Tracking Intraoperative Fluoroscopy Utilization Reduces Radiation Exposure During Ureteroscopy

Abstract

Purpose:

Recent studies have demonstrated deleterious effects of ionizing radiation from diagnostic and therapeutic imaging procedures. One of the barriers to minimizing patient exposure is physician awareness. We prospectively studied whether providing surgeons with feedback on their fluoroscopy utilization would affect intraoperative fluoroscopy times.

Materials and Methods:

In 2007, we prospectively began to track fluoroscopy usage for all urology cases. Nine months later, surgeons started to receive periodic reports with their mean fluoroscopy time compared with their peers. We reviewed all ureteroscopic cases for nephrolithiasis from the date tracking began (2006–2010, n = 311). Using the initial 9-month period as a control, we studied the effect of providing feedback on mean fluoroscopy times in subsequent periods and analyzed patient factors that may affect radiation exposure.

Results:

Mean fluoroscopy times for unilateral ureteroscopy decreased by 24% after surgeons received feedback (2.74–2.08 minutes, p = 0.002). On multivariate analysis, factors that independently predicted decreased fluoroscopy times included female sex (p = 0.02), stones in the distal ureter (p = 0.04), and if the surgeon had received feedback (p = 0.0004). Factors that increased fluoroscopy times included the presence of hydronephrosis (p = 0.001), use of a ureteral access sheath (p = 0.04), ureteral balloon dilation (p = 0.0001), and placement of a postoperative stent (p = 0.002).

Conclusions:

Providing surgeons with feedback on their fluoroscopy usage reduces patient and surgeon radiation exposure. Implementing such a tracking system requires minimal changes to existing operating room staff workflow. Further study is warranted to study the impact of this program on other procedures that utilize fluoroscopy in urology and other specialties.

Introduction

R ecent studies have highlighted the risk of secondary malignancies associated with ionizing radiation from diagnostic imaging, particularly with computed tomography (CT), as its utilization becomes increasingly prevalent. Notably, an estimated 62 million CT scans are performed each year in the United States and, although a large-scale epidemiologic trial to study the cancer risk associated with diagnostic imaging has only recently begun, current models based on atomic bomb survivors estimate that 1.5%–2.0% of cancers in the United States can be attributed to radiation from CT scans.^1,2

Unfortunately, few trials have studied the effect of ionizing radiation used in the diagnosis and treatment of urologic conditions.^3,4 Minimally invasive modalities for the treatment of kidney stones such as ureteroscopy, extracorporeal shockwave lithotripsy (SWL), and percutaneous nephrolithotomy (PCNL) require fluoroscopy in nearly all cases. The radiation exposure to the patient from a typical ureteroscopy or SWL for stone disease is approximately one order of magnitude less than that of CT scan of the abdomen.^5,6 In contrast, the radiation dose from a PCNL is approximately one-third that of a CT scan of the abdomen.⁷ Although small, these exposures become significant when one considers that there was an increased risk of cancer in the subset of atomic bomb survivors who received low doses of radiation (5–150 mSv), that many patients with stone disease require multiple interventions, and that most patients with nephrolithiasis have undergone multiple CT scans by the time they are seen in by a urologist.

The surgeon and operating room staff are placed at risk with continued exposure to ionizing radiation. While most adhere to the ALARA⁸ principal (“As Low As Reasonably Achievable”), over a career a urologist may perform thousands of procedures utilizing ionizing radiation. Little long-term data exist describing the incidence of secondary malignancy in urologists.

Clinicians, for the most part, underestimate the amount of radiation exposure patients experience with many of the common radiographic studies, and many do not even know the units used to measure radiation exposure.⁹ This lack of awareness has been partly blamed for excessive utilization of imaging and the associated iatrogenic exposure to ionizing radiation.

We hypothesized that increased surgeon awareness of their fluoroscopy times, especially in comparison to their peers, would prompt them to take active measures to limit exposure. We prospectively implemented a system to track and report fluoroscopy usage for all urologic cases at a single institution and studied the effect it had on fluoroscopy times over 3 years.

Materials and Methods

In late October 2006, we prospectively began to track fluoroscopy times for all cases performed by the urology staff at a single institution (Santa Clara Valley Medical Center). This change was implemented with the operating room personnel as an additional step in their postprocedure documentation and was not widely publicized. No reports were generated for a 9-month period (10/2006–7/2007) as data were accrued to establish a baseline. After this initial period, surgeons were provided quarterly reports that contained their mean fluoroscopy times as well as the mean times of their peers. For the purposes of this study, the remaining cases were partitioned into 12-month periods (8/2007–7/2008, 8/2008–7/2009, and 8/2009–7/2010). We limited our analysis to cases of unilateral ureteroscopy for stone disease and excluded cases that involved only ureteral stent placement or ureteroscopy for other diagnoses.

After obtaining institutional review board approval, each patient's chart was retrospectively reviewed and preoperative characteristics, including age, sex, ethnicity, diameter of largest stone, stone side, stone location (kidney vs. proximal, mid, or distal ureter), the presence and degree of hydronephrosis at presentation, and medical comorbidities, were tabulated. Intraoperative parameters, including the presence of a preoperative ureteral stent or nephrostomy tube and utilization of the Holmium laser, ureteral access sheath, ureteral occlusion device (Accordion^®; PercSys, Palo Alto, CA), ureteral balloon dilation, and postoperative stent, were also recorded. Those performing the chart review were blinded to fluoroscopy and operating times. These data were tabulated separately and merged into the database at the end of the chart review process.

Statistical analysis was performed using the JMP Statistical Discovery Software (SAS Institute Inc., Cary, NC). Student's t-test was used to compare means (before feedback vs. after feedback). Analysis of variance was used when there were multiple means being compared. Dunnett's multiple comparison test (with a = 0.05) was used to perform posteriori comparisons between each mean to a control (before feedback was provided vs. subsequent periods after feedback was provided). Pearson's χ ² test was used to compare proportions between different groups. The values for fluoroscopy time and operating room time were log transformed to facilitate statistical analysis. All tests for significance were two-sided and a p ≤ 0.05 was considered statistically significant.

Results

A total of 311 cases were included in this study. For the first 9 months, while fluoroscopy time was collected for baseline statistics, 48 cases were performed. Then, the first set of reports was distributed. The remainder of the study period was partitioned into three consecutive 12-month periods, each accounting for 67, 114, and 82 cases, respectively.

Patient demographics are depicted in Table 1. The mean age of the patients in this study was 45.9 years (95% confidence interval, 44.3–47.5), with males accounting for roughly half of the cohort. The mix of ethnicities roughly matches that of the general population at this hospital, with Hispanics and Asians making up two-thirds of our cohort. A significant percentage of these patients have medical comorbidities known to be associated with nephrolithiasis. There were no statistically significant differences with respect to age, sex, race, and medical comorbidities between the cases performed before and after reporting of fluoroscopy time began.

Table 1.

Patient Demographics

	Total (n = 311)	Before reporting (n = 48)	After reporting (n = 263)	p
Age (years)	45.9 ± 0.8	46.3 ± 2.3	45.9 ± 1.0	0.86
Sex (%)				0.16
Female	169 (54)	31 (65)	138 (52)
Male	142 (46)	17 (35)	125 (48)
Race (%)				0.29
Hispanic	137 (44)	20 (42)	117 (45)
White	91 (29)	12 (25)	79 (30)
Asian/Pacific Islander	61 (20)	9 (19)	52 (20)
Middle Eastern	13 (4)	4 (8)	9 (3)
Black	9 (3)	3 (6)	6 (2)
Medical comorbidities (%)
Hypertension	139 (45)	24 (50)	115 (44)	0.43
Smoking	102 (32)	13 (27)	89 (34)	0.41
Hyperlipidemia	83 (27)	11 (23)	72 (27)	0.52
Diabetes	69 (22)	10 (21)	59 (22)	0.99
Obesity	63 (20)	12 (25)	51 (19)	0.43

The perioperative characteristics of the cohort are described in Table 2. The mean stone diameter was 8.2 mm (in patients with multiple stones, the largest one was used). The distribution of left- and right-sided stones was equivalent. One-fourth of the stones were located in the kidney, which reflects a shift in our practice of treating lower pole stones with ureteroscopy instead of SWL. Two-thirds of patients had hydronephrosis at presentation. There were no serious complications requiring unplanned hospitalizations or reoperative surgery. There were three minor ureteral injuries from wire manipulation (two cases of false passage through ureteral mucosa and one case of mild contrast extravasation on retrograde pyelography) that were managed with ureteral stenting. There were not enough incident complications to perform formal statistical testing. There were no statistically significant differences between the perioperative characteristics of the cases performed before and after reporting began except for preplacement of a nephrostomy tube. However, the numbers of those with a preplaced nephrostomy tube were small. More importantly, the stone size, laterality, and location were similar between the cases occurring before and after reporting began.

Table 2.

Perioperative Characteristics

	Total (n = 311)	Before reporting (n = 48)	After reporting (n = 263)	p
Stone diameter	8.2 ± 0.4	8.4 ± 0.5	8.0 ± 0.2	0.50
Multiple stones (%)	70 (23)	7 (15)	63 (24)	0.15
Stone side (%)				0.68
Left	147 (47)	24 (50)	123 (47)
Right	164 (53)	24 (50)	140 (53)
Stone location (%)				0.33
Kidney	78 (25)	13 (27)	65 (25)
Proximal ureter	101 (32)	15 (31)	86 (33)
Mid ureter	43 (14)	3 (6)	40 (15)
Distal ureter	89 (29)	17 (36)	72 (27)
Presence of hydronephrosis (%)	207 (67)	33 (69)	174 (66)	0.89
Degree of hydronephrosis (%)				0.68
Mild	74 (36)	14 (42)	60 (35)
Moderate	107 (52)	15 (46)	92 (53)
Severe	25 (12)	4 (12)	21 (12)
Intraoperative findings (%)
Preplaced stent	72 (23)	10 (21)	62 (24)	0.68
Preplaced nephrostomy	15 (5)	7 (15)	10 (4)	<0.05
Ureteral balloon dilation	37 (12)	7 (15)	30 (11)	0.53
Ureteral access sheath	168 (54)	23 (48)	145 (55)	0.36
Ureteral occlusion device^a	80 (26)	0 (0)	80 (30)	N/A
Holmium laser	277 (89)	42 (88)	235 (89)	0.71
Postoperative stent	282 (91)	42 (88)	240 (91)	0.41

The ureteral occlusion device was not available at our institution during the initial period. Therefore, statistical testing was not performed for this parameter to compare the before and after reporting periods.

N/A = not available.

When the mean fluoroscopy time for cases performed before the reporting period began was compared with that of cases performed after, there was a statistically significant 24% reduction (2.74–2.08 minutes, p = 0.002). However, operating time between these two groups did not significantly differ (70.7 vs. 61.2 minutes, p = 0.10).

When the cases were partitioned into one 9-month period before reporting began and three 12-month periods subsequent to that, the resulting means showed a downward trend and the differences among the groups were statistically significant by analysis of variance (Fig. 1). Defining the initial 9-month period as the control group, Dunnett's multiple comparison test showed that the most recent period was significantly different from the control (Table 3).

FIG. 1.

Fluoroscopy time over several reporting periods.

Table 3.

Univariate Analysis of Fluoroscopy Time as a Function of Reporting Periods

Period	Log (fluoro time)	SE	Significant? ^a
Baseline^b	0.8321	0.0954	—
Year 1	0.6146	0.0748	No
Year 2	0.5995	0.0516	No
Year 3	0.4853	0.0687	Yes

Dunnett's multiple comparison test with α = 0.05.

Control group.

SE = standard error.

Multivariate linear regression identified covariates that independently influenced fluoroscopy time (Table 4). Female sex, distal ureteral stones, and the fact that a surgeon has received feedback independently predicted lower fluoroscopy time in our model. Feedback on prior fluoroscopy time was associated with the greatest reduction in fluoroscopy time (most negative β coefficient). Conversely, using a ureteral access sheath, the presence of hydronephrosis at presentation, placement of a postoperative stent, and requirement for ureteral balloon dilation all independently predicted increased fluoroscopy time. Balloon dilation was associated with the greatest increase in fluoroscopy time (highest positive β coefficient). This may be explained by the fluoroscopy needed to perform the dilation itself or may simply reflect more challenging cases with impacted stones. These factors could not be quantified with the available data.

Table 4.

Multivariate Linear Regression for Fluoroscopy Time as a Function of Perioperative Characteristics

Parameter	Coefficient (β)	SE	p
Patient age (years)	0.0009	0.0020	0.65
Sex (female)	−0.0757	0.0330	0.02
Ethnicity
White	−0.1509	0.0778	0.06
Black	0.1075	0.1504	0.48
Hispanic	0.0150	0.0633	0.81
Asian/Pacific Islander	−0.1485	0.0778	0.06
Middle Eastern	0.1769	0.1273	0.17
Had feedback	−0.1604	0.0450	0.0004
Obese	−0.0010	0.0402	0.80
Stone Size (mm)	0.0052	0.0999	0.96
Stone side (left)	−0.0268	0.0307	0.38
Stone location
Kidney	0.0878	0.0711	0.22
Proximal ureter	−0.0155	0.0538	0.77
Mid ureter	0.0592	0.0679	0.38
Distal ureter	−0.1315	0.0621	0.04
Hydronephrosis	0.1372	0.0417	0.001
Preplaced ureteral stent	−0.0259	0.0365	0.48
Preplaced nephrostomy	−0.0050	0.0694	0.94
Ureteral balloon dilator	0.2928	0.0483	<0.0001
Ureteral access sheath	0.0786	0.0383	0.04
Ureteral occlusion device	0.0348	0.0399	0.38
Holmium laser	−0.0412	0.0529	0.43
Postoperative stent	0.1838	0.0572	0.002

Discussion

Ionizing radiation from both medical diagnostic studies and therapeutic interventions is now recognized as a public health issue. Mainstream media attention to this problem and the associated secondary malignancies mandate that physicians do whatever possible to limit radiation exposure. We predict that patients' cumulative lifetime ionizing radiation doses from all studies will ultimately be tracked and become part of the medical record.

Intraoperative fluoroscopy times are most likely recorded at many institutions. However, we suspect that little, if any, feedback is provided to the surgeons and how they compare to their peers. Based on successes with retrospective audits on overuse in the blood transfusion literature and the psychology underlying it,^10,11 we aimed to provide similar motivation to individual surgeons after establishing a baseline.

We were not certain at the beginning whether our interventions would produce significant reductions in fluoroscopy times because there is only a finite amount of time that can be reduced on a per case basis. This is especially true when the fluoroscopy times are on average 3 minutes and there is an inherent “overhead” of fluoroscopy time associated with every case. We were thus surprised to see a continuous downward trend in the mean fluoroscopy time with each subsequent reporting period. Additionally, it was very telling that the greatest independent predictor of decreased fluoroscopy time on multivariate analysis was the fact that a surgeon has received feedback on his or her fluoroscopy time.

There is a paucity of literature on the risk of secondary malignancies due to ionizing radiation specifically used for urologic procedures.^12,13 A recent study by Krupp and associates modeled organ-specific dosing by exposing cadavers to index fluoroscopy sessions designed to simulate what is typically performed during ureteroscopy and reported organ specific exposure to the ipsilateral kidney (3.55 mGy), ureter (2.72 mGy), gonad (1.88 mGy), and lung (0.42 mGy). The region that absorbed the most radiation was the skin posterior to the ipsilateral kidney (10.49 mGy). From this, they estimated that 1 in 1000 patients undergoing ureteroscopy would develop a cancer at that site using a linear-no-threshold model of radiation induced cellular injury.⁵

Although we have certainly been able to quantify the reduction in fluoroscopy time that resulted from our interventions, it remains difficult at this point to estimate the associated reduction in secondary malignancies based on available data. Although it is tempting to extrapolate our 24% reduction in fluoroscopy time to the exposure data reported by Krupp (their exposure time was 145 seconds), equipment differences and variability in patient and stone size make this challenging. Nevertheless, we beleive that total fluoroscopy time serves as an adequate surrogate marker since it correlates closely with radiation exposure (R ² = 0.84)⁶ and is convenient to measure.

Although the effect of low-dose radiation is still controversial, the trend has been to be conservative and consider it dangerous pending any evidence to the contrary.¹⁴ It would seem prudent to attempt to reduce radiation exposure when possible, especially when the risks remain undefined and doing so is not difficult to implement and does not affect the outcomes of interventions. Recently, Greene and associates reported that their institution was able to reduce mean fluoroscopy times during ureteroscopic lithotripsy by 82% without affecting surgical outcomes by simply implementing a reduced fluoroscopy protocol.¹⁵

The importance of limiting radiation exposure from fluoroscopy applies not only to patients but also to surgeons and operating room staff. Hellawell and associates showed that the cumulative radiation exposure to a urologists performing 50 endourologic procedures is ∼10 mGy.³ Others have reported similar results, each also confirming that although all operating room staff are exposed, the surgeon receives the most exposure from scatter.^13,16 Although in absolute terms, this level of radiation is small, it is instructive to note that the Biological Effects of Ionizing Radiation (BEIR VII) report estimates that a 10 mSv of radiation exposure per year for a 70-year lifetime causes a 5% lifetime increase in fatal cancer.¹⁷

Weaknesses of our study include the lack of randomization, the variability associated with using fluoroscopy time as a surrogate marker for radiation exposure, and the exclusion of other endourologic procedures. Further studies are needed to validate our findings with PCNL and SWL as well as procedures performed by other subspecialties that utilize fluoroscopy (e.g., orthopedics, interventional radiology, cardiology, and vascular surgery). Additionally, there is a need to more clearly define the risk of malignancy associated with low-dose radiation, an undertaking that will require large epidemiologic studies.

Conclusions

We recommend providing surgeons with feedback on their fluoroscopy usage compared with their peers as an inexpensive, easy to implement measure that reduces radiation exposure to patients and operating room personnel during ureteroscopic stone surgery by approximately one-fourth. Hospitals should consider including fluoroscopy usage into their quality metrics and implementing similar reporting programs.

Disclosure Statement

Dr. Shinghal is an advisor for PercSys and an investigator for GlaxoSmithKline.

Footnotes

Abbreviations Used

References

Brenner

, Hall

. Computed tomography-an increasing source of radiation exposure. N Engl J Med, 2007; 357:2277–2284.

Einstein

, Henzlova

, Rajagopalan

. Estimating risk of cancer associated with radiation exposure from 64-slice computed tomography coronary angiography. JAMA, 2007; 298:317–323.

Hellawell

, Mutch

, Thevendran

et al.

Radiation exposure and the urologist: What are the risks?

J Urol, 2005; 174:948–952discussion 952.

Tarin

, Sonn

, Shinghal

. Estimating the risk of cancer associated with imaging related radiation during surveillance for stage I testicular cancer using computerized tomography. J Urol, 2009; 181:627–632discussion 632–633.

Krupp

, Bowman

, Tenggardjaja

et al. Fluoroscopic organ and tissue-specific radiation exposure by sex and body mass index during ureteroscopy. J Endourol, 2010; 24:1067–1072.

Perisinakis

, Damilakis

, Anezinis

et al. Assessment of patient effective radiation dose and associated radiogenic risk from extracorporeal shock-wave lithotripsy. Health Phys, 2002; 83:847–853.

Safak

, Olgar

, Bor

et al. Radiation doses of patients and urologists during percutaneous nephrolithotomy. J Radiol Prot, 2009; 29:409–415.

Musolino

, DeFranco

, Schlueck

. The ALARA principle in the context of a radiological or nuclear emergency. Health Phys, 2008; 94:109–111.

Shiralkar

, Rennie

, Snow

et al. Doctors' knowledge of radiation exposure: Questionnaire study. BMJ, 2003; 327:371–372.

10.

Wilson

, MacDougall

, Fergusson

et al. The effectiveness of interventions to reduce physician's levels of inappropriate transfusion: What can be learned from a systematic review of the literature. Transfusion, 2002; 42:1224–1229.

11.

Francis

, Stockton

, Eccles

et al. Evidence-based selection of theories for designing behaviour change interventions: Using methods based on theoretical construct domains to understand clinicians' blood transfusion behaviour. Br J Health Psychol, 2009; 14:625–646.

12.

Bagley

, Cubler-Goodman

. Radiation exposure during ureteroscopy. J Urol, 1990; 144:1356–1358.

13.

Kumari

, Kumar

, Wadhwa

et al. Radiation exposure to the patient and operating room personnel during percutaneous nephrolithotomy. Int Urol Nephrol, 2006; 38:207–210.

14.

Wall

, Kendall

, Edwards

et al.

What are the risks from medical X-rays and other low dose radiation?

Br J Radiol, 2006; 79:285–294.

15.

Greene

, Bowman

, Tenggardjaja

et al. Implementation of a reduced fluoroscopic protocol during ureteroscopic lithotripsy significantly reduces radiation exposure. J Urol, 2010; 183:e559.

16.

Majidpour

. Risk of radiation exposure during PCNL. Urol J, 2010; 7:87–89.

17.

Shrader-Frechette

. Trimming exposure data, putting radiation workers at risk: Improving disclosure and consent through a national radiation dose-registry. Am J Public Health, 2007; 97:1782–1786.