FLEX November 2023

2295 10970347, 2022, 10, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/hed.27121 by Mount Sinai Health System Icah, Wiley Online Library on [02/08/2023]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License

COSTANTINO ET AL .

3.4 | Sentinel lymph node biopsy

modeling showed a pooled SLN identification rate of 98.8% (95%CI: 97.0 – 99.8; Figure 3A). Significant hetero geneity was measured between studies ( Q = 47.2, p < 0.05). In particular, the between-study variance was estimated at τ 2 = 0.011 (95%CI: 0.00 – 0.031), with an I 2 value of 59.7% (95%CI: 34.2 – 75.3). Baujat plot showing the studies contribution to the overall heterogeneity is shown in Figure 4A. The median number of SLN excised was 3.6 ( n = 371, 95%CI: 2.0 – 4.4). Table 3 summarize the specific nodal basin stratified by primary tumor location, as reported by seven studies ( n = 201) for a total of 542 SLN. The pooled SLNB positive rate measured using a random effect modeling was 5.6% (95%CI: 2.6 – 9.6; Figure 3B), with sub stantial between-study heterogeneity ( Q = 48.2, p < 0.05). In particular, the between-study variance was estimated at τ 2 = 0.012 (95%CI: 0.002 – 0.047), with an I 2 value of 60.6% (95%CI: 35.8 – 75.8). Baujat plot showing the studies contribution to the overall heterogeneity is shown in Figure 4B. The median follow-up time was 25.5 months ( n = 437, 95%CI: 19.9 – 38.5). Given the substantial between-study heterogeneity ( Q = 48.2, p < 0.05), a pooled RRR of 2.9%

Intradermal injection of radioisotope colloid tracer at the site of the tumor or at the prior resection site was used in almost all the included patients ( n = 685, 97.2%). On the other hand, an intraoperative injection of a blue dye was performed in 376 (53.3%) patients. A random effect

FIGURE 2

Funnel plot for evaluation of publication bias using

the effect size for SLNB positive rate

FIGURE 3

Forest plots showing the pooled (A) SLN identification rate, (B) SLNB positive rate, and (C) RRR. The dashed vertical line

represents the overall measure of effect [Color figure can be viewed at wileyonlinelibrary.com]

FIGURE 4

Baujat plots showing the studies contribution to the overall heterogeneity for (A) SLN identification rate, (B) SLNB positive

rate, and (C) RRR [Color figure can be viewed at wileyonlinelibrary.com]