Langenbeck’s Arch Surg (2002) 387:77–83 DOI 10.1007/s00423-002-0290-1
Mitsuo Tachibana Dipok Kumar Dhar Shoichi Kinugasa Hiroshi Yoshimura Toshiyuki Fujii Muneaki Shibakita Satoshi Ohno Shuhei Ueda Hitoshi Kohno Naofumi Nagasue
Received: 18 September 2001 Accepted: 25 February 2002 Published online: 30 April 2002 © Springer-Verlag 2002
M. Tachibana (✉) · D.K. Dhar S. Kinugasa · H. Yoshimura · T. Fujii M. Shibakita · S. Ohno · S. Ueda H. Kohno · N. Nagasue Second Department of Surgery, Shimane Medical University, Enya-cho 89-1, Izumo 693-8501, Shimane, Japan e-mail:
[email protected] Tel.: +81-853-202232 Fax: +81-853-202229
O R I G I N A L A RT I C L E
Esophageal cancer patients surviving 6 years after esophagectomy
Abstract Background: Esophageal cancer is one of the most malignant tumors, with a dismal prognosis in spite of recent advances in early diagnosis and extended lymphadenectomy. These patients need to be stratified according to prognostic variables for precise identification of high-risk group. Material and methods: Seventy-six patients with esophageal carcinoma were uniformly treated with curative intent between 1980 and 1992 with at least 6 years follow-up. Results and prognostic factors of long-term survival were analyzed by univariate and multivariate analyses. Results: Thirty patients (39.5%) survived 6 years, and the remaining 46 patients died within 6 years: recurrent esophageal cancer in 27 and causes unrelated to esophageal cancer in 19. The 1-, 2-, 3-, and 6-year overall survival rates in all 76 patients were 77.6%, 57.9%, 53.9%, and 39.5%, respectively. The factors influencing survival rate verified by univariate analysis were
Introduction Recent advances in early diagnosis and improvements in perioperative management have made esophageal carcinoma a potentially curable disease when diagnosed at early stages (pT1 and pT2) [1, 2]. Irrespective of the treatment modalities, however, long-term survival is disappointing when the disease extends through the esophageal wall, or when it is diagnosed with widespread
Borrmann classification (0, 1 vs. 2, 3), size of tumor (≤3.0 vs. >3.0 cm), depth of invasion (T1, 2 vs. T3, 4), pN category (pN0 vs. pN1), number of lymph node metastasis (≤4vs. >4), metastatic lymph node ratio (≤0.1 vs. >0.1), time of operation (≤480 vs. >480 min), and amount of perioperative blood transfusion given (≤2 vs. >2 U). Among the significant variables independent prognostic factors for survival determined by multivariate analysis were metastatic lymph node ratio and amount of blood transfusion. Conclusions: A significant number of patients can thus apparently be cured of esophageal carcinoma by extensive resection. Patients with many metastatic lymph nodes and much blood transfusion, on the other hand, should receive appropriate treatment against such esophageal carcinoma. Keywords Esophageal cancer · Esophagectomy · Prognostic factor · Long-term result · Clinicopathological features
lymph node involvement [3, 4]. The operative approach for esophageal cancer thus varies from conventional transthoracic esophagectomy, chiefly for palliation [5], to en bloc esophagectomy with a curative intent [6] to limited excision without thoracotomy [7]. Since the early 1980s three-field extensive lymph node dissection during esophagectomy has become a standard surgical procedure to obtain an accurate pathological staging and to improve surgical results in Japan [8, 9, 10], encouraging
78
early introduction of extended esophagectomy in Western countries [11, 12]. Although mortality and morbidity rates after such esophagectomy have been acceptable, the long-term outcomes are not necessarily satisfactory [8, 9, 10, 13, 14]. Stratification of these patients according to prognostic risk factors might be helpful in identifying a high-risk group for designing an appropriate therapy, while the low-risk group could spare unexpected morbidity from extensive treatment protocols for advanced diseases. The present study was designed to clarify the clinicopathological features of patients with squamous cell carcinoma of the esophagus surviving 6 years after curative esophagectomy who had an almost similar therapeutic regimen.
Patients and methods Between January 1981 and December 1992 there were 155 patients with primary squamous cell carcinoma of the esophagus admitted to the Second Department of Surgery, Shimane Medical University. Among these, 123 patients underwent esophagectomy (R0 esophagectomy [15] in 89, 57.4%; R1/2 esophagectomy in 34, 21.9%), and the remaining 32 patients (20.6%) had palliative treatment (bypass in 5, esophagostomy and/or gastrostomy or jejnostomy in 6, and radiochemotherapy alone in 21). Within 90 days after esophagectomy 13 of the 89 patients (14.6%) undergoing R0 esophagectomy died, 3 of 34 (8.8%) with R1/2 esophagectomy, and 4 of 11 (36.4%) with palliative surgery died. To exclude the effects of surgery-related postoperative complications on longterm survival, 13 patients who died within 90 days after R0 esophagectomy were excluded. Therefore 76 patients who had undergone curative esophagectomy (R0) were analyzed. The patients included in this series had at least a 6-year followup because cancer recurrence was noted even after 5.9 years of curative resection in our series. Of interest, three of six patients who died over 5 years after the operation died of recurrent disease. Thirty patients survived longer than 6 years (survivor group) and 46 patients less than 6 years (nonsurvivor group): following radical esophagectomy due to recurrences of the disease in 27 and due to causes unrelated to esophageal cancer in 19. The majority of these patients underwent a right transthoracic subtotal esophagectomy and dissection of the three fields: the cervical (bilateral supraclavicular regions), mediastinal (periesophagus and around the trachea including recurrent laryngeal nerve nodes), and abdominal (perigastric region and around the celiac axis) lymph nodes. Reconstruction was usually carried out with a gastric tube through the retrosternal rout and esophagogastrostomy was performed in the neck under cervical incision and laparotomy. Eight patients with superficial esophageal cancer received no adjuvant treatment, and another eight received none due to aging and/or poor general condition (six) or patient refusal (two). When the tumor invaded the deeper submucosal layer or shallow muscle layer, two patients received one or two cycles of postoperative chemotherapy alone. According to the previous principles for adjuvant treatment of advanced tumor, 26 patients received pre- and postoperative radiochemotherapy, and thereafter 32 patients received two cycles of the same regimen postoperatively. One cycle of chemotherapy consisted of bolus administration of cisplatin (50 mg/m2 per day) on the 1st and 15th days and 5-fluorouracil (300 mg/m2) for 28 days by continuous intravenous infusion. One cycle of radiotherapy consisted of 1.8 Gy daily directed to the tumor bed. The total dose was 30.6 Gy.
Clinicopathological characteristics of these 76 patients were retrospectively investigated based on the TNM classification of the malignant tumors [15]. Patients received a regular follow-up at the outpatient department and all information were entered into a database. An update inquiry about the present status of all surviving patients was carried out by telephone or postal contact in December 1998. The patients who clearly died of recurrence were regarded as tumor-related deaths. The standard χ2 test with or without Yates’ correction was used for comparative analyses. The 6-year survival rates were estimated by the Kaplan-Meier method [16]. The statistical analysis was carried out by the log-rank test to test for equality of the survival curves. Independence of each prognostic factors was checked using Spearman’s correlation test or the Mann-Whitney U test. In multivariate analysis independent prognostic factors were determined using a Cox proportional hazards model [17] (StatView J4.5: Abacus Concepts, Berkeley, Calif., USA). The level of significance was set at P64
41 35
53.9 46.1
3.58 2.08
61.0 54.3
39 40
NS
Tumor location Middle third Nonmiddle
39 37
51.3 48.7
3.64 1.89
64.1 48.6
43.6 35.1
NS
Borrmann classification 0, 1 24 2, 3 52
31.6 68.4
6.0 1.5
87.5 44.2
59.3 30.8
0.0046
Size of tumor (cm) ≤3.0 16 >3.0 60
21.1 78.9
6.0 1.98
87.5 50.0
62.5 33.3
0.0258
Depth of invasion T1, 2 48 T3, 4 28
63.2 36.8
6.0 1.26
72.9 32.1
52.1 17.9
4 15 19.7
5.1 0.92
67.2 13.3
47.5 6.7
0.1 24
68.4 31.6
6.0 1.0
78.8 12.5
55.8 4.2
30 36 47.4
5.1 2.1
62.5 52.8
45.0 33.3
NS
Time of operation (minute) ≤480 60 >480 16
78.9 21.1
5.1 1.4
68.3 18.8
45.0 18.8
0.0078
Amount of blood transfusion (U) ≤2 24 31.6 >2 52 68.4
6.0 1.6
87.5 44.2
66.7 26.9
0.0010
Histological differentiation Well 13 Not well 63
17.1 82.9
2.1 3.6
61.5 57.1
38.5 39.7
NS
Lymph vessel invasion Absent 19 Present 57
25.0 75.0
5.1 1.9
84.2 49.1
36.8 40.4
NS
Blood vessel invasion Absent 30 Present 46
39.5 60.5
5.2 1.5
73.3 47.8
46.7 34.8
NS
Radiotherapy Yes No
76.3 23.7
3.2 5.1
55.2 66.7
39.7 38.9
NS
58 18
80
Fig. 1 Cumulative survival curves in terms of presence or absence of lymph node metastasis. The 2- and 6-year survival rates were 84.4% and 56.2% in those without lymph node metastasis (n=32) and 36.4% and 27.3% in those with diseased node (n=44; log rank, P=0.0013)
metastasis, and metastatic lymph node ratio) were selected. By univariate analysis the distinction levels were defined by the best predictable value of prognosis in a stepwise fashion. Table 1 shows the clinicopathological variables influencing the cumulative survival rates by univariate analysis: Borrmann classification (0, 1 vs. 2, 3), size of tumor (≤3.0 vs. >3.0 cm), depth of invasion (pT1, 2 vs. T3, 4), presence or absence of lymph node metastasis (pN0 vs. pN1), number of lymph node metastasis (≤4 vs. >4), metastatic lymph node ratio (≤0.1 vs. >0.1), time of operation (≤480 vs. >480 min), and amount of perioperative blood transfusion given (≤2 vs. >2 U). Gender, histological differentiation, and vessel invasions did not influence the survival rate. The 1-, 2-, 3-, and 6-year overall survival rates of 76 patients were 77.6%, 57.9%, 53.9%, and 39.5%, respectively. Those of 89 patients including 13 patients who died of postoperative complications during the hospital stay were 69.4%, 51.8%, 48.2%, and 35.3%, respectively. Cumulative survival curves in terms of presence or absence of lymph node metastasis, number of lymph node metastasis, and metastatic lymph node ratio are shown in Figs. 1, 2, and 3, respectively.
Prognostic significance of metastatic lymph node ratio evaluated by multivariate analysis After check of independence of various factors, three factors of lymph node metastasis were found not to be independent of each other (P
