Procedures · Head, Spine and Pelvis

Lumbar Spine Positioning Series

Compare the three routine lumbar spine projections: AP, oblique, and lateral.

Lumbar spine

Right

StudyBucky

Loading model…

Hover a bone, name it, then click to check · drag to rotate

The routine lumbar spine series

The projections this region is examined with. Set them up yourself in Practice positioning.

AP lumbar spine

Anteroposterior projection · central ray 90° to the image receptor

Demonstrates: The lumbar vertebral bodies and their intervertebral joint spaces, the spinous and transverse processes, the sacroiliac joints and the sacrum. Three findings together prove there was no rotation: the sacroiliac joints lie equidistant from the spinous processes, the spinous processes sit in the midline of the vertebral column, and the transverse processes are of equal length on both sides.
Part position: Supine, with the arms at the side and the head on a pillow. Align the midsagittal plane to the central ray and to the midline of the table. Flex the hips and knees to reduce the lordotic curve, and check that neither the thorax nor the pelvis is rotated.
Central ray: 90° to the image receptor, centered to the midsagittal plane at the level of the iliac crest, which is the level of the L4-L5 interspace. On a receptor collimated to the five lumbar vertebrae alone, center instead to L3 at the level of the lower costal margin, 1.5 inches (4 cm) above the iliac crest.
Why it is positioned this way: The lumbar spine is lordotic, so its bodies are tipped relative to a beam sent straight through a patient lying flat, and the disc spaces close. Flexing the hips and knees rolls the pelvis back and flattens that curve against the table, which brings the interspaces parallel to the beam and opens them. Nothing about the tube changes — the patient is what moves the anatomy into line.

Also covered by this projection

Taken PA insteadThe same anatomy can be taken prone. Doing so places the intervertebral spaces more nearly parallel to the diverging rays, so the joint spaces open a little more readily than they do supine, and it moves the more radiosensitive anterior tissues to the exit side of the beam.

Taken erectUseful when the question is how the spine behaves under load, since an erect image shows the natural weight-bearing stance rather than the shape the table imposes.

AP axial L5-S1A separate special projection, not part of this routine, that opens the L5-S1 joint space from the front. Taken supine with the midsagittal plane aligned to the central ray, the ray angled 30° cephalad for a male patient and 35° for a female patient, centered to the midsagittal plane at the level of the ASIS. The angle cancels the tilt of the lumbosacral junction, which on this skeleton leans about 27° — close to the smaller of the two prescribed angles. Note the trap: the AP axial sacroiliac joints uses that same 30-35° cephalad angle but centers 2 inches (5 cm) BELOW the ASIS, so centring, not angle, is what separates the two.

Posterior oblique lumbar spine

Right posterior oblique · 45° · central ray 90° to the image receptor

Demonstrates: The zygapophyseal joints of the DOWNSIDE — the side turned toward the receptor. A right posterior oblique therefore shows the right joints, and a left posterior oblique the left. The rotation is right at 45° when those joints are open and the pedicle sits about halfway between the midline and the lateral border of the vertebral body; a pedicle projected close to the midline means the patient was turned too far, and one out at the lateral border means not far enough.
Part position: Semisupine, rotated 45° from supine so the side under examination lies down against the image receptor, with the head on a pillow. Align the vertebral column to the midline of the table. Turn the shoulders and the pelvis through the same 45° so the trunk is not twisted, flex the upper knee for stability, bring the arm farthest from the receptor across the chest, and support the raised shoulder and hip on radiolucent sponges.
Central ray: 90° to the image receptor, centered to L3 at the level of the lower costal margin — 1 to 2 inches (2.5 to 5 cm) above the iliac crest — and 2 inches (5 cm) medial to the raised ASIS.
Why it is positioned this way: These joints do not face straight forward or straight sideways; they lie obliquely to the midsagittal plane, and a joint only reads as an open space when the beam runs along its plane rather than across it. Turning the patient 45° is what brings the beam into that plane. The same turn also throws the posterior elements into the outline students learn as the Scottie dog, whose neck is the pars interarticularis — which is the part this projection is really taken to inspect.

Also covered by this projection

Anterior obliques insteadTurning the patient the other way, onto the front, reverses which joints are shown: an anterior oblique demonstrates the UPSIDE joints, the ones farthest from the receptor. So a right anterior oblique and a left posterior oblique show the same set. Both sides are imaged whichever way round the patient is turned, because the two sides are compared against each other.

When 45° is not the best angleThe lumbar joints do not all lie at one angle to the midsagittal plane. The upper ones sit closer to 50° and the L5-S1 joint closer to 30°, so 45° is the compromise that opens the column as a whole. When the question is specifically about one end of it, the rotation is adjusted toward that end’s own angle.

The Scottie dog, part by partOn this projection the transverse process reads as the nose, the pedicle seen end-on as the eye, the superior articular process as the ear and the inferior articular process as the front leg. The neck between them is the pars interarticularis, and a lucent break across that neck is spondylolysis — the finding the projection exists for.

Lateral lumbar spine

Left lateral position · central ray 90° to the image receptor

Demonstrates: The lumbar bodies and their intervertebral joint spaces in profile, the intervertebral foramina, and the spinous processes. This is the projection that shows the foramina — the obliques show the joints — so the two are read together rather than as alternatives. The position is proved by the greater sciatic notches and the posterior borders of the vertebral bodies lying superimposed.
Part position: Lateral recumbent on the left side, with the head on a pillow, the knees flexed and a support placed between the knees and between the ankles to hold the pelvis square. Align the midcoronal plane to the central ray and to the midline of the table, and place a radiolucent support under the waist so the long axis of the spine lies near parallel to the table. Check that neither the thorax nor the pelvis has rolled out of a true lateral.
Central ray: 90° to the image receptor, centered to the midcoronal plane at the level of the iliac crest, which is the level of the L4-L5 interspace. On a receptor collimated to the five lumbar vertebrae alone, center instead to L3 at the level of the lower costal margin.
Why it is positioned this way: The waist support is doing radiographic work, not making the patient comfortable. Left unsupported, the span between the ribs and the pelvis lets the column sag toward the table, and a sagging column no longer lines its disc spaces up with a beam sent straight across. Holding the long axis of the spine parallel to the receptor is what lets the ray stay at 90° and still pass through the interspaces.

Also covered by this projection

Lateral L5-S1The fourth projection in this routine, taken on a small receptor collimated closely to the lumbosacral junction. Positioned as above, lateral recumbent with the waist supported, but centered 1.5 inches (4 cm) inferior to the iliac crest and 2 inches (5 cm) posterior to the ASIS — the ASIS being chosen because it is easy to palpate on a patient lying on their side. With the waist supported sufficiently the central ray stays 90° to the image receptor; with less support it is angled 5° to 8° caudad. It shows the body of L5, the first and second sacral segments and the L5-S1 joint space, and it is proved unrotated by the greater sciatic notches lying superimposed.

Why the angle when the waist sagsThe rule is not the number, it is the line. Angling 5° to 8° caudad works because it puts the central ray parallel to the interiliac line — the imaginary line joining the two iliac crests — which is the direction the L5-S1 joint space opens along once the column has sagged. Palpate both crests, picture the line, and set the tube parallel to it; a large waist can need the angle the other way, cephalad, to open the same space.

Scatter at this thicknessThe lateral lumbar spine is one of the thickest parts routinely imaged, so it generates a great deal of scatter. Collimating closely and laying lead masking on the tabletop behind the patient both matter here, and they matter more with a digital receptor, not less, because a digital detector will still produce a viewable image from a badly scattered exposure instead of failing visibly.

Procedures lessonVertebral column