The American Surgeon
April 14, 2008
The gastrointestinal tract is the most frequent site of extranodal lymphoma, representing approximately 6 to 12 per cent of all extranodal lymphomas.1,2 The stomach is the most frequently involved site followed by the small bowel and large bowel, where lymphomas represent 0.2 to 0.4 per cent of all colorectal tumors.3 It is a rare disease. Most of the published series are retrospective and only a few large series are available.
Diagnosis of primary colorectal lymphoma requires histopathologic confirmation of a lymphoproliferative neoplasm confined to the colon and/or rectum with or without regional lymph nodes, without involving other abdominal organs or bone marrow, and without retroperitoneal or mediastinal lymphadenopathy.4
The cecum is the most commonly involved site in most series, ranging from 38 to 71 per cent.5-7 It is believed that primary non-Hodgkin lymphoma (NHL) of the gastrointestinal tract originates from the mucosa-associated lymphoid tissue (MALT) that has unique immunophysiological characteristics differing from those of peripheral lymph nodes.8
The purpose of this study was to evaluate clinical, pathologic, and treatment related variables for primary colorectal lymphoma at a single tertiary care institution.
Patients and Methods
Under an Institutional Review Board approved protocol, from January 1990 to June 2002 a total of 15 patients were identified with a diagnosis of colorectal lymphoma from our tumor registry and retrospectively reviewed. Demographic data, clinical features, type of treatment, pathological findings, recurrence rate, mortality, and morbidity were analyzed.
The results are presented in mean ± SD or median and range. Differences in survival were evaluated by the log-rank test and the interval of disease free survival was calculated by the Kaplan and Meier method. Significance was defined as P < 0.05.
Results
During the study period, 15 patients with large bowel lymphoma were diagnosed and treated. The mean age at presentation was 52 years (range 23-79). Five patients (33%) were men and 10 were women (67%).
No patient presented B-symptoms and the main clinical feature was abdominal pain, which was present in all patients (100%), followed by diarrhea in 54 per cent (8 patients). Lower gastrointestinal bleeding was present in two patients (13%) and a recent history of alterations in bowel habits was also present in two patients. Nausea and vomiting were present in seven patients (47%) secondary to low-grade gastrointestinal obstruction (see Table 1). All patients underwent a CT scan and none had radiographic evidence of systemic disease. Only one patient (6.7%) had a positive history of previous lymphoproliferative disease (lymphoma located in the tonsils) and radiation exposure, which was treated successfully with systemic chemotherapy and external radiotherapy.
One patient had chronic ulcerative colitis managed surgically with total proctocolectomy and J-pouch reconstruction. Mean LDH was 215 u/L (range 129-309), however five patients had levels above the upper normal limit. After treatment, mean LDH value was 140 u/L (98-241) and all patients were in the normal range. The most common disease location was the cecum (9 patients, 60%), followed by the right colon (4 patients, 27%) and sigmoid colon (2 patients, 13%). There was no evidence of synchronous involvement. Preoperative stage was clinically determined according to the clinical classification of primary NHL of the gastrointestinal tract (Consensus Conference in Lugano 1993).1
All patients were clinically classified as stage I (tumor confined to gastrointestinal tract without evidence of abdominal cavity extension from the primary site or metastases). At the time of operation, the tumor was confined to the gastrointestinal tract in six patients (86%). One patient (7%) had evidence of local extension and in six other patients, there was evidence of metastasis (regional nodes). Six patients (40%) had histological evidence of muscular involvement, six tumors penetrated the serosa, and three involved the pericolonic fat. Eighty per cent of the patients (12 patients) had negative nodes, whereas three patients had regional disease.
Histopathological findings were classified according to the working formulation and Kiel Classification (10). Ten patients had T-cell lymphomas (see Table 2).
Thirteen patients were treated with right hemicolectomy, one patient underwent total proctocolectomy with ileal J J-pouch, and another patient underwent a sigmoid colectomy with colorectal anastomosis. Margins were free of tumor in thirteen patients (86.7%) and positive in two (13.3%). Mean tumor size was 7.9 cm ranging from 3 to 30 cm.
The mean hospital stay was 6.4 days ranging from 3 to 26 days. There was no surgical mortality and the morbidity was 20 per cent (3 patients). One patient was treated with right hemicolectomy and postoperatively developed renal insufficiency and abdominal sepsis related to anastomotic leakage, subsequently managed with diversion. The second patient had an incisional hernia. The third patient with a history of ulcerative colitis (UC) treated with total proctocolectomy with ileal J-pouch developed an ileal pouchvaginal fistula manifested by vaginal discharge and recurrent urinary tract infections. She underwent a temporary loop ileostomy. Twelve patients received postoperative chemotherapy (80%). In 8 patients (67%) the regimen was cyclophosphamide, hydrocortisone, Oncovin and cis-platinum. The remaining 4 patients (33%) received Cytoxan, vincristine, and adriamycine.
The follow-up was complete in all patients with a mean of 31 months (median 28 months). During follow-up one patient developed recurrence of lymphoma. This was a 75-year-old female with a lowgrade lymphoma located at the cecum treated with right hemicolectomy with negative margins and postoperative chemotherapy. Recurrence was systemic and was manifested at 4 months postoperatively with pleural effusion diagnosed by malignant cells on thoracocentesis. She subsequently was treated with systemic chemotherapy (fludarabine and mitoxantrone). The median survival of the whole group was 60 months and the interval of disease free survival was 4 months in patients with recurrence.
Discussion
Primary colorectal lymphoma is an uncommon disease. It is less frequent than lymphoma involving the stomach or small intestine and generalized lymphoma involving the colon and rectum. All patients in this study met the criteria of primary colorectal lymphoma with histologic confirmation.
Pandey et al.,5 in their series of 8 cases, found its peak incidence during the second decade of the life. However other authors like Matkovic et al.,7 Herrmann et al.1 and Morton et al.9 do not report a peak age in the incidence of this disease. In our series the mean age at presentation was 52 years (range 23-79). In the current series there were more females (67%) than males, however no data suggesting any gender prevalence is reported in the literature.
No patient presented B-symptoms and the most common symptom was abdominal discomfort presented in every patient followed by diarrhea in 54 per cent. Seven patients had clinical manifestations of lower gastrointestinal obstruction, yet none required urgent laparotomy.
The most common location of the disease was the cecum in 9 patients (60%), the same predominance reported in most studies.5, 10 Rectal involvement is more frequent, representing the second most common site in many series, although in our series no rectal involvement was found. Multiple site involvement is frequently reported in other series (stomach or small bowel) but in the present series there was no evidence of synchronous lesions.
Surveillance was complete in all patients with a mean follow-up of 31 months (median of 28 months). During follow-up one patient developed recurrence 4 months after initial intervention. In our series, the majority of patients were alive at the time of analysis which is more optimistic than reported by some authors10-12 who conclude that colon and rectal lymphoma has a poor prognosis with marginal response to surgical treatment and chemotherapy.
Chemotherapy may be used in the treatment of locally advanced or metastatic disease in an attempt to control residual microscopic disease and prolong survival. Surgical resection remains an important treatment modality, followed by chemotherapy when appropriate; this is supported by the literature.13
Conclusion
Surgical resection of localized, primary colonic lymphoma provides excellent local disease control and should be considered the primary treatment option. The role of chemotherapy remains controversial depending on the grade, stage, and extension of residual disease.
(C) 2008 The American Surgeon. via ProQuest Information and Learning Company; All Rights Reserved
