Slow pull versus suction in endoscopic ultrasound-guided fine-needle aspiration of pancreatic solid masses.
Author(s): Nakai Y, Isayama H, Chang KJ, Yamamoto N, Hamada T, Uchino R, Mizuno S, Miyabayashi K, Yamamoto K, Kawakubo K, Kogure H, Sasaki T, Hirano K, Tanaka M, Tada M, Fukayama M, Koike K
Publication: Dig Dis Sci, 2014, Vol. 59, Page 1578-85
PubMed ID: 24429514 PubMed Review Paper? No
Purpose of Paper
This paper studied the effects of slow-pull versus suction and needle gauge size in endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) of pancreatic lesions on diagnostic sensitivity and accuracy and specimen cellularity and contamination with blood.
Conclusion of Paper
Diagnostic accuracy and sensitivity were significantly better and specimen cellularity higher when specimens were obtained using the slow-pull technique compared to suction in EUS-FNA of pancreatic lesions; however, blood contamination was also increased. The effects of acquisition technique on diagnostic accuracy and sensitivity, specimen cellularity, and blood contamination were only significant among specimens collected with 25-gauge needles and were observed only in histological assessment (not cytological).
Studies
-
Study Purpose
The study compared the effects of slow-pull versus suction and needle gauge size in endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) of pancreatic lesions. Ninety-three patients with suspected pancreatic malignant solid masses underwent biopsy during which the stylet was removed and the needle moved to-and-fro 10-20 times with 10-20 cc suction applied (suction technique) or the needle was moved to-and-fro 10-20 times while slowly and continuously pulling the stylet (slow-pull technique) with either a 22-gauge or 25-gauge needle. Cellularity (0=no cellularity, 1=few cellular aggregates, 2=fair cellularity, 3=abundant cellularity) and blood contamination scores (0=none, 1=few, 2=moderate, 3=high) were determined by cytologic examination of each specimen by two cytotechnologists. Formalin-fixed samples were used for histologic examination and the pathologic diagnosis was determined by a combination of cytological and histological evaluation.
Summary of Findings:
Overall, the slow-pull technique demonstrated better diagnostic accuracy and sensitivity than suction (P = 0.039 and P = 0.025, respectively) and resulted in fewer specimens with moderate or high levels of blood contamination (50.1% versus 60.8%, respectively P = 0.074); however, 71.8% of specimens obtained with the suction method had a cellularity score of ≥ 2 (fairly or abundantly cellular) compared to only 52.2% of specimens obtained with the slow-pull technique (P < 0.001). No statistical difference between the slow-pull and suction techniques was observed in diagnostic yield, cellularity, or blood contamination when only the specimens obtained with a 22-gauge needle were considered, but specimens obtained with a 25-gauge needle by the slow-pull technique rather than the suction technique resulted in more specimens with a cellularity score ≥ 2 (p<0.001), fewer specimens with moderate or high levels of blood contamination (P < 0.001) and increased diagnostic accuracy (91.1% versus 70%, P = 0.001), sensitivity (90.0 versus 67.9%, P = 0.003), and negative predictive value (54.5 versus 18.2%, P = 0.045). The sensitivity of cytological assessment was comparable between specimens obtained by slow-pull and suction (67.3% versus 67.5%, P = 1.00) but the sensitivity of histological assessment was better for the slow-pull technique than the suction technique (90% versus 73.1%, P = 0.012); resulting in a statistically significant higher diagnostic sensitivity for the combination of cytology and histology in specimens obtained with the slow-pull technique (82.5% versus 71.9%, P = 0.026). However, the effect of acquisition technique on the diagnostic sensitivity of the combined cytological and histological analysis was only observed for specimens obtained with a 25-gauge needle (p=0.003), but not a 22-gauge needle. Increased sensitivity by EUS-FNA was associated with slow-pull technique (P = 0.028), tumor size ≥25 mm (P < 0.001), and tumor location in the pancreas body or tail (P < 0.001) and that the use of a 25-gauge needle rather than 22-gauge resulted in a non-significantly greater sensitivity (P = 0.137).
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform Morphology Light microscopy Cell count/volume Light microscopy Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Preaquisition Diagnosis/ patient condition Tumor size <25mm
Tumor size ≥25mm
Biospecimen Acquisition Needle gauge 22-gauge
25-gauge
Biospecimen Acquisition Method of tissue acquisition Ultrasound-guided fine needle aspiration
Biospecimen Acquisition Biospecimen location Pancreas head
Pancreas body
Pancreas tail