Peak Skin Dose Research Articles

Calculating the peak skin dose resulting from fluoroscopically guided interventions. Part I: Methods

While direct measurement of the peak skin dose resulting from a fluoroscopically‐guided procedure is possible, the decision must be made a priori at additional cost and time. It is most often the case that the need for accurate knowledge of the peak skin dose is realized only after a procedure has been completed, or after a suspected reaction has been discovered. Part I of this review article discusses methods for calculating the peak skin dose across a range of clinical scenarios. In some cases, a wealth of data are available, while in other cases few data are available and additional data must be measured in order to estimate the peak skin dose. Data may be gathered from a dose report, the DICOM headers of images, or from staff and physician interviews. After data are gathered, specific steps must be followed to convert dose metrics, such as the reference point air kerma (Ka,r) or the kerma area product (KAP), into peak skin dose. These steps require knowledge of other related factors, such as the f‐factor and the backscatter factor, tables of which are provided in this manuscript. Sources of error and the impact of these errors on the accuracy of the final estimate of the peak skin dose are discussed.PACS numbers: 87.59.Dj, 87.53.Bn

Status of radiation protection in interventional cardiology in four East European countries

Level of staff and patient radiation protection in interventional cardiology in four counties (Bosnia and Herzegovina, Croatia, Montenegro and Serbia) as a part of International Atomic Energy Agency project (RER/9/093) are presented. Patient doses were assessed in terms of air kerma area product (KAP), peak skin dose (PSD) or air kerma at interventional reference point (K(IRP)). Results were available from nine hospitals: 775 patients for KAP, 157 for PSD and 437 for K(IRP). Eight centres reported KAP >100 Gy cm(2) and five centres reported values >200 Gy cm(2). From patients monitored in terms of PSD, 14 (9 %) had PSD >2 Gy and 6 (3 %) patients from those monitored in terms of K(IRP) had value >5 Gy, indicating risk of skin injury. The results indicate need for optimisation and dose monitoring in complex fluoroscopically guided cardiology interventions.

Radiation doses in a newly founded interventional cardiology department

Coronary angiography (CA) and percutaneous transluminal coronary angioplasty (PTCA) radiation doses were investigated in a recently founded Interventional Cardiology (IC) department. The study includes 336 procedures (177 CAs and 159 PTCAs) carried out with a Philips digital flat detector monoplane system. Patient dose was measured in terms of kerma-area product (KAP) and cumulative dose. Using appropriate conversion factors, peak skin dose (PSD) and effective dose (E) were estimated. Median values of KAP (Gy cm(2)), PSD (mGy) and E (mSv) were: 34 478 and 6.1, respectively for CA and 80 885 and 14.4 for PTCA, within European and international reference levels. Only 1.5 % of patients received radiation dose over the 2 Gy threshold (PTCA procedures) for deterministic effects and none reported any skin effect. Radiation doses were within international standards and comparable with other radiological examinations. The percentage of the high-risk patients for radiation skin effects is extremely low.

Reference air kerma and kerma‐area product as estimators of peak skin dose for fluoroscopically guided interventions

To determine more accurate regression formulas for estimating peak skin dose (PSD) from reference air kerma (RAK) or kerma-area product (KAP). After grouping of the data from 21 procedures into 13 clinically similar groups, assessments were made of optimal clustering using the Bayesian information criterion to obtain the optimal linear regressions of (log-transformed) PSD vs RAK, PSD vs KAP, and PSD vs RAK and KAP. Three clusters of clinical groups were optimal in regression of PSD vs RAK, seven clusters of clinical groups were optimal in regression of PSD vs KAP, and six clusters of clinical groups were optimal in regression of PSD vs RAK and K AP. Prediction of PSD using both RAK andKAP is significantly better than prediction of PSD with either RAK or KAP alone. The regression of PSD vs RAK provided better predictions of PSD than the regression of PSD vs KAP. The partial-pooling (clustered) method yields smaller mean squared errors compared with the complete-pooling method. PSD distributions for interventional radiology procedures are log-normal. Estimates of PSD derived from RAK and KAP jointly are mos t accurate, followed closely byestimates derived from RAK alone. Estimates of PSD derived from KAP alone are the least accurate. Using a stochastic search approach, it is possible to cluster together certain dissimilar types of procedures to minimize the total error sum of squares.

SU‐E‐I‐46: Methods to Determine Peak Skin Dose and Organ Doses for Computed Tomography Examinations

Purpose: This work describes methodologies for determining peak skin dose (PSD) and organ doses in computer tomography (CT) examinations. These methods are demonstrated for the routine adult head protocol. This work allows directly relating the widely used CT dose index parameter, CTDIvol, to actual PSD and organ dose values.Methods: Values of the CT dose index, CTDIvol, available in all commercially available CT units tested in this work, were derived independently from ionization chamber measurements. Simultaneously, either, PSDs (first method) or organ doses (second method) were measured as well. In the first method, PSDs were measured using radiochromic film placed on the outer surface of the phantom. Optical densities were measured with a transmission densitometer. In the second method doses to the brain, eye lens and face skin were measured using an anthropomorphic head phantom configured with optical stimulated luminescence (OSL) dosimeters. The ionization chamber, films and OSLs used in these measurements were calibrated at NIST in x‐ray beams that spectrally approximated those typically produced by CT scanners in clinical use.Results: The suggested PSD method is reliable, cheap and fast. PSD values of up to 160 mGy were observed in a single scan in some CT scanners. Variations in the PSD values up to a factor of 2 are observed for a fixed value of CTDIvol, which supports characterizing independently each scanner using the proposed methods. Results show a quite homogenous distribution of organ doses inside the head. Conclusions: Clinical facilities may apply these methods to relate the values of CTDIvol shown on their CT units to actual PSDs and organ doses delivered. Furthermore, this provides a direct method for obtaining PSD and organ dose, as opposed to the more common practice approach of estimating effective dose derived from CT machine parameter values (CTDIvol and scan length).The first author was funded by an ORISE grant at FDA. He is working for Quart GmbH since March 1, 2011. We are very thankful to Chad Mitchell for his initial encouragement on this project, to Dave Spelic for his advice on the manuscript, and to the group of CT technicians at NIH and NNMC.

SU‐E‐I‐83: Filling the Gap: Using Detailed Machine Parameters to Refine Skin Dose Calculations for Fluoroscopic Sentinel Events

Purpose: To examine the role that detailed machine logs, supplementary procedural documentation and vendor‐generated technique and dosimetry information can play in refining calculation of peak skin dose for sentinel event investigations.Methods: Vendor‐specific machine logs, skin‐dose estimations, and procedural observations can supplement the fluoroscopic times and air kerma values that all fluoroscopic machines are required to report since 2006. Information available from vendors varies considerably, and in some cases specific application products must be purchased to access this information. Information embodied may include table position, c‐arm geometry, collimation, individual run and fluoro acquisition dosimetry, technique factors (mA, kVp, pulse configuration), mode of operation and other data. Peak skin dose estimates based on HIS/RIS data and DICOM image headers were compared to calculations supplemented by the detailed machine logs and vendor dosimetry estimates. Skin dose maps were generated from air kerma readings and from the DICOM information associated with each run. Assumptions are necessary in this approach to apportion the fluoroscopic contribution, which can be 80% of the air kerma. Corresponding skin dose maps were generated with inclusion of more detailed machine log data and the resulting peak skin dose location and magnitudes were compared.Results: Significant deviations were observed between peak skin dose calculations based on DICOM information alone with attendant assumptions versus those obtained with more detailed machine log data. Detailed logs provide complementary information that replaces assumptions on fluoroscopic dose contributions with more reliable values. Conclusions: Additional information from vendorˈs logs and dose‐ estimates and from other sources can increase the confidence in skin‐dose calculations, but the specific assumptions made by both the physicist and the vendor in making such estimates must be carefully examined. The comparison enables the stand‐alone assumptions to be validated and also allows evaluation of vendor dosimetry estimates.

TH-A-214-09: Direct Measurement of Skin Dose from CT Brain Perfusion Scans

Purpose: To directly measure skin dose using point‐dosimeters from clinical CTbrain perfusion scans and compare these measurements to estimates using the standard CTDI100 method. Methods: Our clinical brain perfusion scan protocol included 11 axial 80kVp scan groups plus 2 helical 120kVp scan groups. The Landauer OSLnanoDOT dosimeters were calibrated using the same kVp's as the clinical protocol. Using nanoDOT dosimeters, skin doses at several positions on the surface of an anthropomorphic adult head phantom were measured during a perfusion protocol. The head CTDI 100 was also measured using a 100mm‐pencil CTion chamber. The standard CTDI100 method estimates peak skin dose from the brain perfusion scans based on the head CTDI100(top) for axial scans plus the head CTDI100(top)/pitch for helical scans. The peak skin dose estimated from the CTDI100 method was compared with the measured values at different locations on the anthropomorphic phantom. Results: The measured skin doses ranged from 0.11 to 0.95Gy depending on dosimeter positions. The highest dose location was located at the left‐front surface laterally ∼3cm above the ear, and the lowest dose was at the back‐center surface ∼4cm from the top. The surface doses to the locations of the eyes were ∼0.23Gy. Dose origin was ∼89% from the axial scans, ∼11% from the helical scans, and <1% from the scouts. The CTDI100 method overestimated the peak skin dose to a typical adult head size, by ∼20%. No image artifact was caused by presence of nanoDOT dosimeters in scan FOV. Conclusions: Direct measurements of skin dose using nanDOT dosimeters provided accurate and precise skin dose values without any image artifacts. CTDI100 method overestimated the skin doses for our clinical CTbrain perfusion scans. Direct measurements rather than estimates will be useful in terms of: a) documenting real patient dose, and b) guiding future dose optimization efforts.

MO-D-110-02: Calculating Patient-Specific Skin Dose

Records of radiation exposure with fluoroscopic guided interventional (FGI) exposures commonly include fluoroscopy duration, KAP, cumulative AK at the RP but may or may not include fluoroscopic exposure location information. Newer equipment use DICOM Radiation Dose SR (RDSR) which if coupled with simulation tools can re‐create an entire exposure history to calculate patient skin dose distributions and locate the regions of peak skin dose (PSD). A software tool will be described that automatically captures RDSR data to a database. A separate physicist screen (web enabled or laptop) provides prompts to set up a new or upgraded FGI lab. This task would be similair to acceptance testing a complex imaging device. This activity permits corrections for vendor specified AK at the RP, inclusion of backscatter factors, table and pad attenuation, specifying distance from the x‐ray tube, field size corrections, patient positioning (supine or prone, left or right sided, head or feet first), table movement, patient location on the table and field shape. Other fields that address QA are set by the physicist and include the list of authorized users, email alerts and action levels. A separate technologist web screen is used for each patient to choose a correct simulated patient size, positioning on the table and the individuals performing the procedure. At the end of the procedure the technologist receives a record of skin dose mapping, including an Alert if the reference values are exceeded. As the data are stored in the DICOM Tracker database, a number of QA tools can be applied including the ability monitor individual types of procedures and fluoroscopists comparative values. For legacy FGI equipment (having some, but not all of the required DICOM fields) it is possible to estimate PSD by making assumptions about the proportion of fluoroscopic exposure relative to acquistions (digital spots and cine).

WE‐B‐110‐01: Application of Medical Informatics to Patient Dose Management in Interventional Radiology

Tools for measurement and validation of radiation dose, including ALARA Policies for Fluoroscopically Guided Interventions (FGI) have been being developed using medical informatics. Informatics, primarily DICOM and Hl7 are rich content data sources that provide the potential for determination of peak skin dose, as well as a direct and immediate source of conventional metrics such as KAP, cumulative AK at the RP, and fluoroscopy duration. Newly purchased equipment may provide Radiation Dose Structured Reports which includes each radiation episode (fluoroscopy, DSA, Digitial Spots, or Cine) as separate records. For each of these data files, table position, exposure at the RP, collimation, etc. are available for use. A database has been created that uses DICOM exchange to capture dose related information for FGI. These data are used in conjunction with simulated C‐arm devices that more and mimic that which is encountered in a lab. When the desired physics parameters are included (backscatter factor, table attenuation, calibration of the vendors AK, etc) it is possible to re‐ create the exposure irradiations actually used, to compute and locate on a virtual patient the skin dose distribution and determine the Peak Skin Dose. This dose is saved to the data base for future QA uses.

TH-E-211-10: Methods for Converting CT Scanner Reported Dose Values to Dose Threshold Values Specified in Legislation: A Practical Implementation of California Senate Bill (SB) 1237

Purpose: Concern over increasing utilization of X-ray together with several publicized accidents involving medical radiation has led to government action. California Senate Bill 1237 is the first legislation of its kind to mandate event reporting whenever CT patient doses exceed specified thresholds for Effective, Organ, Skin, and Fetal dose. However, CT scanners only report CTDI and DLP values. Therefore the challenge for the Physicist - and the focus of this work - is developing a pragmatic means of converting any pre-determined action limit for patient dose into the phantom based dose metrics actually reported by a CT scanner. Methods: Appropriate to the dose limit of interest, different methods were used for translating patient dose reporting thresholds into scanner reported “trigger” values for CTDIvol or DLP for both axial and helical scans as follows: For Effective Dose, DLP values were calculated using patient age and body part specific k-factors from AAPM report 96. Organ, Fetal, and Peak Skin Dose trigger thresholds are better estimated using CTDIvol, values for which were determined from Monte Carlo simulations considering patient size. Results: For an Effective Dose action limit of 50mSv, calculated DLP values ranged from 440 mGy*cm for a 0 year abdomen/pelvis to 23800 mGy*cm for an adult head. Cumulative CTDIvol values that would result in an Organ or Skin threshold dose of 500mSv ranged from 200mGy for pediatric scans without table motion to 650mGy for adult scans with table motion. Fetal dose of 50mSv could be exceeded with cumulative CTDIvol of 25mGy and 65mGy without and with table movement respectively. Conclusions: Determining CTDIvol and DLP that result in a given patient dose requires methods specific to the dose question asked and knowledge of scan type and body region. The methods demonstrated can be generalized to any predetermined patient dose threshold.

TU‐E‐224‐01: The Role of Voluntary versus Mandatory Regulatory Standards

The patientˈs skin and the lens of the eye are the critical tissues for most fluoroscopically guided interventional procedures. Risks include hair loss, skin toxicity, and radiogenic cataracts. Appropriate dosimetric feedback to the operator contributes to optimizing benefit‐risk. Ideally, interventional fluoroscopes should provide real‐time feedback via a continuously updated dose maps and accurate estimates of peak skin dose. Such technology is not generally available in 2011. However, a variety of real‐time and post‐procedure tools are currently available for managing fluoroscopic skin irradiation. Equipment conforming to IEC 60601‐2‐43 (2000, 2010) displays fluoroscopic time, PKA and Ka,r at the operatorˈs working position. All fluoroscopes sold in the USA after mid 2006 displays fluoro‐time and Ka,r. Various film and detector arrays have been used to obtain post‐procedure dose maps for more than a decade.Newer fluoroscopes and third‐party dose monitoring systems are beginning to offer useful proprietary dose mapping in angular space (defined by gantry angles) in both real‐time and post‐procedure flavors. Initial commercial deliveries of the DICOM Radiation Dose Structured Report (RDSR) provide detailed post‐procedure procedural information in enough detail to facilitate detailed dose reconstruction. Data in the RDSR is in defined public fields to facilitate accessibility. The relevant DICOM and IEC standards include specifications for real‐time streaming of the data. This will facilitate the development of independent real‐time dose modeling. This lecture will review the technologies available in 2011 and will provide an estimate of future time‐lines in this area. It will also include a discussion of the use of complete sets of dosimetric data as part of the quality process.Learning Objectives:1. Understand the advantages and disadvantages of currently available dose monitoring technologies.2. Understand the evolution toward real‐time skin dose mapping.3. Understand how standardized dosimetry can be incorporated into a QA program.

Does “Spreading” Skin Dose by Rotating the C-arm during an Intervention Work?

Purpose To determine if C-arm rotation is beneficial for reducing peak skin dose (PSD) in interventional radiology (IR) and, if so, under what circumstances. Materials and Methods The Monte Carlo method was used to perform ray tracing for detailed analyses of the effect of C-arm rotation on PSD across a range of patient sizes, C-arm configurations, and procedure types. Automatic dose-rate control curves on modern fluoroscopic systems were measured for input into the simulations. Results Rotating the C-arm to reduce the PSD is in most cases contraindicated and results in increased PSD when the C-arm is rotated from an original posteroanterior projection, in some cases resulting in a PSD increase by a factor of 5 or more. When prophylactic rotation was performed before a procedure, however, and the C-arm was rotated between opposed, distinct oblique angles, substantial reduction in PSD was achieved for patients of any size. Conclusions Rotating the C-arm during a procedure with the aim of “spreading” dose on the skin of the patient may not result in a reduction in PSD and may increase PSD. However, when used as a prophylactic measure combined with tight x-ray beam collimation, C-arm rotation can be used as a tool to reduce PSD. Tight collimation greatly increases the benefit of C-arm rotation.

Comparison of indirect radiation dose estimates with directly measured radiation dose for patients and operators during complex endovascular procedures

A great deal of attention has been directed at the necessity and potential for deleterious outcomes as a result of radiation exposure during diagnostic evaluations and interventional procedures. We embarked on this study in an attempt to accurately determine the amount of radiation exposure given to patients undergoing complex endovascular aortic repair. These measured doses were then correlated with radiation dose estimates provided by the imaging equipment manufacturers that are typically used for documentation and analysis of radiation-induced risk. Consecutive patients undergoing endovascular thoracoabdominal aneurysm (eTAAA) repair were prospectively studied with respect to radiation dose. Indirect parameters as cumulative air kerma (CAK), kerma area product (KAP), and fluoroscopy time (FT) were recorded concurrently with direct measurements of dose (peak skin dose [PSD]) and radiation exposure patterns using radiochromatic film placed in the back of the patient during the procedure. Simultaneously, operator exposure was determined using high-sensitivity electronic dosimeters. Correlation between the indirect and direct parameters was calculated. The observed radiation exposure pattern was reproduced in phantoms with over 200 dosimeters located in mock organs, and effective dose has been calculated in an in vitro study. Scatter plots were used to evaluate the relationship between continuous variables and Pearson coefficients. eTAAA repair was performed in 54 patients over 5 months, of which 47 had the repair limited to the thoracoabdominal segment. Clinical follow-up was complete in 98% of the patients. No patients had evidence of radiation-induced skin injury. CAK exceeded 15 Gy in 3 patients (the Joint Commission on Accreditation of Healthcare Organizations [JCAHO] threshold for sentinel events); however, the direct measurements were well below 15 Gy in all patients. PSD was measured by quantifying the exposure of the radiochromatic film. PSD correlated weakly with FT but better with CAK and KAP (r = 0.55, 0.80, and 0.76, respectively). The following formula provides the best estimate of actual PSD = 0.677 + 0.257 CAK. The average effective dose was 119.68 mSv (for type II or III eTAAA) and 76.46 mSv (type IV eTAAA). The operator effective dose averaged 0.17 mSv/case and correlated best with the KAP (r = 0.82, P < .0001). FT cannot be used to estimate PSD, and CAK and KAP represent poor surrogate markers for JCAHO-defined sentinel events. Even when directly measured PSDs were used, there was a poor correlation with clinical event (no skin injuries with an average PSD >2 Gy). The effective radiation dose of an eTAAA is equivalent to two preoperative computed tomography scans. The maximal operator exposure is 50 mSv/year, thus, a single operator could perform up to 294 eTAAA procedures annually before reaching the recommended maximum operator dose.

A study of the relationship between peak skin dose and cumulative air kerma ininterventional neuroradiology and cardiology

A study of peak skin doses (PSDs) during neuroradiology and cardiology interventionalprocedures has been carried out using Gafchromic XR-RV2 film. Use of mosaics made fromsquares held in cling film has allowed doses to the head to be mapped successfully. Thedisplayed cumulative air kerma (CAK) has been calibrated in terms of cumulative entrancesurface dose (CESD) and results indicate that this can provide a reliable indicator of thePSD in neuroradiology. Results linking PSD to CESD for interventional cardiologywere variable, but CAK is still considered to provide the best option for use asan indicator of potential radiation-induced effects. A CESD exceeding 3 Gy isconsidered a suitable action level for triggering follow-up of patients in neuroradiologyand cardiology for possible skin effects. Application of dose action levels definedin this way would affect 8% of neurological embolisation procedures and 5% ofcardiology ablation and multiple stent procedures at the hospitals where theinvestigations were carried out. A close relationship was observed between CESD anddose–area product (DAP) for particular types of procedure, and DAPs of 200–300 Gy cm2 could be used as trigger levels where CAK readings were not available. The DAP valuewould depend on the mean field size and would need to be determined for eachapplication.

Unprotected Operator Eye Lens Doses in Oncologic Interventional Radiology Are Clinically Significant: Estimation from Patient Kerma-area-product Data

To correlate operator lens dose to patient-delivered kerma-area-product (P(KA)) to evaluate the usefulness of P(KA) as a surrogate for operator eye dose if collar monitor readings are unavailable or deemed unreliable, and to evaluate if unprotected lens dose is clinically significant. A retrospective review of peak skin doses for consecutive interventional radiology procedures performed during 2006 that had P(KA) estimates recorded was performed. Unshielded operator lens dose equivalents (LDE) were obtained from dosimetry monitors worn outside the collar shield of operating interventional radiologists. Operator LDE were correlated with patient P(KA). Average LDE for 2006 was 35.7 mSv ± 32.7 (range 5.2-89.9 mSv). Patient-delivered P(KA) correlated directly with LDE, where 1 Gy cm(2) to the patient resulted in an average of 4.2 μSv to the unprotected eyes of the primary operator (r(2) = 0.7). P(KA) may be useful as a surrogate measure of operator LDE if collar monitor readings are unavailable or deemed unreliable. For this study, the dose-effect threshold for cataract formation could be surpassed for some physicians within 11 years if lens dose-mitigating strategies are not routinely employed.

Patient skin dose measurements during coronary interventional procedures usingGafchromic film

Interventional cardiology (IC) procedures are known to give high radiation doses to patientsand cardiologists as they involve long fluoroscopy times and several cine runs. Patients’dose measurements were carried out at the cardiology department in a local hospital inPenang, Malaysia, using Gafchromic XR-RV2 films. The dosimetric properties of theGafchromic film were first characterised. The film was energy and dose rate independentbut dose dependent for the clinically used values. The film had reproducibility within ± 3% when irradiated on three different days and hence the same XR-RV2 dose–responsecalibration curve can be used to obtain patient entrance skin dose on different days. Theincrease in the response of the film post-irradiation was less than 4% over a periodof 35 days. For patient dose measurements, the films were placed on the tableunderneath the patient for an under-couch tube position. This study included atotal of 44 patients. Values of 35–2442 mGy for peak skin dose (PSD) and 10.9–344.4 Gy cm2 for dose–area product (DAP) were obtained. DAP was found to be a poorindicator of PSD for PTCA procedures but there was a better correlation (R2 = 0.7344) forCA + PTCA procedures. The highest PSD value in this study exceeded the threshold dose value of 2 Gyfor early transient skin injury recommended by the Food and Drug Administration.

Measuring and Monitoring Radiation Dose During Fluoroscopically Guided Procedures

The principal problem in measuring patient radiation dose during fluoroscopically guided procedures is that dose is not administered uniformly throughout the patient's body. Four dose metrics have been developed to quantify patient radiation dose for fluoroscopically guided procedures: fluoroscopy time, peak skin dose, reference dose, and kerma-area-product. Each metric must be understood to be used appropriately. Fluoroscopy time correlates poorly with other dose metrics. It should not be used as the sole method to estimate, monitor, or record patient radiation dose unless no alternative is available. Kerma-area-product is a good metric for estimating stochastic risk. Reference dose is a conservative method to estimate peak skin dose and deterministic risk and is recommended for this purpose. Every fluoroscope sold in the USA since mid 2006 is able to measure, display, and record reference dose. Radiation dose should be monitored during fluoroscopically guided procedures, either by the operator or by a designated individual in the procedure room, such as a technologist or nurse. Patient radiation dose should be recorded appropriately in the medical record. Patients who have received a sufficiently large radiation dose should have follow-up at 10-14 days and at 1 month after the procedure for possible deterministic effects.

Automatic monitoring of localized skin dose with fluoroscopic and interventional procedures.

This software tool locates and computes the intensity of radiation skin dose resulting from fluoroscopically guided interventional procedures. It is comprised of multiple modules. Using standardized body specific geometric values, a software module defines a set of male and female patients arbitarily positioned on a fluoroscopy table. Simulated X-ray angiographic (XA) equipment includes XRII and digital detectors with or without bi-plane configurations and left and right facing tables. Skin dose estimates are localized by computing the exposure to each 0.01 × 0.01 m(2) on the surface of a patient irradiated by the X-ray beam. Digital Imaging and Communications in Medicine (DICOM) Structured Report Dose data sent to a modular dosimetry database automatically extracts the 11 XA tags necessary for peak skin dose computation. Skin dose calculation software uses these tags (gantry angles, air kerma at the patient entrance reference point, etc.) and applies appropriate corrections of exposure and beam location based on each irradiation event (fluoroscopy and acquistions). A physicist screen records the initial validation of the accuracy, patient and equipment geometry, DICOM compliance, exposure output calibration, backscatter factor, and table and pad attenuation once per system. A technologist screen specifies patient positioning, patient height and weight, and physician user. Peak skin dose is computed and localized; additionally, fluoroscopy duration and kerma area product values are electronically recorded and sent to the XA database. This approach fully addresses current limitations in meeting accreditation criteria, eliminates the need for paper logs at a XA console, and provides a method where automated ALARA montoring is possible including email and pager alerts.

MO‐E‐204B‐04: Interventional Fluoroscopy Procedure Skin Dose Assessment Using DICOM Dose Structured Reports

Purpose: Patient skin injury due to radiation exposure during interventional fluoroscopy procedures is a concern. The FDA has recommended dose be monitored, along with the dose distribution on the patient's skin. Modern fluoroscopy equipment provides a readout of the accumulated reference point air kerma. However, this parameter is a poor estimate of localized skin dose since it does not take into account beam movement. Acquisition information needed to calculate skin dose can be recorded as a DICOM Dose Structured Report (Dose SR), which has recently become available on new fluoroscopy equipment. We have developed a system that uses the Dose SR to produce a skin dose map, along with determination of the peak skin dose. Method and Materials: The Dose SR provides data for each irradiation event (fluoroscopy footswitch or acquisition): kerma‐area‐product, reference point air kerma, C‐arm angulation and table position. Using the position of the patient on the table, backscatter and table attenuation, the air kerma to 1‐cm2 regions on the patient's surface is calculated. The patient surface is approximated as a mathematical model using SolidWorks simulation software, with model size varied based on patient parameters. Results: The method was validated using a Siemens Axiom Artis angiography system. Measurements made with an ionization chamber and lucite phantom were compared to calculated skin dose values for a range of C‐arm angles, table positions and exposure levels. Errors ranged from 2–10% with higher errors corresponding to increased C‐arm angulation. Conclusion: A system has been developed to estimate skin dose and generate skin dose maps from an interventional procedure Dose SR. The system allows for a more precise determination of localized skin dose and assists with planning appropriate entry ports for future procedures to minimize skin injury. Application to any vendor platform providing a Dose SR is possible.

SU-GG-I-59: A Process to Streamline Patient Skin Dose Estimation-What We Have and What We Do Not Yet Have

Identifying at‐risk patients for adverse bioeffects including cutaneous radiation injury following high‐dose fluoroscopically‐guided interventional procedures was an FDA recommendation in 1994. The Joint Commission's 15‐Gy fluoroscopic peak skin dose sentinel event caused many facilities to consider potentially‐high patient doses proactively. Purpose: We present our strategy for evaluating interventional procedures to identify patients with high doses and our multilevel support for these patients. We describe problems encountered associated with assessments for patients examined on new equipment (i.e., with FDA‐required air kerma meters) and vintage units. We report planned changes in patient tracking, to improve accuracy, timeliness and efficiency. Methods and Materials: We have 3 levels for dose tracking in our department policy, modified from the procedures reported by Mahesh in 2008. We track data sheets from interventional procedures to identify patients with potentially high doses, and we evaluate prior exams for Level 2 and Level 3 patients. Level 2 patients receive cursory initial dose estimates that depend on the unit. Level 3 patients require detailed dose assessments to determine whether the sentinel event threshold is crossed. Vintage units require measurements and study‐reported data to reconstruct doses. We modified annual and acceptance testing to streamline dosimetry for common high‐dose procedures. New units with in‐line dosimetry need fewer measurements and required assumptions, but some necessary information is lacking. Adding estimated doses for multiple exams present special unit‐dependent challenges, related to the beam entrance location. Results: Examples are presented for improving accuracy and timeliness in identifying patients with high levels, preparing cursory estimates to improve timeliness, and modifying annual physics protocols to collect data for dose estimation in advance of specific patient needs. We include a sample dose report. Conclusion: Following patient dose histories for monitoring potential CRI is time‐consuming. We identify our processes for improving accuracy and timeliness and show what is lacking.