D2 lymph node dissection improves staging in patients with gastric adenocarcinoma

Debate continues over the recommended extent of routine lymphadenectomy for gastric cancer. Although evidence of improved locoregional control with extended dissection accumulates, understaging and stage migration continue to confound the issue. Our objective was to determine whether D2 lymph node dissection improves staging compared with D1 in patients with gastric adenocarcinoma. We performed a retrospective study of 79 consecutive patients who underwent resection of gastric adenocarcinoma at a single institution. The American Joint Committee on Cancer (AJCC) 7th edition (2010) was used for TNM staging. Twenty-seven patients (34%) underwent D2 lymphadenectomy; 52 underwent D1 lymphadenectomy. There was no significant difference in age, gender, or operation. Significantly more lymph nodes were removed with a D2 than a D1 lymphadenectomy (mean, 26 vs 9; P < 0.0001). Significantly more patients had at least 15 nodes removed in the D2 cohort (85 vs 17%, P < 0.001). Within the D2 cohort, nine patients (39%) demonstrated additional lymph node metastases on extended dissection. This altered nodal status in five patients (20%) and altered TNM stage in four patients (16%). There was no significant difference in perioperative morbidity. D2 lymphadenectomy significantly increases node retrieval and AJCC compliance for gastric adenocarcinoma, resulting in improved staging.     

Para-aortic lymph node involvement in gastric adenocarcinoma]

AIMS: To determine the significance of superextended lymphadenectomy (D4) in patients with gastric cancer. The incidence of para-aortic lymph node metastases (N4) was analysed as well as its relationship to the site of the tumour. PATIENTS AND METHODS: The frequency of para-aortic lymph node metastases was assessed in 110 patients who underwent gastrectomy with D4 lymphadenectomy during the period from June 1988 to October 1999; five patients with plastic linitis and three with carcinoma of the gastric stump were excluded from the study. RESULTS: The postoperative mortality rate was 2.7% (n = 3) and the postoperative morbidity rate was 29.1% (n = 32). In our experience the most frequent postoperative complications were pancreatic fistulas (7.3%) and respiratory complications (6.4%). Among the 110 patients, the total number of dissected nodes was 5245 and the mean number of dissected nodes per case was 47.7. The total number of retrieved lymph nodes from the para-aortic station level was 639, with a mean number of 5.8 per patient. N4 nodal involvement was found in 20 (18.2%) out of 110 patients: 12 (33%) patients with a carcinoma located in the proximal third, two (6%) with a tumour located in the middle third and six (15%) with a carcinoma of the distal third of the stomach. CONCLUSION: The presence of para-aortic lymph node involvement in 18.2% of the patients suggests that D4 lymphadenectomy should be considered in the curative surgical treatment of advanced gastric cancer, especially if located in the proximal third of the stomach (N4 + in 33% of the patients).     

Randomized clinical trial of morbidity after D1 and D3 surgery for gastric cancer

BACKGROUND: A randomized comparison of D1 (level 1 lymphadenectomy) and D3 (levels 1, 2 and 3 lymphadenectomy) dissection was performed to evaluate morbidity and effects on survival from gastric cancer. METHODS: A total of 221 patients were studied after resection for gastric cancer, 110 after D1 surgery and 111 after D3 surgery. RESULTS: The morbidity rate was higher after D3 than after D1 resection (17.1 (95 per cent confidence interval (c.i.) 10.1 to 24.1) versus 7.3 (95 per cent c.i. 2.4 to 12.2) per cent respectively; P = 0.012). The difference was largely related to abdominal abscess (8.1 per cent after D3 versus none after D1 resection; P = 0.003). The D3 group had an anastomotic leak rate of 4.5 per cent whereas there was no leakage in the D1 group (P = 0.060). All anastomotic leaks were minor and were managed non-operatively with nutritional support. Patients who had D3 resection had longer operating times, greater blood loss and postoperative drain outputs, and more patients needed blood transfusion. There was no death in either group. The hospital stay was longer after D3 than D1 surgery (mean(s.d.) 19.6(13.9) (range 10-98) versus 15.0(4.0) (range 10-30) days; P = 0.001). CONCLUSION: Extended lymphadenectomy for gastric cancer is associated with more complications than limited lymphadectomy but this does not lead to significant mortality.     

Prospective analysis of the diagnostic yield of extended en bloc resection for adenocarcinoma of the oesophagus or gastric cardia

BACKGROUND: The extent of lymph node dissection can affect tumour node metastasis staging. The resulting 'stage migration' might hamper stage-by-stage comparison between different forms of oesophageal resection. The aim of this study was to assess the diagnostic impact of extended en bloc lymphadenectomy in staging (adeno)carcinoma of the mid/distal oesophagus or gastric cardia. METHODS: This was a prospective study of 74 patients (67 men and seven women; median age 63 (range 40-78) years) who underwent extended oesophagectomy between 1994 and 2000. RESULTS: A median of 31 (range 15-78) lymph nodes was resected (and identified), with a median of 5 (range 0-31) positive nodes. Twenty-seven patients (36 per cent) had tumour-positive nodes in extended fields: 15 patients (20 per cent) in the abdomen and 15 patients (20 per cent) in the mediastinum. Subcarinal nodes were most affected (19 per cent). Extended resection led to tumour upstaging in 17 patients (23 per cent); two patients had isolated positive subcarinal nodes and 15 other tumours became M1a owing to positive nodes near the coeliac axis, hepatic artery or splenic artery. Tumour positivity in paratracheal or aortopulmonary nodes occurred in 8 per cent of patients, without influencing staging. CONCLUSION: Extended en bloc lymphadenectomy altered staging in 17 of 74 patients (23 per cent) with adenocarcinoma of the oesophagus or cardia, mainly into M1a owing to positive coeliac nodes (20 per cent).     

进展期胃癌D2根治术中第14v组淋巴结清扫的意义

目的探讨进展期胃癌行D：根治术时第14v组淋巴结清扫的必要性。方法回顾性分析2003至2007年间天津医科大学附属肿瘤医院收治的131例行胃癌根治术（D2或D2＋）并同时加行第14v组淋巴结清扫的胃癌患者的临床病理资料,分析影响第14v组淋巴结转移的临床病理因素以及第14v组淋巴结转移与预后的关系。结果131例患者中24例（18．3％）有第14v组淋巴结转移。原发灶部位、肿瘤大小、浸润深度、淋巴结分期、TNM分期、第1、6、8a组淋巴结转移与第14v组淋巴结转移有关（均P〈0．05）;其中原发灶部位和淋巴结分期是影响第14v组淋巴结转移的独立因素（均P〈0．05）。第14v组淋巴结转移和未转移患者5年生存率分别为8．3％和37．8％,差异有统计学意义（P〈0．01）。多因素预后分析证实。第14v组淋巴结转移是影响进展期胃癌D：根治术后生存的独立危险因素（P=-0．029,RR=1．807,95％CI：1．064-3．070）。结论对于进展期胃中下部癌．尤其是肿瘤体积较大、浆膜受侵犯、第6组淋巴结可疑转移的患者,第14v组淋巴结清扫是必要可行的。     

胃癌淋巴转移规律与淋巴结清扫范围的分析(附326例报告)

目的 探讨胃癌淋巴结转称规律和胃癌根治术的淋巴清扫范围。方法 １９９０年～１９９９年行Ｄ２、Ｄ３、Ｄ３淋巴结廓清术加腹主动脉旁淋巴结廓汪术（Ｄ３加ＰＡＬ）的胃癌患者３２６例,对期临床资料进行回顾性分析。结果 本组总的淋巴结转移率６９．９％,早期与进展期胃癌淋巴结转移率分别为１５．４％和７７．４％。肿瘤浸润深度达Ｔ１的患者,淋巴结转移主要局限于Ｎ１;达Ｔ２的患者淋巴结转移至Ｎ３、Ｔ４的ＫＫＨＮＦＴＪ     

θ���������ܰͽ���ɨ��Χ�����󲢷�֢

??The postoperative complications associated with the extent of lymphadenectomy for gastric cancer SUN Yi-hong. Zhongshan Hospital of Fudan University, General Surgery Research Institute of Fudan University, Shanghai 200032, China
Abstract Lymph node metastasis is an independent prognostic factor in patients with gastric cancer, the reasonable lymph node dissection is critical to radical gastrectomy. Extended nodal dissection is the most important surgical factor related to postoperative risk of complications. For the surgeon with limited experience of lymphadenectomy, D2 or more extensive lymph node dissection is more likely to cause the occurrence of severe postoperative complications. The rigorous training for surgeons and promotion of a standardized treatment concept, can effectively reduce lymphadenectomy-related complications.     

����θ�������������ܰͽ�ת���ۺ����Ʋ���

??Comprehensive treatment strategies of advanced gastric cancer with para-aortic lymph node metastasis SONG Wu??HE Yu-long. Department of Gastrointestinal Surgery??the First Affiliated Hospital of Sun Yat-sen University??Guangzhou 510080??China
Corresponding author??HE Yu-long??E-mail??YLH@medmail.com.cn
Abstract The whole treatment process of advanced gastric cancer patients with para-aortic lymph node (PALN) metastasis requires collaborative multidisciplinary treatment (MDT) group to assist the assessment of preoperative staging, decision of treatment plan, selection of operation timing and the control of operation. The parao-aortic nodal dissection (PAND) is an absolute operation indication under the condition that there was no evidence of distant metastasis, the enlargement lymph nodes limited to the No.16a2 and No.16b1 station lymph nodes, and other incurable factors were excluded. Advanced gastric cancer with PALN metastasis adopted comprehensive treatment strategies with chemotherapy and surgery. The effective neoadjuvant chemotherapy was prerequisite, the surgeon’s abundant experience of D2 lymph node dissection and PAND was safety guarantee, the radical dissection of No.16a2 and No.16b1 station lymph nodes may give survival benefit for selected patients. The highly suspected para-aortic lymph node metastasis groups, such as preoperative imaging evaluation for clinical N2 or N3, highly suspected No.9 lymph node metastasis, duodenum invaded, still need to be careful reassessment of preventive PAND value after comprehensive treatment. Under the guidance of preoperative neoadjuvant therapy and MDT discussion, the selected patients may get benefit from therapeutic or prophylactic PAND.     

Staging accuracy of endoscopic ultrasound based on pathologic analysis after minimally invasive esophagectomy

Endoscopic ultrasonography (EUS) is a common staging modality used in patients with esophageal cancer. The objective of this analysis was to evaluate the accuracy and sensitivity of EUS in determining the depth of penetration (T stage) and nodal status (N stage) in patients with esophageal cancer who underwent minimally invasive esophagectomy (MIE). A retrospective analysis of all patients at a university hospital who underwent preoperative EUS followed by MIE for cancer was performed. We compared the results of preoperative EUS to final pathologic analyses of the esophageal specimen, examining the accuracy of EUS staging. Ninety-five patients with esophageal cancer who underwent MIE had preoperative EUS. Twenty-four of the 95 patients were excluded for lack of a repeat EUS after neoadjuvant therapy before resection. Hence, 71 patients were evaluated for the accuracy of EUS staging. The accuracy of EUS for T0 disease was 80 per cent; T1 disease was 75 per cent; T2 disease was 39 per cent; and T3 disease was 88 per cent. The overall EUS accuracy for T stage was 72 per cent with overstaging occurring mostly for pathologic T1 tumors in 18 per cent and understaging occurring mostly for pathologic T3 tumors in 11 per cent. The sensitivity and specificity for detection of nodal involvement were 79 per cent and 74 per cent, respectively. However the accuracy for T and N staging by EUS after neoadjuvant therapy decreased to 63 per cent and 38 per cent, respectively. Endoscopic ultrasound in the absence of neoadjuvant therapy is a relatively accurate and sensitive modality for determining the depth of tumor penetration and the presence of nodal disease in patients with esophageal carcinoma. The accuracy for T and N staging is less reliable after neoadjuvant therapy.     

Gastric cancer: an audit of 122 consecutive cases and the results of R1 gastrectomy

In a consecutive series of 122 patients with gastric carcinoma, 9 per cent had no operation, 27 per cent had incurable disease at laparotomy, and 64 per cent underwent gastric resection. R1 gastrectomy was performed in 73 of the 78 resections. The operative mortality after gastric resection was 4 per cent, but there were no deaths after potentially curative resections. The actuarial 5-year survival was 20 per cent overall, 60 per cent in patients undergoing a 'curative' resection with N0 disease, and 18 per cent in patients with N1 disease. Local or regional recurrence without evidence of distant metastases was identified in 11 per cent of cases after 'curative' resections. The probability of survival was adversely affected by N1 nodal involvement (P less than 0.005) and by the presence of poorly differentiated or anaplastic tumours (P less than 0.001). Only 6 per cent of patients had early gastric cancer, and absolute curative resections by Japanese criteria were possible in only 5 per cent of cases. The results suggest that the unfavourable presenting pathology is the principal determinant of the poor prognosis of gastric cancer. A more radical or extended lymphadenectomy (R2/3 gastrectomy) might have cured more patients with N1 metastases, but only 12 per cent of potentially curable patients had N1 disease in this study, and it appears that more radical surgery may have little effect on the overall survival rates for gastric cancer.     

Regional and para-aortic lymphadenectomy in radical surgery for advanced gallbladder carcinoma

BACKGROUND: There have been no reports on the routine use of regional and para-aortic lymphadenectomy for gallbladder cancer. The aim of this study was to elucidate nodal status, its prognostic influence and the efficacy of lymphadenectomy. METHODS: A retrospective analysis was made of 60 patients who underwent radical resection and routine regional and para-aortic lymphadenectomy. RESULTS: Of the 60 patients, 73 per cent had node-positive disease and 38 per cent had positive para-aortic nodes. Postoperative survival was extremely poor in patients with minimal distant metastasis, and similarly in patients with para-aortic disease. The survival of patients with metastasis limited to the regional nodes was significantly better than that of those with distant metastasis (P = 0.029) or para-aortic disease (P = 0.017) and was not significantly different from that of patients with no metastasis (P = 0.82). CONCLUSION: Regional and para-aortic lymphadenectomy provides no survival benefit for patients with para-aortic disease, which has an influence on poor prognosis equivalent to that of distant metastasis. It has the potential to bring survival benefit only in selected patients with metastasis limited to the regional nodes. A sampling biopsy of the para-aortic nodes before starting radical surgery is recommended because they are involved more frequently than expected.     

Unilateral lymphadenectomy in intraoperative stage I nonseminomatous germinal testis cancer

Bilateral retroperitoneal lymphadenectomy is mainly a staging procedure in patients with stage I nonseminomatous testis cancer, and it causes permanent loss of antegrade ejaculation in approximately two-thirds of the cases. Between May 1978 and August 1981, 61 consecutive patients with no intraoperative evidence of lymph node involvement underwent unilateral retroperitoneal lymph-adenectomy for nonseminomatous germinal testis tumors. Microscopic metastases were found in 1 to 4 retroperitoneal nodes in 6 cases (9.8 per cent). Antegrade ejaculation was absent postoperatively in 11 patients (18 per cent), with no significant difference between patients who underwent lymph node dissection on the left or right side. Ejaculation returned spontaneously in 3 patients, 1 of whom fathered a child. The disease recurred in 10 patients 3 to 35 months after lymphadenectomy (median 6 months). Disease recurred in 8 of 55 patients (14.5 per cent) with negative nodes and 2 of 6 (33.3 per cent) with positive histological findings. No patient suffered retroperitoneal recurrence. The more than 3-year survival rates free of disease were 96.4 and 83.3 per cent in patients with pathological stages I and II disease, respectively. Unilateral retroperitoneal lymphadenectomy in patients with intraoperative stage I nonseminomatous germinal testis cancer preserves antegrade ejaculation in more than 80 per cent of the cases without apparently compromising the long-term survival.     

贲门癌手术中淋巴结廓清范围的探讨

目的探讨贲门癌淋巴结转移规律及其手术中合理的清扫范围。方法回顾性分析1999年1月至2004年12月间77例贲门癌患者的临床病理学资料。比较淋巴结的不同清扫范围对患者预后的影响。结果（1）本组77例贲门癌患者临床分期以Ⅲ、Ⅳ期为主，共计57例（74．0％），可进行检索的淋巴结转移率：N1 64．9％，N2 14．3％，N3 10．4％；（2）T1及肿瘤最大径小于2．0cm时未检测到淋巴结转移，T2N（＋）1／5，T3N（＋）68．2％，T4N（＋）80．8％；（3）N1主要集中在No．1、3、2组，N2主要集中在No.7、8、10、9组，N3分别为No．5组6．5％、No．6组1．3％、No．16组1．3％、No．107～110组2．6％；（4）不同淋巴结清扫范围者的5年生存率：D1为22．7％，92为31．3％，D2为36．5％，D2、D3与D1术后生存率比较，P〈0．05，差异有统计学意义。结论对进展期贲门癌患者行D2以上淋巴结廓清术。可满足根治性要求，有延长生存时间的作用。     

Surgical management of malignant melanoma using dynamic lymphoscintigraphy and gamma probe-guided sentinel lymph node biopsy: the Emory experience

Sentinel lymph node (SLN) biopsy is revolutionizing the surgical management of primary malignant melanoma. It allows accurate nodal staging which targets patients who may benefit from regional lymphadenectomy and systemic therapy. This is a retrospective review of patients treated at Emory University for stage I and II malignant melanoma with gamma probe-guided SLN biopsy from 1/1/94 to 6/30/98. Three hundred sixty patients (males 228, females 132) were identified. Primary melanoma sites included: head and neck 58, trunk 148, and extremities 154 (upper 71, lower 83). Primary tumor staging was T1 9, T2 134, T3 153, and T4 64. SLNs were successfully identified in 99.7 per cent of patients and 98.9 per cent of nodal basins mapped. In 275 (76.6%) cases a single draining nodal basin was identified. In 84 (23.3%) cases there were multiple draining nodal basins. Positive SLNs were identified in 63 patients (17.5%). SLN positivity by tumor staging was T1 0 per cent, T2 9.0 per cent, T3 22.2 per cent, and T4 26.6 per cent. The overall recurrence rate was 11.9 per cent. Recurrences by SLN status were SLN+, 27 per cent, and SLN-, 8.8 per cent. Regional recurrence occurred in 7 (2.4%) of the 297 with negative SLN biopsies and 7 (11.1%) of the 63 with positive SLN biopsies. Dynamic lymphoscintigraphy and gamma probe-guided SLN localization was successful in more than 98 per cent of cases. Patients with negative SLN biopsies have a low risk of recurrence.     

Aggressive D2 lymphadenectomy is required for accurate pathologic staging of gastric adenocarcinoma

The therapeutic efficacy of aggressive regional D2 lymphadenectomy as an adjunct to gastrectomy for adenocarcinoma of the stomach remains controversial. It is hypothesized that D2 lymphadenectomy compared with limited D1 lymphadenectomy increases nodal yield without adding to operative morbidity or mortality, and is necessary to allow accurate pathologic staging according to current American Joint Committee on Cancer (AJCC) criteria. A 10-year retrospective review of a consecutive series of 105 gastrectomies for adenocarcinoma at an urban public teaching hospital was performed. Of 69 intended curative gastrectomies, 55 (80%) included D2 lymphadenectomies, whereas of 36 palliative gastrectomies, only 9 (25%) included D2 lymphadenectomies (P = 0.0041). Only D2 and not D1 lymphadenectomy achieved the AJCC minimum guideline of the 15 lymph nodes required for accurate pathologic staging (mean 25.2 vs 12.4 nodes, respectively; P = 0.0001). For D2 cases, 86 per cent had greater than 15 nodes excised compared with only 20 per cent for D1 cases (P = 0.0002). The morbidity and mortality rates for D2 and D1 operations were 22 per cent and 2 per cent, and 41 per cent and 2 per cent, respectively. We conclude that there was no increased morbidity or mortality associated with D2 lymphadenectomy; that reliable harvesting of an adequate number of lymph nodes for accurate AJCC pathologic tumor staging requires D2 lymphadenectomy; and that D2 lymphadenectomy should be performed as part of virtually all gastrectomies for invasive adenocarcinoma having curative intent.     

Relevant prognostic factors in gastric cancer: ten-year results of the German Gastric Cancer Study.

OBJECTIVE: In 1986 a prospective multicenter observation trial in patients with resected gastric cancer was initiated in Germany. An analysis of prognostic factors based on the 10-year survival data is now presented. PATIENTS AND METHODS: A total of 1654 patients treated for gastric cancer between 1986 and 1989 at 19 centers in Germany and Austria were included. The resected specimen were evaluated histopathologically according to a standardized protocol. The extent of lymphadenectomy was classified after surgery based on the number of removed lymph nodes on histopathologic assessment (25 or fewer removed nodes, D1 or standard lymphadenectomy; >25 removed nodes, D2 or extended lymphadenectomy). Endpoint of the study was death. Follow-up is complete for 97% of the included patients (median follow-up of the surviving patients is 8.4 years). Prognostic factors were assessed by multivariate analysis. RESULTS: A complete macroscopic and microscopic tumor resection (R0 resection according to the UICC 1997) could be achieved in 1182 of the 1654 patients (71.5%). The calculated 10-year survival rate in the entire patient population was 26.3% +/- 4.7%; it was 36.1% +/- 1.6% after an R0 resection. In the total patient population there was an independent prognostic effect of the ratio between invaded and removed lymph nodes, the residual tumor (R) category, the pT category, the presence of postsurgical complications, and the presence of distant metastases. Multivariate analysis in the subgroup of patients who had a UICC R0 resection confirmed the nodal status, the pT category, and the presence of postsurgical complications as the major independent prognostic factors. The extent of lymph node dissection had a significant and independent effect on the 10-year survival rate in patients with stage II tumors. This effect was present in the subgroups with (pT2N1) and without (pT3N0) lymph node metastases on standard histopathologic assessment. The beneficial effect of extended lymph node dissection for stage II tumors persisted when patients with insufficient lymph node dissection (<15 nodes) were excluded from the analysis. There was no difference in the postsurgical morbidity and mortality rates between patients with standard and extended lymph node dissection. CONCLUSIONS: Lymph node ratio and lymph node status are the most important prognostic factors in patients with resected gastric cancer. In experienced centers, extended lymph node dissection does not increase the mortality or morbidity rate of resection for gastric cancer but markedly improves long-term survival in patients with stage II tumors. This effect appears to be independent of the phenomenon of stage migration.     

Prostatic carcinoma: incidence and location of unsuspected lymphatic metastases.

Prospective pathologic staging by pelvic lymphadenectomy in 60 patients with clinically localized carcinoma of the prostate disclosed a high incidence (35 per cent) of clinically silent and unsuspected lymph node metastases. When present, metastatic disease was frequently bilateral (57 per cent) and most commonly involved the obturator-hypogastric lymph nodes (87 per cent). Micrometastases alone were found in 5 patients and the potential significance of this finding on survival is discussed. Although the presence or absence of metastases could not be accurately predicted by histologic analysis of biopsy or prostatectomy specimens, the finding of undifferentiated tumor, marked anaplasia and penetration through the capsule correlated positively with nodal metastases. Pelvic lymphadenectomy is a safe and important diagnostic tool in the accurate staging of these patients. Its widespread use is advocated in patients with clinical stage B1, B2 and C tumors prior to definitive therapy. Based on the prospective data generated in this study lymphatic metastasis appears to be an early event in the spread of prostatic cancer.     

Adjuvant chemotherapy in non-seminomatous testis cancer: "mini-VAB" regimen: long-term followup

Since 40 to 50 per cent (range 20 to 80 per cent) of patients with stage II non-seminomatous germ cell tumors of the testis suffer relapse after orchiectomy and retroperitoneal lymph node dissection, relatively non-toxic adjuvant chemotherapy (consisting of vinblastine, actinomycin D, bleomycin and chlorambucil) was given to 62 patients after lymphadenectomy. Of these patients 82 per cent remained free of disease with a 4-year median followup and 18 per cent had relapse. Retrospective analysis reveals that no patient (0 of 33) with stage IIA and 38 per cent (11 of 29) with stage IIB disease had relapse. Patients with histologic evidence of extranodal extension of disease (N3 category) had the highest relapse rate (62 per cent). Based on our experience we recommend that patients with resected stage IIB disease, particularly those with extranodal extension of tumor, receive aggressive adjuvant chemotherapy.     

Correlation of serum tumor markers and lymphangiography with degrees of nodal involvement in surgical stage II testis cancer

There were 60 patients at our cancer center who underwent serum tumor marker studies (beta subunit of human chorionic gonadotropins and alpha-fetoprotein) and pedal lymphangiography before retroperitoneal lymph node dissection. Surgical stage II cases were divided according to tumor, node and metastasis staging. Beta-human chorionic gonadotropin and/or alpha-fetoprotein was elevated in 9 per cent (1 of 11) and the N1 cases, 36 per cent (5 of 14) of the N2A cases, 50 per cent (13 of 26) of the N2B cases and 89 per cent (8 of 9) of the N3 cases. Lymphangiography was positive or suspicious in 9 per cent (1 of 11) of the N1 cases, 36 per cent (5 of 14) of the N2A cases, 46 per cent (12 of 26) of the N2B cases and 56 per cent (5 of 9) of the N3 cases. Serum tumor markers and lymphangiography combined suggested lymph node metastases in 18 per cent (2 of 11) of the N1 cases, 50 per cent (7 of 14) of the N2A cases, 73 per cent (19 of 26) of the N2B cases and 100 per cent (9 of 9) of the N3 cases. We conclude that tumor markers and lymphangiography measurements are equally effective in the diagnosis of retroperitoneal lymph node metastases and that diagnostic accuracy is enhanced significantly by combining these 2 modalities. Retroperitoneal lymph node dissection remains the most reliable staging procedure. Reports of the accuracy of clinical staging should be correlated with subcategories of stage II disease.