Stereotactic Radiosurgery

Selected Readings

AAPM TG-42: Stereotactic Radiosurgery (external link)


Stereotactic Radiosurgery, commonly referred to as SRS, delivers a tumoricidal dose to small targets in a small number of fractions. SRS typically refers to a single fraction cranial radiosurgery and may be distinguished from FSRT (fractionated stereotactic radiosurgery) which uses up to 5 fractions for cranial treatments.

SRS was first described in 1951 by a Swedish Neurosurgeon (Lars Leksell) using a stereotactic frame and a 0.2MeV x-ray therapy machine.

Unlike traditional fractionated radiotherapy, which relies partly normal tissue sublethal damage repair to spare normal tissues, SRS relies on tight margins and rapid dose fall off to prevent normal tissue complications.

Key Point: Stereotaxis has historically referred to the use of a 3D coordinate system to localize a target. With modern imaging, however, use of an external localization system is often considered unnecessary.

SRS Treatment Planning

Patient Selection

Treatment of up to 5 lesions is typical although as many as 10 lesions have been treated with SRS.

Target size is often limited to <4-5cm as unacceptable normal tissue dose volumes become difficult to avoid beyond this size.

Other factors such as the shape (circular is preferred) and target distribution (increased inter-target distance is preferred) must be considered when selecting SRS candidates.

Prescription and Prescription Isodose Line

Prescription for tumors and metastasis are usually limited by nearby tissue tolerances, often associated with the volume of normal tissue receiving 12Gy. As such, smaller tumors will receive a higher prescription dose than larger tumors.

Prescription isodose line refers to the dose level of the prescription. For example, a prescription may be written: 21Gy to the 80% isodose line. In this example, the maximum dose would be 21/0.8 = 26.25Gy.

  • Prescription isodose line is chosen to achieve a given coverage level of the target (often >99%).
  • For most 6MV single isocenter C-arm linac delivered SRS plans, the prescription isodose line will be around 80%.
  • Small fields and plans using sphere packing (especially Gamma Knife) may have prescription isodose lines as low as 50%.

Key Point: Low prescription isodose lines are used to optimize normal tissue sparing by improving conformality. However, there is some evidence that the resulting hot spots increase intratarget radionecrosis which is an adverse effect.

Margin Size

Although a large number of studies have indicated the efficacy of a 0mm treatment margin (i.e. PTV = CTV) this decision should be made on a site-by-site and patient-by-patient basis.

Several factors influence choice of margin size including:

  • Expected target and organ motion during treatment
  • Mechanical factors influencing precision and accuracy
    • Isocenter positioning
    • Imaging
  • Expected dose gradient at field edge
  • Contouring uncertainty
    • Human variability
    • Voxel size, image registration, and resampling

General Planning Techniques

Because SRS techniques depend upon highly accurate and precise delivery of a conformal dose distribution, several planning strategies will be nearly universally applied.

Combined MRI/CT Imaging

Magnetic Resonance (MR) imaging is valued in treatment planning as it is able to produce superior soft tissue contrast to CT images. This is especially valuable in SRS planning were targets are of similar density to their surrounding tissue and have a small size.

CT imaging is used to generate the treatment plan because of its superior spatial linearity, high resolution, and direct correlation between CT number and tissue density.

A fusion of the MR and CT images provides the best of both worlds. After image registration, the fused MR of the fused image set is used to generate target contours. The CT, which is also associated with the contours, is then used to generate the treatment plan.

Non-coplanar beam arrangements

Non-coplanar beam arrangements refers to using multiple angles of beam incident which do not exist on the same plan. This technique improves the conformality of the dose distribution and reduces surface dose.

Large number of beam angles

The large number of beam angles additionally serves to improve conformality. For c-arm linear accelerators, arcs are likely to be used in which the beam is continuously on during gantry rotation at a given angle.

Rigid Immobilization

The highly conformal dose distribution presents significant potential for geometric miss if the target is not precisely localized. Historically, this has been achieved using a stereotactic head frame but real time surface tracking is reducing the need for this invasive approach. Today rigid masks are becoming more popular.

Optical Tracking

Optical tracking uses a camera system to precisely localize the patient in real time. Optical tracking may either track to location reflective markers placed on the patient or the directly track the surface of the patient. These systems are extremely valuable in both reducing risk of geometric miss and decreasing treatment time through reduced need for repeat x-ray imaging during setup.

Key Point: Prescription dose is often limited by normal tissue tolerance. As a result, smaller tumors will often have a higher prescription than larger tumors.

Immobilization: Frame Based vs Frameless SRS

Frame based SRS treatment. Image credit: Elekta
Frameless SRS mask, black arms hold reflectors for fiducial tracking. Image credit: Elekta

Frame Based

An SRS frame refers to an immobilization and registration device physically attached, usually by pointed screws, to the patient's cranium.

Historically, frames included a radio-opaque fiducial marking system which allowed orthogonal port films to verify precise 3-dimentional alignment. This function has been largely superseded by the proliferation of high quality on-board cone-beam imaging.

Frame based SRS may be either couch mounted" or "floor stand mounted."

Note that although the screws will penetrate the patient's skin, they do no significantly penetrate the cranium and are generally well tolerated by patients.


  • Highest degree of immobilization
  • Benefits from but does not require intra-treatment position monitoring


  • Requires neurosurgeon to place frame
  • Requires same day placement, simulation, planning, quality assurance, and delivery


Frameless SRS uses an external mask in conjunction with intra-fraction position monitoring to achieve precision.

Initial target alignment is achieved through cone-beam CT (CBCT) imaging. Primary alignment typically focuses on fixed bony anatomy with fine-tuning based on nearby soft tissue.

Intra-fraction position often achieved using an optical camera to monitor patient surface or externally located fiducial markers.

Gating or real time tracking may be used to assure geometric accuracy of treatment delivery.

Gating shuts off the treatment beam when the target is found to be outside of its designated region.

Tracking causes the beam to follow the target as it moves.


  • Allows simulation, planning, and delivery to take place on different days
  • Less invasive to the patient


  • Inherently less immobilizing than frame based deliveries
  • Real-time positioning system requires additional physics quality assurance

Planning Techniques

Fixed Collimator Sphere Packing

Fixed circular collimation, as achieved in Gamma Knife and cone base deliveries, produces a roughly spherical dose distribution. Because many tumors are not spherically shaped, a technique known as sphere packing is used.

Sphere packing fills the target volume with multiple spherical dose distributions of varying sizes to achieve a conformal net dose distribution.

The sphere packing technique is usually achieved by first filling in the largest area with a large dose sphere and proceeding to fill in the remaining area with smaller and smaller spheres.


Cone based plans achieve the sharpest dose fall off and therefore  have the greatest geometric window.


Large irregular volumes often have significant dose heterogeneity within the target.

Complex targets require multiple cone sizes and isocenters. This greatly increased treatment time.

MLC Based Conformal Arc

Conformal arc planning uses an MLC which changes shape to conform to the shape of a target during a delivery arc. Because the MLC can produce larger and more irregular fields than a cone, MLC based plans typically use less isocenters and thereby require less treatment time.


Single isocenter per target

For multiple closely packed targets, a single isocenter may be sufficient.

Simplifies treatment planning

Shortens delivery

Often achieves lower hot spots for irregular targets


Larger penumbra than cone because final collimation is farther from target.

Choosing an MLC or cone: Choice of cone or MLC based planning involves several trade-offs. Cones are most commonly used for small, spherical targets or those which require the sharpest possible dose fall-off such as Trigeminal Neuralgia. Dynamic MLCs are favorable for their ability to treat large or irregular targets and to do so significantly faster than cones.

Treatment Sites

Intracranial Tumors

Primary Cancers/Surgical Cavity

Background: SRS has been used in the management of most intracranial tumors. However, malignant tumors may be poorly suited for SRS because of their high alpha/beta ratios and probability of proliferation. The most common cancers for SRS treatment are benign meningiomas and malignant gliomas.

Common Prescriptions:

Benign Meningiomas: 12-20Gy

Target: MRI enhancing boarder of lesion plus optional PTV margin of about 1mm.

Outcomes: Outcomes vary significantly with disease.

Benign meningiomas experience excellent control (>97% at 5 years) with SRS.

Glioblastoma Multiforme, the most common glioma, experience generally low rates of control.


Background: SRS for brain metastases is an alternative to surgical removal or whole brain irradiation. These metastasis are often from a lung primary tumor.

Common Prescriptions: 12-21Gy

Target: PTV = CTV + (0 to 2mm margin)

Outcomes: Control of metastatic brain tumors, especially those less than 2cm in diameter, is good and comparable to surgical removal. Clinical trials indicate SRS with 3-4 lesions may improve survival and reduce normal tissue toxicity (i.e. memory loss) as compared with whole brain irradiation. Treatment of 5 or more metastases using SRS does not appear to convey a survival advantage over whole brain irradiation but may reduce cognitive losses and allows for future salvage therapy.

Alternative Treatments:

Surgical Resection - May be appropriate for a small number of superficial metastases.
Whole Brain Irradiation - May be appropriate for large lesions or a large number of metastases (>4).

Functional Disorders

Trigeminal Neuralgia

Background: The first SRS treatments administered were for sufferers of trigeminal neuralgia. Trigeminal neuralgia is a disorder characterize by severe, stabbing, facial pain. The disorder is, in most cases, thought to be caused by compression of the trigeminal nerve root. SRS is a second line treatment used for cases resistant to drug treatments.

Common Prescriptions: 50-90Gy; 85Gy (most common)

Target: SRS for trigeminal neuralgia targets a small segment of the trigeminal nerve between the brainstem and Meckel's cave.

A 4mm cone size is often selected.

Outcomes: Pain relief is achieved in approximately 90% of patients with the mean time to relief of about 24 days. The most common toxicity is facial numbness.

Trigeminal nerve innervation. Credit: Bruce Blaus via Wikimedia Commons.

Arteriovenous Malformation

Background: Atriovenous Malformation (AVM) is an abnormal tangle of blood vessels connecting veins and arteries in the brain. The exact cause of AVM formation is not clear but most AVMs are congenital (present at birth) and are not generally hereditary.

Atriovenous malformation may present as headaches, seizures, or other neurological symptoms associated with hemorrhage. Most AVM SRS treatments are used to prevent repeat bleeding after an initial hemorrhage.

Diagnosis: AVM is diagnosed using CT, MRI, or Cerebral Arteriography.

Cerebral Arteriography (Angiography) involved inserting a catheter into the femoral artery and guiding it to the brain using fluorography.  Once in the suspected region, contrast is injected allowing for detailed images of the local vessels.

Common Prescriptions: 16-25Gy

Outcomes: Obliteration of the AVM takes place over 1-3 years following SRS with an ultimate success rate of 54-92%. Potential tissue toxicities include seizure, headache, neurological deficit, necrosis. Reducing the 12Gy volume outside of target can reduce toxicity.

Alternative Treatment Options:

Surgical Resection - Common in superficial and other easily accessed AVM locations.
Enovascular embolization - A glue-like substance is injected into the AVM preventing blood flow.

SRS Specific Treatment Machines

Today SRS is commonly delivered using a standard or modified C-arm linear accelerator. Several specialized machines exist for this purpose however, each having its own advantages.

Gamma Knife

The Gamma knife was first created by the originator of SRS, Neurosurgeon Lars Leksell, and a physicist, Borje Larsson in the late 1960s. Today Gamma Knife is owned by Elekta and several similar products are made by other manufacturers (GammaART by Cancer Care International, GyroKnife by GammaStar).

Key Features

192 hemispherically arranged Cobalt-60 sources

Collimation achieved using small fixed apertures producing dose spheres of 4-18mm.

Targeting Precision < 0.5mm

Elekta Gamma Knife Icon. Image credit: Elekta
Illustration of Gamma Knife Icon Co-60 collimator. Image credit: Elekta

Cyber Knife

The CyberKnife system is a linac based SRS delivery system mounted on a robotic arm.

Key Features

X-band linac mounted on robotic arm

Collimation provided by fixed cone or a variable MLC-type collimator.

Targeting Precision 1mm

Stereotactic Body Radiation Therapy

Selected Readings

AAPM TG-101: Stereotactic Body Radiation Therapy (external link)


Stereotactic Body Radiation Therapy (SBRT), refers to the use of highly conformal low fraction (5 or less) therapy techniques when used outside of the cranium. The term SBRT is increasingly being replaced by the term SAbR (Stereotactic Ablative Radiotherapy) both because the procedure does ablatively destroy tumor tissue and because the pronunciation of the acronym (Pronounced: Sabre) is faster to say, and sounds cooler, than "S-B-R-T."

Treatment Sites


Background: Breast cancer is not commonly treated with SAbR/SBRT as a primary treatment but may be used as a boost. Very little data is available on this topic.

Common Prescriptions: 20-30Gy delivered in 3 to 5 fractions


Background: SAbR/SBRT is not typically used as a primary therapy for gynecological cancers but may be used for slavage/re-irradiation.

Common Prescriptions: 15Gy delivered in 3 fractions

Outcomes: Population size for salvage radiation studies are small but local control appears to be good (~80%). Normal tissue toxicity may be a concern.

Head and Neck

Background: Head and neck cancers are not commonly treated with SAbR/SBRT but data does exist on its use in salvage/re-irradiation.

Common Prescriptions: 18-40Gy delivered in 3-5 fractions

Outcomes: Local control for these salvage cases is generally strong (70-80%).


Background: Both primary and metastases may be treated with SAbR/SBRT in patients who are unfit for surgery (e.g. tumor near major vessels).

Common Prescriptions: Up to 60Gy delivered in 3 to 6 fractions


Background: Lung cancers, both early stage primary and metastasis, are among the most common locations for SAbR/SBRT. Lung SAbR/SBRT may be preferred over surgery for metastatic disease that is centrally located, includes lesions in both lungs, or have previously undergone lobectomy. Centrally located tumors, those within 2cm of the trachea, primary bronchus, or esophagus, require special care as the location induces significant risk of normal tissue toxicity.

Common Prescriptions:

Primary: 45-60Gy in 3-5 fractions

Metastasis: Many schedules are used, 50-55Gy delivered in 5 fractions is common


Primary cancers experience good to excellent local control (up to 90%) at one year.

Metastasis experience good to excellent levels of local control (up to 90%) at three years.


Common Prescriptions: Multiple with 30-40Gy delivered in 3-4 fractions being common.

Outcomes: Excellent local control (98%) is achieved.


Background: SAbR/SBRT provides excellent palliative pain reduction for spinal metastasis.

Common Prescriptions:

16-20Gy in single fraction

30Gy delivered in 3-5 fractions

The Radiobiology of Radiosurgery

Validity of the linear quadratic model (LQM) is generally supported in the dose ranges of 1 to 5Gy. This makes LQM applicable to most  standard fractionation and hypofractionation therapies. Beyond this range, however, debate exists as to the validity of the model.

Non-LQM Radiobiological Mechanisms

Abscopal Effect

The abscopal effect is a phenomenon in which local radiotherapy is associated with the regression of metastatic cancer in unirradiated locations. Abscopal effect is more common in radiosurgery treatments than in conventional fractionation, possibly because of improved immune mediated response and proximal vascular damage.

Vascular Damage

Treatments over 10Gy/fraction result in significant vascular damage. According to the LQ model, this should induce radio-resistance via hypoxia and acidification. However, the larger fraction dose fractions used in radiatiosurgery have been shown to continue cell killing for as long as 3 days after irradiation due to vascular damage.

Immune-Mediated Responses

In contrast to standard fractionation, which induces an immunosuppressive effect, radiosurgery appears to induce local release of tumor specific antigens. These antigens, combined with a pro-inflamitory and pro-axidant cytokines, improve priming of tumor specific T cells. The net effect is increase immune response to combat the cancer.

Physical Parameters Impacting Biological Response

Geometric Window of Opportunity

Geometric Window of Opportunity simply refers to the ability of highly conformal dose distributions achieved in SRS and SAbR/SBRT to avoid critical structure all together. This capacity allows for significant dose escalation improving tumor control probability. Thanks to improvements in real time localization techniques, the geometric window of opportunity is expanding.

Treatment Duration

Sublethal repair begins, to some extent, immediately and the repair times for tumors are generally shorter than those for normal tissue. Therefore, reducing treatment duration in the SRS/SBRT regime is desirable from a tumor killing prospective.

Treatment duration may be subdivided into two categories:

Intra-fraction time is the time it takes for a given delivery. Increased delivery time increases the likelihood of patient/organ motion which can reduce the geometric window of opportunity. Additionally, treatment times exceeding 30 minutes may lead to significant reduction of tumor cell sterilization.

Inter-fraction time is the time between treatments in a given treatment regimen. Increased inter-fraction time allows for proliferation and repopulation of tumor cells, reducing TCP. However, allowing 24 hours or more inter-fraction time allows for re-oxygenation of the tumor which increases its radio-sensitivity and increasing TCP.

Dose Heterogeneity

SRS/SBRT plans typically have significant dose heterogeneity in order to optimize dose fall-off outside the target. This is typically acceptable and may even be desirable as, up to a point, increased dose increases TCP.

Dose heterogeneity also increases the consequences of geometric misalignment which can result in severe overdose of adjacent normal structures or under dose to the target.

Small Field Dosimetry

What is a Small Field?

Small fields are those meeting the following conditions:

1. Lack of lateral charged particle equilibrium, even along the central axis.

When the range of secondary electrons is approximately equal to or greater than the distance from field edge to central axis, lateral charged  particle equilibrium cannot exist.

The below equation is given in TRS-483 for determining the lateral charged particle equilibrium range (rLCPE). Field sizes greater than twice rLCPE will have lateral electron equilibrium at their central axis.

For a 6MV field, rLCPE  is about 1.2cm.

Simplification: A similar result to the above equation can be obtained by recalling that the mean energy of a linac photon beam is about 1/3 to 1/2 of the maximum energy and that an electron range in water is about 0.5cm/MeV. We can then crudely estimate that a 6MV photon would have a mean energy of between 2MV and 4MV. We can then say that  rLCPE is between about 1cm and 2cm.

2. Partial occlusion of the primary photon source.

Partial source occlusion results from a finite source size and causes the machine output to become highly dependent upon collimator position.

3. Detector collecting volume that is similar to, or larger than, the field cross sectional area.

AAPM TG-106  (external link) recommends that the field output be uniform to within 1% over the detectors sensitive volume. This becomes difficult to achieve for very small fields.

Conventional simplification: although not technically a correct definition, a rule of thumb is that fields become "small" when they are less than 3 x 3 cm2.

Key Point: Small fields are defined both by the actual field size and by the size of the detector used to measure the field!

Defining the Size of a Small Field

For large fields, the field size is defined by the 50% isodose line at a given distance from the source. By this method the field size setting is equal to the full width at half maximum (FWHM) of the dose profile at the reference depth. This size is reported on the operator station and approximately coincides with the light field.

For small fields this relationship between field size setting and FWHM does not hold. Because of partial source occlusion, FWHM becomes much greater than the field size setting for small fields. This effect is known as penumbra broadening.

TRS-483 uses the radiation field at FWHM as its definition of field size.

Illustration of penumbra broadening of small fields.
Illustration of the impact of detector volume on profile edge appearance in small field dosimetry.

The Important of Detector Size

Detectors used in small field dosimetry profile acquisition much have a very small collecting volume to prevent  averaging to obscuring the shape of the dose distribution. AAPM TG-106 recommend that the field output be uniform over the collecting volume to +/-1%.

Field Size Uncertainty

For very small fields, where partial source occlusion becomes the dominant factor in penumbra, uncertainty in field size has an exaggerated effect on the shape of the dose distribution.

Uncertainty in the collimation position closest to the source will tend to be dominant.

This is because the positioning error of an object closer to the source is magnified at treatment distance as in the below equation.

It is for this reason that small treatment fields are defined either by a cone or MLC while the primary collimator, which is closest to the source, is retracted slightly from the field edge.

Key Point: Small MLC shaped fields that are defined on the system as square  are often rectangular when measured as FWHM. This effect is caused by leaf design in which the leaf tips have a difference shape than their tongue-and-groove side. As a result, a field defined by the light field as 5x5cm2 may in fact be a rectangular field with an equivalent square field size other than 5x5cm2.

IAEA TRS-483 Methodology

TRS-483 (external link) is a joint effort between the American Association of Physicists in Medicine (AAPM) and the International Atomic Energy Agency (IAEA). Its purpose is to establish a reliable method of determining the output factor for small fields.

Output factor is defined as the ratio of the absorbed dose to water at a point in a non-reference field to absorbed to to water at that point in a reference field.


Field Size: Field size is defined by the full width at half maximum of the radiation field. That is; field size is defined by the radiation field not the jaw setting readout or light field size.

Machine-Specific Reference Field (MSR): An MSR is a field size used as a reference in place of the traditional 10x10cm2 field size used in TG-51 and other dosimetry protocols. MSR is the used for machines not able to create a 10x10cm2 field such as Tomotherapy, Gamma Knife, and CyberKnife. MSR is usually the largest field a machine is able to make.

Hypothetical Reference Field (\( \Omega^{f_{clin}, f_{msr}}_{Q_{clin}, Q_{msr}} \)): This is the hypothetical field that would result if a machine using MSR could produce a 10x10cm2 field.

Field Output Correction Factor (\( k^{f_{clin}, f_{msr}}_{Q_{clin}, Q_{msr}} \)): A correction factor accounting for the difference in detector response for a clinical (non-reference) and msr field.

Volume Averaging Correction Factor (kvol): The ratio of the absorbed dose to water at the reference point in water without the detector  present to the absorbed dose to water over the sensitive volume of the detector, still without the detector present.

Beam Quality Specification in MSR Field (Qmsr): This term represents the energy spectrum of the MSR field. It is either specified as the photon percent depth dose at 10cm depth of the msr field (%dd(10,S)x) or as the ratio of tissue phantom ratios at 20cm to 10cm depth for the msr field (TPR20,10(S)).

Equivalent Square MSR Field Size (S): The size of a square field in which the same amount of phantom scatter is generated as in the non-square field.


Key Point: If a field is smaller than 4cm, it cannot be considered an MSR field for 6MV or higher beams! Source: TRS-483 pg.   e1135.

Quantity IAEA Symbol AAPM Symbol
Beam quality specifier for msr and conventional 10x10cm2 fields TPR20,10(10) %dd(10,10)x
Auxiliary specifier for the equivalent square field size S TPR20,10(S) %dd(10,S)x
Absorbed dose to water calibration coefficient in the machine-specific reference field \( N^{f_{msr}}_{D,w,Q_{msr}} \) \(N^{Q_{msr}}_{D,w}\)
Absorbed dose to water calibration coefficient in a conventional reference field \(N^{f_{ref}}_{D,w,Q_{0}}\)  \(N^{Q_{0}}_{D,w}\)
Corrected chamber reading in machine specific reference field \(M^{f_{msr}}_{Q_{msr}}\) \(M_{Q_{msr}}\)
Beam quality correction factor for the conventional reference field \(k^{f_{ref}}_{Q,Q_{0}}\) \(k_{Q}\)
Beam quality correction factor for the machine specific reference field \(k^{f_{msr},f_{ref}}_{Q_{msr},Q_{0}}\) \(k_{Q_{msr}}\)
Output correction factor \(k^{f_{clin},f_{msr}}_{Q_{clin},Q_{msr}}\) \(k_{Q_{clin},Q_{msr}}\)
Field output correction factor \(\Omega^{f_{clin},f_{msr}}_{Q_{clin},Q_{msr}}\) \(\Omega_{Q_{clin},Q_{msr}}\)

Reference Dosimetry

The goal of reference dosimetry is to establish the output, in absorbed dose per monitor unit (cGy/MU) of the reference field.

The reference field is 10 x 10 cm2 for machines able to produce such a field. For machines unable to produce a 10 x 10 cm2 field, a machine specific reference field (MSR) is chosen most frequently as the largest field size the machine can produce.

TRS-483 uses nearly the same methodology for reference dosimetry as AAPM TG-51 or TRS-398 except that it uses the MSR field as the reference field.

Reference Chamber Calibration

The reference chamber must be calibrated for the MSR field. Three approached to reference chamber calibration are acceptable:

1. Chamber directly calibrated for MSR field by the accredited calibration laboratory.

Although this method would be the simplest, direct calibration factors for machine reference fields are generally not available except for Co-60 machines.

2. Chamber calibrated for standard reference field but used with generic beam quality correction factor.

3. If neither direct calibration nor generic beam quality data exists, a conventional kQ value for a 10 x 10 cm2 field may be combined with a factor correcting from the standard reference field to the machine-specific reference field.

Reference Class Dosimeters

A reference dosimeter for an MSR field must meet the requirements of AAPM TG-51 Addendum or TRS-483. The dosimeter should have the following features:

  • Be an ionization chamber
    • Ionization chambers are well understood devices which take a near primary measurement of absorbed dose by collecting ionization products of air.
  • Approximately water equivalent
    • This minimizes energy dependence.
  • Air communicating
    • This allows the chamber to rapidly reach equilibrium and for the mass of air inside to be quantified with an atmospheric pressure reading.
  • Water-proof

Specifications for reference class ionization chambers

Parameter Specification
Chamber Settling
  • Reaches equilibrium within 5 minutes
  • Initial reading within +/-0.5% of equilibrium reading
  • <0.1% of reading
Polarity Effect
  • <0.4% of reading
  • Energy dependence <0.3% between Co-60 and 10MV
Recombination Correction
  • Linear with dose per pulse
  • <0.2% with applied potential of 300V
  • Plot of 1/M vs 1/V is linear for pulsed beams
  • Plot of 1/M vs 1/V2 is linear for continuous beams
  • Initial recombination withing +/-0.1% for measurements of opposite polarities
Chamber Stability
  • <0.3% calibration coefficient change over 2 year calibration period
Chamber Material
  • Wall material does not exhibit temperature or humidity effects

Source: Table II TRS-483. Adapted from AAPM TG-51 Addendum.

Relative Dosimetry of Small Fields

Relative output factors for small fields down to about 1cm to 2cm may be measured with small ionization chambers meeting the above specifications of a reference class dosimeter. For smaller fields, even the smallest ionization chambers are found to exhibit large perturbations. These perturbations are largely caused by volume averaging.

Small Field Dosimeters

Common Small field Dosimeters

  • Scanning diodes (most common)
  • Diamond detectors
  • Plastic scintillation detectors
  • Micro-ionization chambers with low-Z electrode (for field sizes above 1-2cm)

Detector size

TRS-483 recommends that the volume averaging correction be less than 5%.

Additionally, AAPM TG-106  (external link) recommends that the field output be uniform to within 1% over the detectors sensitive volume.

Energy Spectrum Response

One less intuitive area of concern for choice of detector is the impact that the small field has on the energy spectrum incident upon the detector. There are two options available to correctly measure dose in a field of non-standard energy spectrum.

1. Determine the effect of the changing energy spectrum on the detector and correct for it.

Although there is some specific data available for this, determining a complete correction factor may require Monte Carlo simulation.

2. Use a detector that minimizes the impact of changes to energy spectrum.


Determining Output Factors

An output factor is the ratio of absorbed dose under given conditions for a small field relative to absorbed dose measured under the same conditions for a reference field.

If a detector is chosen which gives stable response for both the MSR and small field, it may be used alone to determine output factors as:

For small fields, a suitable detector may not be available. In this case, a technique known as daisy chain is used.

In a daisy chain measurement, a small field dosimeter is cross calibrated with an ion chamber at an intermediate field.

The intermediate field size should be the smallest field that the ioinization chamber can accurately measure.

A 4x4cm2 field is commonly chosen as the intermediate field.

Another, way to denote the process of daisy chaining (which is not used in TRS-483 and lacks appropriate correction factors) is as:

Finding small field correction factors

MSR to clinical field correction factors (\(k_{Q_{clin},Q_{msr}}\)) for many chambers have been tabulated and are available in TRS-483 and elsewhere.

Most such correction factors have been determined by Monte-Carlo simulation.

For non-square fields, as defined by full-width-at-half-maximum (FWHM), equivalent square field size must be calculated to find the appropriate correction factor.

MSR to intermediate field correction factors (\(k_{Q_{int},Q_{msr}}\)) may not be available. Provided that the intermediate field is not a small field, \(k_{Q_{int},Q_{msr}}\) = 1 may be used. In this way only \(k_{Q_{clin},Q_{int}}\) is required.

Output decreases rapidly for field sizes below 15mm for a 6MV beam.

Knowledge Test

1. Which of the following is not a factor contributing to the difficulty of small field dosimetry?

Question 1 of 3

2. Select the false statement regarding SRS planning.

Question 2 of 3

3. Select all true statements regarding SBRT.

Question 3 of 3


This is the sample version of the full quiz. Log in or register to gain access to the full quiz.

Not a Premium Member?

Sign up today to get access to hundreds of ABR style practice questions.