子宫内膜癌腹膜后淋巴结转移相关因素分析

目的:探讨预测子宫内膜癌腹膜后淋巴结转移的指标,以期为确定子宫内膜癌手术范围提供参考。方法:回顾分析1997年1月至2006年12月初治为手术治疗的641例子宫内膜癌患者的临床与病理资料,单因素分析用χ2检验和Fish确切概率法,多因素分析用Logistic回归模型。结果:经多因素分析显示,病理分级G3、深肌层浸润、附件转移对预测子宫内膜癌盆腔淋巴结(pelvic lymph node,PLN)转移具有统计学意义;盆腔淋巴结转移与腹主动脉旁淋巴结(para-aortic lymph node,PALN)转移显著相关。结论:病理分级G3、深肌层浸润、附件转移是子宫内膜癌盆腔淋巴结转移的重要预测因素;盆腔淋巴结转移对预测腹主动脉旁淋巴结转移具有重要意义。病理分级G3、深肌层浸润、附件转移的子宫内膜癌患者应行盆腔淋巴结清扫术,并根据术中患者的盆腔淋巴结状况决定是否行腹主动脉旁淋巴结清扫术。     

Analysis of clinicopathologic factors predicting para-aortic lymph node metastasis in endometrial cancer

The purposes of this study were to compare the relationships between para-aortic lymph node metastasis and various clinicopathologic factors to evaluate whether para-aortic lymph node dissection is necessary when treating endometrial cancer. A retrospective study was performed on 841 patients with endometrial cancer, who underwent the initial surgery at the Keio University Hospital. Clinicopathologic factors related to para-aortic lymph node metastasis significant on a univariate analysis were analyzed in a multivariate fashion using a logistic model. According to the multivariate analysis, the clinicopathologic factor most strongly related to the existence of para-aortic lymph node metastasis was positive pelvic lymph node metastasis (P < 0.01). Among the 155 patients who underwent pelvic and para-aortic lymph node dissection, the difference of 5-year overall survival by the presence of retroperitoneal lymph node metastasis was examined by Kaplan-Meier method. The prognosis was poor even if para-aortic lymph node dissection was performed in cases of positive para-aortic lymph node metastasis. In conclusion, when deciding whether to perform para-aortic lymph node dissection in patients with endometrial cancer, it is necessary to consider the pelvic lymph nodal status. If there is no pelvic lymph node metastasis, it could not be necessary to perform para-aortic lymph node dissection.     

Pelvic and aortic lymph node metastasis in epithelial ovarian cancer

OBJECTIVE: The appropriate management of advanced ovarian cancer has been controversial in recent years. There are no adequate data about the importance of lymphadenectomy and the appropriate sites for lymph node assessment. We sought to evaluate the distribution, size, and number of pelvic and aortic lymph node metastases in patients with epithelial ovarian carcinoma. METHODS: Retrospective chart review of 116 patients with stage IIIC or IV epithelial ovarian carcinoma treated at Mayo Clinic who underwent systematic bilateral pelvic and aortic lymphadenectomy between 1996 and 2000. RESULTS: Eighty-six (78%) of 110 patients who underwent pelvic lymphadenectomy were found to have nodal metastases in 422 (16%) of 2705 pelvic nodes that were removed. Eighty-four (84%) of 100 patients had documented aortic lymph node metastases in 456 (35%) of 1313 aortic nodes that were removed. Fifty-five (59%) of 94 patients had bilateral metastatic pelvic and aortic lymph nodes and bilateral aortic lymphadenectomy was conducted in 53 (72%) of 74 patients. The most representative group for detection of nodal metastases was the aortic group (83%) followed by the external iliac group (59%) and the obturator nodes (53%). There was no significant difference between the mean size of positive (1.8 cm) and negative nodes (1.6 cm). Thirty-seven patients had unilateral tumor, and 1 patient (7%) had contralateral node metastasis. CONCLUSION: The incidence of positive nodes bilaterally and positive high aortic nodes indicates the need for bilateral pelvic and aortic node dissection (extending above the inferior mesenteric artery) in all patients regardless of laterality of the primary tumor.     

Contralateral pelvic and aortic lymph node metastasis in clinical stage I epithelial ovarian cancer.

Bilateral pelvic and aortic node lymphadenectomy is recommended for clinically localized unilateral epithelial ovarian adenocarcinoma (International Federation of Gynecologists and Obstetricians stage IA). The laterality of nodal metastasis in clinical stage I disease is rarely documented in the literature. Some authors have reported that ipsilateral node dissection is adequate for staging. A patient with contralateral pelvic and aortic lymph node metastasis and clinical stage I epithelial ovarian adenocarcinoma is presented. Pathologic findings were consistent with contralateral-only lymph node metastasis. This case illustrates the importance of bilateral lymph node sampling for appropriate staging of clinically localized epithelial ovarian cancer.     

Prospective evaluation of FDG-PET for detecting pelvic and para-aortic lymph node metastasis in uterine corpus cancer

OBJECTIVES: To estimate the sensitivity and specificity of positron emission tomography (PET) with 2-[(18)F]fluoro-2-deoxy-d-glucose (FDG) for detecting pelvic and para-aortic lymph node metastasis in patients with uterine corpus carcinoma before surgical staging. METHODS: Patients with newly diagnosed FIGO grade 2 or 3 endometrioid, papillary serous, or clear cell adenocarcinoma or uterine corpus sarcoma scheduled for surgical staging, including bilateral pelvic and para-aortic lymphadenectomy, were eligible. PET was performed within 30 days of surgery and interpreted independently by two nuclear medicine physicians. The imaging, operative, and pathologic findings for each patient and each nodal site were compared, and the sensitivity and specificity of FDG-PET in predicting nodal metastasis were determined. RESULTS: Twenty patients underwent FDG-PET before surgical staging. One patient found to have ovarian carcinoma on final pathology was excluded. Of the 19 primary intrauterine tumors, 16 (84%) exhibited increased FDG uptake. One patient did not undergo lymphadenectomy; her chest CT was suspicious for metastatic disease and FDG-PET showed uptake in multiple nodal and pulmonary foci. Metastatic disease was confirmed by percutaneous nodal biopsy. A total of three pathologically positive nodes were found in 2 of the 18 patients (11%). FDG-PET predicted that 3 patients would have positive lymph nodes (2 true positive and 1 false positive). Analyzed by lymph node regions, FDG-PET had 60% sensitivity and 98% specificity. The sensitivity and specificity by individual patient were 67% and 94%, respectively. CONCLUSIONS: FDG-PET is only moderately sensitive in predicting lymph node metastasis pre-operatively in patients with endometrial cancer. This imaging modality should not replace lymphadenectomy, but may be helpful for patients in whom lymphadenectomy cannot be, or was not, performed.     

子宫内膜癌淋巴结切除意义及指征

子宫内膜癌采用手术病理分期,然而是否对所有子宫内膜癌患者都行全面分期手术(全子宫切除术+双附件切除术+双侧盆腔淋巴结及腹主动脉旁淋巴结切除)争议广泛,尤其是对于早期子宫内膜癌患者淋巴结切除的价值值得探讨。文章回顾子宫内膜癌淋巴结切除的最新研究进展,进一步讨论淋巴结切除的意义及指征。     

Retroperitoneal lymph node metastases in Stage I carcinoma of the endometrium: Correlation with risk factors

The present study was undertaken in patients with Stage I carcinoma of the endometrium to correlate risk factors and the prevalence of retroperitoneal lymph node metastases. From January 1975 to April 1983, 202 patients with Stage I disease had initial therapy at the Long Island Jewish-Hillside Medical Center, New Hyde Park, New York. Among these patients, 74 who had total abdominal hysterectomy, bilateral salpingo-oophorectomy, and selective lymph node biopsy without preoperative radiation were included in the study. Results indicate that risk factors associated with nodal metastasis were Grade 3 tumor (42.1%), papillary adenocarcinoma (28.6%), deep myometrial invasion (42.9%), surface extent of tumor growth greater than of the endometrial cavity (31.8%), and a diffuse pattern of tumor growth (17.2%). In Stage 1 endometrial cancer with any of the above 5 risk factors, it is urged that a selective biopsy of para-aortic and pelvic nodes during hysterectomy should be performed.     

Correlation between clinical and histopathologic risk factors and lymph node metastases in early endometrial cancer (a multivariate analysis of 183 cases)

This study includes 183 patients with clinical stage I endometrial cancer subjected to peritoneal cytology, total abdominal hysterectomy, bilateral salpingo-oophorectomy, bilateral pelvic and para-aortic lymphadenectomy and omental biopsy during a 12-year period in a single institution. The factors analyzed were age, menopausal state, cell type, grade, mitotic activity, myometrial invasion, lymphovascular space invasion, cervical involvement, microscopic vaginal metastases, adnexal metastases, peritoneal cytology, presence of concomitant endometrial hyperplasia and lymph node status. The overall incidences of pelvic and para-aortic lymph node metastases were found to be 15.3% (28/183) and 9.3% (17/183), respectively. In five of 17 patients (29.4%) with para-aortic nodal metastases, pelvic nodes were free of tumor. The most significant prognostic factors for positive pelvic and/or para-aortic nodes were found to be the depth of myometrial invasion, grade of tumor and age.     

Lymph node metastasis in ovarian cancer: difference between serous and non-serous primary tumors

OBJECTIVE: To investigate the lymph node sites most susceptible to involvement relative to primary tumor histology in ovarian cancer. METHODS: The locations of metastatic lymph nodes were investigated in 208 patients with primary ovarian cancer who underwent systemic lymphadenectomy covering both the pelvic and para-aortic regions. RESULTS: Lymph node metastasis was present in 12.8% (20/156) of patients with stage I (pT1M0), 48.6% (18/37) with stage II (pT2M0), and 60% (9/15) with stage III (pT3M0) disease, thus in 22.6% (47/208) of all study patients. Isolated para-aortic nodal involvement was present in 23.3% (14/60) of patients with serous tumor and 4.1% (6/148) of those with non-serous tumor (P = 0.00002). In an analysis of 35 positive nodes from 25 patients with up to 3 positive nodes, 86.4% (19/22) of metastatic lymph nodes from patients with serous tumor were found in the para-aortic region, with 14 positive nodes located above the inferior mesenteric artery (IMA) and 5 below it, whereas metastasis to para-aortic lymph nodes accounted for 53.8% (7/13) of metastatic lymph nodes from patients with non-serous tumor (P = 0.0334). CONCLUSIONS: The locations of metastatic lymph nodes in ovarian cancer depend upon the histologic type of the primary cancer. In cases of serous tumor, the para-aortic region, particularly above the IMA, is the prime site for the earliest lymph node metastasis. However, the likelihood of pelvic node involvement is almost equal to that of para-aortic node involvement in cases of non-serous tumor.     

Early-stage cervical adenocarcinoma treated by surgical intent: the role of para-aortic lymph node dissection.

OBJECTIVE: Previous reports suggest that cervical adenocarcinomas have a unique pattern of spread and are more apt to metastasize to para-aortic lymph nodes. The purpose of this study was to further define the node of para-aortic lymph node dissection in early-stage cervical adenocarcinoma treated by surgical intent. METHODS: Institutional review board approval was obtained to perform a computerized search of the data of all women diagnosed with cervical adenocarcinoma between 1982 and 2000. Hospital charts were retrospectively reviewed. Follow-up was obtained from the tumor registry, medical records, and correspondence with health care providers. RESULTS: Three hundred (87%) of 345 early-stage (FIGO IA(1)-IIA) cervical adenocarcinoma patients were primarily treated by surgical intent. Two hundred seventy-six underwent pelvic and para-aortic node dissection (n = 69) or pelvic node dissection only (n = 207); 24 had no lymph node dissection. The median number of lymph nodes removed was 13 pelvic (range, 1-58) and 3 para-aortic (range, 1-17). Three (4%) of 69 patients had para-aortic nodal metastases. Each had either grossly evident para-aortic adenopathy (n = 2) or an adnexal metastasis. Thirty-six of 40 women developing recurrent disease had at least some component of pelvic recurrence; 4 had only extrapelvic disease. Three patients undergoing para-aortic node dissection developed an isolated extrapelvic recurrence despite originally negative para-aortic nodes (n = 2) or treatment by extended-field radiation for para-aortic metastases. One woman undergoing only pelvic node dissection had an isolated extrapelvic recurrence despite originally negative nodes. CONCLUSIONS: Early-stage cervical adenocarcinoma primarily treated by surgical intent has a very low risk of para-aortic metastases. These were detected only when there was gross evidence of nodal or adnexal disease.     

Laparoscopic port-site metastasis of an early stage adenocarcinoma of the cervix with negative lymph nodes.

BACKGROUND: Port-site metastasis (PSM) following laparoscopic surgery for cancer is being increasingly recognized as a potential problem; the majority of cases appear following laparoscopy for a pelvic mass that subsequently proved to be malignant or in the case of a disseminated intraperitoneal disease. The rare cases of PSM following laparoscopy for endometrial and cervical cancer have all been associated with the presence of regional lymph node metastasis or disseminated disease in the peritoneal cavity. We present here a case report of PSM in the absence of spread beyond the primary tumor. CASE: A 48-year-old woman with stage IA1 adenocarcinoma of the cervix was treated with laparoscopically assisted vaginal hysterectomy, bilateral salpingo-oophorectomy, and bilateral pelvic node dissection. The pathologic study revealed an endocervical adenocarcinoma confined to the cervix with negative lymph nodes. Nine months postoperatively, a cutaneous metastasis at the port-site was diagnosed. This was treated with wide local excision of the recurrence and the port-site track. Explorative laparotomy and para-aortic node sampling showed no evidence of recurrence elsewhere. CONCLUSION: This case emphasizes the risk for PSM in laparoscopic surgery performed for early stage disease.     

Sentinel node identification and intraoperative lymphatic mapping. First results of a pilot study in patients with endometrial cancer

INTRODUCTION: To minimize the surgical morbidity after lymphadenectomy, sentinel node biopsy (SLNB) has become fundamental in the management of different malignancies. We decided to evaluate sentinel lymph node (SNL) biopsies also in patients with endometrial cancer undergoing hysterectomy with lymphadenectomy. METHODS: In the setting of a prospective study we developed a technique for sentinel node biopsy of ten patients with histologically confirmed endometrial cancer. Prior to surgery 99m Tc Nanocol was injected in the peritumoral region by hysteroscopy. Six hours later lymphoscintigraphy was performed to identify the draining lymph nodes. During surgery we first detected the sentinel lymph node by a hand-held gamma tracer and then removed it. Surgery was completed by the standard therapy of total hysterectomy, bilateral salpingo-oophorectomy and pelvic and/or para-aortic lymphadenectomy. RESULTS: Scintigraphic identification was possible in eight out of ten patients. Intraoperative identification of sentinel lymph nodes was possible in seven out of eight patients. In five patients we found the sentinel lymph nodes in the pelvic region while the other two patients had bilateral sentinel nodes in the pelvic and para-aortic region. Histologically confirmed microscopic tumor metastases of the SLNs and para-aortic lymph nodes were only found in one case. The sentinel lymph nodes from the other six patients were free of tumor and accurately reflected the pathological status. CONCLUSION: The identification of sentinel lymph nodes in endometrial cancer is a practical and safe method. In order to improve this technique as a standard procedure for staging of endometrial cancer further studies with a larger number of patients have to be done.     

Sentinel lymph node in endometrial cancer

The aim of this study was to evaluate the possibility of identifying the sentinel lymph node and involvement of neoplastic cells in patients with endometrial carcinoma limited to the uterus, and also its correlation with the conditions of other pelvic and para-aortic lymph nodes. Forty patients with endometrial carcinoma, clinical staging I and II, were submitted to complete surgical staging through laparotomy, as recommended by FIGO in 1988. The sentinel node was investigated using patent blue dye in the myometrial subserosa. The sentinel node was excised and submitted to frozen section examination of specimen, stained with hematoxylin and eosin (H&E). Afterward, selective bilateral para-aortic and pelvic lymphadenectomy, total hysterectomy with bilateral salpingo-oophorectomy were performed. The lymph nodes excised were examined by means of paraffin-embedded slices stained with H&E and of imunohistochemistry with antikeratin antibody AE1/AE3. The sentinel lymph node was identified in 77.5% of patients (31/40), and 16.1% (5/31) presented neoplastic involvement in the node. In 25 cases of negative sentinel node, 96% (24/25) had no neoplastic involvement, and 4% (1/25) had other lymph node affected (false negative). In nine cases with no sentinel node identified, 55.5% (5/9) had lymph node involvement. The results of this study allow us to conclude that it is possible to identify the sentinel node using the methods described, and the pathologic examination significantly represents the same conditions of other pelvic and para-aortic lymph nodes.     

Sister Joseph umbilical nodule coexisting with endometriosis from endometrial carcinoma

A 58-year-old woman was diagnosed with endometrial carcinoma. Total hysterectomy, bilateral salpingo-oophorectomy and paraaortic and a pelvic lymph node dissection were performed. The cytology of peritoneal fluid was negative. There was no peritoneal dissemination except umbilical nodule. A peritoneal 2.0×1.5 cm umbilical nodule was also resected. The nodule was identified as a metastasis from endometrial cancer with endometriosis. The pelvic lymph nodes also showed metastatic lesion with endometriosis. Our case showed that endometriosis coexisted with umbilical and pelvic lymph nodal metastatic lesions from endometrial cancer. This fact suggests that the mode of metastasis to the umbilicus via lymph flow from endometrial cancer is the same as that for endometriosis.     

Maylard incision for radical hysterectomy and pelvic and para-aortic lymph node dissection

Angioli R, Koechli OR, Sevin B-U. Maylard incision for radical hysterectomy and pelvic and para-aortic lymph node dissection. Int J Gynecol Cancer 1998; 8 : 274–278.
Although the transverse, muscle-splitting technique for abdominal incision (Maylard incision) has been described for radical hysterectomy and for lymph node dissection of the pelvis and para-aortic area, the adequacy of the procedure performed through this incision has not been assessed. From 1991 to 1994, 205 patients underwent radical hysterectomy with pelvic lymph node (PLN) and para-aortic lymph node (PALN) dissection at the Division of Gynecologic Oncology, Jackson Memorial Hospital/University of Miami School of Medicine. Twenty-four patients with cervical cancer stage IB-IIA underwent radical hysterectomy, pelvic lymph node and para-aortic lymph node dissection through a Maylard incision. Three patients had panniculectomy performed at the same time. Duration of surgery, estimated blood loss, number of pelvic and para-aortic lymph nodes removed and duration of hospital stay were within acceptable ranges. Postoperative and intraoperative complications were minimal. Excellent cosmetic results were obtained. In conclusion, the Maylard incision offers good exposure to the pelvic and para-aortic area for lymph node dissection, good cosmetic result, and can be performed in association with abdominoplasty. Complications are similar to those observed with a vertical skin incision. This type of incision should be considered in selected young patients with early cervical cancer and in obese patients desiring abdominoplasty.     

Pattern and clinical predictors of lymph node metastases in epithelial ovarian cancer

Para-aortic lymphadenectomy is part of staging in early epithelial ovarian cancer (EOC) and could be part of therapy in advanced EOC. However, only a minority of patients receive therapy according to guidelines or have attendance to a specialized unit. We analyzed pattern of lymphatic spread of EOC and evaluated if clinical factors and intraoperative findings reliably could predict lymph node involvement, in order to evaluate if patients could be identified in whom lymphadenectomy could be omitted and who should not be referred to a center with capacity of performing extensive gynecological operations. Retrospective analysis was carried out of all patients with EOC who had systematic pelvic and para-aortic lymphadenectomy during primary cytoreductive surgery. One hundred ninety-five patients underwent systematic pelvic and para-aortic lymphadenectomy. Histologic lymph node metastases were found in 53%. The highest frequency was found in the upper left para-aortic region (32% of all patients) and between vena cava inferior and abdominal aorta (36%). Neither intraoperative clinical diagnosis nor frozen section of pelvic nodes could reliably predict para-aortic lymph node metastasis. The pathologic diagnosis of the pelvic nodes, if used as diagnostic tool for para-aortic lymph nodes, showed a sensitivity of only 50% in ovarian cancer confined to the pelvis and 73% in more advanced disease. We could not detect any intraoperative tool that could reliably predict pathologic status of para-aortic lymph nodes. Systematic pelvic and para-aortic lymphadenectomy remains part of staging in EOC. Patients with EOC should be offered the opportunity to receive state-of-the-art treatment including surgery.     

Ureteric calculi masquerading metastatic lymph nodes in a case of endometrial cancer

Extensive pelvic lymph node metastasis in the absence of risk factors was noticed at surgical staging for endometrial cancer in a 55-year-old postmenopausal women. On exploration there was a dilemma as to the disappearing lymph nodes which subsequently proved to be asymptomatic ureteric calculi. The need to palpate the pelvic side walls at the time of surgery for endometrial cancer even in cases when no nodal enlargement is detected preoperatively is highlighted.     

盆腹腔淋巴取样术在子宫内膜癌手术中的临床意义

目的探讨盆腹腔淋巴取样术在子宫内膜癌的临床意义。方法分析2000年1月-2007年12月上海同济大学附属第一妇婴保健院手术治疗的213例子宫内膜癌患者,其中,86例行盆腹腔淋巴取样术,127例行淋巴结切除术。手术方式根据手术切除淋巴结的情况分为两组。①取样组：淋巴取样术,筋膜外全子宫双附件切除/次广泛子宫切除术＋盆腔/腹主动脉旁淋巴结取样术86例;②切除组：次广泛/广泛子宫切除术＋盆腔淋巴结切除/腹主动脉旁淋巴结切除术127例。结果取样组：切除淋巴结中位数18枚,淋巴结的转移10例。切除组：切除淋巴结中位数32枚,淋巴结转移11例。5年生存率分别为94.2%和94.5%。取样组无病发症发生,淋巴结切除组中有9例,分别是1例术中大出血（〉2000ml）,淋巴囊肿感染6例,淋巴漏2例。结论在子宫内膜癌中淋巴结取样术可准确了解淋巴结的转移情况,适宜手术分期,并不影响生存率,是避免过度手术减少并发症发生的有效方法。     

预测子宫内膜癌患者腹主动脉旁淋巴结转移方法的研究进展

子宫内膜癌是妇科常见恶性肿瘤之一,发病率逐年上升。淋巴结转移为子宫内膜癌患者的主要转移途径,其中盆腔淋巴结转移较为常见,腹主动脉旁淋巴结转移较为少见。但存在腹主动脉旁淋巴结转移的患者预后相对较差。腹主动脉旁淋巴结转移情况可以通过术前、术中相关方法进行预测。预测子宫内膜癌患者是否存在腹主动脉旁淋巴结转移方法的研究近年来发展迅速,但目前尚无预测方法的统一标准。综合分析患者的病理、血清学和影像学检查方法对腹主动脉旁淋巴结转移的预测价值,对指导临床决策,避免不必要的腹主动脉旁淋巴结切除术,减少不良反应,降低手术费用以及选择合适的术后辅助治疗显得尤为重要。     

Interstitial thermoradiotherapy in recurrent gynecologic malignancies

In patients with ovarian carcinoma, the presence of metastatic disease in a retroperitoneal lymph node is indicative of a poor prognosis. Although a “staging laparotomy” is required for proper treatment, definitive information concerning para-aortic and pelvic lymph node metastasis often is not available. To determine the incidence of retroperitoneal lymph node metastases in untreated cases of ovarian carcinoma, a prospective study by selective nodal biopsy was undertaken in 61 unselected patients with the following distribution: Stage I, 11; Stage II, 10; Stage III, 31; and Stage IV, 9. The incidence of para-aortic node metastasis overall was 37.7% and of pelvic node metastasis, 14.8%. Of 23 patients with positive para-aortic nodes, 30.4% had no concomitant pelvic node involvement. Direct relationships between nodal metastasis and clinical stage, tumor grade, and histologic type of tumor were demonstrated. The incidence of positive para-aortic nodes in Stage I disease was 18.2%; in Stage II, 20.0%; in Stage III, 41.9%; and in Stage IV, 66.7%. The corresponding incidence of pelvic node metastasis was 9.1% in Stage I, 10.0% in Stage II, 12.9% in Stage III, and 33.3% in Stage IV. Grade 3 tumors were associated most frequently with nodal involvement, with an incidence of positive para-aortic nodes of 52.5% and of positive pelvic nodes of 15.5%. In patients with a serous type of malignancy, the frequencies of positive para-aortic/pelvic nodes were 44.4%/16.7%, respectively; in the undifferentiated type, 50.0%/10.0%; in the clear cell type, 25.0%/25.0%; and in the mucinous type, 14.3%/ 14.3%. In this small series, 32 patients (52.5%) had positive retroperitoneal nodal involvement. It is concluded that selective biopsies of the para-aortic and pelvic lymph nodes should be part of any “staging laparotomy” for ovarian carcinoma, and that the true incidence of nodal involvement in these patients awaits further investigation.