Follow-up of patients undergoing oncoplastic surgery – more palpable masses and benign biopsies

Introduction

Breast conserving surgery (BCS), that includes lumpectomy with clear margins followed by radiation therapy, is the leading surgery in breast oncology. In some cases BCS can result in distortion of the breast and significant asymmetry, mainly due to a large dead-space that is left in the breast following the surgery. The rational for immediate oncoplastic reconstruction following breast lumpectomy is based on the concept of obliterating the post-operative breast cavity and by that reducing the breast distortion and asymmetry following the radiation therapy. This involves a wide range of techniques including local tissue rearrangement (reduction and mastopexy pattern techniques), or tissue/volume replacement (augmentation pattern technique and local-regional flap transfer) [21]. The benefits of immediate oncoplastic reconstruction are well described. The combined surgery reduces breast deformities after large breast excisions [6,8]; it allows breast conservation in women with relatively large tumors, with improved surgical margins compared to lumpectomy alone [3,7,13,15,16], thus broadening the indications for breast conservation [5,11]; while improving patient quality of life and cosmetic breast outcome [27].

Despite the growing popularity of oncoplastic reconstruction, concerns have been raised that the combined surgery may affect the local recurrence pattern of breast cancer and may encumber the post-operative surveillance, mainly due to breast parenchymal rearrangement [17]. Few studies in the literature focused on the imaging findings and surveillance after oncoplastic reconstruction [9,19,20,22,25]. Whether patients who undergo oncoplastic reconstruction are subject to more frequent imaging and breast biopsies is still a matter of debate.

The goal of this study was to perform a retrospective comparison of long term follow-up of patients who underwent breast conservation with or without immediate oncoplastic reconstruction. Oncoplastic reconstruction included patterns of breast reduction, mastopexy (minimal breast tissue excision following the lumpectomy) and breast augmentation (simultaneous insertion of submuscular breast implants) techniques [1]. Simple local tissue re-arrangement done by the breast surgeons were excluded in this series. We hypothesized that patients undergoing immediate oncoplastic reconstruction would present with more palpable and imaging abnormalities and undergo therefore more biopsies.

Methods

This retrospective study was approved by the institutional review board at our institution.

All consecutive patients undergoing immediate oncoplastic reconstruction for breast cancer in our institution between 2009 and 2014 were included in the study. The control group was defined as the first 4 patients who underwent lumpectomy alone for breast cancer on the same week as the oncoplastic case. Women ultimately undergoing mastectomy for positive lumpectomy margins were excluded from the study. The decision to perform oncoplastic reconstruction was determined by a multidisciplinary team made up of breast surgeons, plastic surgeons and breast radiologists and breast oncologists.

Oncoplastic reconstruction was defined as immediate partial breast reconstruction by the plastic surgery team using local tissue rearrangement with breast reduction and or mastopexy pattern techniques or breast augmentation pattern technique. Cases of local-regional flap transfer reconstruction following the lumpectomy were excluded. The contralateral breast was simultaneously adjusted during the same operation.

Routine surveillance of breast cancer patients treated at our center includes a yearly mammography starting one year after the diagnosis of cancer. Further imaging modalities were used according to the radiological and clinical findings and were reported to the multidisciplinary breast team. Imaging findings were classified according to the American college of radiology breast imaging reporting and data system (BI-RADS [10]). For the purpose of this study, BIRADS 1–2 were considered normal findings, and BIRADS 3–6 were considered abnormal findings.

All biopsies performed during follow up in this study were ultrasound guided core needle biopsies, or stereotactic guided vacuum biopsy.

Patient charts were reviewed for demographics, tumor characteristics, type of surgery (breast conserving surgery with or without oncoplastic reconstruction), volume of breast removed (calculated using 3 dimensions of the specimen as measured by the pathologist), pathological staging, adjuvant treatment, and post-operative follow-up. Details of the post-operative follow-up included patient complaints, findings on physical exam, number, timing and type of breast imaging, and breast interventions performed. Biopsies done for abnormal lymph nodes or contralateral breast abnormalities were excluded.

The oncoplastic reconstruction and the control groups were compared using chi-square test or the Fisher exact test for categorical variables and the student’s t -test for normally distributed continuous variables. As some patients had more than one biopsy done, a multivariate Poisson model was created using number of biopsies done during follow-up as the outcome, and age, use of neoadjuvant treatment, volume of tumor excised and abnormality on exam during follow-up as explanatory variables entered in the model. All tests are two-sided with p < 0.05 set as significant. All calculations were done using IBM SPSS Statistics for Windows, version 22.0 (Armonk, NY). The sample size was calculated to detect an odds ratio of 3:1 in the proportion of patients undergoing at least one biopsy in the oncoplastic reconstruction group compared to the control group with a power of 80% at a significance level of 5% (calculation done using Winpepi downloaded from http://www.brixtonhealth.com/pepi4windows.html).

Results

A total of 2079 BCSs were performed in our institution between 2009 and 2014. Of these, 335 patients were included in the study: 67 consecutive patients who underwent lumpectomy and immediate oncoplastic surgery and 268 control patients who underwent lumpectomy alone and were matched by same week of surgery.

Oncoplastic reconstruction included breast reduction pattern in 43 (64%) patients, mastopexy in 20 (30%), and augmentation in 4 (6%).

Patient demographics and tumor characteristics are summarized in table 1. Patients undergoing oncoplastic reconstruction were significantly younger (mean age 51 versus 62, P < 0.001), and with more advanced disease; these patients had larger mean tumor size (3.1 versus 1.9 cm, P < 0.001) and a higher rate of involved lymph nodes (50% versus 26%, P = 0.001). Consequently, more oncoplastic reconstruction patients underwent neoadjuvant treatment (39% versus 15%, P < 0.001), and had larger volume of tissue removed during surgery (384 cm³ versus 149 cm³ , P < 0.001).

Follow up findings of the two groups are summarized in table 2. Total follow-up time was similar for both groups (925 days in the oncoplastic group versus 963 days, p = 0.6). Women undergoing oncoplastic reconstruction complained more frequently of pain, discomfort, or mass in the operated breast during follow up (16% versus 5%, P = 0.007), and new lumps were more frequently found (18% versus 5%, P = 0.004) on physical examination. Out of 12 patients in the oncoplastic group with new palpable finding on physical examination, 7 had a mass on imaging and consequently underwent US guided biopsy, compared to 7 out of 14 patients with a new palpable mass in the lumpectomy group. Several women with a new palpable finding in each group underwent US guided biopsy despite normal imaging with no malignancy diagnosed.

There was no significant difference in mean time to first imaging study between the two groups (318 versus 336 days, P = 0.43) or in the mean total imaging studies performed during follow up (mean 4.4 versus 4.2, P = 0.61).

In both groups, the most frequent finding on imaging requiring a biopsy was a mass (Fig. 1), followed by suspicious microcalcifications (Fig. 2), and enhancement on MRI.

More women after oncoplastic reconstruction required at least one breast biopsy during follow-up (31% versus 11%, P < 0.001).

In the oncoplastic reconstruction group, 21 (31%) women had a total of 25 biopsies. Twenty-three biopsies (92%) were benign. The most common finding in this group was fat necrosis (15 cases, 60% of biopsies). Two (8% of biopsies) were found to have cancer. Ten (32%) out of 31 biopsies performed in the lumpectomy group (N = 268) yielded infiltrating ductal carcinoma.

On multivariate analysis, (Table 3), the presence of a palpable abnormality and history of oncoplastic reconstruction were significantly associated with undergoing biopsies during follow-up.

Discussion

Immediate oncoplastic reconstruction following breast lumpectomy has gained popularity in oncological breast surgery. This technique has many advantages compared to lumpectomy alone, including wider oncological margins, broadening the indications for breast conservation, as well as improved cosmetic outcome after surgery [3,5,7,8,11,13,15,27]. Patients with hypertrophic breasts undergoing oncoplastic reconstruction in reduction pattern technique were shown to have improved quality of life after surgery when compared to women undergoing lumpectomy without oncoplastic surgery [2]. This technique has also disadvantages, namely longer operation and recovery, scarring on both breasts, potential injury to nipple sensation and fat necrosis.

The aim of our study was to compare the findings on breast exam and breast imaging of patients after oncoplastic reconstruction with those of patients that underwent BCS alone.

In this study we found that patients undergoing BCS with immediate oncoplastic reconstruction presented more often with pain and discomfort and with palpable findings on exam. Consequently, a higher percentage of these women required needle biopsies during follow-up. However, the fraction of women undergoing a biopsy that yielded cancer was identical in both groups (3–4% of the cohort), as most of the biopsies in the oncoplastic reconstruction group were benign, mainly fat necrosis. In fact, 18% of patients in the oncoplastic reconstruction group had a diagnosis of fat necrosis during follow up. This high rate of fat necrosis following oncoplastic reconstruction has been reported by others [4,12,26]. A palpable abnormality with a subsequent mass detected on ultrasound was a common clinical scenario leading to biopsy after oncoplastic surgery in our study.

Previous works have shown that fat necrosis is a challenging diagnosis to establish due to its various appearances on different imaging modalities, particularly on US [14]. In our study, a frequent appearance of biopsy proven fat necrosis on sonography was a hypoechoic solid mass with non-distinct borders, very similar to breast carcinoma (Fig. 1). More specific sonographic indicators of fat necrosis are echogenic internal bands or cystic mass with mural nodule [14]. However, we found that in many of our cases, the imaging finding were not specific enough to safely recommend follow-up without biopsy.

As shown in previous studies [17,20,23], Mammographic microcalcifications do not pose a significant diagnostic dilemma after oncoplastic reconstruction, with only three cases leading to breast biopsy in our study, one of them diagnosed with recurrent disease. However, despite the higher specificity of mammography compared to sonography in diagnosing fat necrosis, the calcifications can sometimes be indistinguishable from those of malignancy, particularly in the early stages [24] (Fig. 2).

The increased use of MRI in this group of patients may obviate the need for biopsies, however, most (N = 12, 57%) of the women undergoing biopsies in the oncoplastic reconstruction group had at least one MRI done during follow-up.

The impact of oncoplastic reconstruction on surveillance of breast cancer patients has been studied by several groups, with conflicting results. Losken [17] compared imaging of patients undergoing breast conservation with simultaneous breast reduction to the imaging of women undergoing breast conservation alone, and found longer times to stability of mammographic findings (26 versus 21 months, not statistically significant). They found significantly higher rates of biopsies during follow-up in the oncoplastic reconstruction group.

Similar to our results, Dolan [9] reported significantly higher rates of ultrasound exams done after oncoplastic reconstruction, for work-up of palpable findings and subsequent higher rates of biopsies, most being fat necrosis. Piper [22] reported no difference in the rates of mammographic findings and unnecessary biopsies in women undergoing lumpectomy versus women undergoing lumpectomy with oncoplastic breast reconstruction, using reduction technique. This study was relatively small (49 patients in each group). Consistently higher rates of imaging abnormalities for which a biopsy was recommended were reported during follow-up, reaching statistical significance only at the first year of follow-up (16 versus 2%, p < 0.05). Most of these patients were observed without biopsy, with no cancer developing during observation with serial imaging. However, high rates of biopsies were reported in the oncoplastic group (25% versus 18% over 5 years), which are comparable to our results (31% versus 11%).

Our study is limited by its retrospective design. The two groups are different; the oncoplastic group presenting with larger tumors and more advanced disease. This is consistent with previous studies [5,18,19]. This is expected, as there is a selection of patients with extensive disease and hence need for extensive excisions for oncoplastic reconstruction. However, we found that after controlling for age, neoadjuvant treatment and volume of tissue removed, the oncoplastic group had more biopsies done during follow-up. Another limitation is the short follow-up. As studies have shown that stability of imaging after oncoplastic reconstruction takes longer (Losken 2009 [17]), we expect that the need for biopsies will subside over time.

In conclusion, we found that women undergoing breast conservation with immediate oncoplastic reconstruction have more palpable abnormalities and biopsies on follow-up. However, most of these findings prove to be benign. This is probably due to the significant tissue rearrangement causing breast lumps which are difficult to differentiate from recurrent disease. Patients and their physicians should be aware of this limitation of oncoplastic reconstruction and more importantly of its benign nature. However the alternative to immediate oncoplastic reconstruction in patients needing a wide lumpectomy is either distortion of the breast and gross asymmetry or mastectomy with immediate reconstruction.