Skip to main content

Endoscopic ultrasound-guided sampling of solid pancreatic masses: 22-gauge aspiration versus 25-gauge biopsy needles.

Yang MJ, Yim H, Hwang JC, Lee D, Kim YB, Lim SG, Kim SS, Kang JK, Yoo BM, Kim JH

Publication: BMC Gastroenterol (2015) • Vol 15 • Page 122PMID: 26419845View on PubMed

Purpose

This paper compared the effects of needle gauge in endoscopic ultrasound-guided (EUS) biopsies of pancreatic lesions on cytological and histological diagnostic accuracy, technical success, needle passes required for diagnosis, and sample quality.

Conclusion

No statistically significant difference between specimens obtained by 22-gauge fine-needle aspiration (FNA) and 25-gauge fine-needle biopsy (FNB) was observed in ability to achieve diagnosis, cytological diagnostic accuracy, histological diagnostic accuracy, rate of acquiring sufficient material for histological evaluation, or number of passes required for diagnosis. The diagnostic accuracy was comparable between conventional smear (CS) and liquid-based preparation (LBP) for each group. Cytological specimen quality scores were comparable in CS specimens obtained by 22G FNA and 25G FNB but LBP specimens obtained with 25G FNB needles demonstrated higher scores for amount of cellular material present and retention of appropriate architecture than 22G FNA samples.

Studies (1)

Purpose

This paper compared cytological and histological diagnostic accuracy, technical success, needle passes required for diagnosis, and sample quality in 22-gauge fine-needle aspiration (FNA) and 25 gauge fine-needle biopsies (FNB) of pancreatic lesions on.  Seventy-six solid pancreatic lesions underwent endoscopic ultrasound-guided biopsy during which the stylet was removed and the needle moved to-and-fro 10-15 times while suction was applied using a 10 mL syringe with either a 22-gauge FNA (June, 2010-June, 2012) or 25-gauge FNB needle (June 2012-October, 2013) (38 specimens per group). Multiple passes were made in each case with first pass specimens mounted onto slides and fixed with alcohol for conventional smear (CS), second pass specimens were placed in an ethanol-based preservative for liquid-based preparation (LBP), and additional passes were fixed in formalin for histology. Cytological assessment of the Papanicolaou-stained CS and LBP samples was performed by a single cytopathologist and sample quality comparisons were scored based on amount of cellular material (0 - minimal to absent, 1- sufficient, or 2 - abundant), retention of appropriate architecture and cellular arrangement (0 - minimal to absent, 1-moderate, or 2 - excellent) degree of cellular degeneration (0 - marked, 1 - moderate or 2 - minimal), degree of cellular trauma (0 - marked, 1 - moderate or 2 - minimal), and amount of background blood or clot (0 - large, 1 - moderate or 2 - minimal). Histological evaluation of H&E stained samples was performed by two pathologists. LBP or histology slides from cases suspected to contain pseudopapillary, neuroendocrine, or metastatic tumors were immunochemically stained for synaptophysin, chromogranin, MIB-1, beta- catenin, CD 10, CD 56, CD 99, and PR.  Matched surgical specimens were available for 10 22G-FNA specimens and 3 25G-FNB specimens. For the remaining cases, only clinical follow-up data for at least 6 months was available.

Summary of Findings

No statistically significant difference between the 22G FNA group and 25G FNB group was observed in the ability to achieve diagnosis (97.4% versus 89.5%,P = 0.358), cytological diagnostic accuracy (89.5% versus 97.4%, respectively P = 0.358), histological diagnostic accuracy (34.2% versus 52.6%, respectively, P = 0.105), rate of acquiring sufficient material for histological evaluation (60.5% versus 68.4%, respectively, P = 0.472), or the average number of passes required for diagnosis (5.05 and 5.55, respectively, P = 0.132). However, the 25G FNB group demonstrated a better diagnostic yield for histological evaluation of specific tumor types than the 22G FNA group (60.6% versus 32.4%, P = 0.018). The average sample quality scores of 22G FNA and 25G FNB specimens were comparable in CS samples for diagnostic cellular material present (1.26 versus 1.55, respectively, P =  0.085), retention of appropriate architecture (1.39 versus 1.63, respectively, P = 0.132), degree of cellular degeneration (1.42 versus 1.58, P = 0.271), degree of cellular trauma (1.26 versus 1.53, P = 0.061), and amount of obscuring background blood or clot (1.29 versus 1.53, P = 0.150) but 25G FNB LBP samples demonstrated higher scores than 22G FNA LBP specimens for amount of cellular material present (0.92 versus 1.32, respectively P=0.030) and retention of appropriate architecture in (0.97 versus 1.42, respectively, P = 0.010). There was no significant difference between the diagnostic accuracy of CS and LBP for either 22G FNA (89.5 % versus 76.3 %, respectively, P = 0.128) or 25G FNB (84.2%, both, P = 1.000) specimens.

  • Biospecimens

    TissuePancreas
  • Preservation Types

    EthanolFormalin
  • Diagnoses

    NeoplasticCarcinoma
  • Platforms

    Cell count/volumeH-and-E microscopy
    MorphologyLight microscopy
    ProteinImmunohistochemistry
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen AcquisitionMethod of cell acquisition
    Fine needle aspiration
    Fine needle biopsy
    Biospecimen AcquisitionNeedle gauge
    22-gauge
    25-gauge
    Immunohistochemistry SpecificTargeted peptide/protein
    Beta-catenin
    CD 10
    CD 56
    CD 99
    Chromogranin
    MIB-1
    PR
    Synaptophysin