Reduction of radiation dose during facet joint injection using the new image guidance system SabreSource™: a prospective study in 60 patients

Abstract

Interventional procedures are associated with high radiation doses for both patients and surgeons. To reduce the risk from ionizing radiation, it is essential to minimize radiation dose. This prospective study was performed to evaluate the effectiveness in reducing radiation dose during facet joint injection in the lumbar spine and to evaluate the feasibility and possibilities of the new real time image guidance system SabreSource™. A total of 60 patients, treated with a standardized injection therapy of the facet joints L4–L5 or L5–S1, were included in this study. A total of 30 patients were treated by fluoroscopy guidance alone, the following 30 patients were treated using the new SabreSource™ system. Thus a total of 120 injections to the facet joints were performed. Pain, according to the visual analogue scale (VAS), was documented before and 6 h after the intervention. Radiation dose, time of radiation and the number of exposures needed to place the needle were recorded. No significant differences concerning age (mean age 60.5 years, range 51–69), body mass index (mean BMI 26.2, range 22.2–29.9) and preoperative pain (VAS 7.9, range 6–10) were found between the two groups. There was no difference in pain reduction between the two groups (60 vs. 61.5%; P = 0.001) but the radiation dose was significantly smaller with the new SabreSource™ system (reduction of radiation dose 32.7%, P = 0.01; reduction of mean entrance surface dose 32.3%, P = 0.01). The SabreSource™ System significantly reduced the radiation dose received during the injection therapy of the lumbar facet joints. With minimal effort for the setup at the beginning of a session, the system is easy to handle and can be helpful for other injection therapies (e.g. nerve root block therapies).

Introduction

Ionizing radiation is widely used in orthopedic surgery and interventional pain therapy. The use of fluoroscopy guided procedures has increased dramatically. The dose arising from these procedures is about 10% of the total medical exposure [28]. Interventional procedures are associated with high radiation doses for both patients and surgeons, who often have to put their hands in the beams of the fluoroscope.

Biologic effects of radiation can be grouped as stochastic or nonstochastic effects. A stochastic effect is one in which the probability of the effect, rather than its severity, increases with dose. Radiation-induced cancer and genetic effects are stochastic. Stochastic effects are believed to lack a threshold dose because injury to few cells or even a single cell could theoretically result in production of the effect [15]. The typical skin entrance exposure rate is 10–50 mGy/min for normal and 100–200 mGy/min for high-dose fluoroscopy [15]. Therefore it is clear that the potential exists for radiation injuries during prolonged and frequent fluoroscopic procedures.

To reduce the risk from ionizing radiation, it is essential to minimize radiation dose. Monitoring, as well as the development of new interventional possibilities like navigation procedures, is of great importance. Recently computer-assisted procedures have been well-established in the field of orthopedics and traumatology. Most often, computer-assisted procedures are used in the placement of pedicle screws in spinal surgery. Different studies have reported of a significant reduction of radiation dose [11, 12] while the preparation of computer-assisted procedures lasted slightly longer compared to conventional operation techniques [3, 11, 12]. Navigation systems are often complicated, expensive and appropriate for difficult procedures. Currently there is no navigation system commonly used for the injection therapy of facet joints. It is necessary to test a new guidance system which is easy to handle and which is applicably for spinal injection therapies and other simpler fluoroscopic procedures.

In our department we annually treat over 1,500 patients with fluoroscopy guided injection therapy. We therefore have a high rate of radiation exposure and continuously evaluate different possibilities to reduce radiation dose like shielding and automatic adjustment of the fluoroscope. Especially shielding and automatic adjustment showed the difficulty in only reducing radiation dose to the practitioner or the patient, not both. Guidance systems are necessary to reduce radiation dose to the practitioner as well as to the patient. Among the possible fluoroscopy guided interventional procedures, we chose the injection therapy of the lumbar facet joints because this therapy is regularly performed and standardized; and it is easy to evaluate its successful application by recording the pain reduction.

Facet joints are considered to be common sources of chronic spinal pain [6]. When facet joints develop degenerative changes, this can produce considerable back pain, especially in motion. Facet joints may be a source of chronic back pain in 15–45% of patients [6, 21]. The capsules contain mechanoreceptors as well as different nociceptors [7], which are well innervated by the medial branches of the dorsal rami [4, 5]. These nerve fibers are the target for different injection therapies to the facet joints (Fig. 2).

Fig. 2.

a Lumbar spine model and X-ray of the lumbar spine in posterior–anterior-view. Note the marking of the right facet joint L4–L5. b Fluoroscopic image before injection of the right facet joint L4–L5. Note the crosshairs. The SabreSource™ system provides visible crosshairs on the fluoroscope screen that can be positioned to target the desired subsurface structure

In this prospective study, we present a new real time image guidance system for interventional fluoroscopy guided surgical procedures and evaluate a possible reduction of radiation dose compared to fluoroscopy guided injection therapy alone.

Materials and methods

Patients

A total of 60 patients were included in this study of injection therapy to the facet joints L4–L5 or L5–S1. All patients had chronic low back pain for at least 6 months and a lumbar facet joint syndrome with pain in the lower lumbar spine (L4–S1). X-ray (ap and lateral view) and MRI (Siemens TRIO; 3T; T1 and T2 weighted and TIRM sequences) of the lumbar spine were carried out before entering the trial and have been evaluated by an independent and experienced radiologist. Degenerative changes of the facet joints were diagnosed in all patients. Other inclusion criteria were absence of neurological deficits and an increase of the pain during movement or excessive stress. Exclusion criteria were any vertebral fractures, radicular neurologic symptoms, coagulation disturbances, rheumatic diseases, allergy against the local anesthesia, pregnancy, any infections (e.g. spondylitis, spondylodiscitis). An informed consent was obtained from all patients enrolled. The first 30 consecutive patients were included in group I and were treated with fluoroscopy-guided injections alone. The following 30 patients were included in group II and treated with the SabreSource™ targeting system. Before inclusion a visual analogue scale (VAS) score regarding low back pain was obtained from each patient [9]. Pain reduction was evaluated 6 h after therapy by the physician using the VAS. Additionally the patients themselves used a special diary for the documentation of pain (documentation of pain every hour using VAS-scale). Demographic data and the body mass index were recorded.

Injection therapy

A single orthopedic surgeon performed all injection techniques. All procedures were performed using a standard portable C-Arm machine (Philips BV 300, Philips Medical Systems; Eindhoven, the Netherlands). The patient was placed in a prone position and the surgeon stood contralateral to the fluoroscopy unit on the left side of the patient. All patients/fields were prepared in a sterile manner. The fluoroscopy machine was aligned with the patient so that the X-ray beam was at a right angle to the facet joints. Because the injection was carried out periarticular, no further tilting of the image intensifier and the tube was necessary. Both facet joints in one level were treated because of the multisegmental innervation of the joints. Thus, 2 injections in every patient and a total of 120 injections were performed. Before starting the injection, a final X-ray scan was performed to document the correct position of the injection needle in all patients.

The injection points were marked under fluoroscopic control. After desinfection and preparation of the field, the facet joints were punctured under fluoroscopic control (0.9 × 90 mm Sterican injection needle; B. Braun Melsungen, Melsungen, Germany). 2.5 ml of lidocaine (1%) was used for injection. All injections were carried out peri-articulary to the region of the medial branch (Fig. 2), not intra-articular. After injection all patients were observed for 6 h and monitored for pain development. There were no restrictions of movement or exercise.

Radiation dose

The C-Arm unit offers automatic adjustment of the fluoroscopic tube potential and current, depending on the anatomical area examined as well as the size of the patient. The radiation dose was monitored by the fluoroscopic unit, which was equipped with a dose-area-product meter incorporated into the tube head assembly. The hands and the head of the surgeon are always near to the tube. Therefore a reduction of radiation dose measured in the tube head assembly means a reduction for the patient and the critical parts of the surgeon (eyes, hands, thyroid gland). During and after the intervention, the number of exposures, time of active radiation and the radiation dose itself were documented.

SabreSource™ system

The SabreSource™ system (Minrad International Inc.; New York, USA) is a real time image guidance system for pain management and other interventional fluoroscopically guided surgical procedures. An FDA-approval is available. The system consists of:

• The SabreSource™—a laser beam unit which is mounted on a fluoroscopy unit (Fig. 1). The SabreSource™ laser beam unit provides visible crosshairs on the fluoroscope screen (Fig. 2) that can be positioned to target the desired subsurface structure. The SabreSource™ targeting system directs a visible laser beam onto the patient (Fig. 3) which is precisely aligned perpendicular to the target marked by the crosshairs on the fluoroscopic image.

• The Light Sabre™ consisting of an injection needle and a special adapter which is a collimating tube (Fig. 4), that lights up when the needle is positioned in the same direction as the course of the X-ray beam. Using a longer tube will result in a more accurate alignment of the tube with the light beam. Depending on the length of the needle different collimating tubes are available. In this study we used a 50 mm collimating tube.

Fig. 1.

The Sabre Source system. 1 The Sabre Source targeting system directs a visible laser beam onto the patient which is precisely aligned with the cross hairs on the fluoroscopic image. 2 Remote control for moving of the targeting system during the intervention. 3 Alignment sabre (calibration device). 4 Sabre Source needle and collimating tube (laser light adapter). 5 Mounting of the Sabre Source unit onto the fluoroscopy machine

Fig. 3.

Note the visible laser beam, pointing onto the right facet joint L4–L5. The Sabre Source targeting system directs a visible laser beam onto the patient which is precisely aligned with the crosshairs on the fluoroscopic image. This provides the correct point of entry and angle of approach to the desired subsurface target

Fig. 4.

The needle (“1”) with the collimating tube (“2”) is introduced. Position and angulation of the needle can be controlled with the collimating tube which only illuminates when the laser beam hits it longitudinal and thus the needle is parallel to the X-ray beam. This provides the correct angle of approach to the desired target

Before treatment, an initial system setup has to be carried out. This calibration procedure must be performed each time the SabreSource™ unit is mounted to the fluoroscope (Fig. 1). A calibration device is positioned on the surface of the image intensifier. By hitting the markers of the calibration device with the laser beam and matching them with the crosshairs on the screen, the laser is calibrated. This takes only a few minutes and does not have to be repeated for each patient. We treated at least six patients per session without repeated calibration. After desinfection and preparation of the field, the therapy with the Sabre Source system followed three steps:

Alignment of the X-ray unit with the mounted SabreSource™unit

The C-Arm unit is aligned at a right angle to the desired facet joint.

Crosshairs movement

The crosshairs on the fluoroscope monitor are directed to the desired facet joint using a remote control. A final fluoroscopic image is carried out to verify the correct positioning of the crosshairs on the desired facet joint. Due to the use of the remote control the physician can stand away from X-ray unit thus minimizing the radiation exposure to himself. Once the crosshairs are matched with the target (here the facet joints) on the screen of the fluoroscope, the green laser light beam can be seen on the skin and marks the point of entry.

Introducing the needle to its target

The laser beam together with the collimating tube help to ensure the correct direction and angle of approach during the introduction of the needle as the collimating tube only illuminates when the laser beam hits it longitudinally and thus the needle is parallel to the X-ray beam (Fig. 4). This allows the physician to visualize that the instrument is “on-line” to the sub-surface target with an accuracy of ±1 mm at a distance of 1 m. An FDA-approval for the system is available, the accuracy of the SabreSource™ was also tested in a cadaver study [25]. The needle has to be introduced carefully to avoid patient movement. If the patient moves, then the crosshairs might not be on the target any more and a re-positioning of the crosshairs must be carried out. This is possible at any time with a control of the crosshairs by activating the fluoroscopy unit once. It might be possible that the patients move due to pain, but not due to the force of the needle. No patients in our study moved due to the force of the needle or pain.

No activation of the fluoroscopy unit is necessary during the introduction of the needle. After correct placement of the needle, the collimating tube is removed and the injection can be carried out as usual.

Statistical evaluation

A sample size calculation was carried out before starting the trial. Reviewing the literature, we postulated a possible reduction of radiation dose of up to 20% when using navigated injection procedures [13, 16, 18, 23, 27, 28, 30]. We planned to carry out a trial that has a 90% power. For a two tailed P value of 5% we calculated the inclusion of 25 patients in each group. Considering a possible exclusion of patients, we decided to treat 60 patients in all (30 patients per group). We performed univariate analysis of independent variables with the t test for independent samples for quantitative variables such as age, sex, BMI, radiation dose and radiation exposure. All tests were two-sided and a P value < 0.05 was considered to be significant. All analyses were conducted with the SPSS statistical software for Windows 14.0 (SPSS, Chicago, IL, USA).

Results

Patients and evaluation of pain

The mean age of the patients was 60.5 years (range 51–69), mean body mass index was 26.2 (range 22.2–29.9) (Table 1). A total of 120 injections to the facet joints were performed. Each patient was given two injections, either to the facet joints of L4–L5 or of L5–S1. No complications occurred. Before therapy, the mean VAS in all patients was 7.9 (range 6–10). After therapy the mean VAS in all patients was 3.1 (range 0–5). There was a significant reduction of pain after therapy (reduction 60.76%; P = 0.001) but no significant difference between group I (fluoroscopy guided intervention) and group II (SabreSource™ guided intervention) (P = 0.09) indicating that the therapy was equally effective.

Table 1.

Patient data and pain evaluation

	Mean age (years)	Mean BMI	Mean pain pre (VAS)	Mean pain post (VAS)	Reduction of pain (%)
Standard deviation s	(s)	(s)	(s)	(s)
Group I: Fluoroscopy alone	60.6 (4.3)	26.23 (2.17)	8.0 (0.48)	3.2 (0.41)	60
Group II: Sabre Source	60.43 (4.7)	26.26 (2.03)	7.8 (0.52)	3.0 (0.44)	61.5

Radiation dose

The range of the tube voltage was 75–110 kV, the range of tube current was 2.7–4.8 mA. In the fluoroscopy guided group the mean number of fluoroscopy exposures was 4.53 (range 4–6) with a mean time of radiation exposure of 3.46 seconds (range 2–4). The mean entrance surface dose (ESD) in this group was 2.01 mGy (range 1.61–2.42). In the Sabre Source group the mean number of fluoroscopy exposures was 3.3 (range 3–4) with a mean time of radiation exposure of 2.33 s (range 1–3). The mean ESD in this group was 1.36 mGy (range 0.91–1.58). The preparation of the SabreSource™ system and the alignment procedure was completed in a mean time of 6.26 min (range 4.34–8.12). The mean ESD for the alignment procedure was 0.4 mGy (range 0.25–0.65).

Compared to the fluoroscopy guided injection the SabreSource™ System significantly reduced the number of fluoroscopy exposures (reduction 27.2%, P = 0.01), the time of radiation exposures (reduction 32.7%, P = 0.01) (Fig. 5) and the mean ESD (reduction 32.3%, P = 0.01) (Fig. 5).

Fig. 5.

Boxplot illustration of radiation doses found in both groups. Note the significant reduction of the mean entrance surface dose (mGy) in the group with SabreSource™ (P = 0.02)

Discussion

Radiation exposure

Approximately 4–10 million interventional pain procedures are performed annually in the USA, with at least 50% of them being performed under fluoroscopy [10, 30]. A 231% increase in facet injection rates between 1994 and 2001 was documented [8, 10]. Fluoroscopy was commonly used by radiologists (in 78% of their procedures) and, increasingly by surgeons (34%). With an increase of fluoroscopy guided interventions, there is also an increase of radiation exposure and applied radiation dose. Due to the cumulative effect, radiation could pose a major threat to patients and physicians who have to stay close to the radiation source during the intervention, often with their hand in the X-ray beam.

Yoshinaga et al. reviewed several epidemiologic studies [29]. The most consistent finding was increased mortality due to leukemia among early wokers employed before 1950 [29]. Several studies provided evidence of a radiation effect for breast cancer and skin cancer [29]. Other studies provided an even higher evidence especially for skin cancer [1, 20]. Berrington de Gonzales et al. attempted to quantify the risk of cancer induction from diagnostic X-ray procedures [2]. Their analysis suggests that e.g. in Australia about 431 cancers per year (1.3% of all cancers) could be attributable to diagnostic X-rays. The corresponding percentages for 14 other countries considered ranged from 0.6% in the UK and Poland to 3.2% in Japan [2]. These studies show that exposure to X-rays, even at low-dose-levels, increases the risk of cancer.

It is essential to minimize exposure time, maximize distance between occupational personnel and radiation sources and to use shielding if possible. Intermittent fluoroscopy and last image hold/electronic collimation have been described as dose reduction and saving features [16, 18, 23] and were used in this trial. Another possible way to reduce exposure time is to use navigation/tracking devices. In this study we used the new image guidance system SabreSource™. The SabreSource™ significantly reduced the exposure time (32.7%) and radiation dose (32.3%) compared to fluoroscopy guided injection alone (Fig. 5). Reviewing the literature, there are various studies presenting different fluoroscopy guided injection procedures and multiple pain assessments [8, 24, 26, 27, 30]. Only few studies present radiation data and can be found especially in vertebroplasty-, and pedicle screw placement-studies [13, 17, 18]. Manchikanti et al. evaluated different radiation exposures in interventional pain management [17]. A total of 1,000 patients were reviewed retrospectively, of which 352 patients were treated with lumbar facet joint injections [17]. Mean time of radiation exposure was 5.7 s (range 1–14). In a second and third study by the same authors, they evaluated 1,656 patients who had a radiation exposure time ranging from 4.9 to 7.5 s (these interventional pain management procedures were performed by experienced physicians) [17, 18] and which is still higher than in our study. (3.46 s). The different radiation dose reported by various authors can be attributed to diverse surgical practices and projections used, as well as different types of X-ray units and different additional protective equipment used. Especially different projections and angulations may increase or decrease radiation dose significantly. The X-ray beam has to pass through more tissue when using a lateral angulation of 20–40° for intraarticular injection to the facet joints leading to a significant increase of radiation dose. Moreover one sometimes needs several shots to find the right alignment of the C-Arm. In this study the medial branch was injected and the needle was not placed into the facet joints Thus no angulation of the fluoroscopy machine was necessary for this procedure. We have used the SabreSource™ system in other interventional therapies such as vertebroplasties and discographies where an angulated view is mandatory. In those procedures it is very beneficial that the entrance point is shown on the skin because especially in obese patients it may lie far laterally. We think that in such procedures the benefit of a targeting system is even greater.

Facet joints may be a source of chronic back pain in 15–45% of patients [6, 21]. Injection of the facet joints is a widely performed technique to treat patients suffering from facet joint syndrome. There are various studies showing a significant reduction of pain from facet joint syndrome after injection therapy [8, 17, 18, 24, 26, 27]. In our study, we saw a significant short-term improvement (60 resp. 61.5% reduction of pain). There were no significant differences between the two groups in terms of pain relief either immediately after injection therapy nor at the follow-up. In our study we used a periarticular injection technique. Some authors prefer an intraarticular technique. Properly performed intraarticular facet injections may be more accurate in diagnosing facet joint pain. However, intraarticular injections can be technically challenging and have there own limitations. An intra-articular injection also bears the risk of a rupture of the capsule due to a distinct joint space capable of only 1 to 1.5 ml of fluid. Additionally, there are often degenerative changes in the facet joints including a reduction of the joint space making it difficult for the needle to enter. In a cadaveric study, Kellegren et al. demonstrated that 0.5 ml injectate spread into an area encompassing 6 cm² of tissue [14]. In view of the close proximity of the medial branch nerves, even the injection of a low-volume of anesthetic is likely to block these nerves [6, 8, 14]. Reviewing the literature there are only a few trials comparing intra-articular versus peri-articular injection techniques. Nash et al. conducted a prospective study in 67 patients [22]. Only 26 patients completed the study, 12 reported the periarticular injection to be more beneficial, 11 patients reported the intraarticular injection to be better, 3 reported no difference. Marks et al. randomly assigned 86 patients to receive either intra-articular or periarticular injection therapy [19]. No difference in immediate pain reduction was found between the two groups.

The SabreSource™ is a real time image guidance system. It represents the step between a navigation system and stand-alonefluoroscopy. A CT based navigation system has the advantages of three dimensional visualisation and immunity to movement errors. But it is very expensive, requires complicated hard- and software, a reference base must be fixed to the patient and a matching procedure is usually necessary. Nobody would do all this for a simple procedure like a facet joint injection. Those percutaneous procedures are usually done by fluoroscopic or CT guidance. The new image guidance system we present here, aims to reduce the radiation dose during the simpler procedures. It does not compete with navigation systems but aims to increase accuracy and lower radiation during fluoroscopic guided interventions.

The accuracy is higher and the radiation lower than with a stand-alone fluoroscopy guided setup. The complexity and the costs of the SabreSource™ is much lower than that of a navigation system, as it works mechanically and does not need any additional software. The hardware can easily be mounted to a conventional C-arm and no further cameras or other hardware is necessary. It is therefore less expensive and easy to handle with a sufficient accuracy for percutaneous infiltrations without extending the of operation time.

Beside the protection with a lead apron, the reduction of the number of exposures as well as the distance to the target are most effective in reducing radiation dose. As explained above, distance is not easy to handle when doing injection therapy and the hands of the surgeon cannot be sufficiently protected. When using lead gloves, the automatic adjustment of the fluoroscope unit increases tube potential and current significantly. An advantage of the new targeting system is that the physician may step back during the radiation exposure and does not need to use radiation while introducing the needle; the physician can rely on the laser to accurately guide the needle to the target at the right angle of approach. Radiation is only used to adjust the laser to the target and not while the needle is introduced along the laser beam.

The system is easy to handle. In our experience, there was only a prologation of operation time due to the setup procedure, which can be done by trained nursing stuff or technicians (no special education necessary). With this procedure a calibration of the system is carried out. It is only necessary to calibrate the system once prior to starting the treatment and not for every consecutive patient. The more patients you treat in a row, the lower the setup-time is per patient. We usually treated at least six patients in a row. Compared to the fluoroscopy guided infiltration there is a possible extension of the procedure time of about 1 min for each patient. There is a significant learning curve with regards to the calibration procedure. When starting to work with the system, the authors needed 10–12 min for the preparation of the system. After about 20 procedures, the setup was only half of the time. The preparation of the Sabre Source system and the general alignment procedure in this study were finished after a mean time of 6.26 min (range 4.34–8.12). This is one possible disadvantage of the system. If only one patient is treated with the system, there will be a significant prolongation of operation time. Another possible disadvantage is the movement error. The needle has to be introduced carefully to avoid a movement of the patient. If the patient moves, the alignment is lost and the crosshairs on the fluoroscope monitor have to be redirected. Due to the automatic adjustment of the fluoroscopic tube potential and current the applied radiation dose during the calibration procedure (0.4 mGy) is minor. However there is a loss of time and there is a minor radiation exposure to the physician during the calibration procedure.

A possible limitation of the study is that we included the patients into consecutive groups. The use of consecutive patient groups instead of a complete randomisation gave us the possibility to use one setup for more than one patient. Otherwise the SabreSource™ would have to be mounted and calibrated for each patient. As we wanted to test the feasibility of this device it was important to use it under realistic conditions.

Conclusion

The SabreSource™ reduces radiation exposure and radiation dose during the injection therapy of the lumbar facet joints. The system is easy to handle and enables medical professionals to improve the reduction of radiation dose of interventional procedures. It may be applied for injection therapies and other percutaneous procedures (e.g. discography and vertebroplasty).