients with recurrent groin symptoms. Acquisition with the
patient in the supine position maximizes reduction of most reducible hernias,
and with the commonly used adynamic scanning, effort-induced changes cannot be
recorded. In addition, in the era of laparoscopic hernia repair, preoperative
differentiation of types of inguinal hernia often is not necessary
[10]. In certain
circumstances, however, preoperative diagnosis with CT may be desirable. These
situations include acute presentations, such as gastrointestinal obstruction
due to an occult cause, in which the operative approach would be different for
different pathologic conditions; acute groin symptoms with no or equivocal
signs; previous surgery in the surrounding region or previous hernia repair;
and morbid obesity, in which clinical signs are limited.
In part because of the aforementioned shortcomings, CT has been of limited
value in the diagnosis of groin hernia. The hernial sac can be directly
visualized on herniography
[11]. Because, however, it is
performed with contrast medium and ionizing radiation, is minimally invasive,
and can have complications
[12], herniography has largely
been superseded by alternative techniques. Sonography, CT, and MRI have been
used. Sonography with high-frequency (7.5–10 MHz) transducers can depict
fascial and muscular layers of the abdominal wall noninvasively but is highly
operator and patient (body habitus) dependent
[13,
14]. The chief advantage of
MRI is the possibility that images can be obtained in any plane. Like
sonography, MRI can be used for dynamic examinations. It also beautifully
depicts layers of the abdominal wall, and coronal MR images show the so far
elusive IL [15]. MRI, however,
is expensive, has limited availability, and is rarely used in emergencies. In
addition, the inferior epigastric vessels can be difficult to identify in some
planes [15].
Until recently, reliance has been placed on axial CT scans in the diagnosis
and classification of groin hernias. Scanner technology during the period of
the initial studies
[3–6,
16,
17] made it impossible to
produce high-resolution sagittal and coronal images that allowed visualization
of relevant anatomic structures. With the currently available volume data sets
from MDCT, key planes such as the coronal and sagittal can be reconstructed
with minimal loss of image quality compared with the axial plane. Our findings
confirm that crucial structures such as the IL are not easily visualized on
axial scans [18], although
axial CT images of the IL have been obtained
[16]. Delabrousse et al.
[4] designed surrogate markers,
such as the pubic tubercle, for differentiating hernia types. In one study
[19], CT was used after
herniography to try to improve sensitivity, which despite the effort was only
75% for detection of hernia. This result might have been due to absence of IV
contrast enhancement and 10-mm slice collimation. An interesting finding in
that study was bilateral bone spurs on the pubic bone, which the authors
extrapolated to be a possible cause of chronic groin pain. We found this spur
on the pubis on axial views of 17 (85%) of our patients, who were undergoing
CT for reasons other than groin pain. We believe the spurs may simply
represent partial ossification of the tendinous insertion of the pectineus
muscle. Study with a larger series of patients is necessary to confirm whether
this finding is prevalent enough to be considered a normal anatomic
feature.
In the axial plane, the round ligament was seen less often in our series
(60%) than was the spermatic cord (90%). This discrepancy may be due to the
smaller size of the ligament compared with the cord; absence of vascularity
within the ligament; and in some part age-related atrophy, because most of the
women in the study were postmenopausal (mean age, 62 years).
Although the femoral canal and triangle were visible in all views, the
coronal sections displayed these structures at their best as triangular spaces
containing mainly fat. We found, however, that the triangular space visible on
CT was not the anatomic femoral triangle but merely a part of it
(Fig. 10). We propose calling
this area the "radiologic femoral triangle." The anatomic femoral
triangle, because it is an anatomic entity that curves around the anterior
aspect of the thigh, was more difficult to visualize in the linear planes
available on CT. With appropriate reconstruction, the radiologic femoral
triangle itself was seen crisply in almost every patient. The medial aspect of
the femoral vein borders the triangle in the lateral aspect, the inferior
border of the IL in the superior aspect, and the lateral border of the
pectineus muscle in the medial aspect. The importance of the radiologic
femoral triangle is that it includes the anatomic femoral space and forms a
canal along with the adjacent fat and therefore is a potential site of a
femoral hernia. This triangle with its clear radiologic borders acts as a
surrogate site marker for the femoral canal, which in our series was not
identified because the femoral sheath that defines it is not visible even on
MDCT. The dimensions of this canal would vary depending on the patient's
habitus. This hypothesis is currently being tested at our institution.
Nevertheless the longitudinal dimension of the radiologic femoral triangle in
our series was normally approximately 4 cm, shorter than the dimension of the
anatomic femoral triangle. The anatomic femoral canal has been quoted as being
1.25–2 cm long [20].
Clearly the radiologic femoral triangle, the anatomic femoral triangle, and
the anatomic femoral canal are different entities and deserve different
nomenclature.
We concede that for the purposes of this study we assumed that the 20
patients had normal anatomic features on the basis of the absence of groin
symptoms and abnormal imaging findings. This study had a number of
limitations, including the relatively small sample size and the absence of a
reference standard to validate the findings. No attempt was made to assess
interobserver agreement, and the value of these observations in clinical
practice was not established.
MDCT produces images of the inguinal region with detail not available with
previous generations of scanners. We achieved 100% identification of key
anatomic structures in our small series when information from all three views
was combined. The IL and IEA, which are vital to evaluation of inguinal
hernias, can be consistently visualized on contrast-enhanced CT
reconstructions in the coronal and sagittal planes. The radiologic femoral
triangle is particularly well visualized on coronal reconstructions, which
should lead to accurate diagnosis of femoral hernias from coronal images.
References
- Arregui ME. The value of ultrasound in the diagnosis of hernias.
In: Arregui ME, Nagan RF, eds. Inguinal hernia: advances or
controversies? New York, NY: Radcliff Medical Press,1994
: 73–79
- Working Party of the Royal College of Surgeons. Clinical
guidelines on the management of groin hernias in adults. London,
UK: Royal College of Surgeons, 1993:2
–7
- Wechsler RJ, Kurtz AB, Needleman L, et al. Crosssectional imaging
of abdominal wall hernias. AJR 1989;153
: 517–521[Free Full Text]
- Delabrousse E, Michalakis D, Sarlieve P, Paratte B, Rodiere E,
Kastler B. Value of the pubic tubercle as a CT reference point in groin
hernias. J Radiol 2005;86
: 651–654[Medline]
- Toms AP, Dixon AK, Murphy MP, Jamieson NV. Illustrated review of
new imaging techniques in the diagnosis of abdominal wall hernias.
Br J Surg 1999;86
:1243
–1250[CrossRef][Medline]
- Stabile Ianora AA, Midiri M, Vinci R, Rotondo A, Angelelli G.
Abdominal wall hernias: imaging with spiral CT. Eur
Radiol 2000; 10:914
–919[CrossRef][Medline]
- McMinn RM. Last's anatomy: regional and
applied, 9th ed. Edinburgh, UK: Churchill Livingstone,1997
- Standring S. Gray's anatomy: the anatomical basis of
medicine and surgery, 39th ed. Edinburgh, UK: Churchill
Livingstone, 2004
- Van den Berg JC, De Valois JC, Go PM, Rosenbusch G. Detection of
groin hernia with physical examination, ultrasound, and MRI compared with
laparoscopic findings. Invest Radiol1999; 34:739
–743[CrossRef][Medline]
- Sayad P, Abdo Z, Cacchione R, Ferzli G. Incidence of incipient
contralateral hernia during laparoscopic hernia repair. Surg
Endosc 2000; 14:543
–545[CrossRef][Medline]
- Sutcliffe JR, Taylor OM, Ambrose NS, Chapman AH. The use, value and
safety of herniography. Clin Radiol 1999;54
: 468–472[CrossRef][Medline]
- Brierly RD, Hale PC, Bishop NL. Is herniography an effective and
safe investigation? J R Coll Surg Edinb1999; 44:374
–377[Medline]
- Lilly MC, Arregui ME. Ultrasound of the inguinal floor for
evaluation of hernias. Surg Endosc 2002;16
: 659–662[CrossRef][Medline]
- Bradley M, Morgan D, Pentlow B, Roe A. The groin hernia: an
ultrasound diagnosis? Ann R Coll Surg Engl2003; 85:178
–180[CrossRef][Medline]
- van den Berg JC. Inguinal hernias: MRI and ultrasound.
Semin Ultrasound CT MR 2002;23
: 156–173[CrossRef][Medline]
- Shadbolt CL, Heinze SB, Dietrich RB. Imaging of groin masses:
inguinal anatomy and pathological conditions revisited.
RadioGraphics 2001;21
: 261–271
- Zarvan NP, Lee FT, Yandow DR, Unger JS. Abdominal hernias: CT
findings. AJR 1995;164
:1391
–1395[Abstract/Free Full Text]
- Stamm ER, Pretorius DH, Olson LK. Abdominal wall CT: a pictorial
essay. Comput Radiol 1985;9
: 271–278[CrossRef][Medline]
- Markos V, Brown EF. CT herniography in the diagnosis of occult
groin hernias. Clin Radiol 2005;60
: 251–256[CrossRef][Medline]
- Skandalakis JE, Gray SW, Skandalakis LJ, Colborn GL, Pemberton LB.
Surgical anatomy of the inguinal area. World J Surg1989; 13:490
–498[CrossRef][Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
