- Email: [email protected]
Pelvic exenterative therapy in recurrent pelvic carcinoma
Pelvic exenterative therapy in recurrent pelvic carcinoma GEORGE S.
JOSEPH Ann
W.
MARTIN
Arbor,
MORLEY,
M.D.
LINDENAUER, C.
CERNY,
M.D. M.D.
Michigan
During the 5 year period from June 1,1965, to June 1, 1970, 37 patients were treated with some type of pelvic exenteration at the University of Michigan Medical Center. Total pelvic exenteration was performed in over 60 per cent of the cases. Of 14 patients operated upon 3 or more years ago, 7 are alive and clinically free of disease. The operative mortality in the cases reported is 2.7 per cent. Of 21 patients treated 2 or more years ago, 6 had regional lymph node involvement and all have died of disease. Twelve of the I5 patients with no nodal involvement are alive and clinically free of disease. The importance of various technical points to include “ski positioning” of the patient on the operating table, development of a peritoneal sling to be used as a pelvic cover, preparation of a ureterosigmoid conduit for urinary diversion, and profihylactic compartmentalization of the inferior vena cava are described in detail. Vaginal reconstruction when desired is performed at a later date.
In addition, a review of 5 year survival statistics suggested that the continued utilization of pelvic exenterative therapy for the treatment of recurrent carcinoma of the cervix and other pelvic malignancies was justified.2t 3 During the 5 year period from June 1, 1965, to June 1, 1970, 37 patients have been treated with some type of pelvic exenteration under our direction at the University of Michigan Medical Center, and these cases are the basis for this report.
S H 0 R T L Y A F T E R Brunschwigl reported his initial evaluation of radical operative treatment of advanced pelvic carcinoma, a program incorporating this form of therapy was established at the University of Michigan Medical Center. During the 12 year period from 1953 to 1965, 30 patients were treated with some type of pelvic exenteration at this institution with over thirty different physicians participating in the procedure. Because of the high operative morbidity and mortality during this period, a multidisciplinary and cooperative approach to this form of ultraradical surgery was instituted in an attempt to improve the postoperative morbidity and operative mortality. From the Department Gynecology and the Surgery, Section of Section of Urology, Michigan Medical
Selection
of cases
Although local excision or local irradiation of a small lesion occasionally has a place in the treatment of local recurrence, the procedure of choice usually involves a more radical operation such as radical hysterectomy and pelvic lymph node dissection, or partial or total pelvic exenteration. For more than 35 years, all cases of gynecologic malignancy at the University of Michigan Medical
of Obstetrics and Department of General Surgery and University of Center.
Presented at the Thirty-eighth Annual Meeting of the Central Association of Obstetricians and Gynecologists, Chicago, Illinois, Sept. 24-26, 1970.
1175
1176
Morley,
Lindenauer,
and Cerny
Amer.
Table I. Type of pelvic neoplasm NO. Cervix
I
25
Vulva
Colon Vagina
34 (4)* 3 1 1
Uterus Urethra
37 ‘Double adenocarcinoma
primary: Squamous of colon.
cell
% 67.6 10.8 8.1 8.1 2.7 2.7 100
carcinoma
of vagina,
Center have been presented to the Gynecology Tumor Conference for final evaluation and outline of treatment. All patients who are potential candidates for pelvic exenteration are considered by this group after thorough physical examination, extensive laboratory investigation, and appropriate consultation with representatives from the allied services. During the five-year period of study, over 1,300 patients were seen at this biweekly conference. Approximately 10 per cent of the patients were evaluated for recurrent or persistent malignant neoplasm, and about 33 per cent of the patients with recurrence were considered candidates for pelvic exenteration. The final selection depended on the patient’s age, the type and extent of the tumor, the patient’s general physical status, and finally her willingness to undergo the procedure after she had been acquainted with the details of the operation. In our experience, only 2 patients have refused the operation. Of the patients who were explored, 66 per cent proved to be candidates for some form of pelvic exenteration. Sixty-two per cent (23/37) of the patients in the operative series were treated with total pelvic exenteration (Table I). Eight patients (22 per cent) were treated with posterior pelvic exenteration and 6 (16 per cent) were treated with anterior pelvic exenteration when the lesion was sufficiently localized to make total pelvic exenteration appear to be too radical. The ages of the patients in this series ranged from 33 to 70, with a median age of 53. In more than two thirds of the cases
April 15, 1971 J. Obstet. Gym.
the primary diagnosis was recurrent carcinoma of the cervix (Table I). In the order of decreasing frequency, the remaining cases involved the vulva, colon, vagina, uterus, and urethra. In one instance there were two primary lesions-squamous cell carcinoma of the vagina and adenocarcinoma of the colon. Operative method In all cases, the preoperative treatment followed a standard regimen that includes all related laboratory and x-ray examinations, appropriate consultations, bowel preparation, and prophylactic antibiotic therapy. Although routine antibiotic prophylaxis may be questionable in some types of surgery, it is generally advisable for procedures where contamination is likely, as with the gastrointestinal and genitourinary tracts.* Just before the operation, a reliable central venous pressure is established so that it may be monitored during and after the procedure. In the operating room, the patient is placed on the table in the “ski” position gives simultaneous access PigI), which to the abdominal and perineal areas, not only for palpation of the lesion but also for the operative procedure itself. The abdominal cavity is entered through a transverse incision with transection of the rectus muscles bilaterally. The upper part of the abdomen is explored thoroughly for evidence of metastatic disease, for if neoplasm is identified there-for example, involvement of the liver, omentum, extrapelvic peritoneum, or periaortic lymph node areas-the procedure is terminated and the abdomen closed. The pelvic contents are evaluated similarly, and should there be any question concerning the resectability of the lesion, a frozen section is prepared for immediate analysis. It is felt strongly that this operation with its subsequent prolonged rehabilitation should not be undertaken as a palliative procedure. Should the frozen section examination reveal malignant disease beyond the plane of dissection, then the lesion is considered unresectable and the procedure is
Volume Number
109 8
Pelvic exenterative
Fig.
1. “Ski
position”
for
pelvic
exenteration
terminated without having made any irreversible commitments. A peritoneal sling or flap to be used in the pelvic floor closure, if the radical resection is to be undertaken, is developed on entrance into the abdominal cavity (Fig. 2). A transverse incision into the peritoneum is made approximately halfway between the transabdominal incision and the level of the umbilicus. The flap is further developed by a linear incision of the peritoneum along the left lateral abdominal wall. The lower transverse incision of the peritoneal flap is made after the upper part of the bladder has been reflected off the peritoneum and this incision is extended across to the right lateral abdominal wall where it remains attached in continuity or “hinged” for later use. The peritoneal flap measures approximately 8 by 20 cm. An en bloc dissection with removal of the female reproductive organs, the vagina, the distal third of the ureters, the bladder anteriorly, and the bowel posteriorly, when indicated, is then performed. The hypogastric arteries are ligated but not transected. The labial tissues are not removed unless indicated by the extent or location of disease. This allows for a better cosmetic result
illustrating
draping
and
therapy
1177
incisions.
especially if one constructs a neovagina at a later date. Once the en bloc dissection has been completed and a bilateral pelvic lymph node dissection has been performed, the inferior vena cava is exposed on three sides at the level of the L, to L, vertebrae. By means of a Codman stapler the inferior vena cava is divided into four compartments, as a prophylactic measure against major pulmonary embolization. Some type of urinary diversion must be established if the bladder has been sacrificed. Currently, a ureterosigmoid conduit is considered the urinary diversion of choice if a total pelvic exenteration has been performed. Both ureters are mobilized and a one-layer mucosa-to-mucosa anastomosis to the antimesenteric border of a 10 to 12 cm. isolated segment of distal sigmoid colon is performed. The proximal end of this isolated segment is closed and the distal end is brought out as an ostium through an incision placed midway between the umbilicus and the right anterior superior iliac spine. A sigmoid colostomy is then prepared in the usual manner with the ostium similarly placed in the left lower quadrant. Before the abdominal wound is closed,
1178
Morley,
Fig. 2. aperture.
Lindenauer,
Peritoneal
and
sling
Cerny
prepared
Amer.
from
anterior
abdominal
wall.
Used
to cover
the
April 15, 1971 J. Obstet. Gyms.
pelvic
the previously prepared peritoneal flap is attached peripherally at the level of the excised levator muscles of the pelvic diaphragm with interrupted silk suture (Fig. 3). A Robinson catheter is placed in the most dependent position for drainage of the supraperitoneal flap area. The perineal vault lying inferior to the peritoneal sling is packed in a routine manner. Results In our series of 37 patients, the average time required for the operation was 5 hours and 45 minutes (Table II), and the median estimated blood loss per patient was 2,800 ml. The median length of hospital stay was 30 days, but it should be noted that with the gradual decrease in incidence of postoperative
Fig. 3. Perineal view operative day. Note area) of the peritoneal
taken on the fifteenth the under surface sling.
post(light
complications
most
patients
are
now discharged from the hospital within 4 weeks. Although meaningful survival statistics are not yet available for this small series, the short-term results are encouraging. Of the 37 patients, 14 were operated upon 3 or more years ago, and 7 of these patients
Volume Number
Pelvic exenterative
109 8
Table II. Operative
( Medcnn
Age of patient Hospital stay Operating time Estimated blood loss
Table III.
statistics
53 years 30 days 5”45’
1
Range 33-70 years 20-170 days 4”15’-8”45’
2,800 C.C. 1,400-9,700 C.C.
have survived. The over-all (3 year) survival rate of 50 per cent suggests significant improvement over the pre-1965 period, when the 5 year survival rate was only 23 per cent. Only one of the 37 patients died during the first 30 days after the operation. In this case, uncontrollable hemorrhage resulted in renal failure and death on the twentysecond day. The operative mortality in the series reported is 2.7 per cent. Since evidence of metastatic disease in the regional lymph nodes affords a very guarded prognosis, the results in such cases are of particular importance. In our group of 21 patients treated 2 or more years ago, 6 showed lymph node involvement and all of these patients have died, whereas 12 of the 15 with no lymph node involvement are alive and clinically free of disease. Of the three deaths among these 15 patients, one occurred in the early postoperative period, as noted above; one of the other 2 patients died of advancing carcinoma of the colon, and one died of extensive adenocarcinoma of the cervix. In the last instance the patient probably would not have been selected for operation under today’s criteria. Significant complications developed in about half of the 37 cases in this series (Table III) ; however, specific treatment for these was successful in most instances. In 12 cases, further operations were necessary to correct fistulas of the bowel or bladder, to close a wound dehiscence, to improve urinary diversion, or to determine whether the neoplastic disease had recurred. A third operation was required for such reasons in 5 cases, and in one case the patient was operated on five times. There were two small bowel fistulas which occurred in the first half of the series, before the peritoneal sling was
Postoperative
1179
complications
Septicemia Bowel fistula Bladder fistula Small bowel prolapse (vaginal dehiscence ) Intestinal obstruction* Pneumonitis Abdominal wound dehiscence Pulmonary embolization Others *Treated medically.
therapy
i 2 5
repositioned. There also were two large bowel fistulas in patients who had been treated with anterior pelvic exenteration. Four of the 8 patients treated with posterior pelvic exenteration developed bladder fistulas. In 4 cases, one or more loops of small bowel prolapsed into or through the perineal vault, leading in essence to a vaginal dehiscence; these cases too occurred before the repositioning of the peritoneal sling. An abdominal wound dehiscence occurred in 2 cases; both cases were complicated by previous radiation therapy, and one by diabetes and obesity. Postoperative septicemia developed in 8 cases, usually along with acute urinary tract infection. All responded well to specific therapy. Inappropriate water diuresis was noted in one case, probably as a result of methoxyfluorane anesthesia. General considerations With respect to the overall problems of pelvic exenterative operation, it is clear that recent developments, both philosophical and procedural, have led to important changes in the traditional approach. No longer is the conference room the “court of last resort,” nor is recurrent or persistent disease of only prognostic significance. On occasion, the so-called “frozen pelvis” is misinterpreted as evidence of recurrent neoplasm, when in fact-as proved at the time of operation-this change is in part only radiation fibrosis which is indistinguishable from recurrent neoplasm on pelvic examination. Moreover, many lesions formerly
1180
Morley,
Lindenauer,
and Cerny
regarded as “inoperable” can now be resected with safety and considerable benefit for the patient. In realizing that this procedure crosses many interdisciplinary lines and since it is uncommon today for one surgeon to be trained in all these areas unless he accepts further specialized training, the multidisciplinary approach to ultraradical operation has gained acceptance in many centers throughout the country.5 In our institution the multidisciplinary approach to pelvic exenteration which includes a gynecologist, general surgeon, and urologist, has been supported over the past 5 years, since it gives the patient specialist consultation from all three specialties involved and it provides the patient with improved techniques specific to each specialty field. The responsibility for the conduct of the operative procedure is alternated between the gynecologist and the general surgeon since they represent the major disciplines involved. These two members of the team are present throughout the procedure and alternate as first assistant. Frequently during the procedure both members of the team work simultaneously in operating transvaginally and transabdominally as well as in preparation of the abdominal stoma. Pathologists emphasize that lesions of the squamous cell variety tend to grow along the planes of least resistance and in continuity into surrounding tissues for some time before they either involve adjacent vital structures or metastasize distantly. The prime indication for pelvic exenteration, therefore, is recurrent or persistent carcinoma of the cervix both because of its overall incidence and its growth pattern. Essentially all patients with previously untreated carcinoma of the cervix, however, are not considered candidates for pelvic exenteration, irrespective of stage of disease, but are managed by more conventional means at the present time. A local but precariously positioned carcinoma of the vulva responds well to either anterior or posterior pelvic exenteration depending upon its location. An extensive
Amer.
April 15, 1971 J. Obstet. Cynec.
lesion of the vulva requiring a total pelvic exenteration is considered far too advanced to warrant ultraradical surgery as a curative procedure. Other lesions that may be considered for partial or total pelvic exenteration are carcinoma of the colon, vagina, endometrium, and urethra. Carcinoma of the ovary, on the other hand, when sufficiently extensive to be considered for this type of operation does not respond well and is therefore contraindicated. Most of the patients who were not considered candidates for pelvic exenteration by members of the Tumor Conference either had evidence of metastasis outside of the pelvis or the local process was extensive enough to cause pain because of sciatic nerve involvement, swelling of one or both lower extremities, and signs of ureteral obstruction. However, ureteral obstruction alone is not considered an absolute contraindication. The patient’s physical, mental, and emotional status should be thoroughly evaluated before a final decision is made to undertake the procedure since significant readjustments must obviously be faced postoperatively. Whereas age itself is not a reason to withhold operative therapy, it should be obvious that with advanced age the presence of other medical ailments often make this procedure unwise. With respect to problems of the surgical technique as such, the method of urinary diversion is especially important. Currently, we favor the ureterosigmoid conduit as the method of choice for urinary diversion in total pelvic exenteration. Caution must be taken in isolating the sigmoid segment since its blood supply has already been compromised by previous radiation therapy. The ureterosigmoid conduit for urinary diversion was utilized in 82 per cent (19/23) of the patients who were treated with total pelvic exenteration and to date we have been pleased with the results. Acute urinary tract infection has been a postoperative comthese infections have plication; however, appropriate responded satisfactorily to chemotherapy. Should the preparation of a ureterosig-
Volume 109 Number 8
moid conduit be considered inadvisable because of inaccessibility of tissue, compromise of blood supply, or type of exenteration performed, a ureteroileal loop then becomes the method of choice. The main disadvantage is the requirement for a small bowel anastomosis to restore intestinal continuity. This can be avoided when a segment of sigmoid colon is used. Other methods of urinary diversion such as cutaneous ureterostomy and ureterosigmoidostomy have been utilized on occasion-especially early in the series; however, the advantages of rapidity and ease of performance are far outweighed by the frequent occurrence of ureteral stricture, prolonged intubation, ascending pyelonephritis and hyperchloremic-hypokalemic acidosis. These forms of urinary diversion have fallen into disfavor among most urologists. As a prophylactic measure against major pulmonary embolization which occurred in 2 of the first 4 patients in the series, the inferior vena cava is compartmentalized by staples. This then divides the inferior vena cava into four compartments of decreased caliber. Postoperative radiographic studies have confirmed the fact that the return venous flow is not impeded by this technique. No major pulmonary embolus has occurred since this technique was adopted. Because of the significant physiologic and pathologic alterations that accompany this procedure, one must utilize many supportive techniques in the appropriate care of these patients. The routine use of central venous pressure monitoring has been most beneficial in assessing adequacy of blood and fluid replacement in the operating room as well as in the early postoperative period. To keep body temperature alteration at a minimum as recorded by an esophageal temperature monitor, we have utilized a warming blanket underneath the patient on the operating table and a blood warmer to raise the temperature of the recently refrigerated blood prior to transfusion. It must be emphasized that the body temperature can be lowered significantly by blood loss and by the protracted exposure of large areas of skin, viscera, and the denuded pelvis.
Pelvic exenterative
therapy
1181
During the early part of this study we were somewhat discouraged by the high incidence of postoperative complications such as intestinal obstruction, intestinal prolapse into the pelvic vault, and the development of bowel fistulas. In the beginning we made no attempt to peritonealize the exposed or denuded area, but stimulated a fibrinous peritonitis by placing a rubber dam high in the pelvic vault with appropriate packing material used as a buttress. Because of the high incidence of postoperative complications accompanying this technique, we abandoned it in preference to a modification of the peritoneal sling or flap, as described by Mattingly. Various techniques described by others have also been considered.?? s At first we considered it important to attach the peritoneal sling to the peritoneum both posteriorly and anteriorly, so that the peritoneum itself would be in continuity, and the intra-abdominal contents would be confined within an intact peritoneal cavity. Because of difficulties encountered, we changed the technique so that we now attach the peritoneal flap at the level of the excised levator muscles of the pelvic diaphragm. The peritoneal sling or hammock divides this large denuded cavity into two compartments acting simultaneously as a floor for the intra-abdominal contents and a roof for the perineal vault which ultimately becomes a neovagina, in those cases where plastic reconstruction is considered. The incidence of the above-mentioned complications has been favorably altered by lowering the peritoneal sling to the level of the pelvic aperture. The perineal vault is packed and repacked frequently throughout the postoperative period. Such packing not only aids in hemostasis initially but later it stimulates the production of granulation tissue which becomes an excellent bed for the development of a neovagina. Some time later a vaginal reconstruction utilizing a split-thickness skin graft is considered for those patients whose sexual rehabilitation is a matter of concern. The postoperative rehabilitation was significantly im-
1182
Morley,
Lindenauer,
and
April 15,1971 Amer. J. Obstet. Gym.
Cerny
proved for these individuals. Certainly not all patients subjected to exenterative operation are candidates for vaginoplasty, for reasons of age, marital status, and personal inclination. Finally, the surgeon who embarks on a program of pelvic exenterative therapy must be ready to accept increased responsibility, not only for the significant operative risks,
the critical and extensive problems of post operative care but also for the long-term rehabilitation of the patient. On the other hand, the rewards are gratifying-pelvic exenteration does afford the patient a second chance at survival and those who have been cured are able to lead active and rewarding lives.
REFERENCES
1. 2.
Brunschwig, A.: Cancer 1: 177, 1948. Ingersoll, F. N., and Ulfelder, H.: New Eng. J. Med. 274: 648, 1966. Brunschwig, A.: J. A. M. A. 194: 160, 1965. Bernard, H. R., and Cole, W. R.: Surgery 56: 151, 1964. Krieger, James S., and Embree, H. Kenneth: Cleveland Clin. Quart. 36: 1, 1969.
3. 4. 5.
Discussion DR.
ROBERT
Memphis, Tennespelvic operations include a variety R.
HUGIIES,
see. Radical of operative procedures, most of which are formidable. Pelvic exenteration is the most radical of these and has the greatest complication and mortality rate and the poorest cure rate. For these reasons, it is utilized only when more conservative treatment has failed or cannot be used. The most common indication for it is recurrent carcinoma of the cervix that extends beyond the cervix, yet is confined to the pelvis.
Often
the extent
irradiation
fibrosis
of the disease is hidden and
is only
discovered
by
during
dissection of the pelvis. The procedure of choice then rests on the capability and experience of the surgeon. surgeon have
Therefore, it the technical
is essential capability
that the to do as
radical a procedure as is required. Before proceeding, he must have determined that there is no extrapelvic disease by exploration of the upper
iliac
abdomen
node
and
a periaortic
dissection
submitted
and
common
for frozen
sec-
tion. This is essential, as there are no survivors with positive nodes. Reporting the 30 day operative mortality rate does not accurately reflect the relative hazard of exenterative procedures. Hospital or 90 day mortality is more precise. Of our 37 exenteration patients in Tennessee from 1962 to 1970, one died at 30 hours postoperatively when she aspirated the vomitus and another one died on
6. 7. 8.
Mattingly, R. F.: Clin. Obstet. Gynec. 8: 705, 1965. Valle, Ginseppe, and Ferraris, Germano: Obstet. Gvnec. 33: No. 6. 1969. Nelson, J&es H., Jr.: A&as of Radical Pelvic Surgery, New York, 1969, Meredith Corporation, Appleton-Century-Crofts Inc.
Day 28 for a 5.4 per cent 30 day mortality. hospital and 90 day mortality were 10.8 per and 13.5 per cent, respectively, and all operative deaths. The method or methods of achieving proved
results
team
approach
are
and
definitely
enhanced
a thorough
study
The cent were
with
ima
of all
aspects of the patient’s care. Personally, it seems success is related to paying close attention to a myriad of details preoperatively, during the procedure, and postoperatively, all of which demand a tremendous amount of doctor time. Successfully threading one’s way through the countless pitfalls of care requires a maximum effort. This can be achieved by a single surgical discipline, whether Gynecology or General Surgery, if a team approach is used, and adequate time is given to the countless details. Dr. Ruta gynecologist, and Dr. Bricker,z a genledge,l eral surgeon, have demonstrated that. It is m*h more important to point out that the casual surgeon or the occasional operator has no place in the execution of radical pelvic surgery. Over a 5 year period from 1960 to 1965, at three
hospitals in Memphis, 17 exenterations were done by ten different surgeons with 11 hospital deaths. Six of these procedures were done by a multidiscipline approach with 6 operative deaths. During that same period 35 radical hysterec-
tomies were done as primary treatment for carcinoma of the cervix by 13 surgeons with 5 deaths.
JOSEPH Ann
W.
MARTIN
Arbor,
MORLEY,
M.D.
LINDENAUER, C.
CERNY,
M.D. M.D.
Michigan
During the 5 year period from June 1,1965, to June 1, 1970, 37 patients were treated with some type of pelvic exenteration at the University of Michigan Medical Center. Total pelvic exenteration was performed in over 60 per cent of the cases. Of 14 patients operated upon 3 or more years ago, 7 are alive and clinically free of disease. The operative mortality in the cases reported is 2.7 per cent. Of 21 patients treated 2 or more years ago, 6 had regional lymph node involvement and all have died of disease. Twelve of the I5 patients with no nodal involvement are alive and clinically free of disease. The importance of various technical points to include “ski positioning” of the patient on the operating table, development of a peritoneal sling to be used as a pelvic cover, preparation of a ureterosigmoid conduit for urinary diversion, and profihylactic compartmentalization of the inferior vena cava are described in detail. Vaginal reconstruction when desired is performed at a later date.
In addition, a review of 5 year survival statistics suggested that the continued utilization of pelvic exenterative therapy for the treatment of recurrent carcinoma of the cervix and other pelvic malignancies was justified.2t 3 During the 5 year period from June 1, 1965, to June 1, 1970, 37 patients have been treated with some type of pelvic exenteration under our direction at the University of Michigan Medical Center, and these cases are the basis for this report.
S H 0 R T L Y A F T E R Brunschwigl reported his initial evaluation of radical operative treatment of advanced pelvic carcinoma, a program incorporating this form of therapy was established at the University of Michigan Medical Center. During the 12 year period from 1953 to 1965, 30 patients were treated with some type of pelvic exenteration at this institution with over thirty different physicians participating in the procedure. Because of the high operative morbidity and mortality during this period, a multidisciplinary and cooperative approach to this form of ultraradical surgery was instituted in an attempt to improve the postoperative morbidity and operative mortality. From the Department Gynecology and the Surgery, Section of Section of Urology, Michigan Medical
Selection
of cases
Although local excision or local irradiation of a small lesion occasionally has a place in the treatment of local recurrence, the procedure of choice usually involves a more radical operation such as radical hysterectomy and pelvic lymph node dissection, or partial or total pelvic exenteration. For more than 35 years, all cases of gynecologic malignancy at the University of Michigan Medical
of Obstetrics and Department of General Surgery and University of Center.
Presented at the Thirty-eighth Annual Meeting of the Central Association of Obstetricians and Gynecologists, Chicago, Illinois, Sept. 24-26, 1970.
1175
1176
Morley,
Lindenauer,
and Cerny
Amer.
Table I. Type of pelvic neoplasm NO. Cervix
I
25
Vulva
Colon Vagina
34 (4)* 3 1 1
Uterus Urethra
37 ‘Double adenocarcinoma
primary: Squamous of colon.
cell
% 67.6 10.8 8.1 8.1 2.7 2.7 100
carcinoma
of vagina,
Center have been presented to the Gynecology Tumor Conference for final evaluation and outline of treatment. All patients who are potential candidates for pelvic exenteration are considered by this group after thorough physical examination, extensive laboratory investigation, and appropriate consultation with representatives from the allied services. During the five-year period of study, over 1,300 patients were seen at this biweekly conference. Approximately 10 per cent of the patients were evaluated for recurrent or persistent malignant neoplasm, and about 33 per cent of the patients with recurrence were considered candidates for pelvic exenteration. The final selection depended on the patient’s age, the type and extent of the tumor, the patient’s general physical status, and finally her willingness to undergo the procedure after she had been acquainted with the details of the operation. In our experience, only 2 patients have refused the operation. Of the patients who were explored, 66 per cent proved to be candidates for some form of pelvic exenteration. Sixty-two per cent (23/37) of the patients in the operative series were treated with total pelvic exenteration (Table I). Eight patients (22 per cent) were treated with posterior pelvic exenteration and 6 (16 per cent) were treated with anterior pelvic exenteration when the lesion was sufficiently localized to make total pelvic exenteration appear to be too radical. The ages of the patients in this series ranged from 33 to 70, with a median age of 53. In more than two thirds of the cases
April 15, 1971 J. Obstet. Gym.
the primary diagnosis was recurrent carcinoma of the cervix (Table I). In the order of decreasing frequency, the remaining cases involved the vulva, colon, vagina, uterus, and urethra. In one instance there were two primary lesions-squamous cell carcinoma of the vagina and adenocarcinoma of the colon. Operative method In all cases, the preoperative treatment followed a standard regimen that includes all related laboratory and x-ray examinations, appropriate consultations, bowel preparation, and prophylactic antibiotic therapy. Although routine antibiotic prophylaxis may be questionable in some types of surgery, it is generally advisable for procedures where contamination is likely, as with the gastrointestinal and genitourinary tracts.* Just before the operation, a reliable central venous pressure is established so that it may be monitored during and after the procedure. In the operating room, the patient is placed on the table in the “ski” position gives simultaneous access PigI), which to the abdominal and perineal areas, not only for palpation of the lesion but also for the operative procedure itself. The abdominal cavity is entered through a transverse incision with transection of the rectus muscles bilaterally. The upper part of the abdomen is explored thoroughly for evidence of metastatic disease, for if neoplasm is identified there-for example, involvement of the liver, omentum, extrapelvic peritoneum, or periaortic lymph node areas-the procedure is terminated and the abdomen closed. The pelvic contents are evaluated similarly, and should there be any question concerning the resectability of the lesion, a frozen section is prepared for immediate analysis. It is felt strongly that this operation with its subsequent prolonged rehabilitation should not be undertaken as a palliative procedure. Should the frozen section examination reveal malignant disease beyond the plane of dissection, then the lesion is considered unresectable and the procedure is
Volume Number
109 8
Pelvic exenterative
Fig.
1. “Ski
position”
for
pelvic
exenteration
terminated without having made any irreversible commitments. A peritoneal sling or flap to be used in the pelvic floor closure, if the radical resection is to be undertaken, is developed on entrance into the abdominal cavity (Fig. 2). A transverse incision into the peritoneum is made approximately halfway between the transabdominal incision and the level of the umbilicus. The flap is further developed by a linear incision of the peritoneum along the left lateral abdominal wall. The lower transverse incision of the peritoneal flap is made after the upper part of the bladder has been reflected off the peritoneum and this incision is extended across to the right lateral abdominal wall where it remains attached in continuity or “hinged” for later use. The peritoneal flap measures approximately 8 by 20 cm. An en bloc dissection with removal of the female reproductive organs, the vagina, the distal third of the ureters, the bladder anteriorly, and the bowel posteriorly, when indicated, is then performed. The hypogastric arteries are ligated but not transected. The labial tissues are not removed unless indicated by the extent or location of disease. This allows for a better cosmetic result
illustrating
draping
and
therapy
1177
incisions.
especially if one constructs a neovagina at a later date. Once the en bloc dissection has been completed and a bilateral pelvic lymph node dissection has been performed, the inferior vena cava is exposed on three sides at the level of the L, to L, vertebrae. By means of a Codman stapler the inferior vena cava is divided into four compartments, as a prophylactic measure against major pulmonary embolization. Some type of urinary diversion must be established if the bladder has been sacrificed. Currently, a ureterosigmoid conduit is considered the urinary diversion of choice if a total pelvic exenteration has been performed. Both ureters are mobilized and a one-layer mucosa-to-mucosa anastomosis to the antimesenteric border of a 10 to 12 cm. isolated segment of distal sigmoid colon is performed. The proximal end of this isolated segment is closed and the distal end is brought out as an ostium through an incision placed midway between the umbilicus and the right anterior superior iliac spine. A sigmoid colostomy is then prepared in the usual manner with the ostium similarly placed in the left lower quadrant. Before the abdominal wound is closed,
1178
Morley,
Fig. 2. aperture.
Lindenauer,
Peritoneal
and
sling
Cerny
prepared
Amer.
from
anterior
abdominal
wall.
Used
to cover
the
April 15, 1971 J. Obstet. Gyms.
pelvic
the previously prepared peritoneal flap is attached peripherally at the level of the excised levator muscles of the pelvic diaphragm with interrupted silk suture (Fig. 3). A Robinson catheter is placed in the most dependent position for drainage of the supraperitoneal flap area. The perineal vault lying inferior to the peritoneal sling is packed in a routine manner. Results In our series of 37 patients, the average time required for the operation was 5 hours and 45 minutes (Table II), and the median estimated blood loss per patient was 2,800 ml. The median length of hospital stay was 30 days, but it should be noted that with the gradual decrease in incidence of postoperative
Fig. 3. Perineal view operative day. Note area) of the peritoneal
taken on the fifteenth the under surface sling.
post(light
complications
most
patients
are
now discharged from the hospital within 4 weeks. Although meaningful survival statistics are not yet available for this small series, the short-term results are encouraging. Of the 37 patients, 14 were operated upon 3 or more years ago, and 7 of these patients
Volume Number
Pelvic exenterative
109 8
Table II. Operative
( Medcnn
Age of patient Hospital stay Operating time Estimated blood loss
Table III.
statistics
53 years 30 days 5”45’
1
Range 33-70 years 20-170 days 4”15’-8”45’
2,800 C.C. 1,400-9,700 C.C.
have survived. The over-all (3 year) survival rate of 50 per cent suggests significant improvement over the pre-1965 period, when the 5 year survival rate was only 23 per cent. Only one of the 37 patients died during the first 30 days after the operation. In this case, uncontrollable hemorrhage resulted in renal failure and death on the twentysecond day. The operative mortality in the series reported is 2.7 per cent. Since evidence of metastatic disease in the regional lymph nodes affords a very guarded prognosis, the results in such cases are of particular importance. In our group of 21 patients treated 2 or more years ago, 6 showed lymph node involvement and all of these patients have died, whereas 12 of the 15 with no lymph node involvement are alive and clinically free of disease. Of the three deaths among these 15 patients, one occurred in the early postoperative period, as noted above; one of the other 2 patients died of advancing carcinoma of the colon, and one died of extensive adenocarcinoma of the cervix. In the last instance the patient probably would not have been selected for operation under today’s criteria. Significant complications developed in about half of the 37 cases in this series (Table III) ; however, specific treatment for these was successful in most instances. In 12 cases, further operations were necessary to correct fistulas of the bowel or bladder, to close a wound dehiscence, to improve urinary diversion, or to determine whether the neoplastic disease had recurred. A third operation was required for such reasons in 5 cases, and in one case the patient was operated on five times. There were two small bowel fistulas which occurred in the first half of the series, before the peritoneal sling was
Postoperative
1179
complications
Septicemia Bowel fistula Bladder fistula Small bowel prolapse (vaginal dehiscence ) Intestinal obstruction* Pneumonitis Abdominal wound dehiscence Pulmonary embolization Others *Treated medically.
therapy
i 2 5
repositioned. There also were two large bowel fistulas in patients who had been treated with anterior pelvic exenteration. Four of the 8 patients treated with posterior pelvic exenteration developed bladder fistulas. In 4 cases, one or more loops of small bowel prolapsed into or through the perineal vault, leading in essence to a vaginal dehiscence; these cases too occurred before the repositioning of the peritoneal sling. An abdominal wound dehiscence occurred in 2 cases; both cases were complicated by previous radiation therapy, and one by diabetes and obesity. Postoperative septicemia developed in 8 cases, usually along with acute urinary tract infection. All responded well to specific therapy. Inappropriate water diuresis was noted in one case, probably as a result of methoxyfluorane anesthesia. General considerations With respect to the overall problems of pelvic exenterative operation, it is clear that recent developments, both philosophical and procedural, have led to important changes in the traditional approach. No longer is the conference room the “court of last resort,” nor is recurrent or persistent disease of only prognostic significance. On occasion, the so-called “frozen pelvis” is misinterpreted as evidence of recurrent neoplasm, when in fact-as proved at the time of operation-this change is in part only radiation fibrosis which is indistinguishable from recurrent neoplasm on pelvic examination. Moreover, many lesions formerly
1180
Morley,
Lindenauer,
and Cerny
regarded as “inoperable” can now be resected with safety and considerable benefit for the patient. In realizing that this procedure crosses many interdisciplinary lines and since it is uncommon today for one surgeon to be trained in all these areas unless he accepts further specialized training, the multidisciplinary approach to ultraradical operation has gained acceptance in many centers throughout the country.5 In our institution the multidisciplinary approach to pelvic exenteration which includes a gynecologist, general surgeon, and urologist, has been supported over the past 5 years, since it gives the patient specialist consultation from all three specialties involved and it provides the patient with improved techniques specific to each specialty field. The responsibility for the conduct of the operative procedure is alternated between the gynecologist and the general surgeon since they represent the major disciplines involved. These two members of the team are present throughout the procedure and alternate as first assistant. Frequently during the procedure both members of the team work simultaneously in operating transvaginally and transabdominally as well as in preparation of the abdominal stoma. Pathologists emphasize that lesions of the squamous cell variety tend to grow along the planes of least resistance and in continuity into surrounding tissues for some time before they either involve adjacent vital structures or metastasize distantly. The prime indication for pelvic exenteration, therefore, is recurrent or persistent carcinoma of the cervix both because of its overall incidence and its growth pattern. Essentially all patients with previously untreated carcinoma of the cervix, however, are not considered candidates for pelvic exenteration, irrespective of stage of disease, but are managed by more conventional means at the present time. A local but precariously positioned carcinoma of the vulva responds well to either anterior or posterior pelvic exenteration depending upon its location. An extensive
Amer.
April 15, 1971 J. Obstet. Cynec.
lesion of the vulva requiring a total pelvic exenteration is considered far too advanced to warrant ultraradical surgery as a curative procedure. Other lesions that may be considered for partial or total pelvic exenteration are carcinoma of the colon, vagina, endometrium, and urethra. Carcinoma of the ovary, on the other hand, when sufficiently extensive to be considered for this type of operation does not respond well and is therefore contraindicated. Most of the patients who were not considered candidates for pelvic exenteration by members of the Tumor Conference either had evidence of metastasis outside of the pelvis or the local process was extensive enough to cause pain because of sciatic nerve involvement, swelling of one or both lower extremities, and signs of ureteral obstruction. However, ureteral obstruction alone is not considered an absolute contraindication. The patient’s physical, mental, and emotional status should be thoroughly evaluated before a final decision is made to undertake the procedure since significant readjustments must obviously be faced postoperatively. Whereas age itself is not a reason to withhold operative therapy, it should be obvious that with advanced age the presence of other medical ailments often make this procedure unwise. With respect to problems of the surgical technique as such, the method of urinary diversion is especially important. Currently, we favor the ureterosigmoid conduit as the method of choice for urinary diversion in total pelvic exenteration. Caution must be taken in isolating the sigmoid segment since its blood supply has already been compromised by previous radiation therapy. The ureterosigmoid conduit for urinary diversion was utilized in 82 per cent (19/23) of the patients who were treated with total pelvic exenteration and to date we have been pleased with the results. Acute urinary tract infection has been a postoperative comthese infections have plication; however, appropriate responded satisfactorily to chemotherapy. Should the preparation of a ureterosig-
Volume 109 Number 8
moid conduit be considered inadvisable because of inaccessibility of tissue, compromise of blood supply, or type of exenteration performed, a ureteroileal loop then becomes the method of choice. The main disadvantage is the requirement for a small bowel anastomosis to restore intestinal continuity. This can be avoided when a segment of sigmoid colon is used. Other methods of urinary diversion such as cutaneous ureterostomy and ureterosigmoidostomy have been utilized on occasion-especially early in the series; however, the advantages of rapidity and ease of performance are far outweighed by the frequent occurrence of ureteral stricture, prolonged intubation, ascending pyelonephritis and hyperchloremic-hypokalemic acidosis. These forms of urinary diversion have fallen into disfavor among most urologists. As a prophylactic measure against major pulmonary embolization which occurred in 2 of the first 4 patients in the series, the inferior vena cava is compartmentalized by staples. This then divides the inferior vena cava into four compartments of decreased caliber. Postoperative radiographic studies have confirmed the fact that the return venous flow is not impeded by this technique. No major pulmonary embolus has occurred since this technique was adopted. Because of the significant physiologic and pathologic alterations that accompany this procedure, one must utilize many supportive techniques in the appropriate care of these patients. The routine use of central venous pressure monitoring has been most beneficial in assessing adequacy of blood and fluid replacement in the operating room as well as in the early postoperative period. To keep body temperature alteration at a minimum as recorded by an esophageal temperature monitor, we have utilized a warming blanket underneath the patient on the operating table and a blood warmer to raise the temperature of the recently refrigerated blood prior to transfusion. It must be emphasized that the body temperature can be lowered significantly by blood loss and by the protracted exposure of large areas of skin, viscera, and the denuded pelvis.
Pelvic exenterative
therapy
1181
During the early part of this study we were somewhat discouraged by the high incidence of postoperative complications such as intestinal obstruction, intestinal prolapse into the pelvic vault, and the development of bowel fistulas. In the beginning we made no attempt to peritonealize the exposed or denuded area, but stimulated a fibrinous peritonitis by placing a rubber dam high in the pelvic vault with appropriate packing material used as a buttress. Because of the high incidence of postoperative complications accompanying this technique, we abandoned it in preference to a modification of the peritoneal sling or flap, as described by Mattingly. Various techniques described by others have also been considered.?? s At first we considered it important to attach the peritoneal sling to the peritoneum both posteriorly and anteriorly, so that the peritoneum itself would be in continuity, and the intra-abdominal contents would be confined within an intact peritoneal cavity. Because of difficulties encountered, we changed the technique so that we now attach the peritoneal flap at the level of the excised levator muscles of the pelvic diaphragm. The peritoneal sling or hammock divides this large denuded cavity into two compartments acting simultaneously as a floor for the intra-abdominal contents and a roof for the perineal vault which ultimately becomes a neovagina, in those cases where plastic reconstruction is considered. The incidence of the above-mentioned complications has been favorably altered by lowering the peritoneal sling to the level of the pelvic aperture. The perineal vault is packed and repacked frequently throughout the postoperative period. Such packing not only aids in hemostasis initially but later it stimulates the production of granulation tissue which becomes an excellent bed for the development of a neovagina. Some time later a vaginal reconstruction utilizing a split-thickness skin graft is considered for those patients whose sexual rehabilitation is a matter of concern. The postoperative rehabilitation was significantly im-
1182
Morley,
Lindenauer,
and
April 15,1971 Amer. J. Obstet. Gym.
Cerny
proved for these individuals. Certainly not all patients subjected to exenterative operation are candidates for vaginoplasty, for reasons of age, marital status, and personal inclination. Finally, the surgeon who embarks on a program of pelvic exenterative therapy must be ready to accept increased responsibility, not only for the significant operative risks,
the critical and extensive problems of post operative care but also for the long-term rehabilitation of the patient. On the other hand, the rewards are gratifying-pelvic exenteration does afford the patient a second chance at survival and those who have been cured are able to lead active and rewarding lives.
REFERENCES
1. 2.
Brunschwig, A.: Cancer 1: 177, 1948. Ingersoll, F. N., and Ulfelder, H.: New Eng. J. Med. 274: 648, 1966. Brunschwig, A.: J. A. M. A. 194: 160, 1965. Bernard, H. R., and Cole, W. R.: Surgery 56: 151, 1964. Krieger, James S., and Embree, H. Kenneth: Cleveland Clin. Quart. 36: 1, 1969.
3. 4. 5.
Discussion DR.
ROBERT
Memphis, Tennespelvic operations include a variety R.
HUGIIES,
see. Radical of operative procedures, most of which are formidable. Pelvic exenteration is the most radical of these and has the greatest complication and mortality rate and the poorest cure rate. For these reasons, it is utilized only when more conservative treatment has failed or cannot be used. The most common indication for it is recurrent carcinoma of the cervix that extends beyond the cervix, yet is confined to the pelvis.
Often
the extent
irradiation
fibrosis
of the disease is hidden and
is only
discovered
by
during
dissection of the pelvis. The procedure of choice then rests on the capability and experience of the surgeon. surgeon have
Therefore, it the technical
is essential capability
that the to do as
radical a procedure as is required. Before proceeding, he must have determined that there is no extrapelvic disease by exploration of the upper
iliac
abdomen
node
and
a periaortic
dissection
submitted
and
common
for frozen
sec-
tion. This is essential, as there are no survivors with positive nodes. Reporting the 30 day operative mortality rate does not accurately reflect the relative hazard of exenterative procedures. Hospital or 90 day mortality is more precise. Of our 37 exenteration patients in Tennessee from 1962 to 1970, one died at 30 hours postoperatively when she aspirated the vomitus and another one died on
6. 7. 8.
Mattingly, R. F.: Clin. Obstet. Gynec. 8: 705, 1965. Valle, Ginseppe, and Ferraris, Germano: Obstet. Gvnec. 33: No. 6. 1969. Nelson, J&es H., Jr.: A&as of Radical Pelvic Surgery, New York, 1969, Meredith Corporation, Appleton-Century-Crofts Inc.
Day 28 for a 5.4 per cent 30 day mortality. hospital and 90 day mortality were 10.8 per and 13.5 per cent, respectively, and all operative deaths. The method or methods of achieving proved
results
team
approach
are
and
definitely
enhanced
a thorough
study
The cent were
with
ima
of all
aspects of the patient’s care. Personally, it seems success is related to paying close attention to a myriad of details preoperatively, during the procedure, and postoperatively, all of which demand a tremendous amount of doctor time. Successfully threading one’s way through the countless pitfalls of care requires a maximum effort. This can be achieved by a single surgical discipline, whether Gynecology or General Surgery, if a team approach is used, and adequate time is given to the countless details. Dr. Ruta gynecologist, and Dr. Bricker,z a genledge,l eral surgeon, have demonstrated that. It is m*h more important to point out that the casual surgeon or the occasional operator has no place in the execution of radical pelvic surgery. Over a 5 year period from 1960 to 1965, at three
hospitals in Memphis, 17 exenterations were done by ten different surgeons with 11 hospital deaths. Six of these procedures were done by a multidiscipline approach with 6 operative deaths. During that same period 35 radical hysterec-
tomies were done as primary treatment for carcinoma of the cervix by 13 surgeons with 5 deaths.