CLINICAL INVESTIGATION

The Anatomic Relationship Between the

Common Femoral Artery and Common
Femoral Vein in Frog Leg Position Versus
Straight Leg Position in Pediatric Patients
Jennifer W. Hopkins, MD, Fred Warkentine, MD, MDc, Edward Gracely, PhD, and
In K. Kim, MD, MBA

Abstract
Background: Overlap of the femoral artery (FA) on the femoral vein (FV) has been shown to occur in
pediatric patients. This overlap may increase complications such as arterial puncture and failed inser-
tions of central venous lines (CVLs). Knowledge of the anatomic relationship between the FV and FA
may be important in avoiding these complications.
Objectives: The objective was to evaluate the anatomic relationship of the FA and FV in straight leg
position and frog leg position.
Methods: This was a prospective, descriptive study of a convenience sample of 80 total subjects
(16 subjects from each of five predetermined stratified age groups). Each subject underwent a
standardized ultrasound examination in both the straight and the frog leg positions. The location of
the FA in relation to the FV was measured at three locations: immediately distal, 1 cm distal, and
3 cm distal to the inguinal ligament. Overlap of the FA on the FV and the diameter of the FV was
noted at each location. Measurements were repeated in both the straight leg and the frog leg
positions.
Results: For the left leg, immediately distal to the inguinal ligament, the FV was overlapped by the FA
in 36% of patients in straight leg position and by 45% of patients in frog leg position. At 1 cm distal to
the ligament, overlap was observed in 75% of patients in straight leg position and 88% of patients in the
frog leg position. At 3 cm distal to the ligament, overlap was observed in 93% of patients in straight leg
position and 86% of patients in the frog leg position. The percentage of vessels with overlap was similar
in the right leg at each location for both the straight and the frog leg positions. Pooled mean (±SD) FV
diameters for the left leg immediately distal to the inguinal ligament were 0.64 (±0.23) cm in the straight
leg position and 0.76 (±0.28) cm in the frog leg position; at 1 cm distal to the ligament, 0.66 (±0.23) and
0.78 (±0.29) cm; and at 3 cm distal to the ligament, 0.65 (±0.27) and 0.69 (±0.29) cm. FV diameters for the
right leg were similar to the left.
Conclusions: A significant percentage of children have FAs that overlap their FVs. This overlap may be
responsible for complications such as FA puncture with CVL placement. Ultrasound-guided techniques
may decrease these risks. Placing children in the frog leg position increases the diameter of the FV visu-
alized on ultrasound.
ACADEMIC EMERGENCY MEDICINE 2009; 16:579–584 ª 2009 by the Society for Academic Emergency
Medicine
Keywords: emergency ultrasound, pediatric emergency ultrasound, patient safety, central line
placement

From Department of Pediatrics, University of Louisville Health Sciences Center, Division of Pediatric Emergency Medicine, Kosair
Children’s Hospital (JWH, FW, IKK), Louisville, KY; and Drexel University School of Public Health, Department of Family, Commu-
nity, and Preventative Health (EG), Philadelphia, PA.
Received November 19, 2009; revisions received February 22 and February 25, 2009; accepted February 25, 2009.
Presented at the Pediatric Academic Societies Annual Meeting, Baltimore, MD, May 2–5, 2009.
Address for correspondence and reprints: Jennifer W. Hopkins, MD; e-mail: jen.hopkins@louisville.edu.

ª 2009 by the Society for Academic Emergency Medicine ISSN 1069-6563

doi: 10.1111/j.1553-2712.2009.00430.x PII ISSN 1069-6563583 579
580 Hopkins et al. • ANATOMIC RELATIONSHIP BETWEEN COMMON FEMORAL ARTERY AND COMMON FEMORAL VEIN
R
apid and reliable intravenous access plays a vital
role in the resuscitation of critically ill children in
the emergency department (ED). Emergency
physicians (EPs) often prefer central venous lines (CVLs)
because they allow rapid administration of large fluid
volumes, vasoactive substances, and antibiotics.1 The
femoral vein (FV) has become the preferred site for CVL
placement in children due to its easy accessibility during
cardiopulmonary resuscitation.2,3
Traditionally, CVLs were placed using landmark-guided
techniques with the child situated in the frog leg position.1
Complications associated with CVL insertion include arte-
rial puncture, catheter malposition, hematoma formation,
and bleeding.1,4 Multiple attempts at one site or the need
to choose an alternate site after a failed first attempt have
been shown to increase the risk of these complications.2
Previous studies have demonstrated significant differ-
ences between external landmarks and internal anatomy.
These differences may lead to increased complication
rates and multiple attempts in CVL placement.5,6
Femoral vein overlap has been shown to occur just
distal to the inguinal ligament in pediatric patients.5,6
Warkentine et al.6 observed that the femoral artery
(FA) partially overlapped and obscured the FV in 12%
of pediatric patients. More importantly, these authors
noted that the FA completely overlapped the FV in 8%
of pediatric patients. These findings suggest that a sub-
set of pediatric patients has an increased risk of arterial
puncture during FV cannulation.
Accepted knowledge of femoral vessel anatomy
states that, as the FV moves distally in the leg, it lies
closer and almost posterior to the FA. The distance
from the inguinal ligament at which this occurs in pedi-
atric patients is not well established.
Werner et al.7 observed in adults that the frog leg
position may improve access to the FV. These authors
observed the percentage of FA overlap of the FV was
greater in the straight leg position compared to the
frog leg position. Also, these authors observed that the
mean diameter of the FV was greater in the frog leg
position than in the straight leg position. This observa-
tion of greater mean diameter suggests that the FV
may be more easily cannulated in the frog leg versus
the straight leg position.
To the best of our knowledge, no pediatric studies
have examined the relationship between the FV and FA
in the frog leg and straight leg positions or the effect of
the frog leg position on FV diameter in pediatric
patients. Knowledge of the femoral vessel anatomy in
the frog leg position versus the straight leg position
may allow practitioners to place femoral CVLs with
fewer complications.
METHODS
Study Design
We conducted a prospective, descriptive study using
ultrasonography (US) to compare the percentage of
children with overlap of FA on FV in both straight leg
and frog leg positions. We also examined the diameter
of the FV in straight leg and frog leg positions. The
institutional review board approved the study. Written
informed consent was obtained from the parent or
guardian of each subject. Written informed assent was
obtained from all subjects 7 years of age or older.
Study Setting and Population
This study was conducted in the ED of a tertiary care
children’s hospital between July 15 and October 15,
2008. Subjects were chosen from a convenience sample
of patients presenting to the pediatric emergency
department. A total of 80 subjects were enrolled. Six-
teen subjects were enrolled in each of the following age
groups: 1 month to <12 months of age, 2 years of age,
4 years of age, 6 years of age, and 8 years of age. Based
on a previous study that detected 8% of children with
complete overlap of the FV by the FA,6 a sample size of
80 subjects (16 per group) was calculated to provide an
80% power for detecting an 8% difference in overlap
between the straight and frog leg positions.
Subjects were invited to participate if they were of
the appropriate age and were euvolemic. Euvolemic
was defined as children with no history of vomiting or
diarrhea, a heart rate <95th percentile for age, and a
blood pressure >5th percentile for age. Patients were
assessed for euvolemia by history and ED attending
exam. Children were excluded from participation if they
were hypovolemic as assessed by history and ED
attending exam, had previous FV cannulation attempts,
could not be easily and comfortably placed in the frog
leg position, or had an allergy to US gel.
Study Protocol
A single SonoSite Titan (SonoSite, Inc., Bothell, WA)
US machine was used for all study participants. Each
subject was placed supine with legs extending straight
for initial femoral vessel US (Figure 1). Ten milliliters of
Aquasonic 100 US gel (R.P. Kincheloe, Dallas, TX) was
applied to the skin over the site, which corresponded to
the femoral vessels. An age appropriate US transducer
was placed at the inguinal ligament over the pulsation
of the FA. The artery and vein were identified by their
relative positions, compressibility of the vein, and
enlargement of the vein by Valsalva maneuver.
Once the FA and FV were identified, the images were
captured and saved to a compact flash card. Due to the
pulsatile nature of the FA, every attempt was made to
capture each image with the FA in diastole or at its
minimum diameter. The presence of overlap of FA and
FV was identified by direct visual estimation as present
or not present for the primary study outcome by the
primary investigator. The percentage of FV overlapped
by FA was also estimated by direct visualization. FV
diameters were measured from the lateral wall to the
medial wall using the machine’s internal calipers.
The measurements were repeated 1 and 3 cm beyond
the inguinal ligament. Both the left and the right FA
and FV were measured for each participant.
All measurements were then repeated in the frog leg
position. Children placed in the frog leg position had
their hips flexed to 40 and externally rotated so that
their knee contacted the bed8 (Figure 1). Children
4 years of age and older maintained this position easily.
Children 2 years of age and younger who had difficulty
remaining in the frog leg position were assisted by
their parents to maintain this pose.
ACAD EMERG MED • July 2009, Vol. 16, No. 7 • www.aemj.org 581

Figure 2. Relative positions of the femoral artery and femoral

vein.

ages of children with femoral vessel overlap were

calculated for each leg in the straight leg and frog
leg positions at 0, 1, and 3 cm distal to the inguinal
ligament.
The secondary outcome measure was measurement
of the diameter of the FV in the straight and frog leg
positions. Diameters were measured in each position
using the internal calipers. This measurement was
recorded in centimeters for both the left and the right
legs at the inguinal ligament and 1 and 3 cm distal to
the inguinal ligament. The Sonosite Titan’s two sets of
internal calipers have a precision of ±2% of the distance
measured plus 1% of the depth of the image.9

Data Analysis
The differences in percentage of children with overlap
between the straight and frog leg positions at each
location were analyzed using a two-tailed McNemar’s
Figure 1. Straight leg (top) and frog leg positions. test. A p-value of <0.05 was considered significant.
Two-way analysis of variance (ANOVA) was used to
assess the relationship between FV diameter and leg
All procedures and measurements were performed position and ultrasound probe location. Analysis of
by a single pediatric EP to minimize interobserver vari- covariance (ANCOVA) was used to assess the interac-
ability. This investigator had 16 hours of classroom tions of leg position and ultrasound location, with age
training, 4 years of clinical US experience, and addi- on FV diameter. All analyses were conducted using
tional vascular US instruction by the division US coor- SPSS 14.0.2 (SPSS Inc., Chicago, IL, 2005).
dinator prior to the beginning of this study. The data
were recorded using a standardized data collection
sheet. RESULTS
A convenience sample of 92 patients was screened for
Measures eligibility. Twelve declined participation. Eighty patients
The US images were displayed on an 8.4-inch square completed the study. No patients were excluded for
screen. Two different transducers were used for this hypovolemia. Sixteen patients were enrolled in each of
study based on age recommendations from the manu-
facturer. For children 1 month to <1 year of age, a
25-mm broadband (10–5 MHz) linear array hockey-stick
probe was used. For children 2 to 8 years, a 38-mm Table 1
Subject Demographics
broadband (10–5 MHz) linear array vascular probe was
used.
The primary outcome measure was percentage of Race:
White ⁄
children with overlap of FA on FV in the straight and
African
frog leg positions (Figure 2). This was recorded as pres- Age Age Weight Sex American ⁄
ent or absent without consideration of percentage over- Categories (yr) (kg) (M ⁄ F) Other
lap. The FA was defined to overlap the FV if any Infant (n = 16) 0.36 (0.27) 5.83 (1.90) 7⁄9 7⁄9⁄0
portion of the FV lumen was obscured by the FA. The 2 yr (n = 16) 2.43 (0.35) 14.12 (1.99) 8⁄8 9⁄7⁄0
percentage of FV overlapped by the FA was then classi- 4 yr (n = 16) 4.48 (0.31) 19.68 (3.04) 9⁄7 9⁄7⁄0
fied by visual estimation into four categories (0%, 1%– 6 yr (n = 16) 6.43 (0.29) 24.02 (3.61) 8⁄8 9⁄7⁄0
8 yr (n = 16) 8.59 (0.27) 32.39 (6.81) 9⁄7 8⁄8⁄0
33%, 34%–66%, and >66%). The presence or absence of
overlap and the percentage FV overlapped by FA were Data are mean values (±SD).
recorded in a standard data collection set. The percent-
582 Hopkins et al. • ANATOMIC RELATIONSHIP BETWEEN COMMON FEMORAL ARTERY AND COMMON FEMORAL VEIN

Table 2 Table 4
Percentage of Children with Overlap of FA on FV Pooled Mean Diameters

Straight Frog p-Value Location

Left Femoral Vessel Left Femoral Vessel Right Femoral Vessel
0 cm 36% (29 ⁄ 80) 45% (36 ⁄ 80) 0.31
1 cm 75% (70 ⁄ 80) 88% (70 ⁄ 80) 0.02 Straight Frog Straight Frog
3 cm 93% (74 ⁄ 80) 86% (69 ⁄ 80) 0.27
0 cm 0.64 (0.23) 0.76 (0.28) 0.69 (0.24) 0.74 (0.28)
Right Femoral Vessel
1 cm 0.66 (0.23) 0.78 (0.29) 0.70 (0.28) 0.70 (0.29)
0 cm 41% (33 ⁄ 80) 41% (33 ⁄ 80) 1.00
3 cm 0.65 (0.27) 0.69 (0.29) 0.67 (0.30) 0.67 (0.29)
1 cm 93% (74 ⁄ 80) 88% (70 ⁄ 80) 0.39
3 cm 98% (78 ⁄ 80) 93% (74 ⁄ 80) 0.29
Data are mean (±SD) in cm.
In parentheses is the number of children with overlap ⁄ total
number of children.
FA = femoral artery; FV = femoral vein.
straight leg position by 0.12 cm (19%) at 0 cm, by
0.12 cm (18%) at 1 cm, and by 0.04 cm (6%) at 3 cm
(p < 0.001). Utilizing two-way ANOVA with position
the five age groups. Demographic characteristics of (frog or straight) and location (0, 1, and 3 cm distal to
each age group are included in Table 1. the inguinal ligament) as the independent variables, the
The percentage of children with overlap of FA on FV FV diameter was significantly greater in the frog leg
is listed in Table 2. This percentage did not vary signifi- position (p < 0.001) and at locations closer to the ingui-
cantly between the straight and frog positions at any nal ligament (p = 0.027). An interaction also existed
location except in the left leg 1 cm distal to the inguinal between position and location to influence FV diameter
ligament (p = 0.02). Here, the percentage of children (p = 0.008). FV diameter was greater in the frog leg
with overlap of FA on FV was greater in the frog leg position at locations closer to the inguinal ligament.
position. In each leg, the percentage of patients with Using ANCOVA to adjust for age as a covariate, the
overlap of FA on FV increased with increasing distance effects of leg position (p = 0.055) and US location
from the inguinal ligament. This observation was noted (p = 0.437) lost their significance. Thus when taken
in both the straight and the frog leg positions. together, a three-way interaction of position by location
Table 3 shows the percentage of FV overlapped by by age (p = 0.003) becomes apparent where FV diame-
FA at 0, 1, and 3 cm distal to the inguinal ligament in ter shows a greater increase in the frog leg position,
both frog and straight leg positions. In each leg, in both closer to the inguinal ligament, but only in children
the frog and the straight leg positions, there was less ‡2 years old.
overlap closer to the inguinal ligament. The associations of FV diameter, leg position, and
Pooled mean diameters for the left and right FVs are measurement location did not hold for the right leg. FV
listed in Table 4. Left leg FV diameter in frog leg diameter did not increase in the frog leg position
position was greater than left leg FV diameter in the (p = 0.42). FV diameter increased moderately at locations
closer to the inguinal ligament (p = 0.05). No interaction
Table 3 was found between position and location (p = 0.27).
Percentage of Children with Overlap of FA on FV, Stratified by Mean FV diameters for three age ranges (<1, 2 to 4, and
Percentage of FV Overlapped 6 to 8 years) are listed in Table 5. The youngest group
of patients (1–12 months) had the smallest maximal
Location increase in FV diameter of 12% in the frog leg position at
3 cm. This increase was 10% less than the maximal FV
Left Leg Right Leg
diameter increase in the 2- to 4-year group and 11% less
Straight Frog Straight Frog than the maximal FV diameter increase in the 6- to
0 cm 8-year group. FVs had larger increases in diameter at
0 64 (51 ⁄ 80) 55 (44 ⁄ 80) 60 (48 ⁄ 80) 60 (49 ⁄ 80) the 0- and 1-cm locations than at the 3-cm location in the
1–33 19 (15 ⁄ 80) 25 (20 ⁄ 80) 18 (14 ⁄ 80) 26 (21 ⁄ 80) 2- to 4- and 6- to 8-year groups.
34–66 8 (6 ⁄ 80) 11 (9 ⁄ 80) 14 (11 ⁄ 80) 8 (6 ⁄ 80)
>66 10 (8 ⁄ 80) 9 (7 ⁄ 80) 8 (7 ⁄ 80) 5 (4 ⁄ 80)
1 cm DISCUSSION
0 25 (20 ⁄ 80) 13 (10 ⁄ 80) 8 (6 ⁄ 80) 14 (11 ⁄ 80)
1–33 21 (17 ⁄ 80) 26 (21 ⁄ 80) 11 (9 ⁄ 80) 31 (25 ⁄ 80) Historically, CVLs have been placed using external
34–66 23 (18 ⁄ 80) 28 (22 ⁄ 80) 36 (29 ⁄ 80) 26 (21 ⁄ 80) landmarks as a guide for underlying femoral vessel
>66 31 (25 ⁄ 80) 34 (27 ⁄ 10) 46 (36 ⁄ 80) 29 (23 ⁄ 80)
location.1 Medical textbooks depict the location of the
3 cm
0 9 (7 ⁄ 80) 15 (12 ⁄ 80) 4 (3 ⁄ 80) 9 (7 ⁄ 80) FV as medial to the FA.1 US studies have documented
1–33 6 (5 ⁄ 80) 15 (12 ⁄ 80) 5 (4 ⁄ 80) 16 (13 ⁄ 80) that the FA frequently overlaps the FV.5,6 Because of
34–66 31 (25 ⁄ 80) 24 (19 ⁄ 80) 19 (15 ⁄ 80) 15 (12 ⁄ 80) their relatively smaller body surface areas, young chil-
>66 54 (43 ⁄ 80) 46 (37 ⁄ 80) 74 (58 ⁄ 80) 60 (48 ⁄ 80) dren have smaller femoral vessel diameters than older
children and adults.6 Overlap of FA on FV, combined
In parentheses is the number of children with overlap ⁄ total
number of children. with smaller femoral vessel diameter, may increase dif-
FA = femoral artery; FV = femoral vein. ficulties associated with CVL placement in the pediatric
population.
ACAD EMERG MED • July 2009, Vol. 16, No. 7 • www.aemj.org 583

Table 5
Age-specific Mean Femoral Diameters

Location
Left Femoral Vessel Right Femoral Vessel

Age Straight Frog % Change Straight Frog % Change

Infant
0 cm 0.34 (0.08) 0.34 (0.09) 0 0.39 (0.10) 0.35 (0.11) 10
1 cm 0.33 (0.09) 0.36 (0.06) 9 0.33 (0.11) 0.34 (0.10) 3
3 cm 0.31 (0.07) 0.35 (0.06) 12 0.33 (0.07) 0.34 (0.06) 3
2–4 yr
0 cm 0.62 (0.15) 0.76 (0.16) 23 0.69 (0.14) 0.72 (0.17) 4
1 cm 0.69 (0.16) 0.79 (0.18) 14 0.70 (0.16) 0.65 (0.17) 7
3 cm 0.64 (0.16) 0.70 (0.20) 9 0.64 (0.17) 0.65 (0.21) 2
6–8 yr
0 cm 0.82 (0.17) 0.97 (0.17) 18 0.85 (0.23) 0.95 (0.20) 12
1 cm 0.80 (0.16) 0.99 (0.21) 24 0.89 (0.24) 0.92 (0.24) 2
3 cm 0.84 (0.25) 0.85 (0.29) 0 0.87 (0.30) 0.86 (0.28) 1

Data are mean (±SD) in cm.

Emergency medicine procedural textbooks suggest
placing patients in the frog leg position to facilitate
placement of femoral CVLs.1 A recent study by Werner
et al.7 supported placing adult patients in the frog leg
position. The authors noted that adults placed in the
frog leg position had an increased percentage of the FV
available for CVL placement.
In our study, children placed in the frog leg posi-
tion compared to those in the straight leg position
did not have an appreciable decrease in the presence
of femoral vessel overlap. The percentage of children
with femoral vessel overlap increased as the measure-
ment location moved distal from the inguinal liga-
ment. Our study examined the difference in FV
diameter in the frog leg and straight leg positions.
We evaluated these differences at 0, 1, and 3 cm
distal to the inguinal ligament. When adjusted for
age, placing children in the frog leg position no
longer made a difference in FV diameter at any loca-
tion. Our data show that the percentage of increase
in FV diameter is largest for children ‡2 years old
and that these changes are most significant at mea-
surements closest to the inguinal ligament. Our data
suggest that clinicians may consider placing children
‡2 years old in the frog leg position with the US
probe at or just distal to the inguinal ligament to opti-
mize the view and approach to the FV.
LIMITATIONS
The percentage of children with FV overlapped by FA
in our study was greater than that previously reported.6
Several factors could have contributed to this differ-
ence. Participants in this study were taken from a con-
venience sample of children presenting to the ED. The
sampling of participants may have differed from previ-
ous studies. Also, US is known to be an operator
dependent skill and differences in technique can
account for a variance in images obtained.10
Our sample size was not powered to determine the
effect of the frog leg position on FV diameter in chil-
dren of different ages. Despite this limitation, we noted
that, when age is considered as a covariate, the effects
of position and location on FV diameters lose their sta-
tistical significance. We observed that children 2 and
4 years of age had similar FV diameters, children 6 and
8 years of age had similar FV diameters, and children 1
to <12 months of age had similar FV diameters. A study
with a larger sample size powered to examine this
effect in each of these three age ranges would be a
future consideration.
Our study used visual estimation to approximate the
percentage of FV overlapped by FA. Caliper measure-
ments would have provided a more precise picture of
the percentage of FV overlapped by FA. Considering
the small size of infant femoral vessels, caliper mea-
surement of overlap would have been imprecise.
CONCLUSIONS
Knowledge of femoral vessel anatomy is recommended
prior to attempting placement of central venous lines in
children.6 Our study demonstrated a population in
which a large percentage of children had femoral veins
with partial overlap by the femoral artery. This femoral
vessel overlap may make accessing the FV for veni-
puncture more difficult. Additionally, placing children
in the frog leg position did not decrease the percentage
of children with FV overlap. However, the frog leg
position did increase the overall diameter of the FV
available for line placement. The percentage increase in
FV diameter is highest closer to the inguinal ligament
in children ‡2 years of age. The clinician may consider
positioning children ‡2 years of age in the frog leg
position for central venous line placement and attempt
placement near the inguinal ligament.
References
1. Roberts JR, Hedges JR. Clinical Procedures in
Emergency Medicine, 4th ed. Philadelphia, PA: Else-
vier Inc., 2004, pp 370–5.
584 Hopkins et al. • ANATOMIC RELATIONSHIP BETWEEN COMMON FEMORAL ARTERY AND COMMON FEMORAL VEIN
2. Haas A. Clinical review: vascular access for fluid
infusion in children. Critical Care. 2004; 9:478–94.
3. Skippen P, Kissoon N. Ultrasound guidance for cen-
tral vascular access in the pediatric emergency
department. Pediatr Emerg Care. 2007; 23:203–7.
4. Karapinar B, Cura A. Complications of central
venous catheterization in critically ill children. Pedi-
atr Int. 2007; 49:593–9.
5. Hughes P, Scott C, Bodenham A. Ultrasonography
of the femoral vessels in the groin: implications for
vascular access. Anaesthesia. 2000; 55:1198–202.
6. Warkentine FH, Pierce MC, Lorenz D, Kim I. The
anatomic relationship of femoral vein to femoral
artery in euvolemic pediatric patients by ultrasono-
graphy: implications for pediatric femoral central
venous access. Acad Emerg Med. 2008; 15:426–30.
7. Werner SL, Jones RA, Emerman CL. Effect of hip
abduction and external rotation on femoral vein
exposure for possible cannulation. J Emerg Med.
2008; 35:73–5.
8. Ballinger PW. Merrill’s Atlas of Radiographic Posi-
tions and Radiologic Procedures, 5th ed. St. Louis,
MO: Mosby-Year Book Inc., 1982, pp 98–9.
9. SonoSite. Titan Ultrasound System User Guide. Bot-
hell, WA: SonoSite Inc., 2004, p 189.
10. Lei C, Baker MD. Novel applications of ultrasound
in pediatric emergency medicine. Pediatr Emerg
Care. 2007; 23:115–23.