In this analysis, we found that surgical management of stage I-III ILC of the breast varied significantly by BMI. While rates of mastectomy overall were similar across BMI groups, the use of oncoplastic approaches, immediate reconstruction, and shave margins differed significantly. Although several series have reported on differences in the surgical approach to breast cancer based on BMI, none have focused specifically on patients with ILC.10–12 Because of its diffuse growth pattern, ILC has distinct surgical challenges, including discordance between tumor size on imaging versus on final pathology, increased rates of positive margins, and increased need for mastectomy compared to patients with the more common IDC, or carcinoma of no special type.5,13,14 BMI itself may impact the tumor features of ILC, a strongly hormone-positive tumor type.4,15
The literature shows an inconsistent relationship between BMI and utilization of various breast surgery procedures. While we found no difference in rates of attempted BCS by BMI group, other investigators have shown that patients with overweight/obese BMI are more likely to undergo BCS than mastectomy.11,12 This could be related to a preference to avoid longer anesthesia times in the setting of co-morbid conditions associated with elevated BMI, but could also reflect a preference for BCS if reconstruction after mastectomy is less likely to be offered.16,17
In our study population, the lack of higher rates of BCS in those with higher BMI could reflect institutional practices, but could also be specific to those with ILC. Because ILC grows in a diffuse pattern and presents at higher stages, mastectomy is more common in this tumor type compared to the more common carcinoma of no special type. Interestingly, Tong et al showed that rates of surgical complications after ORM were lower than after mastectomy with reconstruction in those with obese category BMI.18 These findings suggest that utilization of ORM may allow for avoidance of mastectomy with fewer complications. For patients with ILC, the ability to safely offer ORM can improve outcomes, as positive margin rates are reduced; this was also observed in this analysis across BMI groups.
We and others previously showed that both shave margins and oncoplastic surgery are associated with significantly reduced risk of positive margins specifically for those with ILC.5,19,20 In this analysis stratified by BMI, we found that shave margins were associated with lower incidence of positive margins regardless of BMI. Interestingly, oncoplastic surgery was only associated with lower odds of positive margins for those with group 3 or 4 BMI (classified as overweight or obese). In this group, the use of oncoplastic surgery at the time of lumpectomy resulted in a 58% reduction in the odds of positive margins when adjusting for tumor size. In contrast, for those with group 1 or 2 BMI (classified as underweight or normal weight), the use of oncoplastic surgery at the time of lumpectomy was also associated with lower odds of positive margins, but this did not reach statistical significance.
For those with higher BMI, these findings are reassuring and suggest that shave margins and oncoplastic surgical approaches are helpful tools to reduce positive margin risk. For those with lower BMI, use of both shave margins and oncoplastic approaches were less common. We have not seen this reported in the literature before and hypothesize that oncoplastic approaches may be less common possibly due to smaller breast size and reduced options for approaches such as reduction mammoplasty. One limitation of this study is that breast size was unavailable as a variable to evaluate. Potentially smaller breast size could also have reduced the likelihood of surgeons taking shave margins, which were associated with significantly reduced positive margins in this group.
These findings suggest the need for further investigation into the drivers of these disparate approaches to the surgical management of ILC. For those with overweight/obese BMI, reduced rates of immediate reconstruction warrant additional research into the risks of complications, particularly compared to lumpectomy with oncoplastic approaches. For those with underweight/normal BMI, more study is needed to understand the lower utilization of shave margins and oncoplastic surgery. In this study, it is possible that the lower rates of oncoplastic surgery in this group limited our ability to detect a significant association with positive margin rates.
Additional limitations of this study include the retrospective design and the single institution population, potentially limiting the generalizability of findings. The choice of surgical procedure was also likely subjected to biases from both patients and physicians. Factors such as comorbidities, insurance coverage, and surgeon experience may also have influenced the choice of surgery.