PRELIMINARY INVESTIGATION OF RADIATION DOSE TO PATIENTS FROM CARDIOVASCULAR INTERVENTIONAL PROCEDURES IN TANZANIA.

Although contemporary cardiac X-ray exams are typically set so benefits outweighs the risk, the growing use and increasing complexity of the cardiovascular interventional radiological (CVIR) procedures does increase the risk of radiation-related tissue effects and stochastic effects to the individual patients and the population. In view of these radiological concerns there is a need to investigate factors that influence the doses received by the patients and enable optimisation needed. The air kerma area product (KAP), cumulative air kerma (CAK) and fluoroscopy time (FT) to patients from two major CVIR procedures: coronary angiography (CA) and percutaneous coronary interventions (PCI), were obtained from two major hospitals in Tanzania. The CAK and KAP were determined using ionisation chambers equipped in each angiographic unit. The median values of the KAP, CAK and FT for the CA procedures were 37.8 Gy cm2, 425.5 mGy and 7.6 min, respectively, while for the PCI were 86.5 Gy cm2, 1180.3 mGy and 19.0 min, respectively. The overall differences among individual KAP, CAK and FT values across the two hospitals investigated differed by factors of up to 33.5, 58.7 and 26.3 for the CA, while for the PCI procedures differed by factors of up to 10.9, 25.3 and 13.8, respectively. The mean values of KAP and FT for both CA and PCI were mostly higher than those reported values for Ireland, Belgium, Greece, France, China and Australia. The third quartiles of the KAP, CAK and FT for both CA and PCI were relatively above the preliminary diagnostic reference levels proposed by the IAEA, DIMOND III and SENTINEL. The observed substantial variations of mean values of technical parameters and patient doses (KAP, CAK and FT values) observed for the CA and PCI procedures inter and intra-hospitals were mainly explained by the complexity of the CVIR procedures, the nature of pathology, patient-specific characteristics, the variation in levels of skills and experiences among IC personnel, and the different procedural protocols employed among interventional cardiologists and hospitals. The observed great variations of procedural protocols and patient doses within and across the hospitals and relative higher dose than reported values from the literature call for the need to optimise radiation dose to patient from IC procedures.

Streamlined open-source gel dosimetry analysis in 3D slicer.

Three dimensional dosimetry is being used in an increasingly wide variety of clinical applications as more gel and radiochromic plastic dosimeters become available. However, accessible 3D dosimetry analysis tools have not kept pace. 3D dosimetry data analysis is time consuming and laborious, creating a barrier to entry for busy clinical environments. To help in the adoption of 3D dosimetry, we have produced a streamlined, open-source dosimetry analysis system by developing a custom extension in 3D Slicer, called the Gel Dosimetry Analysis slicelet, which enables rapid and accurate data analysis. To assist those interested in adopting 3D dosimetry in their clinic or those unfamiliar with what is involved in a 3D dosimeter experiment, we first present the workflow of a typical gel dosimetry experiment. This is followed by the results of experiments used to validate, step-wise, each component of our software. Overall, our software has made a full 3D gel dosimeter analysis roughly 20 times faster than previous analysis systems.

Best Practice Recommendations for the Retention of Radiotherapy Records.

This paper offers best practice recommendations for the maintenance and retention of radiotherapy health records and technical information for cancer programmes. The recommendations are based on a review of the published and grey literature, feedback from key informants from seven countries and expert consensus. Ideally, complete health records should be retained for 5 years beyond the patient's lifetime, regardless of where they are created and maintained. Technical information constituting the radiotherapy plan should also be retained beyond the patient's lifetime for 5 years, including the primary images, contours of delineated targets and critical organs, dose distributions and other radiotherapy plan objects. There have been increased data storage and access requirements to support modern image-guided radiotherapy. Therefore, the proposed recommendations represent an ideal state of radiotherapy record retention to facilitate ongoing safe and effective care for patients as well as meaningful and informed retrospective research and policy development.

Initial Investigation of Factors Influencing Radiation Dose to Patients Undergoing Barium-Based Fluoroscopy Procedures in Tanzania.

The aim of this study was to investigate the nature and causes of radiation dose imparted to patients undergoing barium-based X-ray fluoroscopy procedures in Tanzania and to compare these doses to those reported in the literature from other regions worldwide. The air kerma area product (KAP) to patient undergoing barium investigations of gastrointestinal tract system was obtained from four consultant hospitals. The KAP was determined using a flat transparent transmission ionization chamber. Mean values of KAP for barium swallow (BS), barium meal (BM) and barium enema (BE) were 2.79, 2.62 and 15.04 Gy cm2, respectively. The mean values of KAP per hospital for the BS, BM and BE procedures varied by factors of up to 7.3, 1.6 and 2.0, respectively. The overall difference between individual patient doses across the four consultant hospitals investigated differed by factors of up to 53, 29.5 and 12 for the BS, BM and BE procedures, respectively. The majority of the mean values of KAP was lower than the reported values for Ghana, Greece, Spain and the UK, while slightly higher than those reported for India. The observed wide variation of KAP values for the same fluoroscopy procedure within and among the hospitals was largely attributed to the dynamic nature of the procedures, the patient characteristics, the skills and experience of personnel, and the different examination protocols employed among hospitals. The observed great variations of procedural protocols and patient doses within and across the hospitals call for the need to standardize examination protocols and optimize barium-based fluoroscopy procedures.

Leuco-crystal-violet micelle gel dosimeters: II. Recipe optimization and testing.

In this study, recipe optimization of Leuco Crystal Violet (LCV) micelle gels made with the surfactant Cetyl Trimethyl Ammonium Bromide (CTAB) and the chemical sensitizer 2,2,2-trichloroethanol (TCE) was aided by a two-level three-factor designed experiment. The optimized recipe contains 0.75 mM LCV, 17.0 mM CTAB, 120 mM TCE, 25.0 mM tri-chloro acetic acid (TCAA), 4 wt% gelatin and ~96 wt% water. Dose sensitivity of the optimized gel is 1.5 times higher than that of Jordan's standard LCV micelle gel. Spatial integrity of the 3D dose distribution information in 1L phantoms filled with this recipe is maintained for >120 d. Unfortunately, phantoms made using the optimized recipe showed dose-rate dependence (14% difference in optical attenuation at the peak dose using electron beam irradiations at 100 and 400 MU min(-1)). Further testing suggests that the surfactant CTAB is the cause of this dose rate behaviour.

Leuco-crystal-violet micelle gel dosimeters: I. Influence of recipe components and potential sensitizers.

Radiochromic leuco crystal violet (LCV) micelle gel dosimeters are promising three-dimensional radiation dosimeters because of their spatial stability and suitability for optical readout. The effects of surfactant type and surfactant concentration on dose sensitivity of LCV micelle gels are tested, demonstrating that dose sensitivity and initial colour of the gel increases with increasing Triton x-100 (Tx100) concentration. Using Cetyl Trimethyl Ammonium Bromide (CTAB) in place of Tx100 produces gels that are nearly colourless prior to irradiation, but reduces the dose sensitivity. The separate effects of Tri-chloro acetic acid concentration and pH are investigated, revealing that controlling the pH near 3.6 is crucial for achieving high dose sensitivity. The sensitizing effect of chlorinated species on dose sensitivity is tested using 2,2,2-trichloroethanol (TCE), chloroform, and 1,1,1-trichloro-2-methyl-2-propanol hemihydrate. TCE gives the largest improvement in dose sensitivity and is recommended for use in micelle gel dosimeters because it is less volatile and safer to use than chloroform. Preliminary experiments on a new gel containing CTAB as the surfactant and TCE show that this new gel gives a dose sensitivity that is 24% higher than that of previous LCV micelle gels and is nearly colourless prior to irradiation.

Evaluation of the potential for diacetylenes as reporter molecules in 3D micelle gel dosimetry.

Radiochromic micelle gel dosimeters are promising for three-dimensional (3D) radiation dosimetry because they can be read out by optical CT techniques and they have superior spatial stability compared to polymer and Fricke gel dosimeters. This study evaluates the use of diacetylenes as reporter molecules in micelle gel dosimeters. Several gels containing pentacosa-10,12-diynoic acid (PCDA) emulsified using sodium dodecyl sulfate (SDS) changed from colourless to blue upon irradiation. Unfortunately, all phantoms that experienced a colour change were turbid and would be unsuitable for 3D dosimetry. Two techniques (use of organic solvent and aqueous-phase additives) were successful in increasing colloidal stability to prevent the turbidity problem, but none of the resulting transparent gels changed colour in response to radiation. Transparent PCDA emulsions were prepared using NaOH solutions with no SDS or other emulsifier, but these transparent emulsions also did not change colour. Only turbid gels and emulsions with precipitated particles responded to radiation. These results indicate that the colour change was due to the oligomerization within precipitated PCDA crystals, and that liquid-phase emulsified PCDA did not undergo oligomerization. As a result, PCDA is not suitable for use in micelle gel dosimeters, and other radiochromic reporter molecules will need to be identified.

Use of a dual-labelled oligonucleotide as a DNA dosemeter for radiological exposure detection.

A reporter molecule consisting of a synthetic oligonucleotide is being characterised for a novel damage detection scenario for its potential use as a field-deployable, personal deoxyribonucleic acid (DNA) dosemeter for radiation detection. This dosemeter is devoid of any biological properties other than being naked DNA and therefore has no DNA repair capabilities. It supports biodosimetry techniques, which require lengthy analysis of cells from irradiated individuals, and improves upon inorganic dosimetry, thereby providing for a more relevant means of measuring the accumulated dose from a potentially mixed-radiation field. Radiation-induced single strand breaks (SSBs) within the DNA result in a quantifiable fluorescent signal. Proof of concept has been achieved over 250 mGy-10 Gy dose range in radiation fields from 6°Co, with similar results seen using a linear accelerator X-ray source. Further refinements to both the molecule and the exposure/detection platform are expected to lead to enhanced levels of detection for mixed-field radiological events.

Immobilization and catheter guidance for breast brachytherapy.

PURPOSE: Brachytherapy is an important mode of breast cancer treatment; however, improvements in both treatment planning and delivery are needed. In order to meet these specific needs, integration of pre-operative imaging, supplemented by computerized surgical planning and mathematical optimization were used to develop and test an intra-operative immobilization and catheter guidance system. METHOD: A custom template specific to each patient with optimally placed guide holes for catheter insertion was designed and fabricated. Creation of the template is based on a virtual reality reconstruction of the patient's anatomy from computed tomography imaging. The template fits on the patient's breast, immobilizing the soft tissue, and provides pre-planned catheter insertion holes for guidance to the tumor site. Agar-based phantom and target models were used for quantitative validation of the template by ascertaining the precision and accuracy of the templates. RESULTS: Tests were performed on agar-based tissue models using computed tomography imaging for template planning and validation. Planned catheter tracks were compared to post-insertion image data and distance measurements from target location were used to create an error measure. Initial results yielded an average error of 4.5mm. Once the workflow and template design were improved, an average error of 2.6mm was observed, bringing the error close to a clinically acceptable range. CONCLUSION: Use of a patient-specific template for breast brachytherapy is feasible and may improve the procedure accuracy and outcome.

Poster - Thur Eve - 44: Digital tomosynthesis image quality in a Co-60 treatment beam.

Image guidance capability is an important feature of modern radiotherapy machines. Cobalt-60 units will require some form of image guidance if they are to be brought up to modern standards. Imaging in the treatment beam is an appealing option, for reasons of simplicity and cost, but the dose needed to produce cone beam CT images in a Co-60 treatment beam is too high for this modality to be clinically useful. Digital tomosynthesis (DT) offers a quasi-3D image, of sufficient quality to identify bony anatomy or fiducial markers, while delivering a much lower dose than CBCT. A series of experiments were conducted on a prototype Co-60 cone beam imaging system to quantify the resolution, selectivity, geometric accuracy and contrast sensitivity of Co-60 DT. Although the resolution was severely limited by the penumbra cast by the ∼2cm diameter source, it was possible to identify high contrast objects on the order of 1 mm in width, and bony anatomy in anthropomorphic phantoms was clearly recognizable. Low contrast sensitivity down to electron density differences of 3% was obtained, for uniform features of similar thickness. The conventional shift-and-add algorithm was compared to the FDK filtered backprojection algorithm using several different spatial filters. The Co-60 DT images were obtained with a total dose of 5 to 15 cGy. We conclude that, should Co-60 radiotherapy units be upgraded with image guidance capabilities, filtered backprojection DT in the treatment beam is a versatile and promising modality that would be well suited to the task of patient positioning.

Poster - Thur Eve - 18: Characterization of a camera and LED lightbox imaging system for radiochromic film dosimetry.

Radiation therapy treatment modalities continue to develop and have become increasingly complex. With this, dose verification and quality assurance (QA) is of great importance to ensure that a prescribed dose is accurately and precisely delivered to a patient. Radiochromic film dosimetry has been adopted as a convenient option for QA, because it is relatively energy independent, is near tissue equivalent, and has high spatial resolution. Unfortunately, it is not always easy to use. In this study, preliminary work towards developing a novel method of imaging radiochromic film is presented. The setup consists of a camera mounted vertically above a lightbox containing red LEDs, interfaced with computer image acquisition software. Imaging results from this system will be compared with imaging performed using an Epson Expression 10000XL scanner (a device in common clinical use). The lightbox imaging technique with camera readout is much faster relative to a flatbed scanner. The film measurements made using the camera are independent of film orientation, and show reduced artifacts, so that there are fewer corrections required compared to the use of flatbed scanners. Optical scatter also appears to be less of an issue with this design than with the flat bed scanner. While further work needs to be done to optimize the lightbox imaging system, the lightbox system shows great promise for a rapid, simple, and orientation independent setup, improving on existing film scanning systems.

Cosolvent-free polymer gel dosimeters with improved dose sensitivity and resolution for x-ray CT dose response.

This study reports new N-isopropylacrylamide (NIPAM) polymer gel recipes with increased dose sensitivity and improved dose resolution for x-ray CT readout. NIPAM can be used to increase the solubility of N, N'-methylenebisacrylamide (Bis) in aqueous solutions from approximately 3% to 5.5% by weight, enabling the manufacture of dosimeters containing up to 19.5%T, which is the total concentration of NIPAM and Bis by weight. Gelatin is shown to have a mild influence on dose sensitivity when gels are imaged using x-ray CT, and a stronger influence when gels are imaged optically. Phantoms that contain only 3% gelatin and 5 mM tetrakis hydroxymethyl phosphonium chloride are sufficiently stiff for dosimetry applications. The best cosolvent-free gel formulation has a dose sensitivity in the linear range (~0.88 H Gy(-1)) that is a small improvement compared to the best NIPAM-based gels that incorporate isopropanol as a cosolvent (~0.80 H Gy(-1)). This new gel formulation results in enhanced dose resolution (~0.052 Gy) for x-ray CT readout, making clinical applications of this imaging modality more feasible.

Polymer gel dosimetry.

Polymer gel dosimeters are fabricated from radiation sensitive chemicals which, upon irradiation, polymerize as a function of the absorbed radiation dose. These gel dosimeters, with the capacity to uniquely record the radiation dose distribution in three-dimensions (3D), have specific advantages when compared to one-dimensional dosimeters, such as ion chambers, and two-dimensional dosimeters, such as film. These advantages are particularly significant in dosimetry situations where steep dose gradients exist such as in intensity-modulated radiation therapy (IMRT) and stereotactic radiosurgery. Polymer gel dosimeters also have specific advantages for brachytherapy dosimetry. Potential dosimetry applications include those for low-energy x-rays, high-linear energy transfer (LET) and proton therapy, radionuclide and boron capture neutron therapy dosimetries. These 3D dosimeters are radiologically soft-tissue equivalent with properties that may be modified depending on the application. The 3D radiation dose distribution in polymer gel dosimeters may be imaged using magnetic resonance imaging (MRI), optical-computerized tomography (optical-CT), x-ray CT or ultrasound. The fundamental science underpinning polymer gel dosimetry is reviewed along with the various evaluation techniques. Clinical dosimetry applications of polymer gel dosimetry are also presented.

Preliminary investigation of the NMR, optical and x-ray CT dose-response of polymer gel dosimeters incorporating cosolvents to improve dose sensitivity.

This study reports on efforts to increase the dose sensitivity of polymer gel dosimeters used in 3D radiation dosimetry. The potential of several different cosolvents is investigated, with the aim of increasing the solubility of N,N'-methylene-bisacrylamide crosslinker in polymer gel dosimeters. Glycerol and isopropanol increase the limit for the crosslinker solubility from approximately 3% to 5% and 10% by weight, respectively. This enables the manufacture of polymer gel dosimeters with much higher levels of crosslinking than was previously possible. New dosimeter recipes containing up to 5 wt% N,N'-methylene-bisacrylamide were subjected to spatially uniform radiation and were studied using nuclear magnetic resonance (NMR), as well as x-ray and optical CT techniques. The resulting dosimeters exhibit dose sensitivities that are up to 2.7 times higher than measured for a typical dosimeters with 3% N,N'-methylene-bisacrylamide without the addition of cosolvent. Two additional cosolvents (n-propanol and sec-butanol) were deemed unsuitable for practical dosimeters due to incompatibility with gelatin, cloudiness prior to irradiation, and immiscibility with water when large quantities of cosolvent were used. The dosimeters with high N,N'-methylene-bisacrylamide content that used isopropanol or glycerol as cosolvents had high optical clarity prior to irradiation, but did not produce suitable optical CT results for non-uniformly irradiated gels due to polymer development outside of the high dose regions of the pencil beams and significant light scatter. Further experiments are required to determine whether cosolvents can be used to manufacture gels with sufficiently high dose sensitivity for readout using x-ray computed tomography.

Poster - Thurs Eve-15: Comparison of Cobalt-60 and 6 MV linac based tomotherapy: A prostate case study.

Previous work reported by us has shown the potential for Cobalt-60 (Co-60) tomotherapy for sites with small separations such as in head and neck site. In this work we extend our investigations by comparing tomotherapy plans for the treatment of a typical prostate cancer obtained for 6 MV and Co-60 beams. Beam collimation was provided by the MIMiC® (NOMOS Corporation, Sewickly, PA) multi-leaf collimator (MLC). Both plans used 21 beam angles, each utilizing the central 10 leaf-pairs of the MLC for intensity modulation. An in-house inverse treatment planning program, based on the active-set conjugate gradient method, was used for dose-volume optimization. BEAMnrc and DOSXYZnrc Monte Carlo simulated beam and dose data, including inhomogeneity corrections, were used to calculate the optimized tomotherapy dose distributions. Prostate, rectum, and external body contours were outlined and dose-volume optimization objectives were set to deliver a minimum of 95% and a maximum of 105% of the 76 Gy dose prescription to the prostate and limiting only 20% of the rectum volume to receive ⩾ 70 Gy. A quantitative analysis of the dose distributions and dose-area histograms show that both Co-60 and 6 MV plans achieve the initial objectives for target (prostate) and organ at risk (rectum). Although the dose to the body and rectum for Co-60 is slightly higher than that for 6 MV, it satisfies the plan objectives based on the clinical dose tolerance. Our results demonstrate that Co-60 based tomotherapy can provide clinically competitive dose distributions for the treatment of prostate cancer.

Sci-Fri AM: YIS-04: Scatter correction of cone beam optical computed tomography for polymer gel dosimetry.

Optical computed tomography (OptCT) may become the preferred scanning method for gel dosimetry dose validations, due to its high sensitivity and relatively low cost. Cone beam computed tomography (CBCT) arrangements are advantageous because of reduced scan time. However, CBCT arrangements are more sensitive to errors associated with scatter than other CT configurations. Unfortunately in polymer gel dosimetry this problem is amplified as the primary mode of beam attenuation is through scatter. Thus, managing and reducing the effects of scatter remains an important challenge for cone beam OptCT. In this work we examine two schemes for reducing the effects of scatter in the Vista cone beam OptCT system. First, we employed a pair of anti-scatter polarizing planes to reduce the magnitude of stray light reaching the camera. Secondly, we implemented a beam stop array (BSA) sampling method -which has been successful in correcting for scatter in X-ray CBCT- to obtain scatter field measurements that are subtracted from CT projections removing veiling glare. While both implementations reduced scatter related artifacts, the BSA technique yielded greater improvement without obvious image degradation. Comparative studies between absorbing dye standards and colloidal scattering standards with the same spectrophotometric optical attenuation revealed that application of the BSA technique nullified OptCT measurement disagreements between scattering and absorbing systems. When the BSA scatter correction technique was applied to polymer gel dosimetry 3%3mm agreement rose from 79.2% to 99.82%. These findings underscore the strength of the BSA sampling technique and its utility in cone beam OptCT.

Poster - Thurs Eve-42: A revision of the γ-evaluation: Initial interpretation of dose disagreements on γ-vector fields.

Gamma evaluations are a common clinical tool used as a quantitative comparison between dose-distributions, combining both dose difference and distance to agreement criteria. Because gamma evaluations permit rapid analysis of agreement between complex dose distributions, they are often a preferred comparison method for assessing delivery of conformal radiotherapy distributions. Although the comparison provides a useful measure of agreement between distributions when the index is less than one, the scalar gamma value provides little information into the clinical significance or source of disagreements of failing gamma values (i.e., when γ>1). Previously, Stock et al., have presented the gamma angle as an indicator of the relative influence of the distance to agreement versus the dose difference on gamma. We present a modification to the gamma evaluation such that the complete 3D gamma vector information is considered. The predictive nature of each vector component was investigated by simulating various dose disagreements in test distributions. Misalignment tests revealed that the mean gamma vector components indicate the offset direction and relative magnitude for all test distributions. The mean dose component of the gamma vector was prognostic of double Gaussian overdoses and underdoses in a virtual conformal delivery. The response of the vector field depends on properties distinctive to each distribution, such as the local dose gradient. Understanding how these unique properties affect the vector field may permit better diagnosis of dose disagreement sources. Other vector field properties, such as curl and divergence, may yet provide more information for interpreting the cause and significance of γ>1.

Polymer gel dosimeters with reduced toxicity: a preliminary investigation of the NMR and optical dose-response using different monomers.

In this work, three new polymer gel dosimeter recipes were investigated that may be more suitable for widespread applications than polyacrylamide gel dosimeters, since the extremely toxic acrylamide has been replaced with the less harmful monomers N-isopropylacrylamide (NIPAM), diacetone acrylamide and N-vinylformamide. The new gel dosimeters studied contained gelatin (5 wt%), monomer (3 wt%), N,N'-methylene-bis-acrylamide crosslinker (3 wt%) and tetrakis (hydroxymethyl) phosphonium chloride antioxidant (10 mM). The NMR response (R2) of the dosimeters was analysed for conditions of varying dose, dose rate, time post-irradiation, and temperature during irradiation and scanning. It was shown that the dose-response behaviour of the NIPAM/Bis gel dosimeter is comparable to that of normoxic polyacrylamide gel (PAGAT) in terms of high dose-sensitivity and low dependence on dose rate and irradiation temperature, within the ranges considered. The dose-response (R2) of NIPAM/Bis appears to be linear over a greater dose range than the PAGAT gel dosimeter. The effects of time post-irradiation (temporal instability) and temperature during NMR scanning on the R2 response were more significant for NIPAM/Bis dosimeters. Diacetone acrylamide and N-vinylformamide gel dosimeters possessed considerably lower dose-sensitivities. The optical dose-response, measured in terms of the attenuation coefficient for each polymer gel dosimeter, showed potential for the use of optical imaging techniques in future studies.

Cross-relaxation bottleneck in water-lysozyme proton magnetization exchange.

The proton spin-lattice relaxation parameters in natural and deuterated lysozyme solutions have been measured as a function of temperature (0-50 degrees C). The variation of the apparent magnitudes of the water proton magnetizations in the solutions with temperature indicates that magnetic coupling mixes protein and water proton magnetizations. The results are consistent with an exchange cross-relaxation model (Hills, B. P., Mol Phys 1992, 76, 489-508) in which the cross-relaxation acts between the labile and nonlabile protons, rather than between water and protein protons. Although this cross-relaxation pathway clearly affects the observed magnetization fractions in this protein solution, its influence on the relaxation rates is less apparent.

Surface applicators for high dose rate brachytherapy in AIDS-related Kaposi's sarcoma.

PURPOSE: The development of commercially available surface applicators using high dose rate remote afterloading devices has enabled radiotherapy centers to treat selected superficial lesions using a remote afterloading brachytherapy unit. The dosimetric parameters of these applicators, the clinical implementation of this technique, and a review of the initial patient treatment regimes are presented. METHODS AND MATERIALS: A set of six fixed-diameter (1, 2, and 3 cm), tungsten/steel surface applicators is available for use with a single stepping-source (Ir-192, 370 GBq) high dose rate afterloader. The source can be positioned either in a parallel or perpendicular orientation to the treatment plane at the center of a conical aperture that sits at an SSD of approximately 15 mm and is used with a 1-mm thick removable plastic cap. The surface dose rates, percent depth dose, and off-axis ratios were measured. A custom-built, ceiling-mounted immobilization device secures the applicator on the surface of the patient's lesion during treatment. RESULTS: Between November 1994, and September 1996, 16 AIDS-related Kaposi's sarcoma patients having a total of 120 lesions have been treated with palliative intent. Treatment sites were distributed between the head and neck, extremity, and torso. Doses ranged from 8 to 20 Gy, with a median dose of 10 Gy delivered in a single fraction. Treatments were well tolerated with minimal skin reaction, except for patients with lesions treated to 20 Gy who developed moderate/severe desquamation. CONCLUSION: Radiotherapy centers equipped with a high dose rate remote afterloading unit may treat small selected surface lesions with commercially available surface applicators. These surface applicators must be used with a protective cap to eliminate electron contamination. The optimal surface dose appears to be either 10 or 15 Gy depending upon the height of the lesion.