Keywords:
Breast, Lymph nodes, Oncology, Ultrasound, Biopsy, Cancer
Authors:
N. Alberdi1, N. Baraibar Argota1, I. Baraibar1, P. Lopez Sala1, M. Mellado1, S. Cervantes1, P. L. Alejandre García1, I. Sanmartin2, N. Alonso Ordás1; 1Pamplona/ES, 2Tudela/ES
DOI:
10.26044/ecr2019/C-1503
Results
All patients were between 26 and 74 years of age.
Axillary lymph node involvement at diagnosis was found in 91 patients (79.8%) either by ultrasound-guided biopsy or SLNB.
The AU was pathological in 81 cases (71%),
with histological confirmation in 76 (93.8%).
In 33 cases,
ultrasonography did not find any pathological result.
However,
15 of these 33 patients (45.5%) with normal AU who underwent SLNB had histologically confirmed lymph node disease (these results are reported in Fig. 1 ).
The results of axillary ultrasound before neoadjuvancy were: ( Table 2 ).
- Sensibility (S): 83,5%
- Specificity (SP): 78,3%
- Positive predictive value (PPV): 93,8%
- Negative predictive value (NPV): 54,5%
Eighty-four patients went through lymphadenectomy.
One patient refused lymphadenectomy.
Persistence of lymph node involvement (Miller-Payne B or C) after neoadyuvancy was found in 34 (40.5%).
However,
only 9 patients (10.7%) had pathologic post-neoadjuvancy AU,
and only 7 of them were true positive ( Table 3 ).
So,
the results of axillary ultrasound post-neoadjuvancy were ( Table 2 ):
- Sensibility (S)=20.6%,
- Specificity (SP)=96%
- Positive predictive value (PPV): 77,8%
- Negative predictive value (NPV): 64%.
In 50 of the 84 patients (59,5%) who underwent lymphadenectomy no residual axillary disease was found after neoadjuvancy (Miller-Payne A or D).
If each subtype is considered independently,
the percentage of patients with nodal negativization was:
- Luminal A: 66,7%
- Luminal B: 56%
- HER2: 62.5%
- Triple negative: 66.7%
There are some limitations we can find in this study: 6 different radiologists and 5 pathologist participated in the axillary stage by ultrasound and the anatomopathological analysis respectively,
so interobserver variability migh be high.
Besides,
the study has the limitations of a retrospective and unicentric cohort and small sample size.