Gastrostomy tubes in patients with recurrent gynaecological cancer and intestinal obstruction

Transcription

1 British Journal of Obstetrics and Gynaecology September 1999, Vol106, pp Gastrostomy tubes in patients with recurrent gynaecological cancer and intestinal obstruction Efthalia 'hihalina Specialist Registrar, Robert P. Woolas Gynaecological Oncology Fellow, Paul G. Carter Senior Registrar, Felix Chan fisiting Gynaecological Oncology Fellow, Martin E. Gore Consultant Medical Oncologist, Peter M. Blake Consultant Clinical Oncologist, John H. Shepherd Consultant Gynaecological Oncologist, Desmond P. J. Barton Consultant Gynaecological Oncologist The Royal Marsden Hospital, London Objective Women with recurrent gynaecological cancers who are not suitable for exenterative surgery commonly present with gastrointestinal dysfunction. This paper is a retrospective review of the use of gastrostomy tubes in such women. Methods We performed a chart review of women with recurrent gynaecological cancer who had a gastrostomy tube placed between January 1991 and April Results Thirty-nine women (mean age 53.2 years, range 17-82) had a gastrostomy tube placed. liventyeight (72%) had ovarian cancer, eight (21%) had cervical cancer, two had endometrial cancer and one had vaginal cancer. In 14 women a gastrostomy tube was placed as the sole procedure for palliation (1 1 elective, 3 emergency). In the remaining 25 women, who underwent major surgery, a gastrostomy tube was placed in anticipation of, or in the presence of, significant intestinal distension and expected prolonged post-operative ileus. Eleven women (28%) died without leaving hospital after their operation (median 11 days, range 2-36). All but one of the 28 women who left hospital had satisfactory oral intake. Twenty-one women (54%) died with the gastrostomy tube in place (median 28 days, range 2-157) and 18 (46%) had the gastrostomy tube removed (median 14.5 days, range 9-180), 13 of whom (33%) have since died (median 167 days, range 77 days-7 years). Five women (13%) are alive (median 2.2 years, range 10 months45 years). There were no problems which required the gastrostomy tube to be removed. Conclusion Gastrostomy tubes have an important role in the treatment of women with recurrent gynaecological cancer, allowing gastric drainage and decompression without the disadvantages of nasogastric tubes. INTRODUCTION Women with recurrent or progressive gynaecological cancer commonly present with intestinal obstruction characterised by recurrent episodes of nausea, vomiting, abdominal pain, abdominal distension and constipation. Usually subacute or chronic small and large bowel obstruction occurs, rather than acute intestinal obstruction. The initial treatment of the intestinal obstruction involves restriction of oral fluids, rehydration with intravenous fluids and placement of a nasogastric tube. These measures often provide only temporary benefit, and not infrequently the woman returns some weeks later with similar or more severe symptoms. Unlike the intestinal obstruction due to adhesions or to colonic cancer, where there is usually mechanical obstruction at a single site, there are often multiple fac- Correspondence: Dr D. P. J. Barton, Pelvic Cancer Group, The Royal Marsden Hospital, Fulham Road, London SW3 655, UK. tors in the aetiology of the intestinal obstruction in women with recurrent gynaecological cancers, many of which can be present in an individual'. These include prior radiation therapy, multiple sites of partial obstruction-sometimes leading to an obstructed isolated segment of bowel-involvement of the mesentery by cancer, retroperitoneal nodal disease, infiltration of the major retroperitoneal neural plexuses by cancer and, less commonly, benign causes such as adhesions from previous surgery. It is not surprising, therefore, that treatment is often unsatisfactory and that the majority of women will have recurrent bowel symptoms. Although overall 10%-32% of women with cancer (including non-gynaecological cancer) who develop intestinal obstruction have a benign cause for the obstruction, this is rarely found with progressive or recurrent gynaecological cancerz4. It is difficult clinically or radiologically, to determine reliably the underlying causes of the gastrointestinal symptoms and to identify which women may benefit RCOG 1999 British Journal of Obstetrics and Gynaecology

2 G A ST ROS TOM Y, G Y N AEC OLOG IC A L C AN C ER & I NTE ST IN A L OB S T R U CTI ON 965 from major surgical interventi~n.~. There are also women for whom major surgery is not in their best interest, due to the progressive or advanced state of their disease. Sometimes, however, satisfactory palliation can be achieved by a simple surgical procedure. Prolonged nasogastric suction is almost always not acceptable, as this adds to the woman s discomfort and to her overall morbidity from upper respiratory infections, trauma to the nasal septum and aspiration pneumonia. Gastrostomy tubes have been used successfully in patients with non-gynaecological cancer as well as in primary gynaecological cancer. In this report, we describe our experience with the palliative use of gastrostomy tubes in women with recurrent or progressive gynaecological cancer who have acute or chronic intestinal obstruction. METHODS A chart review was carried out of all women with recurrent gynaecological cancer who underwent a palliative gastrostomy. Women who underwent exenterative surgery were excluded. All the women had clinical, radiological and histological evidence of recurrent gynaecological cancer, and a clinical diagnosis of intestinal obstruction supported by plain abdominal X-rays. Initial treatment included restriction of oral fluids, intravenous fluid and electrolyte replacement, nasogastric suction, anti-emetics and analgesics, and sometimes enemas or suppositories. More recently, we have used corticosteroids and octreotide, although in women with gynaecological cancer the efficacy of these drugs is not known. All the women had a general anaesthetic. Where the surgical procedure was placement of a gastrostomy tube only, this was accomplished within 45 minutes. The procedure used during this study was similar to that originally described by Stamm*. Essentially, two concentric purse-string sutures of 2/0 silk were placed in the anterior stomach wall and a small gastrotomy made. A gauge Foley catheter (Bard Ltd, Crawley, West Sussex, UK) was used as the gastrostomy tube and was brought through the upper left quadrant of the abdomen. A gastropexy was performed using the same pursestring sutures and the tube secured to the skin. The catheter was flushed with normal saline to ensure patency and free drainage. On two occasions, because of upper abdominal disease when it was difficult to distinguish the stomach from the colon, methylene blue dye solution was passed through the nasogastric tube by the anaesthetist and an 18-gauge needle used for aspiration-the appearance of the methylene blue dye confirmed the position of the stomach. The women were routinely given ranitidine, 50 mg three times daily intravenously for five days or until they were taking food orally. The nasogastric tube was kept on free drainage in the immediate post-operative period with the gastrostomy tube also on free drainage. The nasogastric tube was then clamped and removed within 12 hours if the gastrostomy tube was draining adequately. The gastrostomy tube was flushed with 50 ml of normal saline every 8 hours during the first 24 hours after surgery. When the gastrostomy tube was temporary, it was not removed within 10 days of placement, by which time the gastropexy should be secure and a track formed, thereby reducing the chance of intraperitoneal spillage when the tube was removed. Women who left hospital with the tube in place were taught to flush and plug the tube, and to attach and remove drainage bags. To reduce skin irritation at the gastrostomy site the women were given liquid antacid solution to apply topically. In the women with a gastrostomy tube in place for prolonged use, the tube was removed and replaced in the clinic after approximately 12 weeks, without anaesthesia, along the track formed by the original tube. RESULTS Thirty-nine women with recurrent or progressive gynaecological cancer and intestinal obstruction who had a gastrostomy tube placed were identified. Twenty-eight women (72%) had ovarian cancer, eight (2 1 %) cervical cancer, two endometrial cancer and one vaginal cancer. All the women had undergone prior surgery for their cancer. Four women also had a fistula-two enterovaginal fistulae, one rectovaginal fistula and one enterocutaneous fistula. One woman was receiving radiation therapy to the para-aortic nodes when she presented with acute intestinal obstruction. The 39 women were divided clinically into two main categories: 1. Women with inoperable disease and recurrent symptoms of nausea and vomiting, in whom palliation with a gastrostomy tube was considered the only possible surgical treatment (n = 11). Three other women had a major operation planned but a gastrostomy only was performed. 2. Women in whom a major surgical procedure was undertaken for relief of intestinal obstruction (n = 25). The surgery undertaken was elective (n = 22; 56%) or emergency (n = 17; 44%). Of the 17 emergency cases, 16 underwent major surgical procedures. Of the 22 elective cases, 13 had a gastrostomy only performed and 9 underwent major surgery. The procedures are shown in Table 1. There were no operative problems with the placement of the gastrostomy tubes, although in two cases methylene blue dye was used to locate the stomach. In five women there was mild skin irritation around 0 RCOG 1999 Br J Obstet Gynaecol 106,

3 966 E. TSAHALINA ET AL. Table 1. Surgical procedures performed for intestinal obstruction in women with recurrent gynaecological cancer. Procedure Gastrostomy only 14 Gastrostomy plus defunctioning stoma 4 Gastrostomy plus bowel resection and anastomosis * stoma 15 Gastrostomy plus bowel resection plus stoma 6 the gastrostomy tube which was treated topically with antacid solution. Two women developed repeated vomiting, one at four days and one at two months after surgery as a result of migration of gastrostomy tube with obstruction of the pylorus by the catheter balloon. This was easily treated by decompressing the balloon of the Foley catheter, withdrawing the gastrostomy tube a few centimetres, re-inflating the balloon and securing the catheter to the skin. No gastrostomy tube was removed because of problems with the tube. In two women the gastrostomy tube was dislodged and was easily replaced. No woman had clinical evidence of intraperitoneal leakage either while the tube was in place or following its removal. In the woman receiving radiation treatment to the para-aortic area, and who underwent an emergency laparotomy, there was continued leakage from the gastrostomy site after the gastrostomy tube was removed on the 58th day. After two weeks of parented nutrition, the site closed. In seven women who had already undergone major surgery the gastrostomy tube was used for enteral feeding. The gastrostomy tubes were removed without difficulty except in one woman in whom local anaesthetic was required. The subsequent clinical course of the women is shown in Table 2. Eleven women died without leaving hospital (median 11 days, range 2-36). Of the 28 women who left hospital, all but one had established oral intake. In 21 of the 34 women who died, the gastrostomy tube was still in place (median 28 days, range 2-157); 13 of them died in a hospital or hospice. Eighteen women (46%) had the gastrostomy tube removed (median 14.5 days, range 9-180), 13 of whom died (median 167 days, range 77 days-7 years). The median survival time in the remaining five women is 2.2 years. n DISCUSSION Persistent or recurrent gynaecological cancer, especially ovarian cancer, frequently results in episodes of intestinal obstruction. Th~s is difficult to treat because there can be more than one cause of obstruction in an individual patient'. Long term nasogastric drainage is neither acceptable nor effective. Small bore (less than 20 French gauge) gastrostomy tubes, often used for feeding purposes, may not provide adequate drainage or decompression in intestinal obstruction, a point often forgotten in the literature'-']; large bore tubes should be used and we recommend general anaesthesia for their insertion. In our study, 14 women (36%) underwent the simple surgical procedure of a palliative gastrostomy with almost immediate relief of the distressing symptoms of nausea and vomiting. The open placement of a gastrostomy tube takes minutes and post-operative recovery is rapid. Eleven of these women had a short life-expectancy. Three other women considered for major surgery had inoperable disease and a gastrostomy tube was the only operative procedure performed. Twenty-five women (64%) underwent major abdominal surgery in addition to a gastrostomy. There are few factors which predict reliably whether major surgery is likely to benefit women with recurrent gynaecological cancer and intestinal obstruction, and the decision to intervene should be individualised. The outcome for women with gynaecological cancer who have acute intestinal obstruction is poor. Krebs and Goplerud5 found that older age, ascites, tumour recurrence, poor nutritional status and prior chemotherapy and radiotherapy were poor prognostic factors in a group of women managed over 20 years from Fernandes et a1.i2 found that a short time interval from original diagnosis to intestinal obstruction was a poor prognostic factor. Neither group included performance status, which we consider an important factor in the decision to intervene surgically. Van Ooijen et ~ 1. rec- ' ~ ommended that surgical intervention should be avoided in women with ascites and palpable masses. In our experience, these criteria do not reliably identify women who would benefit from major surgery, and they do not preclude the placement of a gastrostomy tube. Table 2. Outcome of gastrostomy. survival Outcome n Range Median Died with gastrostomy tube in place 21 2 days-1 57 days 28 days Died with gastrostomy tube removed Alive with gastrostomy tube removed days-7 years 302 days-6.3 years 167 days 1.7 years 8 RCOG 1999 Br J Obstet Gynaecol 106,

4 GASTROSTOMY, GYNAECOLOGICAL CANCER & INTESTINAL OBSTRUCTION 967 As there are numerous factors which can contribute to the development of intestinal obstruction in women with gynaecological cancer, treatment should be individualised6.l4 In women with recurrent ovarian cancer and bowel obstruction, the outcome from surgical and medical treatment is reported to be similar, but there are no randomised trials comparing these treatment^'^.'^. We do not agree that major surgery is only of benefit in women with a benign cause of obstruction-this is rare and cannot be identified pre-operativelyi6. In women who undergo major surgery for intestinal obstruction, it is essential to maintain adequate bowel decompression post-operatively, and this may require many days of drainage, They will generally have had a nasogastric tube in place for many days pre-operatively. When the woman is no longer ventilated, a nasogastric tube is at best a troublesome inconvenience and at worst contributes to complications of the upper and lower respiratory tract. It is our impression (although we have no objective evidence) that the gastrostomy tube assisted recovery and rehabilitation in the postoperative period, permitting long term gastric drainage and decompression while avoiding the discomfort of a nasogastric tube. In addition, as a gastrostomy tube is well tolerated, it can be used to give medication and as a route for enteral feeding. Recent reports suggest that larger bore gastrostomy tubes can be placed endoscopically or radiologically, often without a gastropexy, and there have been no reports of leakage, local peritonitis or skin ulcerati~n~.~'.'~. We routinely performed a gastropexy in our open technique. Although there are numerous data on the safety and efficacy of endoscopically or radiologically placed tubes, in most reports the tubes were of smaller bore and were placed for feeding and not for The women in our study were ill with intestinal obstruction, but there were no surgical problems with the palliative gastrostomy tube alone, the operating time was short, and there was minimal morbidity related to the surgery and anaesthesia. In palliative care it may be that nonsurgical methods (endoscopic or radiological) to place gastrostomy tubes for drainage will become more pop~lar'~*~'. However, in the UK, these procedures are not commonly performed in women with gynaecological cancer. There are also relative contra-indications to the nonsurgical placement of gastrostomy tubes in women with upper abdominal carcinomatosis, intestinal obstruction or with ascites, and we consider that there will continue to be a place for open gastrostomy. Gastrostomy avoids the need for long term nasogastric suction. The women can be taught to flush the tube regularly and to keep the tube clamped or unclamped. as indicated by their symptoms, so that they can eat and drink. Regardless of the amount of nutrition absorbed, the absence of vomiting has an important psychological benefit, reducing the fear of eating and drinking. Gastrostomy tubes can be used for many months and are easily changed. Furthermore, a gastrostomy permits the woman to return to her home and family and reduces the need for further hospitalisation2'. We conclude that in women with gynaecological cancer and intestinal obstruction a gastrostomy is a safe, effective treatment. It is associated with minimal morbidity and is beneficial both during recovery from surgery and for subsequent palliation. References 1 Ripamonti C. Management of bowel obstruction in advanced cancer. Curr Opin Oncoll994; 6: Osteen RT, Guyton S. Stele Jr G. Wilson RE. Malignant intestinal obstruction. Surgery 1980; 87: Tunca TC, Buchler DA, Mack EA, Ruzicka FF, Crowley JJ, Can WF. The management of ovarian-cancer-caused bowel obstruction. Gynecol Oncoll981; 12: Butler JA, Cameron BL, Morrow M, Kahng K, Tom J. Small bowel obstruction in patients with a prior history of cancer. Am I Surg 1991; 1626M28. 5 Krebs HB, Goplerud DR. Surgical management of bowel obstruction in advanced ovarian carcinoma. Obstet Gynecoll983; 61: Rubin SC, Hoskins WJ. Benjamin I, Lewis JL. Palliative surgery for intestinal obstruction in advanced ovarian cancer. Gynecol Oncol 1989; 34: Gleeson NC, Hoffman MS, Fionca JV, Roberts WS, Cavanagh D. Gastrostomy tubes after gynecologic oncologic surgery. Gynecol Oncol1994; Stamm M. Gastrostomy by a new method. Med News 1894; 65: Adelson MD, Kasowitz MH. Percutaneous endoscopic drainage gastrostomy in the treatment of gastrointestinal obstruction from intraperitoneal malignancy. Obstet Gynecoll993; 81: Marks WH, Perkal MF, Schwa& PE. Percutaneous endoscopic gastrostomy for gastric decompression in metastatic gynecologic malignancies. Surg Gynecol Obstet 1993; Campagnutta E, Cannizzaro R, Gallo A et al. Palliative treatment of upper intestinal obstruction by gynecological malignancy: the usefulness of percutaneous endoscopic gastrostomy. Gynecol Oncol 1996; Femandez JR, Seymour RJ, Suissa S. Bowel obstruction in patients with ovarian cancer: a search for prognostic factors. Am J Obstet Gynecoll988; Van Ooijen B, van der Berg MEL, Planting ASTh, Siersema PD, Wiggers T. Surgical treatment or gastric drainage only for intestinal obstruction in patients with carcinoma of the ovary or peritoneal carcinomatosis of other origin. Surg Gynecol Obsret 1993; Clarke-Pearson DL, Chin NO, DeLong ER et al. Surgical management of intestinal obstruction in ovarian cancer. Gynecol Oncoll987; 26: Larson JE, Podczaski ES, Manetta A, Whitney CW, Mortel R. Bowel obstruction in patients with ovarian carcinoma: analysis of prognostic factors. Gynecol 01~011989; 35: Woolfson RG, Jennings K. Whalen GF. Management of bowel obstruction in patients with abdominal cancer. Arch Surg 1997; 132: Malone JM, Koonce T, Larson DM, Freedman RS, Carrasco CHO, Saul PB. Palliation of small bowel obstruction by percutaneous gastrostomy in patients with progressive ovarian carcinoma. Obsret Gynecoll986; 68: Cunningham MJ, Bromberg C, Kredentser DC, Collins MB, Malfetan0 JH. Percutaneous gastrostomy for decompression in patients with advanced gynecologic malignancies. Gynecol Oncol 1995; Bell SD, Carmody EA, Yeung EY, Thurston WA, Simons ME, Chia- Sing Ho. Percutaneous gastrostomy and gastrojejunostomy: addi- 0 RCOG 1999 Br J Obstet Gynaecol 106,

5 968 E. TSAHALINA ET AL. tional experience in 519 procedures. Radiology 1995; 194: Wohan B, D Agostino HB, Walus-Wigle JR, Easter DW, Beale A. Radiologic, endoscopic and surgical gastrostomy: An institutional evaluation and rneta-anaylsis of the literature. Radiology 1995; 197: Fainsinger RL, Spachynski K, Hanson J, Bruera E. Symptom control in terminally ill patients with malignant bowel obstruction (MBO). J Pain Symptom M mge 1994; Gemlo B, Rayner AA, Lewis B et al. Home support of patients with end-stage malignant bowel obstruction using hydration and venting gastrostorny. Am JSurg 1986; 152: Accepted 22April RCOG 1999 Br J Obstet Gynaecof 106,