Rapid Implementation of Intraoperative Ultrasonography to Reduce Wire Localization in The Permanente Medical Group

Tweet

Sharon Chang, MD¹; Magdalene Brooke, MD^2,3; Elizabeth Cureton, MD³; Alice Yeh, MD⁴; Rhona Chen, MD⁴;
Nicole Mazzetti-Barros, MPH/CHES⁴; Reza Rahbari, MD⁵; Sherry Butler, MD⁴; Nicole Hill, MD⁵; Veronica Shim, MD³

Perm J 2019;23:18-073 [Full Citation]

https://doi.org/10.7812/TPP/18-073

E-pub: 06/14/2019

ABSTRACT

Context: Preoperative wire localization (WL), the most common localization technique for nonpalpable breast lesions, has drawbacks including scheduling constraints, cost, and patient discomfort.
Objective: To reduce WL use in our health care system, we investigated using hydrogel clips to facilitate intraoperative ultrasonography-guided lumpectomies.
Design: We retrospectively reviewed electronic medical records of patients with nonpalpable, ultrasound-visible breast lesions who underwent lumpectomy by 7 surgeons at 4 pilot sites in Kaiser Permanente Northern California between January 2015 and October 2015. Hydrogel clips, used for several years before the study period, were placed routinely during core-needle biopsy in all patients with nonpalpable, ultrasound-visible breast lesions.
Main Outcome Measures: Localization method, lesion size, margin positivity, and receipt of neoadjuvant therapy.
Results: One hundred forty-three patients underwent hydrogel clip placement and lumpectomy by pilot-site surgeons. Localization consisted of intraoperative ultrasonography alone, preoperative skin marking, or WL. Of the 143 patients, 71.3% did not need WL (60.8% ultrasonography alone and 10.5% skin marking). The non-WL and WL groups had similarly sized lesions, and the positive margin rate was 7.2% overall, with no significant difference between the non-WL and WL groups (5.9% vs 11.5%, p = 0.33). Of the 12 patients who underwent neoadjuvant chemotherapy, 8 (67%) did not require WL.
Conclusion: A multifacility protocol using intraoperative ultrasonography to visualize hydrogel clips was implemented, which decreased WL procedures and produced no significant difference in margin positivity between the WL and non-WL groups. This technique can be a cost-effective alternative to WL in patients who are candidates for hydrogel clip placement.

INTRODUCTION

The number of nonpalpable breast lesions detected on routine screening has increased with improving imaging technology, creating new challenges as surgeons must localize these lesions before removing them.¹ The most common localization technique is preoperative placement of a wire,² which has major drawbacks, including patient discomfort and delays to the operating room (OR) on the day of surgery.

Alternatives to wire localization (WL) include radioactive seed placement, placement of nonradioactive seeds, intraoperative ultrasonography, and intraoperative ultrasonography aided by an ultrasound-visible clip. Radioactive seeds have been used with excellent results.^3-6 Unfortunately, the resources required to manage radioactive seeds can outweigh the benefits of the technique. Furthermore, seed placement requires a second procedure before surgery.⁷ Several nonradioactive seed products have come onto the market, but data on these products are limited, the seeds can be very expensive, and they too require a second procedure for seed placement before surgery.⁸

Another approach is intraoperative ultrasonography.^9,10 One advantage of this technique is that most hospital systems already have ultrasonography machines, which minimizes additional costs. The main downside is that intraoperative ultrasonography alone is very operator dependent, and this technique can be difficult to implement across large, diverse medical systems.

Several groups have described using hydrogel-based, ultrasound-visible clips (HydroMARK, Leica Biosystems, Cincinnati, OH) to facilitate intraoperative ultrasonography.^11-13 When the clip is placed during biopsy, it looks like a typical metal clip. Within a few days after deployment, the biodegradable hydrogel polymer that surrounds the clip absorbs fluid from adjacent tissue, creating a hypoechoic, ultrasound-visible “bubble.” Reexcision rates, specimen sizes, and complication rates are comparable between this technique and traditional WL.^11,13 Furthermore, this technique does not require extra equipment other than the clip, and patients do not undergo procedures beyond the initial biopsy. Previous studies have described single-center and/or single-surgeon experiences.^11-13 We hypothesized that we could rapidly implement intraoperative ultrasonography aided by hydrogel clips across a large, diverse health care system to decrease WL use without sacrificing quality measures.

METHODS

As part of a performance improvement project to decrease WL, 4 Kaiser Permanente Northern California (KPNC) pilot sites used hydrogel clips in combination with intraoperative ultrasonography for patients undergoing lumpectomy. The breast imaging teams at each of these facilities had used hydrogel clips routinely for several years before initiation of this study to percutaneously mark all lesions undergoing biopsy using ultrasound guidance. Postbiopsy mammograms were performed after each biopsy to assess the accuracy of clip placement. Our study focused on nonpalpable, ultrasound-visible lesions identified during diagnostic workup of abnormalities identified on screening mammography. Patients with malignant, high-risk, or benign but discordant biopsy results were referred to the respective facility’s Surgery Department, and pilot-site surgeons performed office-based ultrasonography during the initial office consultation to determine whether these patients were candidates for intraoperative ultrasonography. If the surgeon could visualize the clip and/or the mass easily with ultrasonography in the office, no additional localization procedures were performed before surgery. If the surgeon could not visualize the clip and/or mass well, preoperative skin marking (SM) or WL was requested. Localization procedures (SM or WL) were performed by breast imagers on the day of surgery, before patients went to the OR. In the OR, the surgeon performed a lumpectomy using intraoperative ultrasonography alone, intraoperative ultrasonography guided by SM, or WL with no intraoperative ultrasonography.

Breast imaging, biopsy, and pathology workflows at each facility followed standard KPNC protocols with comparable equipment at each site. Pilot-site surgeons had different breast surgery volumes and varying levels of experience with breast ultrasonography (Table 1), but all surgeons had access to ultrasonography machines in both the office and the OR.

We performed a retrospective review of the electronic medical records of all candidate patients from January 2015 to October 2015. Inclusion criteria were as follows: Nonpalpable, ultrasound-visible lesion; placement of a hydrogel clip during biopsy; and lumpectomy by 1 of the 7 pilot-site surgeons.

We examined localization type, lesion size, margin positivity, and whether patients received neoadjuvant chemotherapy. Comparisons were made between the group that required WL and the group that did not. Statistical analysis was performed using statistical software (Stata 13, StataCorp, College Station, TX). We used a Pearson c² test for comparison of categorical variables and a t-test for continuous variables, with statistical significance defined as a p value less than 0.05.

This study was conceived and performed as a quality improvement project, and all activities in this study were carried out as part of routine clinical care. Thus, it was deemed by our institution not to require review by the institutional review board.

RESULTS

A total of 143 patients met inclusion criteria. All the patients were women, with ages ranging from 25 to 82 years. The distribution of lesion types is summarized in Table 2. Of the patients, 87 underwent only intraoperative ultrasonography and 15 had preoperative SM by the breast imaging team. In total, 71.3% (n = 102) of the patients avoided WL. Preoperative migration of the hydrogel clip, noted on postbiopsy mammogram, was seen in 3 cases (1.7%). Only 1 of these 3 patients required WL, because of the extent of clip migration and the surgeon’s inability to visualize the lesion under ultrasound guidance.

The mean lesion size on imaging was 1.32 cm³ (range = 0-5.2 cm³). There was no statistically significant difference in lesion size between patients requiring WL and those patients whose lesions were localized with ultrasonography alone (Table 3). There also was no significant difference in mean tissue volume excised between the groups (Table 3). Most importantly, there was no significant difference in the rate of positive margins (Table 3).

Most of the pilot-site surgeons had experience with breast ultrasonography (Table 1). We did not track how many cases each surgeon performed before developing the confidence to use intraoperative ultrasonography instead of WL. Notably, at our South San Francisco site, 1 of the 2 pilot-site surgeons began the study with minimal ultrasonography experience. During the course of 8 months, combining the data for both surgeons, we observed a steady decrease in the number and percentage of WL procedures, with no WL performed for ultrasound-visible lesions in the final 4 months and with a consistent case volume for each surgeon (Figure 1).

Twelve patients underwent neoadjuvant chemotherapy. The mean time from biopsy to surgery was 179.8 days. Only 4 patients (33%) required WL. Those who required WL had a longer interval before surgery (mean = 191 days) than those who did not require WL (mean = 174 days), although this difference was not significant (p = 0.26). One patient did not require WL despite clip placement 209 days before surgery.

DISCUSSION

This study is the first to examine the implementation of an intraoperative ultrasonography/hydrogel clip protocol for breast lumpectomies across a large, integrated health care system, in contrast to the single-surgeon and/or single-institution reports published in the literature.^11-13 Most surgeons in this study had some previous experience with ultrasonography, although the level of training ranged from no formal training to breast ultrasonography certification. For those with less experience, proficiency was achieved in a short time. The surgeon who had the least initial experience in breast ultrasonography was able to move completely away from WL for ultrasound-visible lesions within 5 months of starting the protocol. The combination of intraoperative ultrasonography and hydrogel clips reduced the need for WL in patients with ultrasound-visible, nonpalpable lesions, so that more than 70% of these patients did not need WL. We found no significant difference in lesion size, volume of tissue excised, or margin positivity between those patients who required WL and those who did not. The intraoperative ultrasonography/hydrogel clip technique also was effective in patients who required neoadjuvant therapy. Only 4 (33%) of these 12 patients required WL. This rate is consistent with the findings of other groups.^11,13,14 Surprisingly, we found hydrogel clips to be visible on images for up to 209 days.

Interestingly, many of the pilot-site surgeons reported that the hydrogel clip served as a training tool, which helped increase their confidence with intraoperative ultrasonography in general. However, clip migration can be a concern. In our series, radiographic-evident preoperative clip migration was documented in only 3 instances (1.7%). One of the patients had substantial enough clip migration combined with poor ultrasonographic visualization of the index lesion to require a WL procedure. Klein et al¹² found only 2 cases (6.4%) of true radiographic-proven clip migration, but they also found that extrusion of the biopsy marker intraoperatively during transection of the biopsy tract occurred up to 45.2% of the time. We did not review our clip extrusion rates. However, surgeons reported that although they could visualize the clip well by intraoperative ultrasonography and could remove the tumor successfully, sometimes they did not see the clip on the specimen mammogram. Presumably, this lack of visualization was caused by extrusion or accidental suction catheter removal of the clip. This suggests that another benefit of the technique is to eliminate the use of the hydrogel clip after an initial training period. Placing clips only in lesions that meet appropriate size and visibility criteria can increase the cost-effectiveness of the overall strategy.

The rapid learning curve and the decreasing dependence on the hydrogel clip are important considerations, especially because US surgeons have a wide spectrum of ultrasonography training and experience. Sclafani et al¹⁵ surveyed breast surgeons who completed fellowship training between 2005 and 2009. Even in this highly selected group of fellowship-trained breast surgeons, 31% of those who responded felt “poorly prepared,” 17% felt “moderately prepared,” and only 31% felt “well prepared” to use ultrasonography in practice. Extrapolating to the population of surgeons performing breast surgery in the US, most of whom are not fellowship trained,¹⁶ we can assume the level of comfort with ultrasonography is probably even more variable. We found that one of the strengths of the intraoperative ultrasonography/hydrogel clip technique is the ease and rapidity of adoption, independent of initial ultrasonography experience.

Our study has several limitations. First, the surgeons involved in the pilot study were volunteers interested in using cost-effective alternatives to WL to decrease their WL rates. The general population of surgeons may not be as motivated to try intraoperative ultrasonography because this technique has a learning curve, requires practice, and necessitates coordination with a supportive Radiology Department. In a fee-for-service environment, breast imagers may not be as motivated to minimize WL use, as this could potentially compromise their revenue stream. In addition, we performed retrospective chart review rather than a randomized prospective study, which likely introduced selection bias. Tumors with better-circumscribed margins may have been more amenable to intraoperative ultrasonography, and these features could have contributed to the trends toward smaller lumpectomy volumes and lower rates of positive margins in the ultrasonography group. Notably, we did not attempt to correlate tumor type or other variables such as the depth and location of the tumor with positive margin rates or the need for WL, because the sample sizes were too small to make meaningful comparisons. Finally, ultrasonography is an operator-dependent modality, although the presence of an easily visualized hydrogel clip helps to mitigate the difficulty of identifying a breast lesion. Despite these limitations, the success of our pilot initiative as a proof of concept has led to an expansion of the project throughout KPNC. The next phase will include a formal training course for all breast surgeons in KPNC. We plan to track intraoperative ultrasonography use and outcomes both before and after training, and to develop a prospective database that will help us define variables that contribute to the success or failure of this technique.

CONCLUSION

We implemented an intraoperative ultrasonography/hydrogel clip alternative to WL for breast surgery in a large health care delivery system, with excellent outcomes. This required the coordination of surgeons and breast imagers across multiple facilities but was relatively easy to accomplish because of the strong integration of the KPNC health care system. Although we did not collect data on the patient day-of-surgery experience for our study, eliminating WL for many patients likely improved their surgical experience because WL can be a source of substantial distress and discomfort on the day of surgery.¹⁷ 

This technique is cost-effective. Our teams, using existing equipment, simplified OR scheduling, reduced ultrasound-guided WL procedures by more than 70%, and freed radiology resources for other breast imaging procedures. Collaboration with our breast imaging team was both feasible and beneficial in our integrated group practice, because time saved by not placing wires on the day of surgery allowed breast imagers to perform more screening and/or diagnostic breast examinations. With standard KPNC protocols, the elimination of 1 WL procedure frees resources to perform 4 screening mammograms, 2 breast ultrasonograms, and 1 screening mammogram, or 2 diagnostic workups. Our study demonstrates that intraoperative ultrasonography/hydrogel clip lumpectomy is a useful technique that surgeons across the nation can adopt easily.

Disclosure Statement

The author(s) have no conflicts of interest to disclose.

Acknowledgments

Kathleen Louden, ELS, of Louden Health Communications performed a primary copy edit.

How to Cite this Article

Chang S, Brooke M, Cureton E, et al. Rapid implementation of intraoperative ultrasonography to reduce wire localization in The Permanente Medical Group. Perm J 2019;23:18-073. DOI: https://doi.org/10.7812/TPP/18-073

Author Affiliations

¹ Department of Surgery, Fremont Medical Center, CA

² Department of Surgery, University of California, San Francisco East Bay

³ Department of Surgery, Oakland Medical Center, CA

⁴ Department of Surgery, South San Francisco Medical Center, CA

⁵ Department of Surgery, Fresno Medical Center, CA

Corresponding Author

Veronica Shim, MD (veronica.shim@kp.org)

References

1. Saadatmand S, Bretveld R, Siesling S, Tilanus-Linthorst MM. Influence of tumor stage at breast cancer detection on survival in modern times: Population based study in 173,797 patients. BMJ 2015 Oct 6;351:h4901. DOI: https://doi.org/10.1136/bmj.h4901.
 2. Chan BK, Wiseberg-Firtell JA, Jois RH, Jensen K, Audisio RA. Localization techniques for guided surgical excision of non-palpable breast lesions. Cochrane Database Syst Rev 2015 Dec 31(12):CD009206. DOI: https://doi.org/10.1002/14651858.CD009206.pub2.
 3. Diego EJ, Soran A, McGuire KP, et al. Localizing high-risk lesions for excisional breast biopsy: A comparison between radioactive seed localization and wire localization. Ann Surg Oncol 2014 Oct;21(10):3268-72. DOI: https://doi.org/10.1245/s10434-014-3912-2.
 4. Hughes JH, Mason MC, Gray RJ, et al. A multi-site validation trial of radioactive seed localization as an alternative to wire localization. Breast J 2008 Mar-Apr;14(2):153-7. DOI: https://doi.org/10.1111/j.1524-4741.2007.00546.x.
 5. Jakub JW, Gray RJ, Degnim AC, Boughey JC, Gardner M, Cox CE. Current status of radioactive seed for localization of non palpable breast lesions. Am J Surg 2010 Apr;199(4):522-8. DOI: https://doi.org/10.1016/j.amjsurg.2009.05.019.
 6. Lovrics PJ, Goldsmith CH, Hodgson N, et al. A multicentered, randomized, controlled trial comparing radioguided seed localization to standard wire localization for nonpalpable, invasive and in situ breast carcinomas. Ann Surg Oncol 2011 Nov;18(12):3407-14. DOI: https://doi.org/10.1245/s10434-011-1699-y.
 7. Murphy JO, Moo TA, King TA, et al. Radioactive seed localization compared to wire localization in breast-conserving surgery: Initial 6-month experience. Ann Surg Oncol 2013 Dec;20(13):4121-7. DOI: https://doi.org/10.1245/s10434-013-3166-4.
 8. Hayes MK. Update on preoperative breast localization. Radiol Clin North Am 2017 May;55(3):591-603.
 9. Ngô C, Pollet AG, Laperrelle J, et al. Intraoperative ultrasound localization of nonpalpable breast cancers. Ann Surg Oncol 2007 Sep;14(9):2485-9. DOI: https://doi.org/10.1245/s10434-007-9420-x.
 10. Rahusen FD, Bremers AJA, Fabry HF, van Amerongen AH, Boom RP, Meijer S. Ultrasound-guided lumpectomy of nonpalpable breast cancer versus wire-guided resection: A randomized clinical trial. Ann Surg Oncol 2002 Dec;9(10):994-8.
 11. Blumencranz PW, Ellis D, Barlow K. Use of hydrogel breast biopsy tissue markers reduces the need for wire localization. Ann Surg Oncol 2014 Oct;21(10):3273-7. DOI: https://doi.org/10.1245/s10434-014-3917-x.
 12. Klein RL, Mook JA, Euhus DM, et al. Evaluation of a hydrogel based breast biopsy marker (HydroMARK®) as an alternative to wire and radioactive seed localization for non-palpable breast lesions. J Surg Oncol 2012 May;105(6):591-4. DOI: https://doi.org/10.1002/jso.22146.
 13. Gentile LF, Himmler A, Shaw CM, et al. Ultrasound-guided segmental mastectomy and excisional biopsy using hydrogel-encapsulated clip localization as an alternative to wire localization. Ann Surg Oncol 2016 Oct;23(10):3284-9. DOI: https://doi.org/10.1245/s10434-016-5325-x.
 14. Rubio IT, Esgueva-Colmenarejo A, Espinosa-Bravo M, Salazar JP, Miranda I, Peg V. Intraoperative ultrasound-guided lumpectomy versus mammographic wire localization for breast cancer patients after neoadjuvant treatment. Ann Surg Oncol 2016 Jan;23(1):38-43. DOI: https://doi.org/10.1245/s10434-015-4935-z.
 15. Sclafani LM, Bleznak A, Kelly T, El-Tamer MB. Training a new generation of breast surgeons: Are we succeeding? Ann Surg Oncol 2012 Jun;19(6):1856-61. DOI: https://doi.org/10.1245/s10434-011-2164-7.
 16. Manahan E, Wang L, Chen S, et al. What is a breast surgeon worth? A salary survey of the American Society of Breast Surgeons. Ann Surg Oncol 2015 Oct;22(10):3257-63. DOI: https://doi.org/10.1245/s10434-015-4720-z.
 17. Kelly P, Winslow EH. Needle wire localization for nonpalpable breast lesions: Sensations, anxiety levels, and informational needs. Oncol Nurs Forum 1996 May;23(4):639-45.

Keywords: breast cancer, breast surgery, breast ultrasonography, breast ultrasound, hydrogel clip, intraoperative ultrasonography for breast surgery, localization technique for nonpalpable breast lesions, preoperative wire localization, ultrasound clip, ultrasound-guided ultrasonography with hydrogel clips