Radiation Dose Reduction in CT-Guided Spine Biopsies Does Not Reduce Diagnostic Yield

BACKGROUND AND PURPOSE: CT-guided biopsy is the most commonly used method to obtain tissue for diagnosis in suspected cases of malignancy involving the spine. The purpose of this study was to demonstrate that a low-dose CT-guided spine biopsy protocol is as effective in tissue sampling as a regular-dose protocol, without adversely affecting procedural time or complication rates. MATERIALS AND METHODS: We retrospectively reviewed all patients who underwent CT-guided spine procedures at our institution between May 2010 and October 2013. Biopsy duration, total number of scans, total volume CT dose index, total dose-length product, and diagnostic tissue yield of low-dose and regular-dose groups were compared. RESULTS: Sixty-four patients were included, of whom 31 underwent low-dose and 33 regular-dose spine biopsies. There was a statistically significant difference in total volume CT dose index and total dose-length product between the low-dose and regular-dose groups (P < .0001). There was no significant difference in the total number of scans obtained (P = .3385), duration of procedure (P = .149), or diagnostic tissue yield (P = .6017). CONCLUSIONS: Use of a low-dose CT-guided spine biopsy protocol is a practical alternative to regular-dose approaches, maintaining overall quality and efficiency at reduced ionizing radiation dose.

I maging-guided biopsy is a commonly used method to obtain a tissue diagnosis in suspected cases of malignancy. In particular, CT guidance is often used for precise localization of a lesion before and during biopsy. It provides the operator with great anatomic detail for biopsy planning and execution and allows for confirmation of needle placement into the area of concern. CT guidance is the preferred method of biopsy for osseous lesions within the vertebrae. [1][2][3][4] Even though CT guidance has become increasingly used for various procedures, there is concern over the amount of radiation exposure to the patient. [5][6][7][8] Radiation dose reduction is commonly used in routine diagnostic CT scanning. Pediatric patients and patients who receive multiple scans for acute disease follow-up, chronic conditions, and screening purposes often undergo CT with modified scanning protocols to reduce dose. [9][10][11][12][13][14] This type of protocol modification has also been used in CT-guided interventions to limit radiation dose when performing multiple scans during the procedure. 8,[15][16][17][18] Given the increased desire to reduce radiation dose to patients, we transitioned our protocols for CT-guided spine biopsies to use a lower dose.
The purpose of this study was to demonstrate that a low-dose protocol for CT-guided spine biopsies is as effective in tissue sampling without an increase in procedural time or an increase in complication rates compared with our legacy higher-dose approaches.

MATERIALS AND METHODS
After obtaining Institutional Review Board approval, we retrospectively reviewed all patients who underwent CT-guided spine procedures at our institution between May 2010 and October 2013. The total number of charts reviewed was 132.
Patients who underwent disk space aspirations and biopsies for suspected diskitis/osteomyelitis were excluded because of limited availability of surgical pathology data as most specimens were only submitted for microbiology analysis. CT-guided pain management procedures such as facet cyst ruptures and epidural injections were also excluded. Patients for whom dose reports were not available in our institution's PACS were excluded.
Ultimately, 64 patients were included in this analysis. Two lesions were biopsied in 2 patients and 1 lesion in the remaining 62 patients, yielding a total of 66 lesions. All the biopsies were performed by 1 Certificate of Added Qualification-certified neuroradiologist (A.H.D.) with 6 years of experience. The low-dose protocol was initiated in February 2012 and has been almost exclusively used since November 2012.

Procedure
All CT-guided spine biopsies were performed on a 4-channel CT scanner (Volume Zoom; Siemens, Erlangen, Germany) or 8-section CT scanner (LightSpeed Ultra; GE Healthcare, Milwaukee, Wisconsin) in helical mode based on availability. The 8-section scanner was used to guide 50 biopsies (78.1%) and 4-section scanner for the remaining 14 biopsies (21.9%). CT fluoroscopy was not available. Patients all followed a standard course for these biopsies. Each was positioned prone for the procedures. Vital signs were monitored. Mild to moderate conscious sedation was used in 60 patients (93.8%), monitored anesthesia care in 2 patients (3.1%), and local anesthetic only in 2 patients (3.1%). A Fast Find Grid (Webb Manufacturing, Philadelphia, Pennsylvania) was placed over the general biopsy site for localization. In each patient, 1 preprocedure CT scan was obtained using a regular-dose protocol (120 peak kilovoltage [kVp]) and 200 mAs) for planning. Skin was prepped and draped in normal sterile fashion. One percent lidocaine was infiltrated into tissues for local and deep anesthesia. An 11-, 12-, or 13-gauge bone biopsy needle set (Osteo-Site; Cook, Bloomington, Indiana, or Bonopty; AprioMed, Londonderry, New Hampshire) was advanced into the lesion with CT images obtained after each needle advancement. Once the needle was confirmed within the lesion, CT scans were performed after each biopsy pass. In each patient, 1 final postbiopsy scan was obtained after the needle was removed using regular-dose parameters to assess for postprocedural complications. Patients were then transferred to a recovery area to be monitored before discharge or return to their hospital room.

Data Collection and Scanning Parameters
Data from PACS and dose reports were collected, including age, sex, location, and characteristics of lesion biopsied, kVp, mAs, pitch, volume CT dose index (CTDIvol) per series (milligray [mGy]), CTDIvol total (mGy), scan range (mm), dose-length product (DLP) per series (mGy⅐cm), total DLP (mGy⅐cm), number of biopsy-guiding scans, number of preand postbiopsy diagnostic scans, number of needle passes, total number of scans, duration of each biopsy (defined as time from the first prebiopsy scan to last postbiopsy scan), and complications. Pathology results were obtained for each patient from electronic medical records.
Low-dose biopsies were defined as those with a kVp of 80 and mAs of 40 -60. Regular-dose biopsies were defined as those with a kVp of 120 and mAs Ͼ200. Scans performed at kVp and mAs parameters outside the above-mentioned criteria of low-dose or regular-dose biopsies were classified based on average CTDIvol (CTDIvol Ͻ10 mGy for low dose; CTDIvol Ͼ10 mGy for regular dose) as previously described by Kröpil et al. 19 They defined lowdose CTs as having a CT dose index Ͻ10 mGy. For example, 2 patients whose biopsies were started as low-dose protocol were switched to regular-dose protocol at the operator's discretion because of insufficient conspicuity of subtle lesions and were classified as "regular-dose" because the average CTDIvol was 17.1 mGy in one and 20.3 mGy in the other. Figure 1 demonstrates representative images from regular-dose and low-dose CT-guided spine biopsies.
Diagnostic tissue yield was classified as "positive for malignancy," "specific benign diagnosis," and "negative for malignancy without a specific benign diagnosis." Lesions were classified as lytic, sclerotic, or mixed. The location of lesions was recorded as cervical, thoracic, lumbar, or sacral.
Age, biopsy duration, total number of scans (including prebiopsy and postbiopsy scans), total CTDIvol (including that used for prebiopsy and postbiopsy scans), and total DLP (including that used for prebiopsy and postbiopsy scans) of low-dose and regular-dose groups were compared using an unpaired t test (GraphPad Prism software; GraphPad Software, San Diego, California). Diagnostic tissue yield and the distribution of lesions by type and location of low-dose and regular-dose biopsies were compared using Fisher exact test (GraphPad Software). P value Ͻ .05 was considered statistically significant.

RESULTS
Of the 64 patients who underwent CT-guided spine biopsies from 2010 to 2013, 29 patients (45.3%) underwent the procedure using a low-dose protocol and 35 patients (54.7%) using a regular-dose protocol. Table 1 demonstrates the mean and ranges for age, number of scans, duration of procedure, total CTDIvol, and total DLP for low-dose protocol; Table 2 denotes the same for regulardose protocol.

Demographics
There was no significant difference between the 2 groups in patient age (63.86 Ϯ 13.67 years for low dose versus 59.49 Ϯ 14.6 years for regular dose; P ϭ .2239) or in sex distribution (14 of 29 or 48.3% women for low dose versus 19 of 35 or 54.3% women for regular dose; P ϭ .802).

Dose and Scanning Time
There was a statistically significant difference between low-dose and regular-dose groups in total CTDIvol (69.47 Ϯ 24.76 mGy for low dose versus 285.2 Ϯ 132.6 mGy for regular dose; P Ͻ .0001) and total DLP (601.5 Ϯ 237.7 mGy⅐cm for low dose versus 1541 Ϯ 648.1 mGy⅐cm for regular dose; P Ͻ .0001) (Fig 2).
There was no significant difference in total number of scans obtained (11.38 Ϯ 4.354 for low dose versus 12.46 Ϯ 4.527 for regular dose; P ϭ .3385) and duration of procedure (34.31 Ϯ 12.19 minutes for low dose versus 38.17 Ϯ 8.92 minutes for regular dose; P ϭ .149) between the 2 groups (Fig 2).
Several outliers were noted, falling greater or less than 2 standard deviations from the mean. One patient in the low-dose group who had 2 lesions (one mixed and one sclerotic) biopsied had significantly more scans, longer duration of the procedure and higher total DLP than average. Two patients in the regulardose group had significantly more scans than average (25 and 29) because of difficulty in accessibility of small vertebral body lesions, which resulted in significantly higher than average total CTDIvol (761.86 mGy) in one and total DLP (3062.46 mGy⅐cm) in the other.

Biopsy Results
There was no significant difference between the 2 groups in the proportion of cases positive for malignancy (20 of 29 or 69.0% for low dose versus 21 of 35 or 60% for regular-dose; P ϭ .6017), those with a specific benign diagnosis (2 of 29 or 6.9% for low dose versus 6 of 35 or 17.1% for regular dose; P ϭ .2754), and those whose pathology was negative for malignancy without a specific benign diagnosis (7 of 29 or 24.1% for low dose versus 8 of 35 or 22.9% for regular dose; P ϭ 1.00).
Of the 66 lesions that were biopsied, 39 (59.1%) were lytic, 15 (22.7%) were sclerotic, and 12 (18.2%) were mixed. There was no statistically significant difference in lesion type between the low-dose and regular-dose groups (P values ranging from .1174 to .7694).
Most of the lesions that underwent biopsy were located within the lumbar spine (29 of 66; 44%). This was followed by the thoracic spine (28 of 66; 42.4%), sacrum (7 of 66; 10.6%) and cervical spine (2 of 66; 3%). There was no statistically significant difference in the location of lesions between the low-dose and regular-dose groups (P values ranging from .4334 to 1.00).
There were sufficient specimens for diagnosis in all patients in both biopsy groups. Overall, there was only 1 minor complication characterized by bleeding from the cannula, which was successfully   treated with Gelfoam (Pfizer, New York, New York), in a lowdose group patient whose biopsy yielded metastatic renal cell carcinoma. No major complications were reported.

DISCUSSION
Imaging guidance for biopsy is a commonly used procedure in patients with imaging findings concerning for malignancy. In particular, CT guidance has been used for biopsy of a variety of sites within the body. [1][2][3][4]8,15,16,[20][21][22][23] This is largely attributed to an improved ability of the operator to identify the lesion and plan a trajectory for biopsy. CT-guided biopsy has been shown to be an effective tool in identifying pathology with relatively low risk and cost compared with open biopsy. 4,22,24 However, a frequently cited concern with CT scanning is the potential consequences of ionizing radiation, and there is much emphasis on limiting radiation to as low as reasonably achievable to obtain the necessary results whenever possible. 8,25 Previous studies have demonstrated the utility of a low-dose CT technique for a variety of interventional procedures. Meng et al 15 performed biopsies of lung lesions at lower doses and found that a reduction in the measure of radiation dose, CT dose index, and DLP were possible without sacrificing diagnostic yield. Smith et al 8 were able to reduce the radiation dose to the chest during CT-guided percutaneous lung biopsies by greater than 95% (from DLP of 677.5 mGy⅐cm to 18.3 mGy⅐cm) without decreasing technical success or patient safety. Pediatric CT-guided bone biopsies have been performed using lower mAs and kVp techniques producing acceptable image quality and providing similar diagnostic yield compared with standard techniques. 16 A low-dose CT protocol has also been used in spinal pain interventions. One study found that a change in CT parameters to lower radiation dose resulted in an 86% reduction in total DLP (from 1458 mGy⅐cm to 199 mGy⅐cm) for CT-guided spine injection procedures for pain. 17 Artner et al 18 demonstrated that the dose related to CTguided sacroiliac joint injections can be significantly reduced to levels of pulsed fluoroscopy without compromising needle placement into the joint.
In this study, we found a significantly reduced radiation dose as expressed by CTDIvol and DLP in patients undergoing CTguided spine biopsies using a low-dose protocol compared with a regular-dose protocol. There was no significant difference in the total number of biopsy scans, procedure time, or in the diagnostic yield between the groups. To our knowledge, this is the first study demonstrating a significantly reduced patient exposure to ionizing radiation during CT-guided spine biopsies without sacrificing the quality, efficiency, and diagnostic yield of the procedure.
Although there was a significantly lower radiation exposure in the low-dose biopsy group compared with the regular-dose group, we predict that a DLP might be lowered even further by reducing tube voltage (mAs) and/or current (kVp), increasing the pitch and decreasing the scan range in the z-axis. 20,[26][27][28][29] Intermittent axial scanning mode rather than helical mode and the use of a stationary CT table may further contribute to radiation dose reduction. 24,30 A substantial proportion of radiation exposure comes from pre-and postbiopsy scans because they are designed to optimize soft tissue visualization for needle guidance and to exclude post-biopsy traumatic sequelae. In fact, 1 study showed that up to 90% of the total patient dose during biopsies was administered during the helical planning stage. 29 Therefore, prebiopsy diagnostic imaging should be carefully reviewed beforehand to determine whether repeat conventional dose scanning may be avoided during the procedure. 31 If a prebiopsy scan is necessary, a grid can be placed over the spinal level of interest before the first series of scans based on known anatomic landmarks. 31 Chintapalli et al 31 suggest that in low-risk CT-guided interventions, which may include some spine biopsies, regular-dose postbiopsy scans can be eliminated at the discretion of the radiologist. A low-dose protocol, as well as techniques to further reduce dose, should be familiar to the radiologist performing the procedures and technologist acquiring the images. 31 Newer techniques have recently emerged to address image quality when reducing CT dose. These include iterative reconstruction models such as adaptive statistical iterative reconstruction and model-based iterative reconstruction. [9][10][11][12][13][14][32][33][34][35][36][37][38][39][40] Although these imaging algorithms provide an additional method for dose reduction in CT-guided procedures, their availability is currently limited to newer CT scanners for routine diagnostic CT imaging. The greater availability of the iterative reconstruction software over time may allow for increased operator comfort when evaluating low-dose images during CT-guided procedures, potentially further reducing the radiation dose.
This retrospective study did have some limitations. It was not randomized, and a single operator performed most of the CTguided spine biopsies. Therefore, the reproducibility of the results using the low-dose protocol cannot be fully assessed in this study. In addition, it would be difficult to determine whether increasing comfort with the procedure may have contributed to a slightly greater efficiency of the procedure using a low-dose protocol. The retrospective nature of this study also limits assessment of factors related to operator scanning protocol adjustments in challenging biopsy cases.

CONCLUSIONS
Radiation exposure to patients undergoing CT-guided spinal biopsies can be optimized to reduce the overall dose during the examination. Low-dose CT-guided spine biopsies have a significantly lower total cumulative radiation exposure compared with regular-dose CT biopsies without significantly affecting procedural time or diagnostic tissue yield. A simple dose-reduction protocol can use reduction in mAs and kVp during the procedure. A number of additional modifications to image acquisitions can be made to reduce the dose. Our data show that a low-dose protocol should be considered as an alternative to regular-dose protocol when performing CT-guided spinal biopsies, allowing the operator to reduce ionizing radiation dose while maintaining overall quality and efficiency of the procedure.