Gynecologic malignant tumors most often metastasized to retroperitoneal lymph nodes and can be easy to recur or persist within the pelvis. In early stage cervical cancer, lymphadenectomy is a radical treatment. In addition, in early stage endometrial and ovarian cancers, lymph node sampling is frequently carried out to determine patients with metastatic diseases who may benefit from adjuvant therapy after primary surgery. When the recurrence or progression of gynecological malignant tumor occur, the regional lymph nodes may usually be metastasized. Although confirmation of metastatic lesion allows more accurate therapeutic regimen in advanced patients, treatment usually entails chemotherapy and/or radiation rather than surgery [1-3]. In the past, however, many patients have to choose surgical biopsy to identify the diagnosis of metastatic and recurrent diseases.

In order to avoid the delay of treatment and potential complications of open after biopsy of retroperitoneal lymph nodes, the FNAC is regarded as a non-invasive diagnostic method. The clinical diagnosis of many organs proved that FNAC is a highly sensitive and specific method for the diagnosis of tumors [1-3]. The location of the peritoneum is deep, so many gynecologists believe that the biopsy is more difficult.

Recently, a number of studies have showed that ultrasound and CT are effective methods for the identification of enlarged retroperitoneal lymph nodes [1-3]. In addition, it has been suggested that ultrasound guidance can allow the placement of needle more easily and accurately into the target. In the past few years, we have frequently used US-FNAC rather than surgery for diagnosis of retroperitoneal masses and nodules from gynecologic malignancies. The purpose of the study is to evaluate the usefulness and limitations of this approach for such lesions.

The 235 records of department of ultrasound at our hospital were reviewed who underwent US-FNAC retroperitoneal masses and lymph nodes from January1, 2014, to December 31, 2015. Ethical approval was taken from the institutional ethical committee and all patients provided written informed consent.134 patients were excluded in view of non-gyneco-oncological diagnoses (111 cases), incomplete imaging material (15 cases), and miscellaneous reasons (8 cases). Nine patients (9 cases) were lost to follow-up. During the period, 92 patients entered the study, representing 99 punctures. All patients were treated and examined at the Department of Gynecologic Oncology. The material was analyzed and complete clinical follow-up was obtained. Average age of the included patients was 52 years (23-86y). Puncture of 1 site was performed in 86 patients, puncture of 2 sites in 5 patients, and puncture of 3 sites in 1 patients.

Puncture was carried out by the same radiologist (RB) in 75% of patients, and were treated by the gyneco-oncological center of our hospital. The specimens were stained by haematoxylin and eosin. Unfortunately, we could not have on-site cytopathology service in our department for economical and organizational reasons to get immediate feedback as to the presence or absence of diagnostic material. All original reports were reviewed. Histological reports were reread. Puncture was carried out in 34 masses (36.9%) and 58 retroperitoneal lymph nodes (63.1%).

Aspiration were performed by hospitalization under supervision of staff gynecologic oncologist and radiologist. On the day of biopsy, we use GE-E9 computed sonography system with a 2.5-MHz or 4-MHz (multifrequency) transducer. The punctures were carried out, under local anesthesia by means of PTC biopsy needle (Hakko Co, Japan). After being positioned, the needle was attached to a 5-ml syringe. One to four (mean, two) needle passes were made for each patient. With these needles, Up to none or 1-2 ml of negative pressure was used for aspiration. The specimens were obtained by simultaneously rotating the needle and moving it back and forth under real-time guidance to ensure that the needle tip remained entirely within the lymph node. After withdrawal of the needle, a slide and cell block were prepared with the material. The cytologic material was reviewed on all cases and the diagnoses were compared with the primary histologic finding.

The FNAC results were categorized as positive, negative, unsatisfactory, or suspicious. Negative results were reported when a representative benign cell sample was obtained from the aspirated organ. An unsatisfactory result was defined as a sample consisting exclusively of fresh blood, an acellular specimen or a cell sample not representative for an aspirated organ. Suspicious reports were issued on specimens containing a few atypical cells, not conclusive for malignancy, in an otherwise adequate cell sample. A positive result was reported when malignant cells were observed. Quality of cytological sample was analyzed.

Subsequent clinical follow-up for at least 6-12 months (50 cases) or surgery (42 cases) was used to define true negative and false-negative outcomes. Specificity, sensitivity, accuracy, positive and negative predictive values, and false negative and false-positive rates were calculated according to standard definitions.

The main usefulness for fine needle aspiration cytology were pretherapeutic assessment (25.1%), tumor recurrence (15.2%), and tumor metastasis (59.7%). Ninety-nine fine needle aspiration were performed on 92 patients suspected of having a gynecologic malignancy. Among the 92 patients in the study,62 had cervical cancer (67.4%), 18 had ovary cancer (19.6%), 5 had endometrial cancer (5.4%), 7 had other cancer (7.6% 2 uterine sarcoma, 5 Inflammatory masses of ovarian appendages) (Table 1). There were 50 squamous cell carcinoma, 35 adenocarcinoma, 5 inflammatory disease and 2 uterine sarcoma. Treatment were carried out after punctures by chemotherapy + radiotherapy (45.6%), chemotherapy (17.4%) and surgery (5.5%). Surgery and chemotherapy were combined in 30.4% of cases. However, one case was not treated after puncture (1.1%) (Table 2). The lesion position will affect the choice of route. If the node is located in the shallow layer and there is no blood vessels around it, the needle is easy to enter the target node. In these group, 40.2% cases were located in paraaortic, 40.2% in cavity pelvic, 10.9% in iliac vessels Left, 8.7% in iliac vessels right (Table 3).

The punctures were carried out with Hakko needles, and their size was 21G (BMI>25) or 22G(BMI<25). 22G puncture needle was used for 84 cases and we used the 21G puncture needle in 8 cases. They were performed in puncture (93.5%), 2 puncture (5.4%), 3 puncture (1.1%). A satisfactory sampling was obtained in 91.3% of punctures. A cytological diagnosis was made in 90.2% (Table 5).

Among the 92 aspirates performed, the cytologic interpretation was malignancy in 56 cases, suspicion malignancy 16 cases, and no evidence of malignancy in 20 cases. On review, all 56 aspirates were confirmed to contain malignant cells.The 16 cases aspirates that originally were interpreted suspicious for cancer also were believed to contain suspicious malignant cells. However, in these cases, the aspirate was hypocellular. In the 20 cases that were in the absence of evidence of malignancy, only blood and nentrophils were seen; normal lymphoid elements and cancer cells were absent. According to the histopathological results after surgery and 6-12months of clinical follow-up results, true positive was 72 cases, 0 cases of false positive, false negative was 9 cases (5 cases of ovarian cancer, >20mm, solid part is not; 4 cases of lymph nodes,2 cases about 30mm, necrosis was obvious, 2 cases<20mm), true negative was 11 cases (4 cases of cervical cancer with enteritis; 2 cases were hematoma; 5 cases of adnexal inflammatory masses). According to the four table, the total sensitivity was 88.9%, the specificity was 100%, the accuracy was 90.2%, the positive predictive value was 100% and the negative predictive value was 55% (Table 6).

In these cases, there were 58 cases of lymph node puncture, 34 cases of tumor puncture. In the lymph node puncture, there were 8 cases of short diameter less than 10mm, 28 cases of 10-20mm, 18 cases of 20-40mm, 4 cases of more than 40mm. Mass puncture, the majority is greater than 40mm (32 cases); 2 cases are 20-40mm (Table 4). The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of FNAC in the diagnosis of <20mm were 93.3%, 100%, 94.4%, 100%, 75% and respectively. The sensitivity, specificity, accuracy, positive predictive value and negative predictive value of FNAC in the diagnosis of >20mm were 86.3%, 100%, 90.2%, 100%, 41.7% and respectively (Table 6).

In gynecological tumor neoplastic and non-neoplastic ovarian masses are most common during clinical practice. There are different data concerning the diagnostic accuracy and safety of FNAC for patients with ovarian masses, particularly those having malignant lesions [4-6]. Advancement of imaging guidance modalities have also contributed to the brilliant accuracy of FNAC in recent years [7]. This diagnostic method is applied other medical fields and there have proven value in diseases of thyroid, breast and palpable masses [8]. In gynaecologic malignancy, FNAC has been mainly used both for primary confirmation and clinical management of recurrence in malignant lesions. US guided FNAC is relatively accurate and safe procedure for the diagnosis of tumor that has metastasized to the lymph nodes [9]. Gynaecologists are interested in the safety of this technique and the consequent upstaging of gynecological cancers. Zanetta [10] showed that there were fewer complications in the procedure of aspiration of 838 ovarian cysts. They concluded that surgery can be decreased by using FNAC in many women with ovarian cysts. However, histopathology is still regarded as gold standard for the diagnosis of oncology. But in recent time, Aspiration under US guidance for the assessment of gynaecologic masses is considered as a rapid, versatile, cost and relatively accurate procedure for the primary diagnosis and management of gynaecologic masses with minimal morbidity. Lisa Cole [13] reported there were no false-positive cases. The sensitivity of cytologic evaluation was 50%, the specificity was 100%.

In our study, we also found the high specificity (100%) of cytologic diagnosis and the sensitivity of cytology was high (88.9%), along with poor negative predictive value of cytology (40-60%), which is similar to Sauthier report [3]. In the course of clinical practice, there are few false positive cases, which is the clinical value of FNAC. The sensitivity was significantly different among the research, contributing to several factors: the composition of the diseases, the technical level of the operator, the number of puncture, negative pressure by needle, puncture needle type, ancillary technique for specimens.

Studies have shown that the guidance of cytopathology service on site and the number of puncture are considered to be the main factor that affects the accuracy of the diagnosis [14,15]. In our research, specificity was 100%, although sensitivity and accuracy of FNAC was low at 70.6% and 77.2% in the diagnosis of ovarian tumors. For the cystic and solid ovarian tumor, it is very valuable to carry out cytopathology service on site which is bound to increase the number of puncture, but also increased the potential risk of tumor seeding caused by aspiration cytology [16,17]. Unfortunately, due to economical and organizational factors, we could not have cytopathology service in our ultrasound department, but we hope this type of service would be offered sooner or later. However, there is a high sensitivity and accuracy in PALN (size<20mm), but it is not effective to improve the positive rate of puncture due to the limitation of the puncture technique.

How to select the gauge of the needle, our experience was based on BMI to select. If BMI was greater than 25, we prefered to choose 21G; if BMI is less than 25, we generally choose 22G. But we have discovered that the size and type of the needle did not affect cytological efficiency (21G and 22G). However, it is the number of punctures that increases efficiency of cytological diagnosis and also raises the theoretical risk of complications. In the present study, puncture of 2 site was performed in 5 patients and puncture of 3 site in 1 cases. Fortunately, the positive rate of puncture was improved and may provide the help for clinical decision making. In the procedure, for the puncture of retroperitoneal masses, must pass through the intestine and may encounter small blood vessels, thereby increasing the incidence of complications. Nevertheless, complications that require clinical intervention are rare. There are generally factors accounting for it, such as ultrasound guidance, low BMI population and skilled operator.

The final diagnosis of cytology was directly affected by the degree of satisfaction of the samples. In our series, 7 of 92 cases (7.6%) were acellular and one case was no aspiration. Hypocell, however, was showed in 25% of our cases (23/92) and so most of our aspirations were suboptimal. Consequently, clinicians need caution that in the suboptimal sample, malignancy should not be ruled out. It is significant that, Martinez-Onsurben [18] make attempt to increase diagnostic accuracy by combining cytologic diagnosis with tumor serum markers and anti-inhibin. Athanassiadou [17] also reported that they use ancillary methods such as cytometry and immunocytochemistry to improve the diagnostic yield of ovarian cyst FNA, especially for borderline tumors. In the study, some nodules and masses were added to cell block, but not immunocytochemistry which may be the key point for the future research work.

There were some limitations of our study. First, the included cases was small and spectrum of disease was narrow. Second, the gold standard lymph node excision was not applied in all cases. Third, choice of needle gauge was solely on the basis of BMI. BMI of most included cases was less than 25. Whether it is suitable for the countries and regions with overweight or obesity (especially abdominal type obesity) remains to be further studied.

In summary, aspiration under ultrasound guidance for the assessment of gynecologic malignancy, with minimal morbidity, is considered as a relatively rapid, valuable, cheap and patient-friendly procedure. Despite the potential shortcomings, US-FNAC has an increasingly significant value to play in the diagnosis and management of gynecologic oncology.