1730
BRIEF REPORT
The S1 “Scotty Dog”: Report of a Technique for S1
Transforaminal Epidural Steroid Injection
David E. Fish, MD, MPH, Paul C. Lee, BS, MSIV, Daniel B. Marcus, MD
ABSTRACT. Fish DE, Lee PC, Marcus DB. The S1 “Scotty
dog”: report of a technique for S1 transforaminal epidural
steroid injection. Arch Phys Med Rehabil 2007;88:1730-3.
Objective: To report a technique for needle placement by
using the bony landmark of the “Scotty dog” on an oblique
view for epidural injection of corticosteroid into the S1
foramina.
Design: Brief report on a technique for S1 transforaminal
epidural steroid injection.
Setting: Academic multispecialty spine center.
Participants: Patients with L5 and S1 foraminal and para-
central disk herniation with concurrent L5-S1 radicular pain.
Intervention: Fluoroscopically guided, contrast-enhanced
L5 and S1 transforaminal epidural steroid injections (ESIs).
Main Outcome Measures: Not applicable.
Results: The L5-S1 foramina can be visualized with 1
oblique (and usually caudally tilted) fluoroscopic view. An S1
Scotty dog can be visualized as an anatomic landmark for the
guidance of the needle tip into the S1 foramen. While perform-
ing simultaneous L5 and S1 transforaminal ESIs, 1 view can be
used to guide both needles into the foramen. Thus, the proce-
dure can be completed in less time and potentially with less
radiation exposure than if different views for each foramen
were to be used.
Conclusions: Classic description of the S1 spinal nerve
block uses an anteroposterior approach to the foramen. Look-
ing for an S1 Scotty dog facilitates predictable visualization of
the foramen, medial needle placement with epidural flow of
contrast, and simultaneous visualization for needle placement
to the L5 foramen.
Key Words: Injections, epidural; Radiculopathy; Rehabili-
tation.
© 2007 by the American Congress of Rehabilitation Medi-
cine and the American Academy of Physical Medicine and
Rehabilitation
PIDURAL INJECTION OF corticosteroids is com-
E monly performed for management of radicular pain
syndromes with demonstrated pain relief. Radicular pain is 1
assumed to arise from inflammation of nerve roots or spinal
nerves caused by mechanic or chemical irritation often from
a herniated intervertebral disk. It is postulated that the
application of corticosteroid directly on the inflamed nerve
From the Department of Orthopaedics, David Geffen School of Medicine, UCLA,
Los Angeles, CA (Fish); Albert Einstein College of Medicine at Yeshiva University,
Bronx, NY (Lee); and Santa Cruz Medical Clinic Physical Medicine and Rehabili-
tation Pain Medicine, Santa Cruz, CA (Marcus).
No commercial party having a direct financial interest in the results of the research
supporting this article has or will confer a benefit upon the author(s) or upon any
organization with which the author(s) is/are associated.
Reprint requests to David E. Fish, MD, MPH, Dept of Orthopaedics, UCLA School
of Medicine, 1245 16th St Tower Bldg 7th Fl, Rm 715, Santa Monica, CA 90404,
e-mail: dfish@mednet.ucla.edu.
0003-9993/07/8812-00251$32.00/0
doi:10.1016/j.apmr.2007.07.041
Arch Phys Med Rehabil Vol 88, December 2007
would have greater benefit than a less direct approach (in-
terlaminar or caudal). Furthermore, direct application of
local anesthetic on the spinal nerve has diagnostic and
surgical prognostic benefits. Transforaminal access to the
epidural space requires image guidance.
Fluoroscopic guidance and injection of contrast during trans-
foraminal epidural steroid injections (ESIs) are used to ensure
proper needle placement and medication administration. A
study by White at al2 showed a 25% and 30% incidence of
incorrect blind placement for caudal and lumbar injections,
respectively. Studies of caudal-approach ESIs by Renfrew et
al3 show that only 55 (61.7%) of 89 nonfluoroscopically di-
rected placements were correct and that fluoroscopy revealed
intravenous injection in 29 (9.2%) of 316 procedures in which
no blood had been evident on the Valsalva maneuver or aspi-
ration. In addition, Furman et al4 report that aspiration of blood
or the presence of a flash of blood in the needle hub fail as
alternatives to fluoroscopy in detecting intravascular placement
because of low sensitivity (44.7%) and that the incidence of
intravascular injection at S1 is especially high (21.3% vs 8.1%
at lumbar levels).
However, fluoroscopy exposes the patient and medical staff
to radiation. Mettler et al5 noted that current fluoroscopes are
capable of producing dose rates in the range of 0.2Gy (20rad)
a minute and produce even higher rates when taking serial cine
images. They add that there have been over 100 cases of
documented and likely many more undocumented cases of skin
and underlying tissue injury, including dermal necrosis, with
various fluoroscopic procedures.
Placement of the needle tip into the foramen from which
the L5 spinal nerve exits is classically described using
multiplanar fluoroscopic images. The initial approach uses
an oblique view (usually with caudal tilt) to visualize the
foramen. The L5 vertebra is visualized as a “Scotty dog.”
The eye of the dog is the pedicle and allows for predictable
visualization of the foramen. The needle is advanced toward
the 6-o’clock position under the L5 pedicle. Anteroposterior
(AP) and lateral views are then used to ensure the needle tip
is in the “safe triangle.”6
The foramen from which the S1 nerve root exits is classi-
cally accessed with an approach that differs from the lumbar
foraminal approach. Often an AP view with caudal tilt is used.
The dorsal and ventral foramina are superimposed, and the
lateral view is used to ensure epidural placement of the needle
tip.
The L5-S1 intervertebral disk herniation is the most com-
mon cause of lumbar radicular syndromes. A lateral herni-
ation can cause irritation of the exiting L5 spinal nerve,
whereas a more paracentral herniation will affect the de-
scending S1 spinal nerve. A broad-based, eccentric hernia-
tion often causes an L5 and S1 radiculitis. Furthermore,
many clinicians will inject corticosteroid at the level of the
disk herniation and at the level of the suspected inflamed
nerve. Thus, an L5-S1 herniation with an S1 radicular pic-
ture could reasonably be treated with corticosteroid injection
at the L5-S1 disk and S1 spinal nerve.
 THE S1 “SCOTTY DOG,” Fish 1731

Fig 1. A 58-year-old woman with 9 months of pain in her lower back radiating to her left foot through the back of her leg without
neurologic loss. Magnetic resonance imaging revealed disk herniation at L5-S1 with apex deviated to the left causing minor impinge-
ment of the S1 nerve root. Electromyography suggested an S1 radiculopathy. After conservative options failed, a combined L5 and S1
transforaminal ESI using the procedure described improved symptoms more than 75% at 3 months. (A) The oblique view with and
without highlights of the Scotty dog formation. (B) Simultaneous placement of needles in L5 and S1 foramen in oblique view with and
without highlights of the Scotty dog formation. (C) The lateral view. (D) The AP view. (E) The AP view with contrast.

To perform a 2-level L5 and S1 transforaminal ESI, an METHODS

oblique view would be obtained, the needle advanced to a
comfortable depth, and an AP view would then be obtained to
visualize the S1 foramen. We report that both foramina can be
comfortably and reproducibly visualized with the same oblique
view. Viewing the S1 vertebral segment as a Scotty dog facil-
itates visualization of the S1 foramen. This allows for initial
needle advancement for an L5 and S1 transforaminal ESI using
the same fluoroscopic view.
With the patient in prone position, the skin is prepped and
draped in a sterile fashion. An AP view is obtained. Caudal tilt
is adjusted to line up the L5-S1 endplates. The C-arm is then
rotated ipsilateral oblique to view the L5 vertebral segment as
a Scotty dog. When the L5 Scotty dog is visualized, the
operator should view the S1 segment, and the dog’s “neck” and

Arch Phys Med Rehabil Vol 88, December 2007

1732 THE S1 “SCOTTY DOG,” Fish

Fig 1. (Continued)

forelimb can be visualized as the superomedial landmarks of DISCUSSION

the S1 foramen (even if the rest of the dog is not seen). The S1
superior articular process is the ear of the S1 dog (fig 1A).
The needles are then advanced en point to the respective
foramina (fig 1B). AP and lateral views are obtained, and the
needles are repositioned as appropriate (figs 1C–1D). Con-
trast is then administered using live fluoroscopy (fig 1E) and
digital subtraction angiography is used if there is suspicion
of vascular uptake. Medication is administered, the needles
are removed, and the area is cleaned and dressed.
This approach to the S1 foramen is predictable and effective.
The AP view of the S1 foramen is not always predictable as the
ventral foramen is often the more easily seen structure. Also, when
lining up the ventral and dorsal S1 foramen, misadventure with
placement of the needle tip anterior to the ventral foramen is
possible with gastrointestinal penetration as a possible conse-
quence in the AP view. Instead of advancing the S1 needle in AP
view and the L5 needle in oblique view, an oblique view would be
used to advance both needles. Thus, potential advantages include

Arch Phys Med Rehabil Vol 88, December 2007

 THE S1 “SCOTTY DOG,” Fish
reduced time of procedure and reduced radiation exposure. How-
ever, a head-to-head trial is necessary to establish reduced radia-
tion exposure time. A potential disadvantage of this approach is
that, in the oblique view, the iliac crest may obstruct the pathway
of the S1 foramen. This may be compensated for by varying the
caudal tilt.
CONCLUSIONS
Visualization of the S1 Scotty dog using an oblique and
caudal tilt fluoroscopic view facilitates visualization of the S1
foramen and should be considered as a reasonable alternative to
classic descriptions when the physician plans to inject cortico-
steroid at both L5 and S1 levels.
References
1. Abdi S, Datta S, Trescot AM, et al. Epidural steroids in the
1733
management of chronic spinal pain: a systematic review. Pain
Physician 2007;10:185-212.
2. White AH, Derby R, Wynne G. Epidural injections for the diag-
nosis and treatment of low back pain. Spine 1980;5:78-86.
3. Renfrew DL, Moore TE, Kathol MH, el-Khoury GY, Lemke JH,
Walker CW. Correct placement of epidural steroid injections: flu-
oroscopic guidance and contrast administration. AJNR Am J Neu-
roradiol 1991;12:1003-7.
4. Furman MB, O’Brien EM, Zgleszewski TM. Incidence of intravas-
cular penetration in transforaminal lumbosacral epidural steroid
injections. Spine 2000;25:2628-32.
5. Mettler FA Jr, Koenig TR, Wagner LK, Kelsey CA. Radiation injuries
after fluoroscopic procedures. Semin Ultrasound CT MR 2002;23:428-
42.
6. Bogduk N, Aprill C, Derby R. Epidural steroid injections. In: White
AH, Schofferman JA, editors. Spine care. Vol 1. St Louis: Mosby;
1995. p 322-43.
Arch Phys Med Rehabil Vol 88, December 2007