Background & Aims:Fine-needle aspiration (FNA) of pancreatic neuroendocrine tumors (Pan-NENs) has been proposed to obtain the grading, using Ki-67 proliferation index calculation. Data on reliability of grading (G) and Ki-67 index calculations from FNA cell blocks are controversial as there are potential limitations. Methods:One hundred patients subjected to FNA for a presumed PanNEN and subsequent resection material were evaluated at a single institution. FNA was obtained with endoultrasonography or with a percutaneous approach. Ki-67 calculation was performed using WHO guidelines. A comparison be-tween cytology (c) and histology (h) was performed for grading, Ki-67 values and diagnostic rate. Results:The overall level of agreement for G (n=63) was moderate (Cohen’s k = 0.455, 95%CI from 0.219 to 0.691, p<0.001; r=0.430, 95%CI from 0.204 to 0.613, p<0.001). The overall sensitivity for G was 76.2%, whereas for G1, G2, and G3 it was 76.6%, 84.6% and 33.3% respectively. Mean in-dex values of cKi-67 and hKi-67 were 4.35 ± 9.5% and 5.26 ± 12% respectively (p=0.334). Spear-man’s r for Ki-67 was good and very good for the overall population and for PanNENs of tumor size ≤ 20 mm (respectively r=0.615, 95%CI from 0.434 to 0.749, p<0.001, and r=0.862, 95%CI from 0.718 to 0.935, p<0.001).The Bland-Altman plot showed an agreement between cKi-67 and hKi-67 assessment.The overall concordance rate for diagnosis was 100%. Conclusions:In the presence of a suspected PanNEN, FNA (especially obtained through endoultra-sonography) achieves an excellent diagnostic rate, with satisfactory and reliable results for grading and Ki-67 assessment, particularly for tumors ≤ 20 mm.
Preoperative fine-needle aspiration of pancreatic neuroendocrine tumors: a reliable tool to assess diagnosis and grading - A prospective single-center analysis of 100 cases
Salvatore Paiella
2019-01-01
Abstract
Background & Aims:Fine-needle aspiration (FNA) of pancreatic neuroendocrine tumors (Pan-NENs) has been proposed to obtain the grading, using Ki-67 proliferation index calculation. Data on reliability of grading (G) and Ki-67 index calculations from FNA cell blocks are controversial as there are potential limitations. Methods:One hundred patients subjected to FNA for a presumed PanNEN and subsequent resection material were evaluated at a single institution. FNA was obtained with endoultrasonography or with a percutaneous approach. Ki-67 calculation was performed using WHO guidelines. A comparison be-tween cytology (c) and histology (h) was performed for grading, Ki-67 values and diagnostic rate. Results:The overall level of agreement for G (n=63) was moderate (Cohen’s k = 0.455, 95%CI from 0.219 to 0.691, p<0.001; r=0.430, 95%CI from 0.204 to 0.613, p<0.001). The overall sensitivity for G was 76.2%, whereas for G1, G2, and G3 it was 76.6%, 84.6% and 33.3% respectively. Mean in-dex values of cKi-67 and hKi-67 were 4.35 ± 9.5% and 5.26 ± 12% respectively (p=0.334). Spear-man’s r for Ki-67 was good and very good for the overall population and for PanNENs of tumor size ≤ 20 mm (respectively r=0.615, 95%CI from 0.434 to 0.749, p<0.001, and r=0.862, 95%CI from 0.718 to 0.935, p<0.001).The Bland-Altman plot showed an agreement between cKi-67 and hKi-67 assessment.The overall concordance rate for diagnosis was 100%. Conclusions:In the presence of a suspected PanNEN, FNA (especially obtained through endoultra-sonography) achieves an excellent diagnostic rate, with satisfactory and reliable results for grading and Ki-67 assessment, particularly for tumors ≤ 20 mm.
| File | Dimensione | Formato | |
|---|---|---|---|
|
Tesi finale 13-12.pdf
Open Access dal 01/10/2019
Tipologia:
Tesi di dottorato
Licenza:
Creative commons
Dimensione
268.46 kB
Formato
Adobe PDF
|
268.46 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
