99mTc-labeled colloid SPECT/CT versus planar lymphoscintigraphy for sentinel lymph node detection in patients with breast cancer: a meta-analysis

Abstract

Abstract Background The aim of this meta-analysis was to compare single-photon emission computed tomography (SPECT/CT) and planar lymphoscintigraphy (PL) in patients with primary breast cancer, undergoing lymphoscintigraphy at initial staging. Specifically, we assessed the detection rate (DR) for sentinel lymph node (SLN), the absolute number of detected SLNs by each technique, and the proportion of patients with additional SLNs detected by one technique compared to the other one. Finally, we aimed to evaluate the impact of SPECT/CT on the surgical approach. Methods Original articles, providing a head-to-head comparison between SPECT/CT and PL, including patients with primary breast cancer at first presentation, were searched in PubMed/MEDLINE and Scopus databases through March 31st, 2022. The DR of the imaging techniques was calculated on a per-patient analysis; studies were pooled on their odds ratios (ORs) with a random-effects model to assess the presence of a significant difference between the DRs of SPECT/CT and PL. The number of additional SLNs, calculated as relative risk (RR), and the pooled proportion of patients with additional SLNs using one imaging technique rather than the other one were investigated. The pooled ratio of surgical procedures (SLN harvesting) influenced by the use of SPECT/CT, according to the surgeons, was calculated. Results Sixteen studies with 2693 patients were eligible for the calculation of the DR of SPECT/CT and PL. The DR was 92.11% [95% confidence interval (95% CI) 89.32–94.50%] for SPECT/CT, and 85.12% (95% CI 80.58–89.15%) for PL, with an OR of 1.96 (95% CI 1.51–2.55) in favor of SPECT/CT. There was a relative risk of detection of larger number of SLNs (RR: 1.22, 95% CI 1.14–1.32; 12 studies; 979 patients) for SPECT/CT (n = 3983) compared to PL (n = 3321) and a significant proportion of patients with additional SLNs detected by SPECT/CT, which were missed by PL (18.88%, 95% CI: 11.72%-27.27%; 13 studies). Four articles, with a total number of 1427 patients, revealed that 23.98% of the surgical procedures benefited from the use of SPECT/CT. Conclusions This meta-analysis favors SPECT/CT over PL for the identification of SLN in patients with primary breast cancer at staging due to higher DR, more SLNs depicted, and a significant proportion of subjects with additional detected SLNs by SPECT/CT compared to PL. Furthermore, SPECT/CT positively influences the surgical procedure. However, PL remains a satisfactory imaging option for imaging departments not equipped with SPECT/CT due to its good patient-based DR.