Success of percutaneous biopsy techniques has rapidly increased in the last quarter century along with the rapid advancements in biopsy devices and in US and CT technologies, which are the guided imaging techniques in percutaneous biopsy procedure and this technique has become essential in the clinical practice for tissue diagnosis of hepatic mass lesions 10,11
. In addition, it has been stated that US-guided biopsy enables histological diagnosis with a high accuracy even in small (≤1 cm) lesions 12
. In the present study as well, biopsies were performed under the guidance of US.
Numerous different biopsy devices and needles that provide to obtain high-quality specimens appropriate for histopathological examination have been introduced for the use of operators 10. In general, a side-notch cutting needle in 18G thickness, which is compatible with automatic biopsy gun, is used in our clinic for liver biopsy procedures. In the present study, it was aimed to compare diagnostic success and efficacy of 18G cutting needle and 20G cutting needle, which have the same design, in US-guided liver biopsy.
In the present study, among 60 liver biopsy samples, 54 (90%) had a definite diagnosis of malignancy and six (10%) had a definite diagnosis of benignancy. Appelbaum et al. 13 performed US-guided cutting needle biopsy in 205 liver masses and reported that 176 (85.9%) were diagnosed as malignant and 29 (14.1%) were diagnosed as benign.
There may occur minor (post-procedure pain, temporary hypotension, and bleeding not requiring treatment, etc.) and major complications (bleeding requiring transfusion, adjacent organ injury, pneumothorax, hemothorax, peritonitis, sepsis, death, etc.) associated with US-guided liver needle biopsy. Rivera-Sanfeliz et al. 14 performed 154 liver biopsies using automatic device and reported that there were no major complications and that pain (18.2%) was the most frequent minor complication. Cakmakci et al. 15 reported localized pain, vasovagal syncope, and nausea to be the most common complications in the outpatients (n=1018) undergoing US-assisted needle biopsy performed by tru-cut biopsy gun (18-20G); however, they reported no death. Moreover, they observed all vasovagal syncope episodes during the preparation phase prior to the biopsy. Padia et al. 16 performed biopsies (n=539) using 18G automated needle and reported the complication rate to be 2%. They observed severe post-procedural pain, symptomatic hemorrhage, infection, and rash; however, no sedation-related complications or procedure-related death occurred. Caliskan et al. 17 suggested that complications would be minimized with the use of lesion-focused approach technique. In the present study, no minor or major complications were encountered. We thought that being sensitive about coagulation disorders and performing the procedure in the patients with an INR value of ≥1.5 after adjusting the coagulation values influenced the results critically. Besides, sampling from subdiaphragmatic lesions was avoided as much as possible in the presence of more than a single lesion. Moreover, attention was paid to obtain samples from the lesions that were distant from great vessels and main biliary ducts, if possible. We believe this contributed to low complication rates.
Since US-guided biopsy procedure requires technical skill, the operator may have a role in success rate. Liver biopsies are primarily performed by two groups of specialists; gastroenterologists (hepatologists) and radiologists 18. Anania et al. 18 compared the liver biopsy procedures performed by gastroenterologists (US-guided with 16G needle) and interventional radiologists (US-guided with 18G needle) performing in terms of adequacy of samples and complications and they found no significant difference. Free-hand technique was used in the present study; however, the fact that all biopsy procedures were performed by a single specialist who was experienced in the field of interventional radiology eliminated the operator-related difference.
In the present study, tissue sampling was performed by the same researcher using cutting needles with different thickness but with the same structure and from the same lesion. Accordingly, lesion-related variables (such as size, localization, and distance to the skin of the lesion), patient-related variables (such as patient cooperation and presence of ascites or parenchymal disease), and biopsy procedure-related variables (such as the use of different biopsy devices, the use of different transducers, and obtaining different numbers of samples) which could affect diagnostic success of different needles, were eliminated.
The sensitivity, specificity, PPV, NPV, and diagnostic accuracy in our study were evaluated by using the histopathological results of each single tissue sample obtained with 18G and 20G cutting needles. Yu et al. 19 obtained a mean of 2 tissue samples from each of 137 liver masses under US guidance using 18G cutting needle and reported the sensitivity, specificity, PPV, NPV, and diagnostic accuracy to be 96.4%, 100.0%, 100.0%, 94.6% and 97.8%, respectively. In our study, the histopathological examination of the samples, which were obtained from the liver masses with 18G needle in US guidance, revealed sensitivity, specificity, PPV, NPV, and diagnostic accuracy of 90.7%, 100.0%, 100.0%, 75% and 91.6%, respectively. It was thought that lower NPV was associated with low number of patients with benign lesions. Duysburgh et al. 20 obtained a single tissue sample from each of 77 liver masses of 72 patients using 21G needle. They reported that the technique had a sensitivity of 88%, a specificity of 100%, a NPV of 77%, and a diagnostic accuracy of 91% in distinguishing the malignant lesions from the benign lesions. In our study, the sensitivity, specificity, PPV, NPV, and diagnostic accuracy were found to be 87.0%, 100.0%, 100.0%, 66.7%, and 88.3%, respectively, for 20G needle. The results of the two studies were quite close to our study.
In the present study, comparison of the data from the histopathological examination of the samples obtained using 18G cutting needle and 20G cutting needle revealed no significant difference between the two techniques in terms of accuracy rates (p=0.540). The results of the present study, which was designed to compare the diagnostic success of only the needle types by eliminating all variables related to the lesion undergoing biopsy, to the patient, and to the operator performing the procedure, are worth to consider. Limited number of patients and the number of patients with benign diagnosis being very small within the study population were the limitations of the present study.