Diagnostic Reference Levels for Common Orthopedic Hand and Wrist Procedures Using Intraoperative Mini C-arm Fluoroscopy

Abstract

Background:

Diagnostic reference levels are radiation dose levels in medical radiodiagnostic practices for typical examinations for groups of standard-sized individuals for broadly defined types of equipment. This study aimed to contribute to national diagnostic reference levels for common hand and wrist procedures using mini C-arm fluoroscopy. Small joint and digital fracture procedure diagnostic reference levels have not been reported in significant numbers previously with procedure-level stratification.

Methods:

Data were collected from fluoroscopy logbooks and were cross-referenced against the audit log kept on fluoroscopy machines. A total of 603 procedures were included.

Results:

The median radiation dose for wrist fracture open fixation was 2.73 cGycm², Kirschner wiring (K-wiring) procedures was 2.36 cGycm², small joint arthrodesis was 1.20 cGycm², small joint injections was 0.58 cGycm², and phalangeal fracture fixation was 1.05 cGycm².

Conclusions:

Wrist fracture fixation used higher radiation doses than phalangeal fracture fixation, arthrodeses, and injections. Injections used significantly less radiation than the other procedures. There are significant differences in total radiation doses when comparing these procedures in hand and wrist surgery. National and international recommendations are that institutional audit data should be collected regularly and should be stratified by procedure type. This study helps to define standards for this activity by adding to the data available for wrist fracture diagnostic reference levels and defining standards for digital and injection procedures.

Introduction

Fluoroscopy is a commonly used x-ray imaging modality in orthopedic surgery providing intraoperative radiographs. This allows real-time visualization of the anatomy and implant placement on a digital monitor informing timely decision-making. ¹ While this is crucial for modern orthopedic surgery, fluoroscopy is a source of radiation exposure to both staff and patients. This is important because excessive exposure to ionizing radiation may have biological effects such as opacification of the lens of the eye or cause malignancies, particularly hematological or solid tumors, for example, of the thyroid.^2,3 Bioeffects from ionizing radiation are directly related to both the magnitude of the radiation dose (measured in Gray) and the amount of tissue irradiated. It is therefore important to limit doses to both patients and staff as both one-off and cumulative exposures. For staff, current effective occupational dose levels in the United Kingdom and Europe should not exceed 2 cGy annually; however, up to 5 cGy can be accepted provided that the 5-year annual average does not exceed 2 cGy. ⁴ This is comparable to the annual limits set out by the U.S. Nuclear Regulation Committee (USNRC), ⁵ which set a total annual limit of 5 cGy. These international regulations enshrine the “as low as reasonably possible” (ALARP) principle in medical exposures, that is, the minimum possible radiation dose should be utilized for any given examination.

Clinically, x-ray doses to patients are routinely measured as the product of radiation absorption and area irradiated, known as the dose-area product (DAP), which reflects the biological importance of the exposure. The units for DAP are Gray meter square (Gym²) which may be expressed as decimal subdivisions, for example, cGycm². The Gray is defined as 1 J/kg and represents the average energy deposited per unit mass in the absorber. ⁶ Many fluoroscopy units contain an ionization chamber attached to the x-ray source collimator that directly measures the DAP, making this a readily available parameter. ⁷ Sequential radiographs are taken during each procedure and cumulatively contribute to the total dose area product (tDAP) for that procedure.

Current UK legislation, as set out by the Ionizing Radiation Medical Exposure Regulations, require that diagnostic reference levels (DRLs) are established for commonly carried out procedures. ⁸ Similarly, regular audit of DRLs are also required in all healthcare settings in Europe and United States to ensure optimal patient safety, as set out by the European Directive 2013/59/Euratom and USNRC regulations, respectively.^4,5 DRLs are defined as “. . . dose levels in medical radiodiagnostic or interventional radiology practices, or, in the case of radio-pharmaceuticals, levels of activity, for typical examinations for groups of standard-sized individuals or standard phantoms for broadly defined types of equipment.” ⁸ Institutions should frequently review DRLs in order to ensure patients are not being exposed to unacceptably high radiation doses. In the absence of any nationally set reference ranges, it is the responsibility of the local organization to set these. It also stipulates each DRL should be procedure specific due to their variability in complexity. ⁹

The primary aim of this study was to contribute to setting national DRLs for commonly performed procedures in orthopedic upper limb extremity. In addition, the study also looks to discern any significant differences in total radiation doses between commonly performed procedures.

Materials and Methods

Two models of mini C-arm units are in use at our institution and the data from both are included in this study. One is a Hologic (Bedford, California) InSight 2 with an analogue detector. The other is a Hologic (Bedford) InSight FD with a digital detector. Both are equipped with a Thermo Scientific (Scotts Valley, California) model PSX11-100-35 x-ray source. Only suitably trained and accredited surgical staff are permitted to use the machines. Both units are used with automatic exposure control and medium noise suppression. Wherever practical, collimation is set to the smaller (4-inch, 10 cm) field, for example, for digital surgery and joint injections. The larger (6-inch, 15 cm) field is routinely used for wrist fracture surgery. This is in line with the ALARP safety principle. In addition, all staff members present in the operating theater at the time of radiation, regardless of distance from the C-arm, are required to wear full-length lead aprons of at least 0.25 mm thickness, to ensure radiation exposure is reduced.

Data were extracted from the mini C-arm logbooks, which included all fluoroscopic procedures between February 18, 2016 and September 5, 2019. Following every procedure that utilizes the mini C-arm, the operator manually documents key information from the mini C-arm into the attached logbook. The logbooks, which are used for auditing purposes, record patient demographics, procedure type, operating Consultant, and tDAP values. These data are also replicated on the individual mini C-arm hard-drive via the audit log mechanism. Total DAP values collected from the logbook were cross referenced against those stored on the mini C-arms to ensure accurate recording.

In order to identify whether the anatomical location had an effect on tDAP, data were grouped into the most commonly performed procedures within the institution. These groups included: (1) open reduction and internal fixation (ORIF) of wrist fractures; (2) phalangeal fracture ORIF; (3) carpal injections; (4) small hand joint arthrodesis procedures (distal interphalangeal joint [DIPJ]/proximal interphalangeal joint [PIPJ]/metacarpophalangeal joint [MCPJ]); and (5) hand and wrist K-wiring. Procedures that had fewer than 20 recorded tDAP doses were excluded from the analysis as it was deemed that these did not have significant numerical validity. Procedures that had missing data in the logbook or were illegible were also excluded from the study. In the uncommon scenario where 2 or more procedures were performed at the same sitting, the tDAP was averaged across the procedures.

The data were checked for normal distribution with the D’Agostino and Pearson normality test. Data are presented as median (inter-quartile range) for nonnormally distributed data and mean (standard deviation) for normally distributed data. The difference in radiation exposure between the groups was compared using the Kruskal-Wallis test corrected for multiple comparisons. Results were considered significant for P < .05.

Results

A total of 603 procedures were recorded during the study period. There were 17 procedures that were excluded due to missing data in the logbooks and/or illegible documentation. The total number of procedures included in data analysis was 586.

Procedure Breakdown

All of the wrist fracture ORIF group had open surgery with plate fixation. In the injection group (total 167), 85 had scaphotrapeziotrapezoid joint (STT) injections, 62 carpometacarpal joint (CMCJ) injections, 12 pisotriquetral (PT) injections, 5 distal radio-ulnar joint (DRUJ) injections, 2 PIPJ injection, and 1 wrist injection. Within the K wiring group, 41 procedures were for metacarpal trauma and 15 for finger injuries. In the arthrodesis group, there were 40 DIPJ arthrodeses, 15 PIPJ arthrodeses, 15 thumb MCPJ arthrodesis, and 5 thumb IPJ arthrodeses.

Radiation Dose by Procedure

None of the procedures observed displayed a normal distribution, all being right skewed. The median tDAP for wrist ORIF and k-wire procedures were 2.73 cGycm² and 2.36 cGycm² respectively. DIPJ/PIPJ/MCPJ arthrodeses recorded a median of 1.20 cGycm² and phalanx ORIF 1.05 cGycm². Carpal injections demonstrated the lowest radiation exposure with a median of 0.58 cGycm² (Table 1).

Table 1.

Collated Results of Median tDAP by Procedure Type.

Procedure	Number of patients	Median total DAP (cGycm2)	IQR	10%-90% centile	Normal distribution? a
Wrist ORIF	212	2.73	1.75-4.40	1.12-7.57	No
Phalanx ORIF	76	1.05	0.66-2.22	0.33-4.83	No
Injection	167	0.58	0.33-0.91	0.24-2.11	No
DIPJ/PIPJ/MCPJ arthrodeses	75	1.20	0.55-2.92	0.40-4.35	No
K-wiring	56	2.36	1.30-3.90	0.36-6.40	No

Note. tDAP = total dose area product; DAP = dose area product; IQR = interquartile range; ORIF = open reduction and internal fixation; DIPJ = distal interphalangeal joint; PIPJ = proximal interphalangeal joint; MCPJ = metacarpophalangeal joint; K-wiring = Kirschner wiring.

^aNormal distribution was assessed using the D’Agostino and Pearson normality test (P < .05 for all procedures).

Analysis of Radiation Exposure Between Groups

Wrist fracture ORIF used significantly higher doses of radiation than small joint arthrodesis, phalangeal ORIF and injections (P < .0001). Kirschner wiring surgery observed higher radiation exposure than phalangeal ORIF (P = .03) and injection (P < .001) but similar doses to small joint arthrodesis and wrist fracture ORIF (P = .14 and .24, respectively). On further analysis of k-wire procedures by anatomical location, phalangeal procedures resulted in significantly smaller radiation doses (1.51 cGycm²) as compared to metacarpal K-wiring procedures (3.41 cGycm²) (P < .05). Injections used a significantly lower dose of radiation than the other procedures (P < .001). Interphalangeal joint arthrodesis and phalangeal ORIF used similar radiation doses (P = .99).

Discussion

In this study, we present DRLs for wrist and phalangeal fracture ORIF, small joint arthrodesis in the digits, K-wiring, and joint injections in the hand and wrist. The study demonstrates a significant difference in radiation exposure as measured by tDAP between procedure groups. Wrist fracture ORIF and K-wiring result in larger radiation doses when compared to all of the other procedures examined. This is likely due to their higher procedural complexity, number of inserted implants, and number of images required to demonstrate bony anatomy, ensure satisfactory reduction of fractures, and correct implant placement.

Carpal injections had the lowest levels of radiation exposure as anatomical surface markings can be used to guide injections, using fluoroscopy only to confirm location of the injection. ¹⁰ Radiation doses for phalangeal ORIF and DIPJ/PIPJ/MCPJ arthrodesis were higher on average than those of injections. However, there was no statistical difference identified between phalangeal ORIF and small joint arthrodesis, with low radiation exposure for both procedures. We therefore recommend grouping these procedures together for dose reference level purposes.

The United Kingdom, European, and U.S. medical radiation regulations, recommend DRLs should be locally audited against nationally set DRLs, or in the absence of these, the local organization should set DRLs.^4,5,8 To our knowledge, there are no current published DRLs for the small joint and phalangeal procedures included within this study by organs of the UK government such as Public Health England. Hardman et al ¹¹ conducted a retrospective, multi-center study of radiation exposure in commonly performed orthopedic trauma cases. For a dose comparison, the national DRLs state a median tDAP for a plain chest radiograph is 15 cGycm² and a lumbar spine plain radiographic series is 400 cGycm². ¹² A tDAP of 20.71 cGycm² was reported for wrist fracture ORIF; the intra-operative fluoroscopy equipment used was not stated. ¹¹ One of the 4 centers included in the study produced a significantly increased mean radiation dose, which was 14 times higher than the center with the lowest radiation exposure for wrist ORIF. It is not certain what may have caused this variation; however, it may be attributed by differences in the grade of performing surgeon or fluoroscopic equipment used at that center. This resulted in a deviation from the mean at the other 3 centers, which were more similar to the data set in this study with mean values of 3.22 cGycm², 6.55 cGycm², and 6.74 cGycm². Lee et al ¹³ quoted a figure of 40 cGycm2 using standard mobile fluoroscopy equipment for the same procedure. Rashid et al ¹⁴ reported on DRLs for wrist fracture surgery. This study did not report on the type of fluoroscopy equipment used for the individual procedure types in their study, which certainly included some standard mobile fluoroscopy machine data as hip fracture procedure doses were also reported. A median tDAP of 2.735 cGycm² was reported for 50 wrist fracture plating, which correlates well with the median tDAP for wrist fracture ORIF of 2.73 cGycm² from our study across 212 cases. For wrist fracture K-wiring, Rashid et al reported a median dose of 2.52 cGycm² (139 cases), which again compares to our finding of a median of 2.36 cGycm² in 56 cases.

A more recent study by McCann et al ¹⁵ aimed to set local DRLs for orthopedic upper limb procedures. Their data were grouped according to type of procedure (open/closed) and anatomical area, that is, digits, thumb/hand and wrist/forearm but did not identify DRLs for individual procedures. Procedures were grouped into closed procedures, open procedures, steroid injections, and unknown procedures. In the published open procedure group, the authors recorded a median tDAP of 1.45 cGycm². In our dataset, combining all of our open procedures (wrist fracture ORIF, phalangeal ORIF, and arthrodesis surgery) gives a median tDAP of 1.645 cGycm². As the open procedures in the McCann et al study were not broken down by type, it is difficult to directly compare to our results. For steroid joint injections, McCann et al report a median tDAP of 0.45 cGycm² of 871 cases which is comparable to a median of 0.58 cGycm² for 167 joint injections in our study. Our higher median tDAP may be explained by our current practice. In our institution, most trapeziometacarpal joint injections and digital joint injections are routinely performed in the out-patient clinic using a landmark technique. The majority of joint injections performed in the operating theater under fluoroscopic guidance are to the more technically challenging scaphotrapeziotrapezoid or pisotriquetral joints, and injections where there is a prosthetic joint replacement present or where the injection is combined with an examination under anesthesia or fluoroscopy.

In our study, we excluded 17 cases due to illegibility and/or missing information from the physical logbooks. This gives a drop-out rate of 2.8% indicating a robust data set. Limitations of the study include no assessment of the grade of surgeon, which may affect the tDAP due to differences in experience. Data recording of multiple procedures in 1 case meant we had to average the tDAP per procedure, for example, 2 injections on 1 sitting recorded as a single tDAP datum. This may have had an effect on results when combined with single injections episodes. At our institution, injections are administered using a landmark technique rather than fluoroscopically guided, which limited the available sample size for injection procedures in particular.

Although the mini C-arm unit is thought to reduce radiation exposure when compared to its standard counterpart, ¹⁶ Singer et al ¹⁷ showed that the mini C-arm is associated with 53% to 70% greater exposure than the large C-arm due to its lesser source-to-intensifier distance. However, the study only used cadaveric prosections of the hand, wrist and elbow which is unlikely to account for pathologies such as fractures, requiring multiple and higher magnification images. There may also be significant differences in the absorption of radiation between living and cadaveric tissue meaning that these findings may not be directly applicable to clinical practice. Contrastingly, Giordano et al ¹⁸ reported significantly lower radiation doses when using mini C-arm fluoroscopy in comparison to standard fluoroscopy when imaging cadaveric ankle specimens. Specific conditions were tightly controlled in the study such as specimen distance to radiation source, exposure length, and sensor positioning. Images were subsequently taken in identical conditions on both fluoroscopes; the mini C-arm produced an exposure of 0.31 cGycm² and the standard C-arm 0.85 cGycm² when placed at the optimum distance from the radiation source. Both mini C-arm and standard fluoroscopy radiation exposure increased with reducing the distance to the radiation source. This suggests that the preference of mini C-arm fluoroscopy results in lower radiation doses; however, it can be significantly effected by the discontinuity in size and radiation distances between individual devices.

Larger radiation doses not only pose a risk to patients but also contribute to significant occupational radiation exposure to staff in the operating theater. This study did not measure specific occupational radiation exposure; however, it’s importance should not be overlooked and is a topic that should be considered within the wider scope of radiation exposure. The similarities in legislation between United Kingdom, Europe, and the United States with regard to procedure-specific DRLs and regular audit of such, suggests that the findings this study may be used for comparisons at an international level.

In conclusion, this study has demonstrated there are significant differences in total radiation doses when comparing these 5 procedures. This study will allow our unit to set local DRLs for individual procedures and contribute to setting nationally accepted DRLs, particularly as there is a high degree of alignment between our recorded tDAPs and those previously published. It is important to regularly audit this data set, in line with statutory regulations, to ensure patients and staff receive as little radiation as clinically possible during orthopedic surgery. We would recommend arthrodesis of the interphalangeal joints of the fingers are considered as a group with phalangeal fracture surgery for DRL purposes. Furthermore, this study confirms the use of a mini-C-arm fluoroscopy units for upper limb extremity procedures is in line with the ALARP principle enshrined in the international statutory regulations.