- Email: [email protected]
Increased Fluoroscopy Time for Central Venous Catheter Placement by Radiology Residents versus Staff Radiologists
Increased Fluoroscopy Time for Central Venous Catheter Placement by Radiology Residents versus Staff Radiologists Baogang J. Xu, PhDa, Richard Duszak Jr, MDa,b, Robert S. McGinnis, MDb, John G. Stanfill, MDb, Jeff O’Rear, MDb, Angel Q. An, MSc
Purpose: To evaluate differences in interventional radiology procedural fluoroscopy time (FT) for radiology residents versus staff radiologists, using central venous catheter (CVC) placement as an index service. Methods: To minimize interservice and complexity variables, stand-alone temporary internal jugular CVC procedures were targeted for analysis. Reports and images from 1,067 temporary CVC services from 2 hospitals over 2 years were reviewed as part of a quality improvement initiative. Insertion site, catheter type (eg, smaller triple lumen versus larger hemodialysis), resident identifier, staff identifier, and documented FT were compiled and analyzed. Results: Applying clinical (eg, concomitant venous angioplasty) and anatomic (eg, femoral access) exclusions, 537 cases with complete CVC procedure records were available for analysis. Radiology residents and staff radiologists were primary operators in 128 and 409 procedures, respectively. Distribution of resident procedures (82% right, 66% large lumen) was similar to that of staff (79% right, 63% large lumen). Mean FT of resident services was twice as long as that of staff services (1.24 minutes versus 0.63 minutes, P ⬍ .0001). Resident FT was independent of supervising staff radiologist. Increasing years of training for residents did not significantly reduce FT. Conclusions: When CVCs are placed by radiology residents, FT is double that for identical procedures performed by staff radiologists. Similar discrepancies likely exist for other interventional radiologic procedures. Residency training programs should initiate measures to monitor and manage fluoroscopy during interventional procedures to minimize radiation dose to patients, trainees, and other staff. Key Words: Fluoroscopy time, radiation exposure, radiology resident training, jugular central venous catheterization J Am Coll Radiol 2013;10:518-522. Copyright © 2013 American College of Radiology
INTRODUCTION
Increased professional and public scrutiny has led to continued efforts to monitor and minimize medical radiation [1-4]. Various short- and long-term stochastic and deterministic effects associated with prolonged medical radiation are well recognized. Fluoroscopically guided interventional radiology (IR) procedures have been reported to cause radiation-induced injuries to skin and
a
University of Tennessee Health Science Center, Memphis, Tennessee. Baptist Memorial Hospital, Memphis, Tennessee. c St. Jude Children’s Research Hospital, Memphis, Tennessee. Corresponding author and reprints: Baogang J. Xu, PhD, 920 Madison Ave, College of Medicine, University of Tennessee Health Science Center, Memphis, TN 38163; e-mail: [email protected]. b
hair when high doses of radiation are delivered [5-7]. Cancer risk from medical radiation remains a concern, despite the advancement of digital equipment and newer techniques [8-10]. At the same time, IR continues to serve as a dynamic field with increasing volume and diversity, with meaningful contributions to patient care. Balancing the treatment and potential adverse effects of medical radiation in various settings, especially in academic training programs, is of vital importance to both patients and operators. At training institutions, radiology residents frequently serve as the primary operators for various, often minor, interventional procedures. Anecdotally, central venous catheter (CVC) procedures are among the most common. The associated radiation exposure to patients undergoing procedures by residents compared with ex-
518 0091-2182/13/$36.00
●
© 2013 American College of Radiology http://dx.doi.org/10.1016/j.jacr.2012.12.023
Xu et al/Increased Fluoroscopy Time 519
posures by staff radiologists has received little attention. Similarly, minimization of radiation exposure to trainees themselves might not be a high priority in many programs. For some general surgical procedures, such as in laparoscopic appendectomy, laparoscopic cholecystectomy, or open inguinal hernia repair, the impact of surgical residents on operative time has been examined [11,12]. To our knowledge, however, no study has been performed in the field of radiology to quantitatively measure differences in operative time and related fluoroscopy time (FT) associated with invasive procedures performed by radiology residents. Furthermore, it is not known if the duration of FT would change for radiology residents at different levels of training or supervised by different staff radiologists. A clear understanding of such information would aid residency training programs in appropriately monitoring and potentially minimizing FTs during interventional procedures. This knowledge could thus advance both training and policies to minimize dose to patients, trainees, and other hospital staff. In this study, we aim to evaluate differences in FT for IR procedures performed by radiology residents versus staff radiologists. Because imaging-guided CVC placement by radiologists has increased dramatically over recent years [13] and is commonly performed by residents in our community, we used temporary jugular CVC procedures as our model. The FT differences among residents with different years of training and under different supervising staff radiologists were also investigated.
confounding procedural complexity variables, clinical exclusions were implemented. We excluded the following cases from analysis: (1) implanted devices (eg, tunneled catheters, implanted subcutaneous venous access ports); (2) cases performed in conjunction with other procedures (eg, IVC filter); and (3) cases with documented venous occlusion or stenosis requiring angioplasty, stenting, or other forms of recanalization. Statistical Analysis
Reported FTs between the resident and staff group were first compared using Wilcoxon’s rank-sum test. Wilcoxon’s rank-sum test is a preferred nonparametric alternative to the two-sample t test when data distribution is not normal and is based on the rank order in which the data from the 2 groups fall. We next aimed to test whether increasing the number of years in residency training reduces FT. A linear regression was fitted for FT using the years of residency training to evaluate possible correlation. Finally, because the procedures performed by the residents were supervised by different staff radiologists, we tested whether particular supervising staff radiologists impacted trainee FTs using the Kruskal-Wallis rank-sum test. As an extension of Wilcoxon’s rank-sum test, Kruskal-Wallis rank-sum test allows the comparison of more than 2 independent groups [15]. All of the statistical analysis was performed using open-source R software (version 2.15.1). RESULTS
METHODS Data Source and Included Cases
The index cases analyzed in this study were obtained from a Protected Health Information (PHI)–redacted database, which was initially established as part of a professional quality improvement project by interventional radiologists for ABR Maintenance of Certification [14]. At our institution, retrospective analysis of PHI-redacted quality improvement files is exempt from the Institutional Review Board. All CVC cases performed in IR suites at 2 large metropolitan teaching hospitals in 2008 and 2009 were included in the professional quality improvement database. Database fields included PHI-redacted patient numerical identifiers and identifiers for hospital site, resident, and staff. Also included were documented FT, catheter insertion vein site, side, and type of catheter (eg, large lumen hemodialysis versus smaller triple lumen). To minimize service heterogeneity, anatomic exclusions were implemented. Only isolated temporary jugular central venous catheterization procedures were included for this analysis (eg, direct inferior vena cava, hepatic venous, femoral venous, and subclavian venous access were excluded). Additionally, to remove
Case Files and Data Collection
A total of 1,067 temporary central venous access cases were identified. Applying the described clinical and anatomic exclusions criteria, 537 cases with sufficiently complete records (ie, procedure report with documented FT) were available for analysis. Of these cases, 128 cases were performed by 19 residents who were supervised by 5 different staff radiologists; 409 cases were performed by 7 staff radiologists with no resident assistance. Similar distributions of operative approaches were observed among the resident and staff radiologist groups. Specifically, resident cases included 105 (82%) right and 23 (18%) left approaches, whereas staff cases included 324 (79%) right and 85 (21%) left approaches. In addition, resident cases included 85 (66%) larger lumen and 43 (34%) smaller lumen catheters, whereas staff cases included 258 (63%) larger lumen and 151 (37%) smaller lumen catheters. FT Comparison of Residents versus Staff Radiologists
FT for resident procedures ranged from .03 to 8.20 minutes (mean, 1.24; SD, 1.39). FT for staff radiologist procedures was significantly shorter (P ⬍ .0001) and ranged from .01 to 5.80 minutes (mean, .63; SD, .76).
8
8
520 Journal of the American College of Radiology/ Vol. 10 No. 7 July 2013
P = 0.24
0
0
2
2
Minutes 4
Minutes 4
6
6
P < 0.0001
Staff Radiologists
Radiology Residents
Fig. 1. The box-and-whisker plot of the FT for the staff radiologist group versus resident group. A significant difference was observed in the FT between the 2 groups (P ⬍ .0001). The median FT for the staff radiologists (.4 minutes) was significantly lower than that of the residents (.8 minutes).
A
B
C
D
E
Staff Radiologists Fig. 3. The box-and-whisker plot of the FT for the resident groups with 5 different supervising staff radiologists. No significant differences in FT were observed among the different supervising staff radiologists (P ⫽ .24), suggesting that the required FT in the residents group was independent of the supervising staff radiologist.
Box-and-whisker plots of the FT for the resident group versus staff group are illustrated in Figure 1. FT Comparison of Residents With Different Postgraduate Year Levels
8
The number of procedures performed by residents in postgraduate years 2 through 5 were 37, 17, 27, and 47, respectively. The mean FTs were 1.16, 1.45, 1.25, and 1.21, respectively. As illustrated in Figure 2, no linear
Minutes 4
6
P = 0.96
correlation was observed between FT and residents at different levels of training (P ⫽ .96), indicating increasing years of training for residents did not change FT. Comparison of Resident FT by Different Supervising Staff Radiologists
The number of resident-performed procedures supervised by the 5 different staff radiologists were as follows: 28 by staff member A, 2 by staff member B, 19 by staff member C, 8 by staff member D, and 71 by staff member E. The mean FTs were 1.12, .70, .80, 1.08, and 1.43, respectively. As shown in Figure 3, no significant differences in FT were observed among procedures by different supervising staff radiologists (P ⫽ .24). The result suggests that FTs were independent of supervising staff radiologist.
0
2
DISCUSSION
2
3
4
5
Resident PGY Years Fig. 2. The box-and-whisker plot of the FT for the resident groups with 4 different postgraduate year (PGY) levels. No linear correlation was observed between the FT and the resident groups with different years of training (P ⫽ .96), showing increasing years of training for residents did not change the FT.
We have demonstrated that the involvement of radiology residents as primary operators doubled the FT associated with CVC placement procedures compared with the FT associated with the same procedures completed by attending radiologists. Among residents, however, FT was independent of specific supervising staff radiologists. Furthermore, no linear correlation between FT and resident seniority was observed, suggesting increasing years of training for residents does not reduce the FT. In explaining the differences between resident and staff FT, other factors such as fellowship training, experiencebased increased awareness of radiation dose tactics, or overall higher career procedure volumes may all contrib-
Xu et al/Increased Fluoroscopy Time 521
ute. Future analyses will be necessary to more definitively explain our observations. For fluoroscopic-guided procedures, FT is likely a surrogate of overall procedure time. The finding of longer FT for resident procedures when compared with staff radiologist procedures without resident involvement is not unexpected. Nonetheless, this study represents the first such report of increased FT (and thus, perhaps radiation dose) in services performed by residents in radiology. As trainees develop procedure expertise, their postgraduate medical training necessarily consumes more operative time than that of more experienced physicians. For radiologists, who perform nearly all procedures with imaging guidance, their “operative time” often involves ionizing radiation, as in the case of CVC placement. In an era of enlightened concerns for radiation exposure to patients and staff, those increased FTs could conceivably carry with them a very small increased risk. Obtaining quantitative data on FT associated with resident training would be necessary to more completely address such issues with regard to residency program training policy development. Furthermore, we point out that having different supervising staff radiologists did not alter the fact that residents require longer operative time when residents were the primary operators. Therefore, increased FT by residents appears to be related to the process of training itself and does not seem to be dependent on particular supervising staff radiologists. We also identified that there was no linear correlation between the length of FT and the residents’ seniority, suggesting resident seniority did not reduce the FT. This finding appears in contrast to the concept that more experienced surgeons usually operate with higher efficiency and increasing speed. However, the finding could be due to the supervising radiologists spending fixed amounts of time in teaching, regardless of the residents’ seniority levels (ie, perhaps attending physicians are more hands-on with junior residents and grant more senior residents more individual procedural autonomy). Our finding is consistent with previous studies indicating no significant differences in operative time for junior and senior surgical residents performing laparoscopic surgical procedures [12,16]. Four patient radiation dose metrics for interventional radiology fluoroscopy have been described [17]: (1) peak skin dose, (2) reference air kerma, (3) kerma-area product, and (4) FT. Of these, FT is most widely reported. This may be related to the fact that it is the simplest metric to acquire, and also because its reporting has been used as the basis for Medicare physician pay-for-performance bonuses [18]. By itself, FT is insufficient to determine patient or operator radiation dose but nonetheless serves as a common analogue of dose [17]. For these reasons, FT was a practical metric for this initial report. However, we would anticipate that future research into differential radiation exposure associated with radiology
resident training would include additional more rigorous metrics, when available. By design, we focused on only a very specific service— the fluoroscopic-guided IR procedure performed most commonly by residents in our training program. This represents the experience at only a single residency program, which reflects the biggest limitation of our initial analysis. Nonetheless, we believe that the findings are important, given increased emphasis in recent years on patient radiation dose reduction and trainee safety. Although we suspect that our findings may be generalized to CVC placement by residents at other facilities, and to other fluoroscopic-guided procedures performed by radiology residents in general, larger scale analyses will be necessary. Such studies will require considerably more laborious report mining. At the time our database was created, natural language processing algorithms for mining large volumes of radiology reports were largely proprietary [19], but as these become increasingly commercially available and more widely used within the radiology domain [20], they may considerably facilitate future analyses. Overall, the present study found that when CVCs are placed by radiology residents, associated FT is double that for identical procedures performed by staff radiologists. Although not part of our analysis, similar discrepancies likely also exist for other IR procedures. Because of patient, trainee, and overall staff safety implications, residency training programs should initiate measures to monitor and manage FT during interventional procedures, to minimize dose as much as possible, in keeping with the “as low as reasonably achievable” principle. TAKE-HOME POINTS ●
● ●
●
When radiology residents place CVC, the associated FT is double that for identical procedures performed by staff radiologists. Increasing years of training for residents does not significantly reduce FT for CVC procedures. Supervision by different staff radiologists does not alter the fact that residents require longer FT as primary operators. Because of patient, trainee, and overall staff safety implications, residency training programs should initiate measures to monitor and manage FT during interventional procedures to minimize dose as much as possible, in conformance with “as low as reasonably achievable” principles.
REFERENCES 1. Vano E. Global view on radiation protection in medicine. Radiat Prot Dosimetry 2011;147:3-7. 2. Rehani MM, Holmberg O, Ortiz Lopez P, Mettler F. International action plan on the radiation protection of patients. Radiat Prot Dosimetry 2011; 147:38-42. 3. Archer BR. High-dose fluoroscopy: the administrator’s responsibilities. Radiol Manage 2002;24:26-32; quiz 33-5.
522 Journal of the American College of Radiology/ Vol. 10 No. 7 July 2013 4. Archer BR, Wagner LK. Protecting patients by training physicians in fluoroscopic radiation management. J Appl Clin Med Phys 2000;1:32-7. 5. Koenig TR, Wolff D, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures: part 1, characteristics of radiation injury. AJR Am J Roentgenol 2001;177:3-11. 6. Koenig TR, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures: part 2, review of 73 cases and recommendations for minimizing dose delivered to patient. AJR Am J Roentgenol 2001;177: 13-20.
13. Reeves AR, Seshadri R, Trerotola SO. Recent trends in central venous catheter placement: a comparison of interventional radiology with other specialties. J Vasc Interv Radiol 2001;12:1211-4. 14. Strife JL, Kun LE, Becker GJ, Dunnick NR, Bosma J, Hattery RR. The American Board of Radiology perspective on maintenance of certification: part IV–practice quality improvement in diagnostic radiology. J Am Coll Radiol 2007;4:300-4. 15. Kruskal WH, Wallis WA. Use of Ranks in One-Criterion Variance Analysis. J Am Stat Assoc 1952;47:583-621.
7. Balter S, Hopewell JW, Miller DL, Wagner LK, Zelefsky MJ. Fluoroscopically guided interventional procedures: a review of radiation effects on patients’ skin and hair. Radiology 2010;254:326-41.
16. Tseng WH, Jin L, Canter RJ, et al. Surgical resident involvement is safe for common elective general surgery procedures. J Am Coll Surg 2011; 213:19-26; discussion 26-8.
8. Einstein AJ. Beyond the bombs: cancer risks of low-dose medical radiation. Lancet 2012;380:455-7.
17. Miller DL, Balter S, Dixon RG, et al. Quality improvement guidelines for recording patient radiation dose in the medical record for fluoroscopically guided procedures. J Vasc Interv Radiol 2012;23:11-8.
9. Linet MS, Kim KP, Miller DL, Kleinerman RA, Simon SL, Berrington de Gonzalez A. Historical review of occupational exposures and cancer risks in medical radiation workers. Radiat Res 2010;174:793-808. 10. Ron E. Cancer risks from medical radiation. Health Phys 2003;85:47-59.
18. Duszak R, Jr. P4P: pragmatic for practice. J Am Coll Radiol 2009;6: 477-8.
11. Babineau TJ, Becker J, Gibbons G, et al. The “cost” of operative training for surgical residents. Arch Surg 2004;139:366-9; discussion 369-70.
19. Sistrom CL, Dreyer KJ, Dang PP, et al. Recommendations for additional imaging in radiology reports: multifactorial analysis of 5.9 million examinations. Radiology 2009;253:453-61.
12. Papandria D, Rhee D, Ortega G, et al. Assessing trainee impact on operative time for common general surgical procedures in ACS-NSQIP. J Surg Educ 2012;69:149-55.
20. Duszak R, Jr., Nossal M, Schofield L, Picus D. Physician documentation deficiencies in abdominal ultrasound reports: frequency, characteristics, and financial impact. J Am Coll Radiol 2012;9:403-8.
Purpose: To evaluate differences in interventional radiology procedural fluoroscopy time (FT) for radiology residents versus staff radiologists, using central venous catheter (CVC) placement as an index service. Methods: To minimize interservice and complexity variables, stand-alone temporary internal jugular CVC procedures were targeted for analysis. Reports and images from 1,067 temporary CVC services from 2 hospitals over 2 years were reviewed as part of a quality improvement initiative. Insertion site, catheter type (eg, smaller triple lumen versus larger hemodialysis), resident identifier, staff identifier, and documented FT were compiled and analyzed. Results: Applying clinical (eg, concomitant venous angioplasty) and anatomic (eg, femoral access) exclusions, 537 cases with complete CVC procedure records were available for analysis. Radiology residents and staff radiologists were primary operators in 128 and 409 procedures, respectively. Distribution of resident procedures (82% right, 66% large lumen) was similar to that of staff (79% right, 63% large lumen). Mean FT of resident services was twice as long as that of staff services (1.24 minutes versus 0.63 minutes, P ⬍ .0001). Resident FT was independent of supervising staff radiologist. Increasing years of training for residents did not significantly reduce FT. Conclusions: When CVCs are placed by radiology residents, FT is double that for identical procedures performed by staff radiologists. Similar discrepancies likely exist for other interventional radiologic procedures. Residency training programs should initiate measures to monitor and manage fluoroscopy during interventional procedures to minimize radiation dose to patients, trainees, and other staff. Key Words: Fluoroscopy time, radiation exposure, radiology resident training, jugular central venous catheterization J Am Coll Radiol 2013;10:518-522. Copyright © 2013 American College of Radiology
INTRODUCTION
Increased professional and public scrutiny has led to continued efforts to monitor and minimize medical radiation [1-4]. Various short- and long-term stochastic and deterministic effects associated with prolonged medical radiation are well recognized. Fluoroscopically guided interventional radiology (IR) procedures have been reported to cause radiation-induced injuries to skin and
a
University of Tennessee Health Science Center, Memphis, Tennessee. Baptist Memorial Hospital, Memphis, Tennessee. c St. Jude Children’s Research Hospital, Memphis, Tennessee. Corresponding author and reprints: Baogang J. Xu, PhD, 920 Madison Ave, College of Medicine, University of Tennessee Health Science Center, Memphis, TN 38163; e-mail: [email protected]. b
hair when high doses of radiation are delivered [5-7]. Cancer risk from medical radiation remains a concern, despite the advancement of digital equipment and newer techniques [8-10]. At the same time, IR continues to serve as a dynamic field with increasing volume and diversity, with meaningful contributions to patient care. Balancing the treatment and potential adverse effects of medical radiation in various settings, especially in academic training programs, is of vital importance to both patients and operators. At training institutions, radiology residents frequently serve as the primary operators for various, often minor, interventional procedures. Anecdotally, central venous catheter (CVC) procedures are among the most common. The associated radiation exposure to patients undergoing procedures by residents compared with ex-
518 0091-2182/13/$36.00
●
© 2013 American College of Radiology http://dx.doi.org/10.1016/j.jacr.2012.12.023
Xu et al/Increased Fluoroscopy Time 519
posures by staff radiologists has received little attention. Similarly, minimization of radiation exposure to trainees themselves might not be a high priority in many programs. For some general surgical procedures, such as in laparoscopic appendectomy, laparoscopic cholecystectomy, or open inguinal hernia repair, the impact of surgical residents on operative time has been examined [11,12]. To our knowledge, however, no study has been performed in the field of radiology to quantitatively measure differences in operative time and related fluoroscopy time (FT) associated with invasive procedures performed by radiology residents. Furthermore, it is not known if the duration of FT would change for radiology residents at different levels of training or supervised by different staff radiologists. A clear understanding of such information would aid residency training programs in appropriately monitoring and potentially minimizing FTs during interventional procedures. This knowledge could thus advance both training and policies to minimize dose to patients, trainees, and other hospital staff. In this study, we aim to evaluate differences in FT for IR procedures performed by radiology residents versus staff radiologists. Because imaging-guided CVC placement by radiologists has increased dramatically over recent years [13] and is commonly performed by residents in our community, we used temporary jugular CVC procedures as our model. The FT differences among residents with different years of training and under different supervising staff radiologists were also investigated.
confounding procedural complexity variables, clinical exclusions were implemented. We excluded the following cases from analysis: (1) implanted devices (eg, tunneled catheters, implanted subcutaneous venous access ports); (2) cases performed in conjunction with other procedures (eg, IVC filter); and (3) cases with documented venous occlusion or stenosis requiring angioplasty, stenting, or other forms of recanalization. Statistical Analysis
Reported FTs between the resident and staff group were first compared using Wilcoxon’s rank-sum test. Wilcoxon’s rank-sum test is a preferred nonparametric alternative to the two-sample t test when data distribution is not normal and is based on the rank order in which the data from the 2 groups fall. We next aimed to test whether increasing the number of years in residency training reduces FT. A linear regression was fitted for FT using the years of residency training to evaluate possible correlation. Finally, because the procedures performed by the residents were supervised by different staff radiologists, we tested whether particular supervising staff radiologists impacted trainee FTs using the Kruskal-Wallis rank-sum test. As an extension of Wilcoxon’s rank-sum test, Kruskal-Wallis rank-sum test allows the comparison of more than 2 independent groups [15]. All of the statistical analysis was performed using open-source R software (version 2.15.1). RESULTS
METHODS Data Source and Included Cases
The index cases analyzed in this study were obtained from a Protected Health Information (PHI)–redacted database, which was initially established as part of a professional quality improvement project by interventional radiologists for ABR Maintenance of Certification [14]. At our institution, retrospective analysis of PHI-redacted quality improvement files is exempt from the Institutional Review Board. All CVC cases performed in IR suites at 2 large metropolitan teaching hospitals in 2008 and 2009 were included in the professional quality improvement database. Database fields included PHI-redacted patient numerical identifiers and identifiers for hospital site, resident, and staff. Also included were documented FT, catheter insertion vein site, side, and type of catheter (eg, large lumen hemodialysis versus smaller triple lumen). To minimize service heterogeneity, anatomic exclusions were implemented. Only isolated temporary jugular central venous catheterization procedures were included for this analysis (eg, direct inferior vena cava, hepatic venous, femoral venous, and subclavian venous access were excluded). Additionally, to remove
Case Files and Data Collection
A total of 1,067 temporary central venous access cases were identified. Applying the described clinical and anatomic exclusions criteria, 537 cases with sufficiently complete records (ie, procedure report with documented FT) were available for analysis. Of these cases, 128 cases were performed by 19 residents who were supervised by 5 different staff radiologists; 409 cases were performed by 7 staff radiologists with no resident assistance. Similar distributions of operative approaches were observed among the resident and staff radiologist groups. Specifically, resident cases included 105 (82%) right and 23 (18%) left approaches, whereas staff cases included 324 (79%) right and 85 (21%) left approaches. In addition, resident cases included 85 (66%) larger lumen and 43 (34%) smaller lumen catheters, whereas staff cases included 258 (63%) larger lumen and 151 (37%) smaller lumen catheters. FT Comparison of Residents versus Staff Radiologists
FT for resident procedures ranged from .03 to 8.20 minutes (mean, 1.24; SD, 1.39). FT for staff radiologist procedures was significantly shorter (P ⬍ .0001) and ranged from .01 to 5.80 minutes (mean, .63; SD, .76).
8
8
520 Journal of the American College of Radiology/ Vol. 10 No. 7 July 2013
P = 0.24
0
0
2
2
Minutes 4
Minutes 4
6
6
P < 0.0001
Staff Radiologists
Radiology Residents
Fig. 1. The box-and-whisker plot of the FT for the staff radiologist group versus resident group. A significant difference was observed in the FT between the 2 groups (P ⬍ .0001). The median FT for the staff radiologists (.4 minutes) was significantly lower than that of the residents (.8 minutes).
A
B
C
D
E
Staff Radiologists Fig. 3. The box-and-whisker plot of the FT for the resident groups with 5 different supervising staff radiologists. No significant differences in FT were observed among the different supervising staff radiologists (P ⫽ .24), suggesting that the required FT in the residents group was independent of the supervising staff radiologist.
Box-and-whisker plots of the FT for the resident group versus staff group are illustrated in Figure 1. FT Comparison of Residents With Different Postgraduate Year Levels
8
The number of procedures performed by residents in postgraduate years 2 through 5 were 37, 17, 27, and 47, respectively. The mean FTs were 1.16, 1.45, 1.25, and 1.21, respectively. As illustrated in Figure 2, no linear
Minutes 4
6
P = 0.96
correlation was observed between FT and residents at different levels of training (P ⫽ .96), indicating increasing years of training for residents did not change FT. Comparison of Resident FT by Different Supervising Staff Radiologists
The number of resident-performed procedures supervised by the 5 different staff radiologists were as follows: 28 by staff member A, 2 by staff member B, 19 by staff member C, 8 by staff member D, and 71 by staff member E. The mean FTs were 1.12, .70, .80, 1.08, and 1.43, respectively. As shown in Figure 3, no significant differences in FT were observed among procedures by different supervising staff radiologists (P ⫽ .24). The result suggests that FTs were independent of supervising staff radiologist.
0
2
DISCUSSION
2
3
4
5
Resident PGY Years Fig. 2. The box-and-whisker plot of the FT for the resident groups with 4 different postgraduate year (PGY) levels. No linear correlation was observed between the FT and the resident groups with different years of training (P ⫽ .96), showing increasing years of training for residents did not change the FT.
We have demonstrated that the involvement of radiology residents as primary operators doubled the FT associated with CVC placement procedures compared with the FT associated with the same procedures completed by attending radiologists. Among residents, however, FT was independent of specific supervising staff radiologists. Furthermore, no linear correlation between FT and resident seniority was observed, suggesting increasing years of training for residents does not reduce the FT. In explaining the differences between resident and staff FT, other factors such as fellowship training, experiencebased increased awareness of radiation dose tactics, or overall higher career procedure volumes may all contrib-
Xu et al/Increased Fluoroscopy Time 521
ute. Future analyses will be necessary to more definitively explain our observations. For fluoroscopic-guided procedures, FT is likely a surrogate of overall procedure time. The finding of longer FT for resident procedures when compared with staff radiologist procedures without resident involvement is not unexpected. Nonetheless, this study represents the first such report of increased FT (and thus, perhaps radiation dose) in services performed by residents in radiology. As trainees develop procedure expertise, their postgraduate medical training necessarily consumes more operative time than that of more experienced physicians. For radiologists, who perform nearly all procedures with imaging guidance, their “operative time” often involves ionizing radiation, as in the case of CVC placement. In an era of enlightened concerns for radiation exposure to patients and staff, those increased FTs could conceivably carry with them a very small increased risk. Obtaining quantitative data on FT associated with resident training would be necessary to more completely address such issues with regard to residency program training policy development. Furthermore, we point out that having different supervising staff radiologists did not alter the fact that residents require longer operative time when residents were the primary operators. Therefore, increased FT by residents appears to be related to the process of training itself and does not seem to be dependent on particular supervising staff radiologists. We also identified that there was no linear correlation between the length of FT and the residents’ seniority, suggesting resident seniority did not reduce the FT. This finding appears in contrast to the concept that more experienced surgeons usually operate with higher efficiency and increasing speed. However, the finding could be due to the supervising radiologists spending fixed amounts of time in teaching, regardless of the residents’ seniority levels (ie, perhaps attending physicians are more hands-on with junior residents and grant more senior residents more individual procedural autonomy). Our finding is consistent with previous studies indicating no significant differences in operative time for junior and senior surgical residents performing laparoscopic surgical procedures [12,16]. Four patient radiation dose metrics for interventional radiology fluoroscopy have been described [17]: (1) peak skin dose, (2) reference air kerma, (3) kerma-area product, and (4) FT. Of these, FT is most widely reported. This may be related to the fact that it is the simplest metric to acquire, and also because its reporting has been used as the basis for Medicare physician pay-for-performance bonuses [18]. By itself, FT is insufficient to determine patient or operator radiation dose but nonetheless serves as a common analogue of dose [17]. For these reasons, FT was a practical metric for this initial report. However, we would anticipate that future research into differential radiation exposure associated with radiology
resident training would include additional more rigorous metrics, when available. By design, we focused on only a very specific service— the fluoroscopic-guided IR procedure performed most commonly by residents in our training program. This represents the experience at only a single residency program, which reflects the biggest limitation of our initial analysis. Nonetheless, we believe that the findings are important, given increased emphasis in recent years on patient radiation dose reduction and trainee safety. Although we suspect that our findings may be generalized to CVC placement by residents at other facilities, and to other fluoroscopic-guided procedures performed by radiology residents in general, larger scale analyses will be necessary. Such studies will require considerably more laborious report mining. At the time our database was created, natural language processing algorithms for mining large volumes of radiology reports were largely proprietary [19], but as these become increasingly commercially available and more widely used within the radiology domain [20], they may considerably facilitate future analyses. Overall, the present study found that when CVCs are placed by radiology residents, associated FT is double that for identical procedures performed by staff radiologists. Although not part of our analysis, similar discrepancies likely also exist for other IR procedures. Because of patient, trainee, and overall staff safety implications, residency training programs should initiate measures to monitor and manage FT during interventional procedures, to minimize dose as much as possible, in keeping with the “as low as reasonably achievable” principle. TAKE-HOME POINTS ●
● ●
●
When radiology residents place CVC, the associated FT is double that for identical procedures performed by staff radiologists. Increasing years of training for residents does not significantly reduce FT for CVC procedures. Supervision by different staff radiologists does not alter the fact that residents require longer FT as primary operators. Because of patient, trainee, and overall staff safety implications, residency training programs should initiate measures to monitor and manage FT during interventional procedures to minimize dose as much as possible, in conformance with “as low as reasonably achievable” principles.
REFERENCES 1. Vano E. Global view on radiation protection in medicine. Radiat Prot Dosimetry 2011;147:3-7. 2. Rehani MM, Holmberg O, Ortiz Lopez P, Mettler F. International action plan on the radiation protection of patients. Radiat Prot Dosimetry 2011; 147:38-42. 3. Archer BR. High-dose fluoroscopy: the administrator’s responsibilities. Radiol Manage 2002;24:26-32; quiz 33-5.
522 Journal of the American College of Radiology/ Vol. 10 No. 7 July 2013 4. Archer BR, Wagner LK. Protecting patients by training physicians in fluoroscopic radiation management. J Appl Clin Med Phys 2000;1:32-7. 5. Koenig TR, Wolff D, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures: part 1, characteristics of radiation injury. AJR Am J Roentgenol 2001;177:3-11. 6. Koenig TR, Mettler FA, Wagner LK. Skin injuries from fluoroscopically guided procedures: part 2, review of 73 cases and recommendations for minimizing dose delivered to patient. AJR Am J Roentgenol 2001;177: 13-20.
13. Reeves AR, Seshadri R, Trerotola SO. Recent trends in central venous catheter placement: a comparison of interventional radiology with other specialties. J Vasc Interv Radiol 2001;12:1211-4. 14. Strife JL, Kun LE, Becker GJ, Dunnick NR, Bosma J, Hattery RR. The American Board of Radiology perspective on maintenance of certification: part IV–practice quality improvement in diagnostic radiology. J Am Coll Radiol 2007;4:300-4. 15. Kruskal WH, Wallis WA. Use of Ranks in One-Criterion Variance Analysis. J Am Stat Assoc 1952;47:583-621.
7. Balter S, Hopewell JW, Miller DL, Wagner LK, Zelefsky MJ. Fluoroscopically guided interventional procedures: a review of radiation effects on patients’ skin and hair. Radiology 2010;254:326-41.
16. Tseng WH, Jin L, Canter RJ, et al. Surgical resident involvement is safe for common elective general surgery procedures. J Am Coll Surg 2011; 213:19-26; discussion 26-8.
8. Einstein AJ. Beyond the bombs: cancer risks of low-dose medical radiation. Lancet 2012;380:455-7.
17. Miller DL, Balter S, Dixon RG, et al. Quality improvement guidelines for recording patient radiation dose in the medical record for fluoroscopically guided procedures. J Vasc Interv Radiol 2012;23:11-8.
9. Linet MS, Kim KP, Miller DL, Kleinerman RA, Simon SL, Berrington de Gonzalez A. Historical review of occupational exposures and cancer risks in medical radiation workers. Radiat Res 2010;174:793-808. 10. Ron E. Cancer risks from medical radiation. Health Phys 2003;85:47-59.
18. Duszak R, Jr. P4P: pragmatic for practice. J Am Coll Radiol 2009;6: 477-8.
11. Babineau TJ, Becker J, Gibbons G, et al. The “cost” of operative training for surgical residents. Arch Surg 2004;139:366-9; discussion 369-70.
19. Sistrom CL, Dreyer KJ, Dang PP, et al. Recommendations for additional imaging in radiology reports: multifactorial analysis of 5.9 million examinations. Radiology 2009;253:453-61.
12. Papandria D, Rhee D, Ortega G, et al. Assessing trainee impact on operative time for common general surgical procedures in ACS-NSQIP. J Surg Educ 2012;69:149-55.
20. Duszak R, Jr., Nossal M, Schofield L, Picus D. Physician documentation deficiencies in abdominal ultrasound reports: frequency, characteristics, and financial impact. J Am Coll Radiol 2012;9:403-8.