Sequelae of Fat Grafting Postmastectomy: An Algorithm for Management of Fat Necrosis

Demographics

This is an institutional review board—approved, retrospective review of a prospective breast reconstruction database. At Moffitt Cancer Center, from 2005 to 2010, 1289 patients had breast reconstruction. A search was conducted to identify patients having autologous fat grafting procedures and identified 286 patients who had fat grafting by a single reconstructive surgeon (P.D.S.). Patients having lumpectomy and fat grafting were excluded from this study. Patient factors hypothesized to impact development of fat necrosis, such as body mass index, history of smoking, comorbid diabetes and/or hypertension, and family or personal histories of breast cancer were recorded. In addition, cancer characteristics, including stage, tumor type, and receipt of postmastectomy radiation, were documented. Patient demographic and clinical data are presented in Table 1.

Surgical technique

All fat grafting procedures were performed or supervised by a single plastic surgeon (P.D.S.). Areas of contour deformity in the reconstructed breast were marked preoperatively. Fat was most commonly harvested from the lateral thigh, abdomen, or flank. Prior to lipoaspiration, the donor sites were infiltrated with tumescent solution (1000 mL of saline, 10 mL of 1% lidocaine, and 1:100000 epinephrine). Fat was aspirated by conventional liposuction technique utilizing a blunt cannula attached to a liposuction machine. Harvested fat was immediately transferred into 10-mL syringes, capped and allowed to purify by gravity sedimentation. After discarding oil and debris, small aliquots of fat were injected subdermally through small incisions in the premarked areas of the reconstructed breast. The location and amount of fat injected were recorded in the patient's operative report.

Postoperative surveillance and follow-up

Patients return to Moffitt Cancer Center after breast reconstruction at regular intervals to be evaluated by both the breast and plastic surgery teams. Any small palpable nodule or cyst found on physical examination was appropriately documented. If nodules were subcutaneous, small (<5 mm), not associated with any skin changes or lymphadenopathy, and found in the area of fat grafting, a decision was made to follow the lesion clinically or to have an ultrasound. Patients with any suspicious lesion—change in skin color, induration, larger nodules, lymphadenopathy, or lesion in location other than fat grafting—were immediately referred to targeted ultrasound by a fellowship trained breast radiologist and further workup was performed as deemed necessary. For this study, the number, size, orientation, location, and breast imaging-reporting and data system (BI-RADS) classification of nodules were recorded. Histopathology was recorded for all patients undergoing core or excisional biopsy. Any local, regional, or distant recurrences of disease were similarly documented.

Statistical analysis

Along with descriptive statistics, we evaluated the association between various clinical and breast reconstruction variables and the development of a palpable mass. The unit of this analysis was at the patient level, with development of a palpable mass being considered true if either breast developed a palpable mass. Because many patients had bilateral mastectomy with reconstruction, and multiple procedures, fat graft data were evaluated and presented per breast mound. Breast reconstruction-specific variables, such as amount of fat and tumescent injected, were averaged across the reconstructions for each patient. Wilcoxon rank sum test and exact chi-square test using Monte-Carlo estimation approach were used to compare the development of a palpable mass with continuous and categorical variables, respectively. All P values are 2-sided tests and declared as significant at a .05 level. SAS 9.3 (SAS Institute Inc, Cary, NC) was used for the analysis.

Complications

As with any surgical procedure, autologous fat grafting has associated complications. While major complications are exceedingly rare, minor complications of fat necrosis/oil cysts are consistently reported. Fat necrosis is a complication of several breast procedures, including biopsy, reduction, augmentation, and reconstruction.^20-23 The most accepted theory is that insufficient vascularity and ischemia are primarily responsible for the development of fat necrosis. Other complications associated with fat grafting are low and include seroma, infection, and liposuction-related pneumothorax, none of which were found in our study.^12,13

Rates of graft-related fat necrosis have been reported between 2% and 18%.^3,4,9-14 In our series, the incidence of fat necrosis was notably higher at 23%. This may be attributed to our increased postoperative surveillance and criteria for fat necrosis reporting. In our institution, patients are monitored at routine intervals by their breast and plastic surgeon providers (every 6 months for 5 years, and every 6 months for first year and then yearly, respectively). Once palpable nodules or cysts were compared to the locations fat grafted, the fat necrosis rate directly related to grafting was 16.9%, as not all nodules were found at sites grafted. Other palpable nodules are attributed to necrosis secondary to other breast procedures including mastectomy, autologous muscle flap, or expander placement; all of these patients required ultrasound or biopsy to rule out cancer.

Understanding that fat necrosis and oil cysts can occur after fat grafting procedures, the next step was to analyze whether we could differentiate these minor complications from recurrent cancer. Of specific concern was our ability to differentiate palpable nodules as sequelae of fat grafting from palpable nodules related to recurrent disease. There are currently no National Comprehensive Cancer Network guidelines that define a screening protocol for the reconstructed breast. Physical examination is the most accepted method to reliably screen for recurrent cancer.^24,25 Clinical assessment has also been suggested to be an accurate method for detecting fat necrosis presenting as palpable lesions.²⁶ However, a strict protocol must be followed if clinical evaluation is the primary determinant of fat necrosis. Patients must return for follow-up examinations at regular intervals, and examinations must be performed by an experienced breast surgeon. The presence, size, and location of fat necrosis should be consistently documented. New nodules, nodules occurring in areas that were not injected with fat, nodules increasing in size, or suspicious nodules detected on clinical examination should undergo further workup. We, therefore, propose the algorithm in Figure 2 as a reference to guide physicians in the care of a patient with a palpable nodule after fat grafting for reconstruction.

Figure 2. Algorithm for management of a palpable nodule. BI-RADS indicates breast imaging-reporting and data system; FNA, fine needle aspiration; US, ultrasound.

Clinical-patholgic and technical variables

In our study, clinical-pathologic and technical data were assessed to identify any risk factors that may complicate results of fat grafting, aiding in our future patient selection. Diabetes, hypertension, and smoking are established risk factors for poor microcirculation and therefore, in theory, should contribute to an increased incidence of fat necrosis. However, our results challenge the veracity of this assumption in the clinical population. We found that diabetes, hypertension, and smoking were not significant risk factors for complications (Table 5). This is consistent with prior studies that show that medical comorbidities make no contribution to complication rates.^9,10 Our results show that patients with these comorbidities can benefit from fat grafting procedures and should not be excluded when selecting patients.

The impact radiation therapy has on the outcome of fat grafting is controversial. Some purport that a previously irradiated breast creates a “poor recipient bed” for fat grafting.² Rietjens et al¹¹ report a higher incidence of complications with previously irradiated patients. However, clinical studies have shown no significant association between a history of radiation therapy and complications. We observed no correlation between a history of breast radiation and complications (Table 5). Experimental studies investigating the effects of radiation on fat grafts histology are warranted to lend clarity to these findings.

It has been suggested, though not clinically proven for the breast in vivo, that injecting large quantities of fat to a deficit or “overcorrection” may contribute to fat necrosis because the graft exceeds the recipient vascular supply. To counteract this, we utilize the well-established principle of injecting small aliquots of fat in multiple passes.⁹ Of note, we found that neither the volume of tumescent solution used nor the amount of fat injected to a particular site was significantly associated with an increased incidence of fat necrosis (Table 5).

Oncologic safety

In patients with a history of breast cancer, oncologic safety is paramount. Concerns have been propagated from in vitro studies suggesting a potentially deleterious interplay between grafted fat and the recipient tissue. In 2009, the American Society of Plastic Surgeons reported findings from a task force investigating the safety and efficacy of fat grafting. Although unable to make specific recommendations for the use of fat grafting, the task force concluded that there is a risk of the procedure interfering with breast cancer detection.⁹ The task force recommended concentrating future research efforts on studies that identify risk factors for fat grafting procedures and studies that assess the effect of fat grafting on breast cancer detection and treatment.⁹ Lohsiriwat et al²⁷ recently summarized a plethora of experimental studies and found that adipocyte, adipokines, and other adipocyte secretions can potentially promote breast cancer tumorgenesis by enhancing tumor-stroma interactions. Despite this, no in vivo studies demonstrate the effect of fat grafting on breast cancer development, progression, or local-regional recurrence. In our experience, only 1 patient was found to have recurrent breast cancer after fat grafting, and the recurrence was remote from the fat graft site. All other nodules occurring in areas of fat grafting were ultimately found to be benign, either by biopsy or by demonstrating clinical and radiographic stability over time. The incidence of local-regional recurrence in breast cancer treatment/reconstructive patients is reported to be between 2.2% and 8.1%.^{12,13,15,25,28,29} The local-regional recurrence rate in our study was low at 2.2%. Therefore, we found no association between fat grafting and increased occurrence or recurrence of cancer. Of note, however, is that our study is limited by a median follow-up of 28 months.