1

Taisa Polishchuk
DOS 531 Clinical Oncology for Medical Dosimetry
6/17/2015
Craniospinal Irradiation (CSI) Assignment
1). If the patient were positioned prone, a board is usually placed under the chest to build
up the lower torso. What does this accomplish?
Placing the board underneath the length of the body such that the body is elevated from
the table at a greater distance relative to the head will level the spine with the head, and build up
patient to reduce cervical and thoracic spine curvature.1
2). How do you match the spine and head ports for a craniospinal setup? Be specific. Give
me the formulas used to determine any angles and give an example of using the formula(s).
Provide a diagram or drawing.
The image below shows the cranial field (in pink) perfectly matched to the upper spine
field (in yellow). We can see that both the collimator and the table have been rotated to match the
divergence.

The next image displays no rotations and a mismatch of the fields (the cranial pink field
does not match the spinal yellow field, whereas the blue cranial field matched the yellow spinal
field perfectly).

In order to accomplish the match between cranial and upper spinal fields we need to
perform multiple calculations. The inferior border of the cranial field is matched to the
divergence of the superior border of the upper spine field. This match is dependent on the cranial
field collimator rotation, the angle of which gets determined utilizing the following formula
tan-1= (1/2 spinal filed length)/(SAD).1For example, on a 100SAD linac, the collimator rotation
required to align 25x25cm cranial fields with a direct spinal axis field of 36 cm in length is 10
degrees. We were able to calculate this value by using the above formula tan-1= (36/2)/(100)=
10degrees.
In addition, eliminating the divergence between the cranial port and spinal port, the table
is rotated through a floor angle, which gets calculated by the following formula tan-1= (1/2
cranial field length)/(SAD).1 For example, on a 100SAD linac, the degree of the table, rotated
towards the side of the 25x25cm cranial field, is measuring 7 degrees. We were able to calculate
this value by using the above formula tan-1= (25/2)/(100)= 7 degrees. Below we can see the
match through the couch rotation.

3). If you wanted to remove any divergence from the eyes in the cranial port, how would
this be accomplished? Why would you do this? Show a formula and how it can be used.
Provide a diagram or drawing.

Two techniques could be utilized in order to eliminate the divergence into the anterior
globe or lens of the eye in the cranial port. We can place the isocenter in the middle of the brain
and rotate the gantry to be non-divergent at the level of the eyes or we can place it midline in the
brain at the level of the bony outer canthus of the eye. In the event the isocenter is placed in the
middle of the brain, the degree of the gantry rotation will be determined based on the formula
tan-1 = (length of anterior jaw/SAD). The image below shows the application of the formula, and
the gantry rotation of 95 for the right lateral and 265 for the left lateral fields.

The possibility of cataract occurrence raises the importance of keeping the dose to the eye
lens as low as possible, and not reaching the tolerance dose of 1000cGy.2 Taking the total dose in
some instances to as high as 3600 cGy when treating CSI, the cataract occurrence could be of
concern and therefore must be evaluated when creating a treatment plan.
4). In your own words describe the setup for a CSI adult patient (specify prone or supine)
where 2 spine ports must be matched that extend to the bottom of S-2.
For craniospinal irradiation delivered in the prone position, the patient must lie on thick
enough styrofoam to build up patient to reduce cervical and thoracic spine curvature, with the
arms down by the sides and the face down in the constructed prone face mask. The spine should
be as parallel to the table top from the skull down to the sacrum. A vac-lock bag should be placed

under the patients upper body or the full body depending on the size of the bag. This will help
immobilize the patients upper body and assist in daily reproducibility. In addition we have to
make sure the shoulders are down with a minimum of 3-4 cm between the chin and the
shoulders. In doing so we have to make sure there are no skin folds posteriorly on the neck to
reduce the skin reaction. A final mask is placed over the back of the skull to assure the skull is
immobilized as well.
Matching the upper and the lower spinal fields could be accomplished by utilizing the
gap calculation. Gap=s1+s2, where s1= (1/2L1)x(d1/SSD1) and s2= (1/2L2)x(d2/SSD2)
Matching upper and lower spine fields and performing the gap calculation could be avoided by
kicking the table to 90 or 270 degrees and rotating the gantry to match the divergence of the
upper spine field. In this case, the length of the upper spine field will determine the gantry angle.
5). Pretend you are telling the therapist everything that is needed during the CT
simulation. (How the patient should be positioned. Do not forget to include all devices used,
head position, chin position, arms).
The patient will be simulated prone. The prone position will allow direct visualization of
the light field from the linear accelerator on the patient thereby allowing daily adjustments of the
junctions. In addition, the prone position will let the therapists palpate the spinal processes to
make sure the patient is strait and aligned for the treatment fields. The patient will be positioned
on a styrofoam board with arms down by side, and with the face down in a partial aquaplast
mask. This part of the mask is constructed in the supine position with the neck somewhat flexed
(the mandible, maxilla, and oral cavity should not be in the exit of the upper spine field) position
with the B-adult head-holder under the head. The aquaplast mask is attached to the q-fix prone
cranial spinal axis kit (displayed in the image below) in order to make sure that the mask is
draped over the face, and the blue horn part of the mask does not cross into the treatment field. If
the mask is pushed onto the patients face too deep, the horn part of the mask will be in the path
of the treatment beam and will attenuate it changing the dose delivered to the brain of the patient
creating a cold spot anteriorly.3

We will make sure the head is not rotated by using the lateral laser on the outer canthus of
the eyes and the sagittal laser should be running down the midline of the patients face. Once the
mask has been made and cooled slightly, then the soft moleskin material will be adhered into the
mask at the location of the chin and forehead for patients comfort. The next step is to have the
patient lie prone on the styrofoam and a vac-lock bag under the patients upper body to help
immobilize the patients body and assist on daily reproducibility, with the arms down by the side
and the face down into the newly constructed prone face mask. Making sure the shoulders are
down, we will place a final mask over the back of the skull to assure the skull is immobilized as
well. The cushion could be placed under the patients ankles for comfort.
6). For treatment planning, approximately where will you place the isocenter for each field
for the patient above, will the isocenters be moved? Why or why not? What are the
approximate field borders?
When setting up the isocenter for the upper spinal field we want to make sure to keep the
same x and y coordinates (right/left and ant/post) as the cranial field. This way we will only have
to change one coordinate - the z coordinate, determining our superior/inferior offset. Performing
the shift for the upper spinal field and only shifting in one direction (inferior from the cranial
isocenter) can eliminate setup errors and uncertainties. Even though there are no set rules on
where to split the upper and lower spine fields, some dosimetrist prefer to set the lower border of
the upper spinal field at the bottom of L2. L2 is a good area to do the matching of the fields, due
to it being a less critical part of the spinal cord since cauda equina starts from that point. When
creating a lower spinal field, maintaining the same x and y coordinates as in the upper spinal
field, and just shifting the z coordinate will again allow for the easier set up and less error

occurrence. Matching the upper and the lower spinal fields could be accomplished by utilizing
the gap calculation. Gap=s1+s2, where s1= (1/2L1)x(d1/SSD1) and s2= (1/2L2)x(d2/SSD2)
Matching upper and lower spine fields and performing the gap calculation could be avoided by
kicking the table to 90 or 270 degrees and rotating the gantry to match the divergence of the
upper spine field. In this case, the length of the upper spine field will determine the gantry angle.
The field borders for the upper spinal cord:
The superior border of the field starts as soon as we can clear the shoulders (approximately at the
level of C3-C5), and terminates at L2. Right to left, the field should extend 2 cm lateral of the
vertebral body.
The field borders for the lower spinal cord:
The superior border starts at the L2 level and terminates below the S2-S3 level. When it comes to
determine how wide the lower spinal field should be, some physicians use a spade shape field,
increasing the width of the sacral portion of the spine field to encompass the sacroiliac joints and
cover the sacral nerve roots. Other physicians feel that straight field borders throughout the
whole spine are adequate.
The field borders for the cranium:
Lateral portal fields are used for treating the whole brain. The target volume in the lateral fields
includes the entire brain and the cervical spinal cord to the junction with the posterior thoracic
field.4 The inferior margin of the field is at the level of the superior border of the upper spinal
field. An equal amount of flash (usually 2 cm) surrounds the cranium on the anterior, superior,
and posterior margins.
7). Pretend that you must give the therapist a detailed description for treatment (feathering
the gaps) for the patient above. How will the fields be feathered during treatment?
Technique utilized in our clinic does not involve the gap between the upper and the lower
spine fields.5 We usually angle the gantry and the couch on the lower spinal field in order to
match the divergence of the upper spinal field. When listing a detailed description for the
treatment and the feathering between the cranial and upper spinal field I would note these details
in the time out form, which gets checked by the therapists prior to the treatment start.6
For cranial fields I would note in the time out form:
Setup to marks on mask. No shifts. These fields are 1.1 and 1.2.
Matchline changes after 5th fraction 0.5 cm decrease in the brain block. 1.1a, 1.2a are the 1st
matchline fields, 1.1b, 1.2b are the 2nd matchline fields, with the additional 0.5 cm decrease in
the brain block to a total of 1.0 cm from the original matchline. In addition, make sure spine
tattoos are also in alignment with marks on mask.
For upper spine field I would note in the time out form:

Shift IN (inferior) 36 cm from brain isocenter to PA isocenter. No other shifts. This field is 2.1.
Matchline changes after the 5th fraction (field 2.1a) superior border will increase by 0.5 cm and
match to the new (1.1a and 1.2a) inferior cranial borders. Matchline changes after the 10th
fraction (field 2.1b) superior border will increase by 0.5 cm and match to the new (1.1b and
1.2b) inferior cranial borders. The inferior border of the upper spinal field will stay the same and
not shift throughout all of the treatment fractions.
For lower spine field I would note in the time out form:
Shift IN 19 cm from the upper spine field (total of 55cm from the brain isocenter). Match on skin
with the upper spine field with couch rotation and gantry angle field 3.1. There are no
feathering or matchline changes between the upper spine field and the lower spine field.
Matching will be accomplished by utilizing half-beam block and matching the divergence with
the upper spine field through the couch and gantry rotation.

References
1. Scott RL. An overview of craniospinal axis fields and field matching. Med
Dosim.2013;38(4):424-429.
2. Vonkadich AC. Overview of radiobiology. In: Washington CM, Leaver D, eds. Principles
and Practice of Radiation Therapy. 3rd ed. St. Louis, MO: Mosby-Elsevier; 2010: 57-85.
3. Discussion with Ed Bump, Medical Dosimetrist at the Virginia Commonwealth
University Health System Massey Cancer Center. June 15, 2015.
4. Bentel GC. Radiation Therapy Planning. 2nd ed. Columbia: McGraw-Hill; 1996.
5. Discussion with Ruth Ann Good, Medical Dosimetrist at the Virginia Commonwealth
University Health System Massey Cancer Center. June 17, 2015.
6. Discussion with Louise Francis, Medical Dosimetrist at the Virginia Commonwealth
University Health System Massey Cancer Center. June 16, 2015.