Radiation Exposure and Case Characteristics in National Sample of Female Orthopaedic Trauma and Arthroplasty Surgeons

Abstract

Background:

The risks of radiation exposure in orthopaedic surgery have become a topic of increasing interest in the setting of widespread fluoroscopy use and concern for an increased prevalence of breast cancer among female orthopaedic surgeons. The aim of this national study of 31 female orthopaedic surgeons was to achieve a deeper understanding of fluoroscopic use in the OR and its associated exposure to radiation, by comparing female orthopaedic trauma and arthroplasty surgeons.

Methods:

A total of 31 surgeons wore dosimeters for 10 operating days each to track cumulative radiation exposure. Surgeons were not asked to modify their practice in any way, with no requirement that the operating days had to be chosen with the knowledge that fluoroscopy would be used. Participants were also asked to fill out a form at the end of each day, detailing the number of cases that day, the number of hours spent in the OR, and the total amount of time using fluoroscopy.

Results:

Trauma surgeons received significantly higher radiation doses in the OR (p=0.01) and reported longer use of fluoroscopy (p<0.001). Trauma surgeons also spent more time per day in the OR and had more cases per day compared to arthroplasty surgeons, but this difference was not significant. Radiation dose penetrating through protective equipment remained minimal.

Conclusion:

Although the female trauma surgeons in the study operated longer and performed more procedures per day, the higher radiation exposure was best explained by the amount of time fluoroscopy is used in the OR. The fluoroscopic times in this study therefore may be a useful self-assessment tool for attending trauma and arthroplasty surgeons. Awareness of these differences will hopefully increase an individual surgeon’s mindfulness toward the length of fluoroscopy use in each case, regardless of orthopaedic subspecialty.

Level of Evidence: IV

Introduction

The widespread adoption of fluoroscopy in orthopaedic procedures has led to increasing interest in evaluating the risks of occupational radiation exposure to the surgeon. Although evidence directly linking cancer prevalence to radiation exposure among orthopaedic surgeons is minimal, it is well known that a dose-dependent relationship exists between radiation exposure and malignancy.^1-3 Female orthopaedic surgeons, in particular, have a 2.9 fold higher prevalence of breast cancer compared to the general US population, as well as an increased prevalence relative to surgeons in other surgical specialties.^4,5 Similarly, a 3-fold higher risk of breast cancer has been reported in female radiographic technologists with exposure to long-term, low-dose radiation.⁶

Practice guidelines have been developed to mitigate this exposure, with maximum acceptable radiation limits set by the National Council on Radiation Protection and the United States Nuclear Regulatory Commission (USNRC). Methods of reducing radiation risk include wearing protective equipment such as lead aprons and thyroid shields, using optimal intra-op fluoroscopic positioning, using smaller fluoroscopy machines, and standing a safe distance from the source. Many institutions require surgeons to wear dosimeters clipped above lead aprons to track intra-operative radiation exposure, although the recorded doses are not a true reflection of the actual radiation that is received underneath the lead. Indeed, a recent study that used several standard lead apron sizes on an anthropomorphic model demonstrated substantial scatter radiation to the breast region.⁷

Several studies track radiation exposure secondary to specific procedures or fluoroscopic techniques, but differences between practices and individual surgeons also limits the generalizability of these studies.^8-11 Gausden et. al. tracked cumulative radiation dose in one year among various orthopaedic subspecialties and across several training levels at a single institution, demonstrating that trauma surgeons were exposed to the highest amounts of radiation and arthroplasty surgeons the least.¹² The relationship between operating room (OR) data such as total fluoroscopy time, number of procedures performed per day by a surgeon, and radiation exposure, was explored in one previous study but focused on hand radiation exposure and was limited to one institution.¹³ Given the increased prevalence of breast cancer in female orthopaedic surgeons relative to the general US population and even other surgical specialties, radiation exposure in the context of operating room characteristics has significant implications for limiting occupational risk in this population. Therefore, further research is needed to gain a deeper understanding in why radiation exposure varies among female orthopaedic surgeons in different specialties, specifically how OR data translate into fluoroscopy use.

Thus, the purpose of this study was to achieve a deeper understanding of fluoroscopic use in the OR and its associated exposure to radiation comparing female orthopaedic trauma and arthroplasty surgeons. The hypothesis was that female trauma surgeons would spend more time in the OR with surgical procedures, and have increased fluoroscopy exposure leading to overall higher radiation exposure compared to arthroplasty surgeons. The primary outcome measure was the radiation dose measured by dosimeters, as well as the practice parameters of the surgeons involved in the study.

Methods

Data Collection

After obtaining Institutional Review Board (IRB) approval, 40 female orthopaedic trauma and arthroplasty surgeons in various practice settings across the United States were recruited using personal contacts and word-of-mouth recommendations from colleagues in the field. A few participants also reached out directly to ask to be enrolled. A total of 31 surgeons competed the study (Figure 1).

Figure 1.

Institutions represented in a sample of 31 female orthopaedic surgeons in the US.

All participants were mailed two dosimeters provided by Stanford University’s Environmental Health and Safety Department. The dosimeters were from Mirion Technologies (Model: Genesis Ultra TLD) and were the same type worn by surgeons at Stanford University Hospital, who use fluoroscopy in their practice, to track cumulative radiation exposure. Participants were asked to wear one dosimeter on the outside of their lead apron, and one dosimeter on the inside of the lead apron clipped to the pocket of the scrub top. These dosimeters were worn for a total of 10 operating days. With the exception of wearing the dosimeters, surgeons were not asked to modify their practice in any way. There was no requirement that the operating days had to be chosen with the knowledge that fluoroscopy would be used.

Participants were also asked to fill out a paper form at the end of each day, detailing the number of cases that day, the number of hours spent in the OR, and the total amount of time using fluoroscopy. To gather the most accurate data, the fluoroscopy time was recorded directly from the device used for fluoroscopy in each case. Demographic information on the type of institution (academic versus private) and use of personal protective equipment (PPE), i.e. axillary protection wings, was also collected. At the end of data collection, participants mailed back both dosimeters and paper forms.

Statistical Analysis

Frequency and percentage or means and standard deviation were calculated for each measure, separately by surgeon type (trauma versus arthroplasty). Differences between surgeon type were tested using Fisher’s exact test for personal protective equipment and Student’s t-tests for cases/day, operating time/day, fluoroscopy use/day, and outside dosimeter reading. Correlation tests were used to examine if operating time correlated with fluoroscopy use and if fluoroscopy use correlated with outside dosimeter reading. Correlations were calculated separately for each surgeon type and a Fisher’s (1925) z test was used to examine if the correlation coefficients were significantly different between the two groups. Exploratory univariate and multivariate linear regression tests were used to examine surgeon characteristics of surgeon type, practice setting, axillary protection wing use, number of procedures, and operating time/day associated with fluoroscopy use and outside dosimeter reading. Fluoroscopy use was also included as an independent variable in the model examining outside dosimeter. R version 3.6.1 and Rstudio version 1.2.1335 and R packages psych and lme4 were utilized, and p<0.05 denoted statistical significance.

Results

Patient Population

A total of 20 female orthopaedic trauma surgeons and 20 female orthopaedic arthroplasty surgeons were recruited, representing 8% and 20% of all U.S. female orthopaedic surgeons in trauma and arthroplasty, respectively. Out of these, 15 trauma surgeons and 16 arthroplasty surgeons completed the study by wearing the dosimeters for 10 days in the OR and returning both dosimeters and paper forms. Nine surgeons were excluded, as the remaining participants lost their dosimeters (two), returned the dosimeters but not the paper forms (two), or were lost to follow-up (five). All surgeons were attendings at their respective institutions, with the exception of one surgeon who at the time of data collection was a trauma fellow.

Descriptive statistics for participating surgeons are reported in Table 1. Surgeons were located all across the United States, representing a total of 30 institutions in 18 states (Figure 1). In our study population, trauma surgeons practiced mostly in an academic setting, while arthroplasty surgeons practiced mostly in a private setting. All participants reported using a lead apron in every case in which they used either fluoroscopy. Only two participants in each group reported using axillary protection wings attached to their lead apron.

Table 1.

Descriptive Statistics of Sample of 31 Female Orthopaedic Surgeons

Variables	Trauma Surgeons (n=15)	Arthroplasty Surgeons (n=16)	T-test
	n (%) or Mean (SD)	n (%) or Mean (SD)	p-value	Effect Size
Practice Setting
Academic	12 (80%)	4 (25%)
Private	3 (20%)	12 (75%)
Personal Protective Equipment
Lead Apron	13 (87%)	14 (88%)
Axillary Wing	2 (13%)	2 (12%)
Cases/Day	2.55 (0.83)	2.40 (0.70)	.59	0.20
Operating Time (hours)/Day	4.51 (1.69)	3.80 (1.55)	.23	0.44
Fluoroscopy Use (seconds)/Day	213.54 (142.40)	25.21 (48.64)	<0.001*	1.77
Outside Dosimeter Reading (millirem)	76.60 (96.35)	5.00 (7.97)	0.01*	8.01

SD = standard deviation, * = statistically significant

Across the 31 participants with 10 OR days per participant, a total of 759 cases were reported (n=381 for arthroplasty, n=378 for trauma). Surgeons used fluoroscopy on a total of 199 days (n=51 for arthroplasty, n=148 for trauma). The average length of fluoroscopy time in cases utilizing fluoroscopy was 66.80 seconds. Description of procedures was not elicited.

Radiation Exposure

Trauma surgeons used fluoroscopy 213.54 seconds per day ± 142.40 (mean ± standard deviation [SD]), which was significantly more than arthroplasty surgeons (25.21 seconds ± 48.64, p < 0.001, Figure 2A). Outside dosimeter readings were significantly higher for trauma surgeons (76.60 millirems ± 96.35) than arthroplasty surgeons (5.00 millirems ± 7.97, p = 0.01, Figure 2B). Inside dosimeter readings were all zero millirem, except for two surgeons (one trauma, one arthroplasty) who had a reading of two mrem and one surgeon (trauma) who had a reading of nine mrem. Given the very low radiation dose penetrating through the lead apron, no further analysis was performed on the inner dosimeter. Years of experience as an attending surgeon were not elicited, but of note, the dosimeter from the trauma fellow in the study had the highest reading at 273 millirems.

Figure 2A.

Outside dosimeter reading by specialty, in ascending order. (mrem = millirem(s); ID = identification number)

Figure 2B.

Average fluoroscopy use per day by specialty, in ascending order. (sec = second(s); ID = identification number)

Trauma surgeons on average spent more time per day in the OR and had more cases per day compared to arthroplasty surgeons, but this difference was not significant. Longer operating time was significantly correlated with more fluoroscopy use for trauma surgeons (r = 0.53, p = 0.04), but not arthroplasty surgeons (r = -0.08, p = 0.77). However, the correlation coefficients were not significantly different between groups for operative time and fluoroscopy use (p = 0.09). Longer use of fluoroscopy significantly correlated with higher outside dosimeter reading among both trauma (r = 0.59, p = 0.02) and arthroplasty surgeons (r = 0.93, p <0.001), and the correlation coefficients were significantly different (p = 0.02).

In univariate linear regression models testing the association between surgeon factors and fluoroscopy use, being a trauma surgeon (b = 188.83, standard error [SE] = 37.72, p < 0.001) and having longer operating time/day (b = 34.12, SE = 14.66, p = 0.03) had significantly more fluoroscopy use (Table 2). A multivariate linear regression model found the same results and indicated that being a trauma surgeon (b = 176.98, standard error [SE] = 44.72, p < 0.001) and having longer operating times (b = 31.91, SE = 14.62, p = 0.04) were independently associated with more fluoroscopy use after adjusting for other factors (Table 3). Univariate linear regression models testing the association between surgeon factors and outside dosimeter reading found that being a trauma surgeon (b = 71.60, SE = 24.15, p = 0.006) and having longer fluoroscopy use/day (b = 0.37, SE = 0.07, p < 0.001) had significantly higher outside dosimeter readings. A multivariate linear regression model indicated that longer fluoroscopy use (b = 0.42, SE = 0.12, p = 0.001) was independently associated with higher outside dosimeter reading, but the type of surgeon was no longer significant.

Table 2.

Univariate Linear Regression Models Examining Surgeon Characteristics Associated with Fluoroscopy Use and Outside Dosimeter Reading

	Fluoroscopy Use				Outside Dosimeter Reading
	b	SE	t	p-value	b	SE	t	p-value
Surgeon Type	188.83	37.72	4.99	< 0.001*	71.60	24.15	2.97	0.006*
Practice Setting	-86.52	48.86	-1.77	0.09	-10.42	27.50	-0.38	0.71
Axillary Wing	68.17	75.62	.90	0.37	35.71	40.56	0.88	0.39
Number of Procedures	35.36	33.92	1.04	0.31	4.14	18.50	0.22	0.83
Operating Time/Day	34.12	14.66	2.33	0.03*	4.82	8.51	0.57	0.58
Fluoroscopy Use/Daya					0.37	0.07	5.26	< 0.001*

^aFluoroscopy use was the dependent variable in the first linear regression model and an independent variable in the second linear regression model. SE = standard error, * = statistically significant

Table 3.

Multivariate Linear Regression Models Examining Surgeon Characteristics Associated with Fluoroscopy Use and Outside Dosimeter Reading

	Fluoroscopy Use				Outside Dosimeter Reading
	b	SE	t	p-value	b	SE	t	p-value
Surgeon Type	176.98	44.72	3.96	< 0.001*	16.03	32.79	0.49	0.63
Practice Setting	20.42	49.40	0.41	0.68	31.25	28.50	1.10	0.28
Axillary Wing	95.87	61.92	1.55	0.13	-24.23	37.27	-.065	0.52
Number of Procedures	-6.06	29.80	-0.20	0.84	0.32	17.15	0.02	0.99
Operating Time/Day	31.91	14.62	2.18	0.04*	-8.77	9.17	-0.96	0.35
Fluoroscopy Use/Daya					0.42	0.12	3.65	0.001*

^aFluoroscopy use was the dependent variable in the first linear regression model and an independent variable in the second linear regression model. SE = standard error, * = statistically significant

Discussion

This study found that trauma surgeons had significantly higher radiation exposure levels based on the dosimeter placed outside the lead apron compared to arthroplasty surgeons. Very little to no radiation penetrated through the lead apron in either the trauma or arthroplasty groups, regardless of operating time or fluoroscopy use. Trauma surgeons used fluoroscopy significantly more than arthroplasty surgeons, with the type of surgeon also having a large effect size on fluoroscopy use.

Dosimeter Readings

The finding that trauma surgeons have significantly higher radiation exposure levels according to the dosimeter outside the lead apron aligns with our hypothesis, and previous findings describing higher radiation exposure levels among trauma surgeons.¹² Given the large effect size and population coverage of our sample (6% and 16% of female trauma surgeons and arthroplasty surgeons, respectively), this result helps confirm on a national level that the differences in radiation exposure between these two specialties are reproducible. Radiation exposure increases cancer risk, and doses accumulated over the course in career could be a concern for contributing to the increased prevalence of breast cancer among female orthopaedic surgeons. The proportion of women in orthopaedic surgery training programs has also increased over the past few decades, and female orthopaedic surgeons under the age of 40 represent a higher proportion of all orthopaedic surgeons compared to older age groups (15.9% in age<40, 8.1% in age 40-49, 6.7% in age 50-59).^14,15 As the proportion of female orthopaedic surgeons increases and trainees choose among various subspecialties for post-residency fellowships, detailed investigation into radiation-associated occupational risk is warranted.

Our study reported radiation doses as 10-day cumulative totals, to control for variation in OR days per week. Therefore, there is not a perfect comparison for the dose levels reported in our study compared to other studies on radiation exposure, which are usually grouped by exposure per month or per case. Nevertheless, the dose levels grouped by specialty are still in a similar range, with one study reporting an average of 52.7 mrem/ month among trauma surgeons and 4.0 mrem/month among arthroplasty surgeons.¹²

According to the dose readings from the dosimeters placed under the lead apron on the chest, almost no radiation penetrated through the lead aprons. Considering every surgeon in this study reported wearing a lead apron when fluoroscopy was used, the overall risk of radiation penetrance to the surgeon’s chest in the area detected by the dosimeter is reassuring. Several studies have shown that orthopaedic surgeons are exposed to radiation levels lower than the recommended yearly limit¹⁶ given proper use of protective equipment. As with these previous studies, the authors recommend the use of a lead apron any time there is the possibility of fluoroscopy use during the surgical procedure.

Surgeon Factors

Fluoroscopy use was significantly higher in trauma surgeons compared to arthroplasty surgeons, aligning with the study’s hypothesis. Interestingly, while trauma surgeons on average also spent more time per day in the OR and had more cases per day compared to arthroplasty surgeons, the difference was not significant. When including other variables in a multivariable regression, and after adjusting for operating time and number of procedures, fluoroscopy use was still significantly associated with higher outside dosimeter readings. Therefore, regardless of surgeon type, fluoroscopy use was still the most important explanation of higher dosimeter readings. Average fluoroscopy use per case was overall lower than existing reports describing a range of 1.5-6.3 minutes/case in common orthopaedic trauma procedures, although our study did not elicit specific procedure descriptions.¹⁷ Efforts to decrease radiation exposure should thus be focused in ways to decrease the amount of fluoroscopy use, rather than limiting OR time or the number of procedures that surgeons perform per day. This strategy can be seen in the introduction of various technologies aimed toward decreasing fluoroscopy time such as computer-aided intra-op visualization technology. In one such study on computer-aided surgical navigation in an orthopaedic trauma OR, the use of fluoroscopic time per procedure was lowered to 30 seconds per case.¹⁸ However, computer navigation equipment has overall not been widely adopted and at present is not a part of routine practice at most institutions.

Several outside dosimeter readings had a reading of 0 millirems despite the surgeon reporting using fluoroscopy for every case, which did not align with the finding overall that fluoroscopy use was the most important factor in radiation dose. The importance of a well-maintained lead apron that is new, well cared for, and is without cracks cannot be overstated.¹⁹ It is likely that the high quality of the protective equipment used by these surgeons is partially responsible for the results. Additionally, research into scatter radiation shows that doses exponentially decrease with even short distances standing away from the source of the fluoroscopy.^20,21 It is possible that surgeons may have stood away from the radiation field, given the association between radiation exposure and breast cancer in women, although this was not assessed. It is also possible that surgeons may have used other protective equipment, such as lead shields. Therefore, while the results of this study suggest that duration of fluoroscopy use is the most important factor explaining radiation exposure, there may be other surgeon factors that affect dosage read on the dosimeter.

Lastly, all surgeons in this study were practicing orthopaedic surgeons at their respective institutions with the exception of one trauma fellow. Although this study did not elicit years of experience, previous studies at the resident level have shown that fluoroscopy use decreases as residents advance in their training.^12,22 It has been proposed that fluoroscopy use times could be a useful assessment tool for resident proficiency at certain procedures. In this study, the outside dosimeter reading from the trauma fellow was the highest in the study in both the trauma and arthroplasty group. However, this should be interpreted not necessarily as being a reflection of the level of training, given that trauma fellows tend to operate more than the average practicing orthopaedic surgeon. Overall, taking into account variations in practices and case load, as well as case variety, the fluoroscopic times in this study therefore may be a useful self-assessment tool for attending trauma and arthroplasty surgeons.

Strengths and Limitations

The strengths of this study lie primarily in its breadth of 31 surgeons across 30 institutions in the US who completed 759 cases. To the authors’ knowledge, this is the largest multi-institutional study completed analyzing radiation exposure and OR characteristics in female orthopaedic surgeons and aligns with existing studies utilizing dosimeters both above and below the lead apron to observe penetrance of radiation through protective equipment. This study builds off existing literature describing both increased radiation exposure among orthopaedic trauma surgeons as a whole, as well as previous studies showing increased prevalence of breast cancer among female orthopaedic surgeons.^4,5,12 Therefore, these results can contribute to advancing knowledge of radiation risk factors and also present an opportunity for surgeons to evaluate their own fluoroscopy use in the context of this study’s data. The finding that fluoroscopy use is the single most important factor for radiation exposure can guide female surgeons in decreasing radiation exposure. It may be prudent to consider using fluoroscopy more selectively each case, comparing one’s own fluoroscopy times per day to the results in this study as a self-assessment tool, or incorporating the use of technologies aimed at decreasing the need for fluoroscopy such as computer aided navigation systems.

Although this is the largest study of its kind, the regressions should still be considered exploratory given the small sample size relative to the total number of orthopaedic surgeons in the US. Factors that affect radiation exposure on an individual surgeon’s level such as distance from fluoroscopy source, angle relative to fluoroscopy source, and type of procedures were not captured. Also, it is theoretically possible that the dosimeters did not function properly and that surgeons may have deviated from the protocol. All dosimeters were acquired from the same source to minimize this risk. Other limitations include the selection of the participant population and operating days, which may be comprised of surgeons particularly interested in radiation exposure and thus already more likely to take additional safety precautions, as well as the fact that we did not ask for information regarding the surgeon’s role in each case. Additionally, there are far fewer female arthroplasty surgeons in the US compared to female trauma surgeons, thus this study captured a greater fraction of all female trauma surgeons compared to arthroplasty surgeons. Lastly, the dosimeter placed under the lead in this study’s protocol may not have captured radiation to the axillary region, and very few surgeons in this study noted wearing axillary protection wings. Previous publications have described the axilla as being most susceptible to scatter radiation exposure,^7,23 therefore it is recommended that axillary protection wings be included in a surgeon’s set of standard radiation protection equipment.