Published online before print January 19, 2006, 10.1148/radiol.2382042070
(Radiology 2006;238:1044-1050.)
Vascular and Interventional Radiology |
Radiation Doses from Venous Access Procedures1
Erik S. Storm, DO, CPT, MC, USAR,
Donald L. Miller, MD,
Laurie Jean Hoover, RN,
Jeffrey D. Georgia, MD, CAPT, MC, USN and
Tara Bivens, RN
1 From the Department of Radiology, National Naval Medical Center, Bethesda, Md (E.S.S., D.L.M., L.J.H., J.D.G., T.B.); and Department of Radiology and Radiological Sciences, F. Edward Hébert School of Medicine, Uniformed Services University of the Health Sciences, Bethesda, Md (D.L.M., J.D.G.). Received January 4, 2005; revision requested March 10; revision received April 26; final version accepted June 3.
Address correspondence to D.L.M., Department of Radiology, Albert Einstein Medical Center, 5501 Old York Rd, Philadelphia, PA 19141.
 |
ABSTRACT
|
|---|
Purpose: To retrospectively analyze radiation dose data for six common venous access procedures.
Materials and Methods: Institutional review board approval was obtained for this HIPAA-compliant study; informed consent was not required. Data review was limited to a quality assurance database. Patient medical records were not reviewed. We retrospectively analyzed radiation dose data from a prospective quality assurance program. Dose data were analyzed for 1010 instances of six different venous access placement procedures performed between February 1998 and July 2004. Radiation dose measurements were generated automatically by the interventional fluoroscopy units and were recorded at the conclusion of each procedure. Descriptive and summary statistical analyses were performed to determine median, minimum, and maximum values of radiation dose for each procedure. A P value of less than .05 indicated a significant difference. Because the data distribution was highly skewed, logarithmic transformation was performed. Dose data for four different venous access procedures (excluding chest port placement and peripherally inserted central catheter placement) were compared with a one-way analysis of variance. Pairwise comparisons with the Tukey honestly significant difference test were subsequently performed for each analogue where analysis of variance demonstrated a significant result.
Results: No procedure yielded a cumulative dose of more than 950 mGy or a peak skin dose of more than 760 mGy. The highest mean cumulative dose (ie, 88 mGy), mean dose-area product (ie, 873 cGy · cm2), and mean peak skin dose (ie, 43 mGy) were observed for tunneled dialysis catheter placements. Significant differences in dose were observed for tunneled catheter placement versus nontunneled catheter placement (<.001 to .027). No significant differences in dose were observed for larger-diameter versus smaller-diameter catheters.
Conclusion: Radiation doses from venous access procedures are low. Even extreme outlier cases are unlikely to produce doses high enough to cause skin effects, especially when knowledgeable operators using well-calibrated equipment perform the procedures.
|
INTRODUCTION
|
|---|
In general, the risk of radiation injury to the skin during most fluoroscopically guided procedures is low. The majority of cases of skin injury reported in the literature or to the U.S. Food and Drug Administration have been caused by cardiac radiofrequency ablation or coronary angioplasty (1,2). Some reported skin injuries have been associated with transjugular intrahepatic portosystemic shunt creation, renal angioplasty, multiple hepatic or biliary procedures, or embolization (1–5). In interventional radiology practice, embolization procedures, angioplasty in the abdomen and pelvis, and transjugular intrahepatic portosystemic shunt creation have been shown to potentially cause high radiation doses to the skin (6). In general, radiation doses from fluoroscopically guided venous access procedures have not been studied because of the clinical impression that the radiation doses that may be caused by these procedures are relatively low (7).
Absorbed dose in the skin can be difficult to measure; therefore, dose has been reported in terms of the following dose analogues: fluoroscopy time, dose-area product (DAP), cumulative dose (CD), and peak skin dose (PSD) (6,7). The purpose of our study was to retrospectively analyze radiation dose data for six common venous access procedures.
|
MATERIALS AND METHODS
|
|---|
Study Design
This study was performed to retrospectively evaluate prospectively collected radiation dose data available from a quality assurance database. The quality assurance program was designed to track radiation dose from all procedures performed by either the interventional radiology section or the interventional neuroradiology section of our institution's (National Naval Medical Center) radiology department. The database was created in 1998 and is compliant with the Health Insurance Portabilty and Accountability Act. This database includes the following information: procedure type, fluoroscopy time, CD, DAP, and PSD. No patient identifiers or protected patient information were included. All dose metrics were not available for every procedure. Our institutional review board approved the retrospective data analysis and did not require patient informed consent or notification. Data review was limited to the quality assurance database. Patient medical records were not reviewed.
The radiation dose data in the quality assurance database were analyzed to determine if procedures involved the placement of any of the following types of venous access devices: (a) peripherally inserted central catheters (PICCs), (b) chest wall ports, (c) nontunneled (ie, temporary) dialysis catheters, (d) tunneled (ie, long-term) dialysis catheters, (e) nontunneled (ie, temporary) central venous catheters (CVCs), and (f) tunneled (ie, long-term) CVCs. Tunneling refers to the creation of a short subcutaneous track through which the proximal portion of the catheter travels prior to its insertion into the vein. Tunneled catheters usually have a cuff of Dacron (Du Pont, Wilmington, Del) or a similar material within the tunnel exit. Procedures involving exchange, repositioning, or removal of a venous access device were excluded from analysis, as were procedures involving placement of a venous access device and any additional procedure.
For the six types of central venous access procedures, we identified 1010 fluoroscopically guided percutaneous venous access procedures performed at our institution between February 1998 and July 2004.
Catheter Placement
Tunneled and nontunneled dialysis catheters and tunneled and nontunneled CVCs were typically placed in the venous system via an internal jugular vein. Jugular vein access was obtained with ultrasonographic (US) guidance. The remainder of the procedure was performed with fluoroscopic guidance. The axillary vein was used for more than 95% of chest wall port placements. Fluoroscopy was used to access the axillary vein and provide imaging guidance for the remainder of the procedure. PICCs commonly were placed in a brachial vein; venous access was guided with US, and the remainder of the procedure was performed with fluoroscopic guidance. For all venous access procedures, dose data included the dose used to obtain digital chest images at the conclusion of the procedure; these images were used to document the position of the catheter tip.
During the time period covered by this study, a number of different catheters were used for each of the different venous access procedures we describe. Typical catheters and sizes were as follows: (a) placement of 5-F PICCs (Vaxcel no. 45–432, Boston Scientific, Natick, Mass; Bard no. 4235105, Bard Access, Salt Lake City, Utah), (b) nontunneled placement of 7-F CVCs (Arrow triple-lumen no. AK-14703; Arrow, Reading, Pa), (c) tunneled placement of 9.5-F CVCs (Bard no. 7726954; Bard Access), (d) nontunneled dialysis with 14-F catheters (Schon XL; AngioDynamics, Queensbury, NY), (e) tunneled dialysis with 14.0–15.5-F catheters (More-Flow or Dura-Flow; AngioDynamics), and (f) placement of a chest wall port with a 6.5-F catheter (Cook no. 6113; Cook, Bloomington, Ind).
Our institution is a teaching hospital. In the majority of cases, portions of the catheter placement procedure were performed by 2nd-year radiology residents in the 1st or 2nd month of interventional radiology training, with faculty supervision.
Dose Measurement
All procedures were performed in one of two interventional radiology suites: a Multistar single-planar unit (Siemens Medical Systems, Malvern, Pa) or a Neurostar biplanar unit (Siemens Medical Systems). Both systems are compliant with the dosimetry portion of International Electrotechnical Commission standard 60601–2-43 (8). Each fluoroscopic unit contains an integrated dosimeter. The fluoroscopic units are used to automatically measure exposure. Dosimetry information—including fluoroscopy time, DAP, and CD at the interventional reference point—is displayed automatically on the console in the control room. Fluoroscopy time is displayed and recorded in units of 0.1 minute, DAP is displayed and recorded in centigrays times square centimeters, and CD is displayed and recorded in milligrays.
Each fluoroscopic unit is also equipped with an additional dose measurement system (CareGraph; Siemens Medical Systems). This skin dose mapping software uses a mathematical model of the patient's height and weight to model surface shape. This is used, along with the location of the patient on the procedure table, to calculate x-ray entrance field size, location, and air kerma incident on the skin (9). The measured DAP, collimator field size, tabletop position, and C-arm angulation are used to monitor skin irradiation in real time. The PSD without backscatter and the spatial distribution of the dose on the skin are displayed on a skin-dose map that represents the skin surface. Both single-planar and biplanar fluoroscopic units generate a single value for PSD. The measurement of skin dose is not applicable to PICC placement, as the software is not used to calculate skin dose for the extremities.
All collected data were analyzed. For individual cases where one or more dose measurements were not recorded, these data points were considered "missing at random." When the fluoroscopy equipment indicated fluoroscopy time, CD, or PSD to be 0, the measurement was recorded as 0.1 minute for fluoroscopy time and 1 mGy for CD and PSD; these are the minimum values that can be displayed by the equipment. This was done because the fluoroscopic unit dose display truncates the dose measurements, and some radiation was used for each case. No dose measurements were greater than the measurement display capability of the fluoroscopic units.
Operators routinely used reduced-dose pulsed fluoroscopy (15 pulses per second), which is one of eight standard modes available on each fluoroscopic unit.
Dosimeter Calibration
Each of the fluoroscopic units used in this study was used in previous clinical dose measurement studies that included rigorous medical physics evaluations of the equipment and integrated dosimeters (6,7). Calibrations and dosimeter checks were performed between April 1999 and January 2002, and no clinically important drift (<10%) was seen during this interval. Procedural details are described elsewhere (10). An initial comprehensive medical physics evaluation of each fluoroscopic unit was conducted to confirm that its dosimeter was functioning properly. In this comprehensive evaluation, the internal reference air kerma readout was compared with the values of an external ionization chamber over a range of exposure conditions. The comprehensive evaluation was repeated after any major equipment modifications and at the end of the studies. Periodic consistency checks of each unit were performed every 1–2 weeks to verify the stability and consistency of the reference air kerma readout and the automatic brightness control. We estimate that the overall error in clinical CD measurements for these units, defined as the standard deviation of all calibration and consistency measurements obtained during the course of the clinical dose measurement study, was ± 24% (10).
Statistical Analysis
Data from the quality assurance database (ie, a handwritten log maintained by nurses permanently assigned to the interventional radiology department) were entered into a computerized database (Access 2000; Microsoft, Redmond, Wash). Descriptive and summary statistics (minimum, maximum, and median values) were calculated and graphs (histograms and box plots) were created with SPSS software (version 12.0; SPSS, Chicago, Ill). Because the data were skewed and none of the data were distributed normally, the data were transformed with a log transformation (11). Means and confidence intervals were calculated with the transformed data, and the results were then back transformed to yield geometric means and corresponding 95% confidence intervals (12). Data transformation and calculation of geometric means and 95% confidence intervals were performed with SPSS software. Additional statistical analyses were also performed with SPSS software. Radiation doses for the four clinically related procedures (placement of tunneled dialysis catheters, nontunneled dialysis catheters, tunneled CVCs, or nontunneled CVCs) were compared by using a one-way analysis of variance on the log-transformed data for each dose analogue (fluoroscopy time, CD, DAP, and PSD). For each dose analogue where the result of the one-way analysis of variance was statistically significant, post hoc pairwise comparisons of the four procedures were performed by using the Tukey honestly significant difference test. A P value of less than .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
The number of cases of each type of procedure is given in Table 1. Table 1 also provides descriptive and summary statistics for the dose analogues of fluoroscopy time, CD, DAP, and PSD. It should be noted that PSD was not measured for PICC placement.
The distribution of fluoroscopy time and other dose analogues for each procedure was non-Gaussian and highly skewed; however, it was consistent among procedures and for all dose analogues. A typical histogram is shown in Figure 1. Cases with high dose measurements were infrequent and were extreme outliers. This is evident from box plots of fluoroscopy time, CD, DAP, and PSD (Figs 2–5). The mean CD for each procedure was less than 100 mGy, and the mean PSD for each procedure was less than 50 mGy. No cases in this study had a CD of more than 950 mGy or a PSD of more than 760 mGy.
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2: Box plot of fluoroscopy time for venous access procedures. The horizontal line within the box is the median. The lower and upper ends of the box are the first and third quartiles. The height of the box is the interquartile range. The vertical lines extending above and below the box terminate on the data points furthest from the mean but within a distance of 1.5 times the interquartile range from the end of the box. Data points indicated by  are outliers and are between one and a half to three times the interquartile range from the end of the box. Data points indicated by * are extremes and are more than three times the interquartile range from the end of the box.
|
|
For all dose analogues, a statistically significant difference existed among the four clinically related procedures (ie, placement of tunneled and nontunneled dialysis catheters, placement of tunneled and nontunneled CVCs) (Table 2). A significant difference between tunneled dialysis catheter placement and nontunneled CVC placement was present for each of the four dose analogues (Table 2). No significant difference between tunneled dialysis catheter placement and tunneled CVC placement or between nontunneled dialysis catheter placement and nontunneled CVC placement was observed for any of the dose analogues. This suggests that differences in catheter diameter within the range we studied (ie, 9.5 F vs 14.0–15.5 F for tunneled catheters and 7 F vs 14 F for nontunneled catheters) did not affect radiation dose.
View this table:
[in this window]
[in a new window]
|
Table 2. Comparison of Differences in Dose for Placement of Tunneled and Nontunneled Dialysis Catheters and Tunneled and Nontunneled CVCs
|
|
Significant differences in fluoroscopy time, CD, and DAP were observed between tunneled and nontunneled dialysis catheter placement (Table 2). Significant differences in fluoroscopy time and CD were also observed between tunneled and nontunneled CVC placement. This suggests that the radiation dose for these venous access procedures is affected by whether or not the catheter is tunneled.
|
DISCUSSION
|
|---|
During fluoroscopy, the highest radiation dose occurs at the skin at the entrance site of the radiation beam. Radiation-induced skin effects are deterministic: Below the threshold for deterministic effects, dose-specific effects do not occur; likewise, above this threshold, dose-specific effects are certain to occur, and their severity increases with increasing dose. This is analogous to sun exposure and the development of sunburn. The threshold absorbed dose for transient skin erythema (ie, the initial effect observed when dose threshold is exceeded) is typically estimated at about 2 Gy (14,15). Some patients may have more severe reactions at the same or a lower dose because of biologic variation (16). Typical manifestations of radiation injury to the skin range from transient erythema at low doses to dermal necrosis or chronic ulceration at high doses (17).
Concern over the potential for high radiation doses as a result of interventional radiology procedures prompted the Society of Interventional Radiology to conduct the Radiation Doses in Interventional Radiology Procedures, or RAD-IR, study (6,7,10). The procedures included in the RAD-IR study were those, on the basis of the length and complexity of the procedure, for which there was a general expectation that the 2-Gy threshold dose could be approached or surpassed. In the RAD-IR study, venous access procedures were not investigated because of the general assumption that these procedures had little chance of imparting a radiation dose sufficient to induce deterministic skin effects (7). To our knowledge, this assumption is reasonable; however, it has not been investigated in detail. Fletcher et al (18) included chest port placement as one of the procedures analyzed in their comparison of dose analogues; the 57 cases in their study comprised 19% of the 303 instances of chest port placement in our study.
Fluoroscopically guided venous access procedures are a common part of interventional radiology practice. While the typical radiation dose for a single venous access case is relatively low and was below the threshold dose for deterministic skin effects in all cases we studied, these procedures are often repeated in the same patient within a short period of time. The radiation dose from multiple procedures, although fractionated, is nevertheless cumulative. Thus, knowledge of the radiation doses from these procedures is desirable. An additional consideration that was not included in this analysis is the risk of stochastic effects, principally cancer, from low-dose radiation exposure. The risk of stochastic effects from low doses of radiation is controversial and has been addressed in detail (19,20).
We were surprised to observe no effect on radiation dose as a function of catheter size. Placement of large-diameter catheters is often more difficult than placement of small-diameter catheters, particularly when a left internal jugular vein approach is used. We also did not expect to see significant differences in radiation dose between tunneled and nontunneled catheter placements. The tunneling process adds additional time to the procedure, but it does not require fluoroscopic guidance. We expected that the radiation doses would be similar.
None of the procedures we investigated resulted in a radiation dose to the skin sufficient to induce a deterministic effect. Even extreme outlier doses are well below the commonly accepted threshold dose of 2 Gy for radiation-induced transient erythema.
The highest skin dose we observed was less than 1 Gy, which was half of the minimum threshold dose. This case was an extreme outlier. The majority of procedures yielded substantially lower skin doses. The shape of the dose distribution histogram (Fig 1) is similar to the shape of curves in studies of radiation dose from other interventional radiology and cardiac catheterization procedures (7,13).
Our institution is a teaching facility; therefore, most of the venous access procedures are performed in part by radiology residents with faculty supervision. Venous access procedures performed by experienced operators will likely yield lower radiation doses.
The major limitation of this study is that the calculated dose data are directly applicable to only those operators who use the same equipment and techniques that we used and who have undergone similar training. Users of other types of fluoroscopic devices, such as portable C-arm units, and users of different equipment may realize substantial differences in typical dose. Some fluoroscopic equipment may not incorporate similar dose-reducing technology; even if this technology is available, operators may not effectively use these features.
It is unlikely that any fluoroscopically guided venous access procedure performed by a reasonably well-trained operator will result in a dose high enough to cause concern for skin injury. Nevertheless, operators should remain cognizant of the cumulative effects of radiation, including the potential risk of stochastic effects.
|
FOOTNOTES
|
|---|
Abbreviations: CD = cumulative dose • CVC = central venous catheter • DAP = dose-area product • PICC = peripherally inserted central catheter • PSD = peak skin dose
The opinions expressed herein are those of the authors and do not necessarily reflect those of the United States Navy, the Department of Defense, or the Department of Health and Human Services.
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, D.L.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, E.S.S., D.L.M., L.J.H., J.D.G.; clinical studies, D.L.M., J.D.G.; statistical analysis, D.L.M.; and manuscript editing, E.S.S., D.L.M.
|
References
|
|---|
- Shope TB. Radiation-induced skin injuries from fluoroscopy. RadioGraphics 1996;16:1195–1199.[Abstract]
- Koenig TR, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures. II. Review of 73 cases and recommendations for minimizing dose delivered to the patient. AJR Am J Roentgenol 2001;177:13–20.
- Huda W, Peters KR. Radiation-induced temporary epilation after a neuroradiologically guided embolization procedure. Radiology 1994;193:642–644.[Abstract/Free Full Text]
- Lichtenstein DA, Klapholz L, Vardy DA, et al. Chronic radiodermatitis following cardiac catheterization. Arch Dermatol 1996;132:663–667.[Abstract/Free Full Text]
- Vañó E, Arranz L, Sastre JM, et al. Dosimetric and radiation protection considerations based on some cases of patient skin injuries in interventional cardiology. Br J Radiol 1998;71:510–516.[Abstract]
- Miller DL, Balter S, Cole PE, et al. Radiation doses in interventional radiology procedures: the RAD-IR study. II. Skin dose. J Vasc Interv Radiol 2003;14:977–990.
- Miller DL, Balter S, Cole PE, et al. Radiation doses in interventional radiology procedures: the RAD-IR study. I. Overall measures of dose. J Vasc Interv Radiol 2003;14:711–727.
- International Electrotechnical Commission. Medical electrical equipment: part 2–43—particular requirements for the safety of x-ray equipment for interventional procedures. Report 60601–2-43. Geneva, Switzerland, International Electrotechnical Commission, 2000.
- den Boer A, de Feijter PJ, Serruys PW, Roelandt JR. Real-time quantification and display of skin radiation during coronary angiography and intervention. Circulation 2001;104:1779–1784.[Abstract/Free Full Text]
- Balter S, Miller DL, Schueler BA, et al. Radiation doses in interventional radiology procedures: the RAD-IR study. III. Dosimetric performance of the interventional fluoroscopy units. J Vasc Interv Radiol 2004;15:919–926.
- Bland JM, Altman DG. Transforming data. BMJ 1996;312:770.[Free Full Text]
- Bland JM, Altman DG. Transformation, means, and confidence intervals. BMJ 1996;312:1079.[Free Full Text]
- Paisley EM, Eatough JP, Mountford PJ, Frain G, Pickerill J. Patient radiation doses during invasive cardiac procedures categorised by clinical code. Br J Radiol 2004;77:1022–1026.[Abstract/Free Full Text]
- Wagner LK, Eifel PJ, Geise RA. Potential biological effects following high x-ray dose interventional procedures. J Vasc Interv Radiol 1994;5:71–84.[Medline]
- International Commission on Radiological Protection. Avoidance of radiation injuries from medical interventional procedures. ICRP publication no. 85. Ann ICRP 2000;30:7–67.[Medline]
- Wagner LK, McNeese MD, Marx MV, Siegel EL. Severe skin reactions from interventional fluoroscopy: case report and review of the literature. Radiology 1999;213:773–776.[Abstract/Free Full Text]
- Koenig TR, Wolff D, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures. I. Characteristics of radiation injury. AJR Am J Roentgenol 2001;177:3–11.
- Fletcher DW, Miller DL, Balter S, Taylor MA. Comparison of four techniques to estimate radiation dose to skin during angiographic and interventional radiology procedures. J Vasc Interv Radiol 2002;13:391–397.[Medline]
- Dendy PP, Brugmans MJ. Low dose radiation risks. Br J Radiol 2003;76:674–677.[Free Full Text]
- Cohen BL. Cancer risk from low-level radiation. AJR Am J Roentgenol 2002;179:1137–1143.[Free Full Text]
TOP
ABSTRACT
INTRODUCTION
