Carcinoma of the Colon and Rectum A Review of Results of Don R.
Surgical
Miller, MD, Frank F. Allbritten Jr,
Treatment in 164 Patients
MD
report is a retrospective review of the authors' experiwith colonic and rectal carcinoma in 164 patients from 1954 to 1968 with a five-year follow-up. The operative approach utilized principles of early vascular isolation, wide-colon and mesenteric resection, and minimal tumor manipulation in curative operations; palliative procedures were aggressively \s=b\ This
ence
employed.
Of the 164 patients, 104 with lesions almost equally divided between colon and rectum had resections with curative intent. The overall five-year survival rate after curative resection was 55%, and included eight hospital deaths that were usually related to cardiopulmonary complications. Lymph node involvement appreciably decreased the survival rate from 72% without node involvement to 39% with node involvement. Patients with rectal lesions experienced more favorable results than those with colonic lesions. Palliative procedures in 46 patients enhanced comfort, but few patients survived longer than one year. Wound and anastomotic complications were infrequent.
(Arch Surg 111:692-696, 1976)
of the colon and rectum continues to be a of death, and although operability and major operative mortality have continued to improve, the end results group data suggest that overall five-year survival rates in this disease have not altered appreciably in recent years.' Changing approaches to diagnosis and manage¬ ment of colorectal cancer make information about currently used methods of treatment essential for proper
Carcinoma
cause
comparison. Accepted
for publication Feb 12, 1976. From the Department of Surgery, University of Kansas, School of Medicine, Kansas City. Dr Miller is now with the University of California, Irvine, and Dr Allbritten is residing in Cunningham, Kan. Read before the annual meeting of the Southern California Chapter of the American College of Surgeons, Newport Beach, Calif, Jan 18, 1976. Reprint requests to Department of Surgery, University of California School of Medicine, Irvine, CA 92717 (Dr Miller).
This report documents the authors' operative experience with carcinoma of the colon and rectum from July 1, 1954, through July 1, 1968, with five years of follow-up at the University of Kansas Medical Center. Management utilized principles of wide resection of colon and mesen¬ tery, minimal tumor manipulation, and early vascular isolation in curative operations; palliative procedures were aggressively employed. Patients were followed up by personal interview or by physician contact by the Univer¬ sity Tumor Registry. All patients could be followed up. SUBJECTS AND METHODS of 164 consecutive personal patients reviewed retrospectively; the data were recorded on charts and subsequently analyzed. Particular reference was made to the symptoms and their duration, patho¬ logic findings, operative procedure, complications, and the subse¬ quent course. A pathologic diagnosis of invasive adenocarcinoma was made in each instance; patients in whom a histologie diagnosis was not established were eliminated from the study. Pathologic reports of biopsy specimens were studied for the extent of lymph node extension, and the level of wall and vascular invasion. In some patients, removal of tissue for biopsy only was done. Preoperative evaluation consisted of complete history and phys¬ ical examination, proctoscopy, barium enema, and routine and indicated special laboratory studies. Carcinoembryonic antigen determination and colonoscopy were not available during this period of study. A routine for bowel preparation consisting of dietary restriction, mechanical cleansing, and orally administered neomycin sulfate and phthalylsulfathiazole for 24 hours preceding the operation was used throughout most of the study. The operative procedure was conducted under general endotra¬ cheal anesthesia, and usually a midline incision was employed. If, after initial assessment of the primary lesion and extent of spread, it appeared that all gross tumor could be removed, a curative resection was begun. Palliative resection to remove the primary lesion was done when it could be accomplished through relatively uninvolved supple tissues that could be expected to heal satisfactorily. Abdominoperineal resection was used for palliation,
The hospital records managed by the authors
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/19/2015
were
for curative resection. Operations to only bypass actual obstructions were done when the incurable disease was extensively spread or fixed, and expectancy was believed to be short, or, in the case of the acutely obstructed colon, in preparation for a later definitive procedure. Curative operations were designed to initially control the arterial supply and venous drainage of the involved area, to avoid manipulation of the primary lesion, and to remove a large segment of colon and its related mesentery. In the case of rectal lesions, the dissection was on the lateral pelvic wall employing the principles outlined by Miles.7 Patients treated before 1958 had more limited resection, but standard operations usually consisted of right hemicolectomy and ileotransverse colostomy for lesions of the cecum and ascending colon, transverse colectomy for lesions of the transverse colon, and left hemicolectomy for lesions of the descending colon, sigmoid colon, and rectum. Lesions at or below the peritoneal reflection were usually removed by abdominoperi¬ neal resection with end colostomy. Recent primary maturation of the colostomy has virtually eliminated the development of colos¬ tomy stricture and the need for finger dilation. When it was possible to achieve a tumor-free margin of approximately 5 cm below the rectosigmoid lesion and a technically satisfactory endto-end anastomosis, so-called anterior resection was done. This was constructed in two layers with special attention to adequate blood supply. Contiguously involved structures were occasionally removed when required to accomplish a curative resection. Fat and nodal tissues overlying the aorta and vena cava and iliac vessels were removed. When the Miles technique was chosen, the abdominal procedure consisted of isolating the inferior mesenterie vascular supply, freeing up the colon to be resected and the rectum, constructing the colostomy and closing the distal part of the colon. The pelvic peritoneum was not closed; the colon, rectum, and related mesen¬ tery were not divided. Using the prone jackknife approach to the perineum, the perineal portion of the operation was completed, widely excising perianal fatty tissues after removal of the coccyx. The exposure provided by this approach, as compared with the lithotomy approach, simplifies the detachment of the levator muscles and permits precise dissection of tissues from the prostatic area or vagina. When possible, the peritoneum over the bladder was approximated to the anterior sacral tissues to reduce the pelvic space, but when this was not possible, the closure consisted only of approximation of pararectal fat and skin around a polyvinyl tube for dependent drainage. Prompt primary healing usually followed. Following operation, antibiotics were used prophylactically. Although tapes were not applied to the colon above and below the lesion, the distal part of the open colon segment was usually irrigated with normal saline solution to reduce both spillage contamination and the possibility of implan¬ tation recurrence at the anastomotic suture line. Local intraperi¬ toneal nitrogen mustard and systemic nitrogen mustard were as
well
potential
as
or
RESULTS
The age of patients Age, yr 20-40 40-50 50-60 60-70 70-80
was as
follows: No. of Patients
3 11 35 55 52
8
Total
164
equal sex distribution (84 men, 80 age of patients was 64.8 years, with a range of 29 to 86 years (median age, 65 years). Approxi¬ mately two thirds of the patients were in the seventh and
There
almost
was
women). The
mean
eighth decades.
The distribution of lesions is
'
occasionally employed. Following recovery, patients were followed up at intervals. Adjunctive postoperative radiotherapy and systemic chemothera¬ py were infrequently used for palliation for recurrent disease.
80-90
as
follows: No. (%) of Patients
Location
78 44 33
(47) (27) (20) 9 (6)
Rectum Left colon Right colon Transverse colon
Î64 (100)
Total
Almost half the lesions were located in the rectum, and the left colon had more lesions than the right. The average duration of symptoms was 6.5 months. There was no direct correlation between the duration of symptoms and the evidence of advanced disease. Two thirds of the patients with symptoms less than six months had local or node extension, while only half the patients with symptoms longer than six months had evidence of similar extension. Twelve patients, including three who refused operation, had no operation or a procedure with biopsy only without definitive or palliative measures because of extensive disease. Death followed progression of disease in one of these patients. The age of onset of rectal carcinoma and time of death in an identical twin who refused operation were similar to those of his 38-year-old twin brother. Curative or palliative procedures were done in the remaining 93% of patients. Palliative Procedures
Palliative procedures were aggressively employed when indicated in 45 patients, and resection of colon or rectum was done in 27 of those (Table 1). In spite of advanced disease in many patients, operative mortality was 8.8%. Causes of operative death after pallia¬ tive procedures were as follows: Year 1961 1958 1957 1963
Palliative Procedure
Ueotransverseostomy
Cause of Death
Pulmonary embolus
Anastomotic leak Colocolostomy Anteroposterior resection Enterocolitis Shock Ileocolostomy
Palliative
abdominoperineal resection was done in seven
death was due to enterocolitis. An anasto¬ motic leak followed by peritonitis was fatal in one patient after colocolostomy. Unfortunately, survival following palliative procedures was usually for a short time only (Table 2). The majority of patients died in less than one year. However, palliative procedures were believed to be worthwhile in extending comfort. Some believe that abdominoperineal resection has no place as a palliative procedure for carcinoma of the rectum. Four of the seven patients survived from one to six years after palliative abdominoperineal resection. Pallia-
instances;
one
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/19/2015
Table 1.—Carcinoma of the Colon and Rectum: Palliative Procedures
(%) of Operative
No. No. of Procedures
Operations Rectum (N = 15) Anteroposterior resection
6
Deaths
Table 2.—Survival With Palliative Procedures Procedure and Survivors Rectum (N = 14)
Duration of
Survival, yr
Surgical
Miller, MD, Frank F. Allbritten Jr,
Treatment in 164 Patients
MD
report is a retrospective review of the authors' experiwith colonic and rectal carcinoma in 164 patients from 1954 to 1968 with a five-year follow-up. The operative approach utilized principles of early vascular isolation, wide-colon and mesenteric resection, and minimal tumor manipulation in curative operations; palliative procedures were aggressively \s=b\ This
ence
employed.
Of the 164 patients, 104 with lesions almost equally divided between colon and rectum had resections with curative intent. The overall five-year survival rate after curative resection was 55%, and included eight hospital deaths that were usually related to cardiopulmonary complications. Lymph node involvement appreciably decreased the survival rate from 72% without node involvement to 39% with node involvement. Patients with rectal lesions experienced more favorable results than those with colonic lesions. Palliative procedures in 46 patients enhanced comfort, but few patients survived longer than one year. Wound and anastomotic complications were infrequent.
(Arch Surg 111:692-696, 1976)
of the colon and rectum continues to be a of death, and although operability and major operative mortality have continued to improve, the end results group data suggest that overall five-year survival rates in this disease have not altered appreciably in recent years.' Changing approaches to diagnosis and manage¬ ment of colorectal cancer make information about currently used methods of treatment essential for proper
Carcinoma
cause
comparison. Accepted
for publication Feb 12, 1976. From the Department of Surgery, University of Kansas, School of Medicine, Kansas City. Dr Miller is now with the University of California, Irvine, and Dr Allbritten is residing in Cunningham, Kan. Read before the annual meeting of the Southern California Chapter of the American College of Surgeons, Newport Beach, Calif, Jan 18, 1976. Reprint requests to Department of Surgery, University of California School of Medicine, Irvine, CA 92717 (Dr Miller).
This report documents the authors' operative experience with carcinoma of the colon and rectum from July 1, 1954, through July 1, 1968, with five years of follow-up at the University of Kansas Medical Center. Management utilized principles of wide resection of colon and mesen¬ tery, minimal tumor manipulation, and early vascular isolation in curative operations; palliative procedures were aggressively employed. Patients were followed up by personal interview or by physician contact by the Univer¬ sity Tumor Registry. All patients could be followed up. SUBJECTS AND METHODS of 164 consecutive personal patients reviewed retrospectively; the data were recorded on charts and subsequently analyzed. Particular reference was made to the symptoms and their duration, patho¬ logic findings, operative procedure, complications, and the subse¬ quent course. A pathologic diagnosis of invasive adenocarcinoma was made in each instance; patients in whom a histologie diagnosis was not established were eliminated from the study. Pathologic reports of biopsy specimens were studied for the extent of lymph node extension, and the level of wall and vascular invasion. In some patients, removal of tissue for biopsy only was done. Preoperative evaluation consisted of complete history and phys¬ ical examination, proctoscopy, barium enema, and routine and indicated special laboratory studies. Carcinoembryonic antigen determination and colonoscopy were not available during this period of study. A routine for bowel preparation consisting of dietary restriction, mechanical cleansing, and orally administered neomycin sulfate and phthalylsulfathiazole for 24 hours preceding the operation was used throughout most of the study. The operative procedure was conducted under general endotra¬ cheal anesthesia, and usually a midline incision was employed. If, after initial assessment of the primary lesion and extent of spread, it appeared that all gross tumor could be removed, a curative resection was begun. Palliative resection to remove the primary lesion was done when it could be accomplished through relatively uninvolved supple tissues that could be expected to heal satisfactorily. Abdominoperineal resection was used for palliation,
The hospital records managed by the authors
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/19/2015
were
for curative resection. Operations to only bypass actual obstructions were done when the incurable disease was extensively spread or fixed, and expectancy was believed to be short, or, in the case of the acutely obstructed colon, in preparation for a later definitive procedure. Curative operations were designed to initially control the arterial supply and venous drainage of the involved area, to avoid manipulation of the primary lesion, and to remove a large segment of colon and its related mesentery. In the case of rectal lesions, the dissection was on the lateral pelvic wall employing the principles outlined by Miles.7 Patients treated before 1958 had more limited resection, but standard operations usually consisted of right hemicolectomy and ileotransverse colostomy for lesions of the cecum and ascending colon, transverse colectomy for lesions of the transverse colon, and left hemicolectomy for lesions of the descending colon, sigmoid colon, and rectum. Lesions at or below the peritoneal reflection were usually removed by abdominoperi¬ neal resection with end colostomy. Recent primary maturation of the colostomy has virtually eliminated the development of colos¬ tomy stricture and the need for finger dilation. When it was possible to achieve a tumor-free margin of approximately 5 cm below the rectosigmoid lesion and a technically satisfactory endto-end anastomosis, so-called anterior resection was done. This was constructed in two layers with special attention to adequate blood supply. Contiguously involved structures were occasionally removed when required to accomplish a curative resection. Fat and nodal tissues overlying the aorta and vena cava and iliac vessels were removed. When the Miles technique was chosen, the abdominal procedure consisted of isolating the inferior mesenterie vascular supply, freeing up the colon to be resected and the rectum, constructing the colostomy and closing the distal part of the colon. The pelvic peritoneum was not closed; the colon, rectum, and related mesen¬ tery were not divided. Using the prone jackknife approach to the perineum, the perineal portion of the operation was completed, widely excising perianal fatty tissues after removal of the coccyx. The exposure provided by this approach, as compared with the lithotomy approach, simplifies the detachment of the levator muscles and permits precise dissection of tissues from the prostatic area or vagina. When possible, the peritoneum over the bladder was approximated to the anterior sacral tissues to reduce the pelvic space, but when this was not possible, the closure consisted only of approximation of pararectal fat and skin around a polyvinyl tube for dependent drainage. Prompt primary healing usually followed. Following operation, antibiotics were used prophylactically. Although tapes were not applied to the colon above and below the lesion, the distal part of the open colon segment was usually irrigated with normal saline solution to reduce both spillage contamination and the possibility of implan¬ tation recurrence at the anastomotic suture line. Local intraperi¬ toneal nitrogen mustard and systemic nitrogen mustard were as
well
potential
as
or
RESULTS
The age of patients Age, yr 20-40 40-50 50-60 60-70 70-80
was as
follows: No. of Patients
3 11 35 55 52
8
Total
164
equal sex distribution (84 men, 80 age of patients was 64.8 years, with a range of 29 to 86 years (median age, 65 years). Approxi¬ mately two thirds of the patients were in the seventh and
There
almost
was
women). The
mean
eighth decades.
The distribution of lesions is
'
occasionally employed. Following recovery, patients were followed up at intervals. Adjunctive postoperative radiotherapy and systemic chemothera¬ py were infrequently used for palliation for recurrent disease.
80-90
as
follows: No. (%) of Patients
Location
78 44 33
(47) (27) (20) 9 (6)
Rectum Left colon Right colon Transverse colon
Î64 (100)
Total
Almost half the lesions were located in the rectum, and the left colon had more lesions than the right. The average duration of symptoms was 6.5 months. There was no direct correlation between the duration of symptoms and the evidence of advanced disease. Two thirds of the patients with symptoms less than six months had local or node extension, while only half the patients with symptoms longer than six months had evidence of similar extension. Twelve patients, including three who refused operation, had no operation or a procedure with biopsy only without definitive or palliative measures because of extensive disease. Death followed progression of disease in one of these patients. The age of onset of rectal carcinoma and time of death in an identical twin who refused operation were similar to those of his 38-year-old twin brother. Curative or palliative procedures were done in the remaining 93% of patients. Palliative Procedures
Palliative procedures were aggressively employed when indicated in 45 patients, and resection of colon or rectum was done in 27 of those (Table 1). In spite of advanced disease in many patients, operative mortality was 8.8%. Causes of operative death after pallia¬ tive procedures were as follows: Year 1961 1958 1957 1963
Palliative Procedure
Ueotransverseostomy
Cause of Death
Pulmonary embolus
Anastomotic leak Colocolostomy Anteroposterior resection Enterocolitis Shock Ileocolostomy
Palliative
abdominoperineal resection was done in seven
death was due to enterocolitis. An anasto¬ motic leak followed by peritonitis was fatal in one patient after colocolostomy. Unfortunately, survival following palliative procedures was usually for a short time only (Table 2). The majority of patients died in less than one year. However, palliative procedures were believed to be worthwhile in extending comfort. Some believe that abdominoperineal resection has no place as a palliative procedure for carcinoma of the rectum. Four of the seven patients survived from one to six years after palliative abdominoperineal resection. Pallia-
instances;
one
Downloaded From: http://archsurg.jamanetwork.com/ by a New York University User on 06/19/2015
Table 1.—Carcinoma of the Colon and Rectum: Palliative Procedures
(%) of Operative
No. No. of Procedures
Operations Rectum (N = 15) Anteroposterior resection
6
Deaths
Table 2.—Survival With Palliative Procedures Procedure and Survivors Rectum (N = 14)
Duration of
Survival, yr
