Is the Vectra 3D Imaging System a Reliable Tool for Predicting Breast Mass?
Kasey L. Wood, BS1, Yasmina Zoghbi, MD1, Ilana G. Margulies, MS1, Anthony H. Bui, BS1, Andrew Y. Ashikari, MD2, Jordan Jacobs, MD1, C. Andrew Salzberg, MD1.
1Icahn School of Medicine at Mount Sinai, New York, NY, USA, 2St. John’s Riverside Health System, Yonkers, NY, USA.
BACKGROUND: In selecting breast implants for breast reconstruction, current preoperative planning largely relies on two-dimensional measurements, which are often limited in suboptimal accuracy and objectivity. While the introduction of 3D imaging modalities has further improved preoperative planning, they require in-depth analysis of accuracy if they are to be considered as a standardized part of preoperative planning. Thus, the present study analyzes the reliability of the Vectra 3D Imaging System (Canfield Scientific) in predicting breast mass and explores potential confounding variables that may limit its accuracy.
METHODS: A retrospective review of 202 breasts that received direct-to-implant reconstruction by a single surgeon between February 2015 and February 2019 was conducted. Variables recorded included Vectra predicted mass (VPM, grams), mastectomy mass (MM, grams), ptosis grade, and body mass index (BMI). BMI was classified as: underweight (BMI<20), normal (20≤BMI<25), overweight (25≤BMI<30), and obese (BMI>30). Cup size was approximated as: A and smaller (MM≤250g), B (250g<MM≤450g), C (450g<MM≤600g), and D and larger (MM≥600g). Correlation between MM and VPM was evaluated using 2-tailed Pearson correlation coefficients (r) and associated formula was derived from a linear model. Equality of variances was assessed with Bartlett’s test. Correlation coefficients calculated for ptosis and BMI categories were then compared to the overall correlation coefficient.1 Significance was set at α=0.05 and analyses were conducted in R 3.6.0, version 1.70.
RESULTS: There was a strong correlation between MM and VPM (r=0.90, p<0.0001). The following equation was derived to predict MM: [MM] = 0.8 x [VPM] + 32 (adjusted r2=0.81). Bartlett’s test indicated VPM varies significantly across cup sizes (p<0.0001). Comparison of correlation coefficients for ptosis and BMI categories revealed a significantly reduced correlation coefficient for pseudo-ptosis (0.90 vs. 0.75, p=0.0425). CONCLUSIONS: The present study suggests that the reliability of Vectra in predicting breast mass varies across cup sizes and that there exists a significantly decreased association between VPM and MM among pseudo-ptotic breasts. These are important considerations when utilizing this technology in surgical planning.
1. Preacher KJ. Calculation for the test of the difference between two independent correlation coefficients. 2002.
Table 1. Comparison of correlation coefficients.
| Breast Cohort | N | Correlation Coefficient | p-value |
| All | 202 | 0.90 | Ref. |
| Non-Ptosis | 124 | 0.92 | 0.3109 |
| Grade 1 Ptosis | 33 | 0.92 | 0.5509 |
| Grades 2 & 3 Ptosis | 23 | 0.82 | 0.1788 |
| Pseudo-Ptosis | 21 | 0.75 | 0.0425** |
| Underweight | 29 | 0.80 | 0.0732 |
| Normal Weight | 96 | 0.87 | 0.2680 |
| Overweight | 57 | 0.84 | 0.1018 |
| Obese | 19 | 0.87 | 0.5923 |
Back to 2019 Abstracts