Anda di halaman 1dari 6

Suzuki et al.

Abdominal Imaging • Original Research

CT of Femoral and Inguinal
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

Differentiation of Femoral Versus

Inguinal Hernia: CT Findings
Shigeru Suzuki1 OBJECTIVE. The purpose of our study was to investigate the CT findings of femoral her-
Shigeru Furui1 nias, focusing on their differentiation from inguinal hernias.
Kota Okinaga2 MATERIALS AND METHODS. We reviewed the records of 46 femoral hernias in
Tsutomu Sakamoto3 seven centers (review of femoral hernias) and those of 215 groin hernias (femoral hernias, 11;
Jun Murata4 inguinal hernias, 204) in one center (review of groin hernias). We evaluated the presence of her-
nia, extent of hernia sac based on the relationship between the hernia sac and the pubic tubercle
Akira Furukawa5
(localized sac: sac was localized lateral to the pubic tubercle; or extended sac: sac extended me-
Yasuo Ohnaka6
dial to the pubic tubercle), and compression of the femoral vein on CT images. The chi-square
Suzuki S, Furui S, Okinaga K, et al. test was used to assess the relationship between the CT findings and femoral versus inguinal
hernias in the review of groin hernias.
RESULTS. In the review of 46 femoral hernias, the lesions were detected on CT in 45. In the
45 lesions, all hernia sacs were localized, and 42 lesions showed venous compression. In the re-
view of 215 groin hernias, all 11 femoral hernias had localized sacs with venous compression on
CT. Of the 204 inguinal hernias, 98 lesions were detected on CT, 65 had extended sacs, and only
10 showed venous compression. Localized sacs with venous compression were seen much more
often in the femoral hernias (11/11, 100%) than in the inguinal hernias (1/92, 1.1%) (p < 0.0001).
Keywords: abdominal imaging, CT, femoral hernia, hernia, CONCLUSION. CT images are useful to differentiate femoral hernias from inguinal hernias.
inguinal hernia

DOI:10.2214/AJR.07.2085 or the evaluation of groin hernias, for our retrospective investigation. This investigation

Received January 1, 2007; accepted after revision

March 15, 2007.
F physical examination is most im-
portant in clinical practice [1], al-
though it is sometimes difficult to
consisted of two reviews: a review of groin hernias in
one center to compare the CT findings of femoral
hernias and those of inguinal hernias, and a review of
distinguish femoral from inguinal hernias [2]. femoral hernias in multiple centers.
of Radiology, Teikyo University School of
Medicine, 2-11-1 Kaga, Itabashi-ku, Tokyo 173-8605, Japan. CT played a minor role until recently, although
Address correspondence to S. Suzuki. it is useful for the identification of groin her- Review of Femoral Hernias
nias and their contents [3]. Some authors have We retrospectively reviewed the CT scans and
2Department of Surgery, Teikyo University School of reported that CT was useful to distinguish be- surgical reports of 46 consecutive patients who un-
Medicine, Tokyo, Japan.
tween direct and indirect inguinal hernias derwent surgery for femoral hernias in seven centers
3Department of Radiology, Kohga Public Hospital, [4–6]. On the other hand, some have noted that (May 1995–July 2006): Teikyo University Hospital
Shiga, Japan. the differentiation between inguinal and femo- (n = 11), Asakadi Central General Hospital (n = 11),
ral hernias was difficult [7, 8], and femoral her- Osaka-hu Saiseikai Noe Hospital (n = 8), Kohga
4Department of Surgery, Asakadai Central General
nias have received only brief mention in the lit- Public Hospital (n = 6), Shiga University of Medical
Hospital, Saitama, Japan.
erature of the CT findings [4, 7]. We Science Hospital (n = 4), Hoyu Hospital (n = 3), and
5Department of Radiology, Shiga University of Medical investigated the CT findings of femoral her- Toyosato Hospital (n = 3). Of these 46 patients, nine
Science, Shiga, Japan. nias, focusing on their differentiation from in- were men and 37 were women. Their mean age was
guinal hernias. To our knowledge, this is the 73 ± 11 (SD) years (range, 44–92 years).
of Radiology, Osaka-fu Saiseikai Noe
Hospital, Osaka, Japan.
first report referring to the compression of the Twenty-three patients underwent both unen-
femoral vein in the differential diagnosis be- hanced and contrast-enhanced CT, and the others
WEB tween femoral and inguinal hernias. underwent only unenhanced CT. The slice thick-
This is a Web exclusive article. ness and slice intervals were 5–10 mm. All patients
AJR 2007; 189:W78–W83
Materials and Methods were symptomatic when CT was performed.
The investigation was approved by the institu- Thirty-three of the 46 femoral hernias were incar-
tional review board of Teikyo University School of cerated lesions, which were determined on the basis
© American Roentgen Ray Society Medicine. Informed patient consent was not required of physical examination findings. In all the patients,

W78 AJR:189, August 2007

CT of Femoral and Inguinal Hernias

years). Fourteen of the 190 patients with inguinal

hernias had bilateral lesions, and all 11 patients
with femoral hernias had unilateral lesions. All 11
femoral hernias (right side, 7 [63.6%]; left side, 4
[36.4%]; incarcerated lesions, 7 [63.6%]; nonincar-
cerated lesions, 4 [36.4%]) and a total of 204 in-
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

guinal hernias (right side, 115 [56.4%]; left side, 89

[43.6%]; incarcerated lesions, 11 [5.4%]; nonincar-
cerated lesions, 193 [94.6%]) were assessed. This
patient group included the 11 femoral hernias in the
review of femoral hernias.
CT was performed with a LightSpeed Plus-R (GE
Healthcare) scanner in 114 patients and with a Hi-
Speed QX/i (GE Healthcare) scanner in 87 patients.
Scanning parameters were 4 × 2.5 mm collimation,
120 kVp, 400 mAs, and 15-mm table speed per gan-
try rotation. Images were reconstructed with a slice
thickness of 7.5 mm and slice interval of 10.0 mm.
One hundred fifty-four patients underwent both un-
enhanced and contrast-enhanced CT, and the others
underwent only unenhanced CT. For enhanced CT, a
total of 85–130 mL of iohexol with an iodine con-
centration of 350 mg I/mL (Omnipaque, Daiichi
A B Pharmaceutical) was given with a power injector at
Fig. 1—Extent of hernia sac was evaluated visually based on relationship between hernia sac and pubic tubercle a rate of 1.0–1.5 mL/s. The delay between the start of
on axial CT images. contrast material administration and the start of heli-
A, Extended sac (arrow) extends medial to pubic tubercle (arrowhead) in 72-year-old man. cal scanning was 90 seconds.
B, Localized sac (arrow) is located lateral to pubic tubercle (arrowhead) in 59-year-old man.

CT Findings
In each review, two radiologists with 12 and 8
Fig. 2—Compression of years of experience in abdominal CT independently
femoral vein on CT scans reviewed the CT findings on axial images. Both re-
through acetabula and
pubic symphysis in 66- viewers were unaware of the surgical findings.
year-old woman. Each Studies were evaluated for the presence of hernia,
lower image is the extent of the hernia sac based on the relation-
magnification of upper
ship between the hernia sac and pubic tubercle, and
portion on each side.
Right femoral vein is compression of the femoral vein (venous compres-
elliptic, and hernia sac sion). Differences in opinions between the two ra-
(arrowhead) lies in diologists regarding the CT findings were settled by
direction of minor
diameter of right femoral a third radiologist with 29 years of experience in
vein (solid white line). abdominal CT. If a hernia was detected, the extent
Minor diameter of right of the hernia sac was evaluated visually and catego-
femoral vein (a) is less
than two thirds of rized as extended sac (the sac extended medial to
diameter of femoral vein the pubic tubercle) or localized sac (the sac was lo-
(b) in symmetric direction calized lateral to the pubic tubercle) (Fig. 1). The
(dotted white line).
femoral vein was considered compressed by the
hernia sac when the following three conditions
were fulfilled: the femoral vein adjacent to the her-
the hernias were unilateral (right side, 30; left side, hundred one patients who underwent abdominal nia sac was elliptic, the hernia sac lay in the direc-
16). The contents of the sac were small bowel CT within 30 days before surgery were selected. tion of the minor diameter of the femoral vein, and
(n = 29, including five Richter’s hernias), omentum All patients were symptomatic when CT was per- the minor diameter of the femoral vein was less
(n = 8), appendix (n = 1), and ascites only (n = 8). formed. As a reference standard, we used surgical than two thirds the diameter of the femoral vein on
findings [9]. Eleven (five men, six women) of the the contralateral side in the symmetric direction
Review of Groin Hernias 201 patients with groin hernias had femoral her- (Fig. 2). In the case of bilateral lesions, only the
We retrospectively reviewed the records of 296 nias, and the other 190 (166 men, 24 women) had femoral vein with a smaller minor diameter was
consecutive adult patients who underwent surgery inguinal hernias. The mean age of the former was evaluated. In addition, one radiologist measured the
for groin hernias in Asakadai Central General Hos- 67.9 ± 11.9 years (range, 44–87 years), and that of maximum minor diameter of the sac on axial im-
pital between January 2003 and March 2006. Two the latter was 58.4 ± 14.5 years (range, 20–85 ages when the sac was detected.

AJR:189, August 2007 W79

Suzuki et al.

TABLE 1: Diagnostic Accuracy of CT Findings in Differentiating Femoral from formed. In the other 45 lesions, the hernias had
Inguinal Hernias localized sacs. The average maximum minor
Positive Rate (%) diameter of the 45 sacs was 29 ± 9 [SD] mm
CT Findings Femoral Hernias Inguinal Hernias p (range, 16–56 mm). The femoral vein was
compressed in 42 (93.3%) of all 45 lesions.
Incarcerated lesions (femoral, 7; inguinal, 11)
Among the three femoral hernias without
Presence of hernia 100 (7/7) 100 (11/11) NA
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

venous compression, all of which were incar-

Extended sac 0 (0/7) 100 (11/11) < 0.0001 cerated lesions, two lesions were Richter’s her-
Venous compression 100 (7/7) 45.5 (5/11) 0.0167 nia and the other was detected in a patient with
Localized sac with venous compression 100 (7/7) 0 (0/11) < 0.0001 severe deformity of the pelvic girdle.
Nonincarcerated lesions (femoral, 4; inguinal, 193)
Review of Groin Hernias
Presence of hernia 100 (4/4) 45.1 (87/193) 0.0292
The presence of each CT finding is summa-
Extended sac 0 (0/4) 62.1 (54/87)a 0.0135 rized in Table 1.
Venous compression 100 (4/4) 6.2 (5/81)a,b < 0.0001 Entire groin hernias—For all 11 femoral
Localized sac with venous compression 100 (4/4) 1.2 (1/81)a,b < 0.0001 hernias, localized sacs with venous compres-
Overall (femoral, 11; inguinal, 204) sion were detected on CT (Fig. 3), and the av-
erage maximum minor diameter of the 11
Presence of hernia 100 (11/11) 48.0 (98/204) 0.000786
sacs was 23 ± 4 mm (range, 17–31 mm). For
Extended sac 0 (0/11) 66.3 (65/98)a < 0.0001 the 204 inguinal hernias, 98 (48.0%) were de-
Venous compression 100 (11/11) 10.9 (10/92)a,b < 0.0001 tected on CT, and the average maximum mi-
Localized sac with venous compression 100 (11/11) 1.1 (1/92)a,b < 0.0001 nor diameter of the 98 sacs was 31 ± 15 mm
Note—Numbers in parentheses are those used to calculate percentages. NA = not applicable. (range, 12–100 mm). Sixty-five (66.3%) of
a Evaluated only for visible hernias. the 98 lesions had extended sacs.
b Six lesions were excluded from evaluation in six cases of bilateral lesions.
Ninety-two of the 98 visible inguinal her-
nias were evaluated for venous compression
(one of the two lesions was excluded from the
evaluation in six cases of bilateral lesions).
Venous compression was seen in 10.9%
(10/92) of the inguinal hernias with localized
sac (n = 1) or extended sacs (n = 9) (Figs. 4
and 5). The details of the 10 inguinal hernias
with venous compression were as follows:
Fig. 3—Femoral hernia
One lesion was an unusual type with a local-
with localized sac and
venous compression in ized sac, and the sac of the indirect inguinal
82-year-old woman. On hernia protruded from the inguinal canal to
contrast-enhanced CT the femoral vein through an anomalous open-
image through
acetabula, hernia sac is ing of the inguinal canal. One lesion had an
localized lateral to pubic extended sac and contained the left ovary and
tubercle (arrowhead). fallopian tube. One lesion was a recurrent her-
Compression of right
femoral vein (arrow) is
nia after surgery and had a narrow orifice and
seen. an extended sac. In one lesion with an ex-
tended sac on the right, not only the right fem-
oral vein but also the right external iliac vein
Statistical Analyses formed for these two subgroups in addition to the had a smaller diameter than the veins on the
In the review of groin hernias, statistical analyses analyses for the entire group. left. The other six lesions had large extended
were performed using a commercially available statisti- To evaluate interobserver agreement, a kappa sacs, with their maximum minor diameters
cal software program (SPSS for Windows [Microsoft], value was calculated for the four CT findings. Values greater than 4 cm on axial CT images.
release 14.0, SPSS). The chi-square test was used to as- of 0–0.20 were considered to represent slight agree- Localized sacs with venous compression
sess the relationship between femoral and inguinal her- ment; 0.21–0.40, fair agreement; 0.41–0.60, moder- were seen much more often in the femoral
nias, and the CT findings of presence of hernia, extended ate agreement; 0.61–0.80, good agreement; and (11/11, 100%) than in the inguinal (1/92,
sac, venous compression, and localized sac with venous 0.81–1.00, almost perfect or perfect agreement [10]. 1.1%) hernias (p < 0.0001).
compression. The number of hernia lesions rather than Incarcerated lesions—Seven of the 11 fem-
the number of patients was used for these analyses be- Results oral hernias and 11 of the 204 inguinal hernias
cause some patients had bilateral lesions. Review of Femoral Hernias were incarcerated lesions (Table 1). All seven
We divided each type of groin hernia lesion into In only one of all 46 patients was a hernia incarcerated femoral hernias had localized
two subgroups: incarcerated and nonincarcerated sac not detected. The clinical diagnosis was sacs, and all incarcerated inguinal hernias had
lesions. The statistical analyses were also per- that the lesion was reduced when CT was per- extended sacs. The average maximum minor

W80 AJR:189, August 2007

CT of Femoral and Inguinal Hernias
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

Fig. 4—Inguinal hernia with extended sac and without venous compression in 37-year-old man.
A and B, Unenhanced CT images through acetabula show hernia sac extends medial to pubic tubercle (arrowhead, B). Compression of left femoral vein (arrow, A) is not seen.

Fig. 5—Inguinal hernia with extended sac and venous compression in 73-year-old woman.
A and B, Contrast-enhanced CT images through acetabula show compression of left femoral vein (arrow, A). Hernia sac extends medial to pubic tubercle (arrowhead, B).
Maximum minor diameter of sac on axial images is 5.3 cm.

diameters of the seven femoral hernia sacs and four femoral hernia sacs and 87 inguinal hernia Interobserver Agreement
the 11 inguinal hernia sacs were 24 ± 3 mm sacs were 21 ± 3 mm (range, 17–23 mm) and Results of interobserver analyses for the
(range, 21–31 mm) and 48 ± 29 mm (range, 28 ± 8 mm (range, 12–53 mm), respectively. four CT findings indicated almost perfect
24–100 mm), respectively. Venous compres- None of the four nonincarcerated femoral her- agreement (Table 2).
sion was seen in 100% (7/7) and 45.5% (5/11) nias had extended sacs, whereas 62.1% (54/87)
in incarcerated femoral and inguinal hernias, of the nonincarcerated inguinal hernias had ex- Discussion
respectively. Localized sacs with venous com- tended sacs. Eighty-one of the 87 visible in- The lifetime incidence of spontaneous ab-
pression were seen in 100% (7/7) and in 0% guinal hernias were evaluated for venous com- dominal hernias is approximately 5% in the
(0/11) of the incarcerated femoral and inguinal pression (because one of the two lesions was
hernias, respectively. excluded for the evaluation in six cases of bilat-
Nonincarcerated lesions—Four of the 11 eral lesions). Venous compression was seen in TABLE 2: Interobserver Reliability
femoral hernias and 193 of the 204 inguinal her- 100% (4/4) and 6.2% (5/81) of the nonincarcer- CT Findings κ
nias were nonincarcerated lesions (Table 1). The ated femoral and inguinal hernias, respectively. Presence of hernia 0.953
hernias were detected on CT in 100% (4/4) of the Localized sacs with venous compression were Extended sac 0.862
nonincarcerated femoral hernias and 45.1% seen in 100% (4/4) and 1.2% (1/81) of the non-
Venous compression 0.922
(87/193) of the nonincarcerated inguinal hernias. incarcerated femoral hernias and nonincarcer-
The average maximum minor diameters of the ated inguinal hernias, respectively. Localized sac with venous compression 0.954

AJR:189, August 2007 W81

Suzuki et al.

world population [11, 12]. Approximately 80% sions. CT is useful for distinguishing these is useful in differentiating a femoral hernia
of abdominal wall hernias are inguinal hernias conditions from a groin hernia [3, 4, 19]. from an inguinal hernia on CT images. These
and 5% are femoral hernias [11]. The other Surgeons differentiate a femoral hernia findings can be evaluated with high agree-
15% include incisional, umbilical, epigastric, from an inguinal hernia by ascertaining the ment even on unenhanced CT images of 10-
and a host of miscellaneous hernia types [11]. A relation of the neck of the sac to the medial mm thickness.
male predominance of about 7:1 is seen with in- end of the inguinal ligament and the pubic tu- When interpreting a CT scan in a patient
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

guinal hernias, whereas a female predominance bercle [13]. The neck of the hernia sac is be- suspected of having a groin hernia, one may
of about 1.8:1 is seen with femoral hernias [11]. low and lateral to the medial end of the in- use the following algorithm: When the hernia
Femoral hernias affect the right side more often guinal ligament in a femoral hernia and is sac extends medial to the pubic tubercle, the
(2:1) [7, 13]. The results of this investigation above and medial to the ligament in an in- diagnosis of inguinal hernia can be made with
agree with those of earlier reports. In our review guinal hernia [13]. Therefore, Wechsler et al. confidence. If the hernia sac is located lateral
of groin hernias, the ratio of prevalence of fem- [4] suggested that a femoral hernia might be to the pubic tubercle, the presence of venous
oral hernias to inguinal hernias was about 1:17. distinguished from an inguinal hernia on the compression suggests the diagnosis of femo-
Inguinal hernias had a male predominance of basis of the relationship between the hernia ral hernia with a high probability. It is ex-
about 7:1, whereas femoral hernias had a fe- sac and pubic tubercle on CT images. The pected that the CT diagnosis has high repro-
male predominance of about 1.2:1 in our re- present data of incarcerated groin hernias are ducibility and objectivity because almost
view. A right-side predominance of about 1.8:1 consistent with that suggestion. The sacs ex- perfect interobserver agreement was obtained
was seen with femoral hernias in this review. tended medial to the pubic tubercles in all the for these CT findings.
The inguinal hernia often reduces when incarcerated inguinal hernias, whereas the As mentioned previously, preoperative dif-
the patient lies down, as mentioned by Rich- sacs were localized lateral to the pubic tuber- ferentiation of a femoral hernia from an in-
ards et al. [11]. In our review of groin hernias, cles in all incarcerated femoral hernias. How- guinal hernia is important clinically, espe-
more than half of the inguinal hernias re- ever, 37.9% (33/87) of the nonincarcerated in- cially in a nonincarcerated case, because it
duced when CT was performed. For in- guinal hernias in our investigation had can affect the indication for surgery. All 12
guinal hernias, major complications, such as localized sacs. Distinguishing femoral from visible nonincarcerated femoral hernias had
incarceration, obstruction, or strangulation, inguinal hernias only by the relationship be- localized sacs with venous compression (re-
are rare [11]. A previous report showed that tween the hernia sac and the pubic tubercle is view of femoral hernias). On the other hand,
the lifetime risks of strangulation of the in- difficult, especially in nonincarcerated cases. only one of the 81 visible nonincarcerated in-
guinal hernia are 0.272 and 0.034 for an 18- In our study, compression of the femoral guinal hernias had a localized sac with venous
year-old man and 75-year-old man, respec- vein was seen in all 11 femoral hernias (re- compression (review of groin hernias). There-
tively [11, 14]. As for the appropriate view of groin hernias) and in 42 (93.3%) of fore, these CT findings are useful in differen-
approach to asymptomatic or minimally the 45 visible femoral hernias (review of tiating femoral hernias from inguinal hernias
symptomatic patients with inguinal hernia, femoral hernias), compared with only 10 in nonincarcerated cases.
surgeons hold two opinions: surgical repair (10.9%) of the 92 visible inguinal hernias Our study has some limitations. First, the
and watchful waiting [15]. In a randomized (review of groin hernias). Because the fem- selection of the patients in the comparison be-
clinical trial, Fitzgibbons et al. [16] con- oral canal is narrow, the femoral vein can tween femoral and inguinal hernias might
cluded that watchful waiting was an accept- easily be compressed by the contents of the have a bias. Among 296 consecutive patients
able option for men with asymptomatic or hernia. On the other hand, the orifice of the who underwent surgery for groin hernias, 201
minimally symptomatic inguinal hernias. inguinal hernia is wide, and the inguinal lig- underwent abdominal CT and only these 201
On the other hand, approximately 40% of ament lies between the hernia sac and femo- were selected for this investigation. Second,
femoral hernias present with incarceration ral vein. Therefore, venous compression is no attempt was made to directly identify the
or strangulation [13]. The high incidence of seldom seen in an inguinal hernia. type of hernias by identification of the hernia
incarceration or strangulation is sufficient Some points should be considered when orifice on CT images. The slice thickness and
reason to recommend surgery, which should the venous compression sign is used for eval- interval affect the difficulty of identifying the
be performed soon after the diagnosis [2, uation of groin hernias. The compression sign hernia orifice. CT images with a thinner slice
13]. Therefore, preoperative differentiation was not seen in two of the three cases of Rich- thickness on MDCT may permit the direct
of a femoral hernia from an inguinal hernia ter’s-type femoral hernias. In this type, the identification of the type of hernia based on
is important clinically, especially in a non- compression of the femoral vein does not oc- its orifice. Coronal and sagittal reconstruc-
incarcerated case. cur because the volume of the hernia content tions might be helpful in the future in differ-
The preoperative diagnosis of a femoral is small. An inguinal hernia with a large con- entiating groin hernias. Third, it is difficult to
hernia is not easy in an asymptomatic patient tent can compress the femoral vein by mass evaluate venous compression in patients who
because palpating the sac is difficult [2]. Even effect. However, the sac of a large inguinal have preexisting collapsed femoral veins or
in a patient with a bulge in the groin, a femo- hernia protruded through the inguinal canal femoral veins with laterality in diameter or
ral hernia may resemble an inguinal hernia and extended medial to the pubic tubercle, deformity of the pelvic girdle.
[17, 18]. Besides inguinal hernia, the differ- whereas that of a femoral hernia was local- In conclusion, the extent of the sac based
ential diagnosis of a femoral hernia based on ized lateral to the pubic tubercle. Therefore, on the relationship between the hernia sac and
clinical findings includes inguinal lymphade- the combination of venous compression sign pubic tubercle and compression of the femo-
nopathy, lipoma, femoral artery aneurysm, and the extent of the sac based on the relation ral vein on CT images are the keys to the dif-
psoas abscess, hydrocele, and cutaneous le- between the hernia fundus and pubic tubercle ferentiation of femoral from inguinal hernias.

W82 AJR:189, August 2007

CT of Femoral and Inguinal Hernias

References Abdominal hernias: CT findings. AJR 1995; surgery, 23rd ed. London, England: Arnold Pub-
1. van den Berg JC, de Valois JC, Go PM, Rosenbusch 164:1391–1395 lishers, 2000:1143–1162
G. Radiological anatomy of the groin region. Eur 8. Hojer AM, Rygaard H, Jess P. CT in the diagnosis 14. Metropolitan Life Insurance Company. Expecta-
Radiol 2000; 10:661–670 of abdominal wall hernias: a preliminary study. Eur tion of life and mortality rates at single years of
2. Naude GP, Ocon S, Bongard F. Femoral hernia: Radiol 1997; 7:1416–1418 age, by race and sex: United States, 1991. Statis-
the dire consequences of a missed diagnosis. Am 9. Bossuyt PM, Reitsma JB, Bruns DE, et al. Toward tics Bulletin of the Metropolitan Insurance Com-
Downloaded from by on 11/04/15 from IP address Copyright ARRS. For personal use only; all rights reserved

J Emerg Med 1997; 15:680–682 complete and accurate reporting of studies of diag- pany 1996; 75:16
3. Lee GH, Cohen AJ. CT imaging of abdominal her- nostic accuracy: the STARD initiative. AJR 2003; 15. Flum DR. The asymptomatic hernia: “if it’s not
nias. AJR 1993; 161:1209–1213 181:51–55 broken, don’t fix it.” (commentary) JAMA 2006;
4. Wechsler RJ, Kurtz AB, Needleman L, et al. Cross- 10. Kundel HL, Polansky M. Measurement of observer 295:328–329
sectional imaging of abdominal wall hernias. AJR agreement. Radiology 2003; 228:303–308 16. Fitzgibbons RJ Jr, Giobbie-Hurder A, Gibbs JO, et
1989; 153:517–521 11. Richards AT, Quinn TH, Fitzgibbons RJ Jr. Abdom- al. Watchful waiting vs repair of inguinal hernia in
5. Hahn-Pedersen J, Lund L, Hojhus JH, Bojsen-Mol- inal wall hernias. In: Greenfield LJ, ed. Surgery, 3rd minimally symptomatic men: a randomized clinical
ler F. Evaluation of direct and indirect inguinal her- ed. Philadelphia, PA: Lippincott Williams & trial. JAMA 2006; 295:285–292
nia by computed tomography. Br J Surg 1994; Wilkins, 2001:1185–1223 17. Ponka JL, Brush BE. Problem of femoral hernia.
81:569–572 12. Rand Corporation. Conceptualization and mea- Arch Surg 1971; 102:417–423
6. Shadbolt CL, Heinze SB, Dietrich RB. Imaging of surement of physiologic health of adults. Santa 18. Rhind JR. Lateral femoral hernia. J R Coll Surg Ed-
groin masses: inguinal anatomy and pathologic Monica, CA: Rand Corporation, 1983:15 inb 1971; 16:299–300
conditions revisited. RadioGraphics 2001; 21[spec 13. Kingsnorth A, Bennet DH. Hernias, umbilicus, ab- 19. Ianora AA, Midiri M, Vinci R, Rotondo A, An-
no]:S261–S271 dominal wall. In: Russel RCG, Williams NS, Bul- gelelli G. Abdominal wall hernias: imaging with
7. Zarvan NP, Lee FT Jr, Yandow DR, Unger JS. strode CJK, eds. Bailey & Lovee’s short practice of spiral CT. Eur Radiol 2000; 10:914–919

AJR:189, August 2007 W83