- Email: [email protected]
Management of bowel obstruction in advanced and terminal cancer patients
Annals of Oncology 4: 15-21, 1993. © 1993 Kluwer Academic Publishers. Printed in the Netherlands.
Special article Management of bowel obstruction in advanced and terminal cancer patients C. Ripamonti,1 F. De Conno,1 V. Ventafridda,1 B. Rossi2 & M. J. Baines3 'Pain Therapy and Palliative Care Division, National Cancer Institute, 2Surgical Division, Desio Hospital, Milan, Italy; 3Sl. Christopher's Hospice, Sydcnham, London, U.K. Summary Background: Bowel obstruction is a common and distressing outcome in patients with abdominal or pelvic cancer. Patients and methods: Patients may develop bowel obstruction at any time in their clinical history, with an incidence ranging from 5.5% to 42% in ovarian carcinoma and from 10% to 28.4% in colorectal cancer. The causes of the obstruction may be benign postoperative adhesions, a focal malignant or benign deposit, relapse or diffuse carcinomatosis. The symptoms which are almost always present are intestinal colic (reported in 72%-76% of patients), abdominal pain due to distension, hepatomegaly or tumor masses (in 92% of patients) and vomiting (68%-10O%) of cases.
Introduction
Bowel obstruction is a common clinical complication in patients with abdominal or pelvic cancers such as colonic, ovarian and gastric. Moreover, extra-abdominal cancers, such as lung, breast, and melanoma can spread to the abdomen, causing secondary bowel obstruction. Bowel obstruction can be caused by intestinal muscle paralysis (paralytic ileus) or occlusion of the lumen (mechanical ileus), or both, leading to the blocking of faeces and gas through the intestinal passage. The obstruction can be partial or complete, single or multiple, due to benign causes (adhesions, post-irradiation bowel damage, inflammatory bowel disease, hernia) or malignant causes (previous or new tumor, recurrence, carcinomatosis). Pseudo-obstruction is caused by tumor infiltration of the mesentery or bowel muscle or, rarely, involvement of the coeliac plexus. It may also occur as a paraneoplastic neuropathy in patients with lung cancer [1,2]. Incidence
Cancer patients may develop bowel obstruction at any time in their clinical history, more quickly at the advanced stage [3]. The duration between initial tumor diagnosis and development of bowel obstruction is significantly longer between intra-abdominal (mean 22.4 Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
Conclusion: While surgery must remain the primary treatment for malignant obstruction, it is now recognised that there is a group of patients with advanced disease or poor general condition who are unfit for surgery and require alternative management to relieve distressing symptoms. A number of treatment options are now available for the patient with advanced cancer who develops intestinal obstruction. In this review of the literature, the indications for surgery will be examined, the use of nasogastric tube and percutaneous gastrostomy evaluated and the place of drugs for symptom control described.
Key words: bowel obstruction, advanced and terminal cancer patients
months) and extra-abdominal (mean 57.5 months) tumors [4]. In patients with ovarian cancer, the incidence of obstruction ranges from 5.5% to 42% and from 10% to 28.4% in patients with colorectal malignancies [3, 5-13]. Gastrointestinal obstruction occurs in about 3% of advanced cancer patients who are receiveing hospice treatment [5]. Symptoms
The diagnosis of bowel obstruction is based on clinical findings and radiological examination [14-16]. In advanced and terminal cancer patients, the onset of obstruction is rarely an acute event. Symptoms gradually worsen until they become continuous, and their presence and intensity depends on the level involved. Vomiting develops early and in large amounts in gastric, duodenal and small bowel obstruction, and develops later in large bowel obstruction. The incidence of vomiting is 68% and 100% according to two different studies [5, 17]. Vomiting can be intermittent or continuous. Pain is due to distension and abdominal colic in small and large bowel obstruction. It can also be due to the tumor masses present. The incidence of intestinal colic is about 75% [5, 17], while continuous abdominal pain is present in more than 90% of the patients [5, 17]. In cases of complete obstruction, constipation is present for faeces and flatus. Sometimes overflow diarrhoea results from bacterial liquefaction of the
16
shows the rate of bowel obstruction due to benign causes. In patients with a history of colorectal carcinoma developing into intestinal obstruction, it is more probable that a benign cause will be found [21]. Not all patients are suitable candidates for surgery. The rate of inoperable patients ranges from 6% to 50% according to different authors [3, 7, 12, 22-27). The most frequent causes reported in the literature are: 1) extensive rumor; 2) multiple partial obstructions; 3) surgical correction impossible. Benefit from surgery is defined as at least 60 days of survival after operation [6]; however, this definition does not take into account the well-being of the patient, the presence or absence of symptoms, or post-operative complications, nor how long the patients had to remain in the hospital. In Lund's study [7], for example, 56% of the patients operated on survived 60 days after the operation, but 43% of them manifested intermittent symptoms of incomplete and complete intestinal obstruction until death. Surgery does not always solve the problem of controlling obstruction and symptoms in advanced cancer patients, and, in a large percentage of cases, it can lead to further complications. Table 2 shows the rate of operative mortality (defined as death from any cause within 30 days of the operation), the morbidity and the length of survival after surgery for bowel obstruction in cancer patients. The most frequent complications are: wound infections and/or dehiscence, sepsis, enterocutaneous fistula, further obstruction, peritoneal abscess, anastomosis dehiscence, gastrointestinal bleeding, pulmonary embolus, deep venous thrombosis. Several authors have pointed out the prognostic criteria needed to help doctors in selecting patients who are likely to benefit from surgical intervention. Poor prognostic factors are:
faecal material blocked in the colon or rectum. Abdominal distension, visible peristalsis, intermittent borborygmi and anorexia are commonly manifested in patients with bowel obstruction. Diagnosis of constipation and simple paralytic ileus should be ruled out. Treatment Surgery
Different surgical techniques appropriate to the type and site of obstruction are available [18]. In advanced cancer patients, guidelines for conservative vs. surgical treatment are still conflicting. According to Ketcham [19] curative or palliative surgical intervention is necessary even when multiple episodes are experienced. According to Osteen [20], if obstruction due to recurrent cancer has been established, the chances of relieving a second obstruction are poor. The decision is further complicated by the fact that the causes of the obstruction may be benign. Table 1 Table I. Rate of bowel obstruction due to benign causes in advanced cancer patients. Author
Soo|12] Clarke-Pearson [30] Tunca [3| Gallick [4] Osteen [20] Spears (21] Aabo [28] Annest [25]
patients
Primary cancer
Benign causes
64 49 127 50 66 62 41 34
Gynaecologic Ovary Ovary Various Vanous Colorectal Vanous Various
34.0
No.
6.1 9.4
26.0 31.8 48.0 12.0 3.0
Table 2. Complications and length of survival after surgery. Author
No. pts
Primary cancer
30-Day mortality (%)
Other operative complications (%)
Survival months
Soo [12[ Lund [7] Rubin |26| Castaldo |6] Clarke-Pearson [30| Krebs [24] Tunca [3| Piver [231 Beattie [91 Walsh [42] Aranha [43] Aranha [43] Osteen |20] Aabo [28] Chan [44] Annest [25] Turnbull [22]
64 25 43 23 49 98 90 60 11 36 40 26 32 41 10 34 89
Gynaecol. Ovary Ovary Ovary Ovary Ovary Ovary Ovary Ovary Various Various Various Various Various Various Various Abdominal
11 32 9 13 14 12 14
15.5 32.0 11.5
2.5 median 2.0 median 6.8 mean 1 7% 1 year 4.5 median 3.1 median 7.0 mean 2.5 median 7.0 mean 11 median 7.0 mean 4.5 mean 3 median 4.5 median 2 median 4.0 mean 4.5 mean
* not reported.
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
16.5 9
43 49 12 a
31 9
19
27.5 46
22.5 15.0 a
—•
•
24.4 40 18 13
80 44 44
17
1. intestinal motility problems due to diffuse intraperitoneal carcinomatosis (4, 24, 28, 29] 2. patient age over 65 years in association with severe nutritional deprivation [24] 3. ascites [4,29] requiring frequent paracentesis [24] 4. previous radiotherapy of the abdomen or pelvis [6,24] 5. combination chemotherapy [24] 6. palpable intra-abdominal masses [3, 4, 24, 29] and liver involvement, or distant metastases, pleural effusion or pulmonary metastases 7. multiple partial bowel obstruction with prolonged passage time on x-ray examination [3] 8. obstruction of either the small or small and large intestines has an even higher complication rate than patients with only obstruction of the large intestine [6, 28, 30] 9. end-stage disease and poor performance status [22]
Table 3. Nausea and vomiting: pharmacological therapy. 1. Hyoscine butylbromide 60-120 mg/day (SC) [32] It is possible to increase the dosage until xerostomia and drowsiness are tolerable for the patient. 2. Haloperidol 5-15 mg/day SC 117] It can be associated with Hyoscine butylbromide in the same syringe. 3. Metoclopramide 60-240 mg/day SC |311 It was not used in Baines' study |5] as it may increase colic. 4. Hyoscine hydrobromide 0.8-2 mg/day SC |45) 5. Cyclizine 100-150 mg/day SC or 50 mg 8 hourly rectally It can be added to prochlorperazine suppositories and to haloperidol SC. Crystallisation may occur at higher dosages or in association with other drugs |45|. 6. Methotrimeprazine 50-150 mg/day SC |45| It is very effective but it causes sedation and SC irritation. 7. Prochlorperazine 25 mg 8 hourly rectally 8. Chlorpromazine 50-100 mg 8 hourly rectally or SC |8, 45] SC — subcutaneously. Table 4. Colicky Pain: pharmacological therapy.
According to Rubin [26], the clinical variables analyzed (i.e., patient age, interval after cancer diagnosis, previous RT, mean number of prior laparotomies, use of perioperative total parenteral nutrition (TPN), or site of intestinal obstruction) were not correlated with operability or survival following definitive surgery. Since surgical palliation in advanced cancer patients is a complex issue, the decision to proceed with surgery must be carefully evaluated for each individual patient. The decision should be made by the doctors together with the patient and family members. Surgery has to be considered for patients with a life expectancy > than two months [3, 22, 24]. Nasogastric suction and intravenous fluids This is a useful treatment for decompressing the stomach and/or intestine, and for correcting fluid and electrolyte imbalance before surgery or while a decision is being made. Prolonged nasogastric suction and intravenous fluids for symptomatic treatment of inoperable patients is not suggested. This treatment can create discomfort in patients who are generally already distressed by previous anticancer and surgical therapies. Such methods are regarded by the patient as barriers between him/her and family members, as they often require hospitalization. These patients need to find an alternative therapy for symptom management. Pharmacological treatment The few published studies in which symptom control is successfully achieved by administration of analgesic, anticholinergic and antiemetic drugs, without the use of decompressive tube or intravenous fluids, come from hospice [5, 31] and palliative care unit experiences [8, 17, 32]. Table 3 shows the drugs used to control nausea and Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
1. 2. 3. 4. 5.
Hyoscine butylbromide starting with 40 up to 380 mg/day |17) Hyoscine hydrobromide 0.8-2.0 mg/day SC |5] Hyoscine hydrobromide 0.3—0.6 mg sublingually prn |5] Loperamide 2 mg four times daily |5] Morphine SC starting with 2.5 mg/h increasing the dose until relief is achieved [31] 6. Transdermal scopolamine 1.5-3 mg every 3 days [8] 7. Coeliac plexus block with alcohol
Gastrokinetic antiemetics like metoclopramide and domperidone and stimulant laxatives should be discontinued.
vomiting and Table 4 shows the treatment of colicky pain. To control continuous abdominal pain, strong opioids such as morphine, hydromorphone or diamorphine have to be administered subcutaneously using the equianalgesic dose switching from oral to parenteral route and increasing the dose until symptom control. In patients with symptoms due to bowel obstruction, the route of drug administration must be personalized. As vomiting is a very frequent symptom, the oral route is not always possible. Although rectal and sublingual routes are very safe and useful in a home setting [33], only a few drugs are available for these routes of administration. Sometimes the patients have a central venous catheter previously inserted which can also be used to administer drugs. Continuous subcutaneous infusion of drugs through a portable syringe pump allows for the parenteral administration of different combinations of drugs, minimal discomfort for the patient and ease of use in a home setting. Some years ago, Baines et al. [5], showed the possibility of controlling nausea, vomiting and pain by means of a pharmacological approach. Drugs were administered either by mouth, by continuous subcutaneous infusion, or by the rectal route. In 68% of the patients, intestinal colic was completely controlled
18
using smooth muscle relaxants, such as scopolamine Inoperable patients were treated with a pharmacologi(hyoscine), atropine and loperamide. Continuous ab- cal combination of morphine hydrochloride to reduce dominal pain due to distension, tumor mass or hepato- pain (initial dosage of 0.5 mg/kg), hyoscine butylbromegaly was treated using strong opioids, namely dia- mide to control colicky abdominal pain and to reduce morphine, morphine and oxycodone, and it was re- gastrointestinal secretions (initial dosage of 1 mg/kg) lieved in 89% of the patients. Nausea and vomiting and haloperidol as an antiemetic (initial dosage of 0.05 were the most difficult symptoms to manage; only 13% mg/kg). of the patients were rendered free of this symptom. The dosage and combination of drugs were changed Prochlorperazine, chlorpromazine and haloperidol by the hour or day by day until the symptoms were were effective antiemetics while metoclopramide was under control. The drugs were administered by continnot found to be effective and in fact worsened colicky uous subcutaneous infusion via a syringe pump or pain. intravenously only when a central venous catheter had The average survival rate of inoperable patients previously been inserted. treated with drugs was 3.7 months after the onset of Placement of the nasogastric tube was considered obstruction. Seven patients (18%) lived longer than 7 only in patients in whom control of the symptomatolmonths. The survival rates of this series of patients ogy with drugs proved ineffective, whereas intravenous were greater than the survival rates reported by other fluids were administered when patients complained of authors using a similar approach [17, 34]. This differ- thirst, despite the fact that drinking water or ice cubes ence is probably due to the fact that some of Baines' had been provided. series of patients had subacute occlusions as opposed The period of treatment of symptoms lasted 2 to 50 to other patient populations in which complete intes- days with a mean value of 13.4 days and a SD = 12.1. tinal occlusion was reported. Seventeen patients were treated at home and 5 in the Isbister et al. [31] treated 24 patients with advanced hospital. In both cases, treatment continued until death. abdominal malignancy and previous operative or Table 5 shows the pharmacological combination of radiological evidence of intestinal obstruction without drugs, dosages and routes of administration for sympoperation. Morphine was administered to control back- tom control in all 22 patients under study. ground pain and intestinal colic (mean dose of 9.2 mg/h) and metoclopramide to control vomiting (mean Table 5. Drug associations, dosages and routes of administration. dose of 6.9 mg/h). A useful starting dose of morphine was found to be 2.5 mg/h, increasing by 2.5 mg/h until No. Morphine Scopolamine Halopendol Route of butylbromide (mg/day) administration relief was obtained. The starting dose of metoclopra- PTS (mg/day) Mean + SD Mean + SD (mg/day) mide was 5 mg/h increased by 5 mg/h until control was Mean + SD (range) (range) achieved. A combination of drugs was administered (range) through continuous subcutaneous infusion via a syringe 47.7 + 21.0 103.3 + 88 5.8 + 3.6 SC driver. The patients were encouraged to take free fluids 9 (20-80) (40-320) (2-15) and a low-fibre diet while treated at home. The average survival rate after the onset of complete obstruction 5 38+ 16.4 144+135 SC — (20-60) (60-380) was 29.2 days. No patients needed to use nasogastric tubes to control symptoms. 4 47.5+ 15.0 — 8.0 + 3.5 SC (30-60) (5-12) Ventafridda et al. [17] carried out a prospective study to assess vomiting and pain control in terminal 2 80 5.0 SC — (40-120) cancer patients with inoperable gastrointestinal obstruction, using a pharmacological symptomatic treat- 2 IV 45 130 — (30-60) (60-200) ment. After diagnosis of obstruction had been established by the examination of clinical signs, symptoms and radiographic results, the opinion of the referring From Ventfridda et al. [ 17|. surgeon was requested. Twenty-two patients (16 women and 6 men) were judged as inoperable, (age An analysis of the patients' self-descriptive records range, 40 to 80 years; mean ± SD, 57.9 ± 10.6). showed that the pain score, after 2 days of treatment, had decreased by more than 80% as compared to the initial value (p < 0.001). A further decrease was reContraindications for surgery: corded 2 days before the patients' deaths (p < 0.05). - previous laparatomy for obstruction at Fifteen of 22 self-descriptive records reported vomwhich no procedure was possible 8 patients iting. At baseline time, 2/3 of the patients had had 4 or - previous palliative surgery 3 patients more vomiting episodes daily. After 2 days of treat- Performance Status < 30 4 patients ment, the symptom completely disappeared in 8 pa- multiple sites of obstruction 4 patients tients. In 4 patients it was reduced to a single episode - surgery refused 3 patients daily. For all patients, symptom control was maintained until death. Three patients with gastric, pancreatic and Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
19
hepatic carcinoma had nasogastric tube placements because after 2 days of treatment the drug therapy failed to reduce the number of vomiting episodes. Dry mouth showed an upward trend throughout this observation period (p < 0.05), but was successfully treated by administration of liquids by mouth or ice cubes to suck. Drowsiness was also on the rise between TO to two days before death (p < 0.001). Only one of 16 patients who reported being thirsty required intravenous hydration. Thanks to the pharmacological approach, it was possible to avert placement of a nasogastric tube in 19 of 22 patients treated. In 3 patients who presented gastric, pancreatic and hepatic carcinoma, nasogastric tubes were inserted because antiemetic drug therapy had proven ineffective. A common feature of these cases was the presence of upper bowel obstruction, in contrast with the other patients. Thus, the results of the study showed that, in cases of inoperable lower abdomen obstructions, placement of a nasogastric tube to control vomiting and pain until death is not necessary, whereas it is required in instances of upper abdominal obstructions. Infusion therapy was not necessary to control dry mouth, but was needed in one case to relieve thirst. Most patients were able to remain at home until death. De Conno et al. [32] described the antiemetic role of scopolamine butylbromide administered subcutaneously by means of a syringe pump in 3 women with bowel obstructions due to advanced ovarian cancer. All patients were judged inoperable because of multiple sites of obstruction and were treated with total parenteral nutrition through a central venous catheter (3000 cc/ day) and with nasogastric drainage. During daily administration of 120, 120 and 80 mg of scopolamine butylbromide, the quantities of gastrointestinal fluids drained through the tube were significantly reduced. Moreover, during the first week of treatment, patients reported a good relief of colicky pain thanks to administration of scopolamine butylbromide alone. Later, morphine hydrochloride was added to the other drugs to control continuous abdominal pain. In all patients, it was possible to remove the tubes after the first week of treatment with the anticholinergic drug. Therapy continued until the death of the patients, having lasted 55, 172 and 54 days. Dry mouth was the most frequently reported side effect of therapy, but patients tolerated it by sucking ice cubes and drinking small sips of water. Several authors recommend the use of corticosteroids because they can reduce peritumoral inflammatory oedema, thus improving intestinal passage. No controlled clinical trials have been carried out and administration routes and dosing of these drugs have as yet not been standardized. MacDonald et al. [35] suggest either dexamethasone (8 mg/day) or prednisone (25 mg every 12 h for five days), thereby gradually reducing the dose when response is positive. When there
was no response to dexamethasone with increased doses (5 to 10 mg a day subcutaneously), Steiner et al. [8] administered a single dose intravenously of 20-50 mg, then 100 mg. Percutaneous gastrostomy
Percutaneous gastrostomy (PG) is a technique whereby a tube is introduced into the stomach through the abdominal wall without requirement of an operative procedure or general anesthesia [34, 36]. For this reason, PG can be performed safely in advanced cancer patients, as a venting procedure in patients suffering from nausea and vomiting due to bowel obstruction. In addition, PG can serve to provide enteral nutrition if the patient obtains partial relief from intestinal obstruction. PG is superior to both nasogastric suction and operative gastrostomy for palliation of small bowel obstruction in terminal ovarian cancer [36]. In Malone's study, the most severe complication occurred in one of the ten women submitted to PG after gastric secretions spilled out of a small tube, causing autodigestion of the abdomen wall. This technique requires the patient's hospitalization for an average of 4.6 days. After PG, the average length of survival was 35 days (ranging 26-56 days) for seven patients, while two patients were still alive after five months. Gemlo et al. [37] reported the experience of 27 inoperable terminally-ill patients with abdominal cancer and partial or complete bowel obstruction. A central venous catheter was placed in all patients for administration of 2—3 liters of standard electrolyte solution and drugs (opioids, antiemetics, antispasmotics) when oral and rectal routes of administration were not viable. In 13 patients, a venting gastrostomy (9 percutaneous and 4 operative) connected to passive drainage, was performed to control intractable vomiting due to complete obstruction. All patients were cared for at home by nurses after an average stay in the hospital of 11.2 days. During this period the patients and their families underwent extensive training in the use of these catheters, infusion pumps and gastrostomy tubes. There were no complications with gastrostomies. Vomiting was under control and, thanks to intravenous infusion, few patients complained of hunger and none complained of thirst. The average length of survival in these patients was 64 days (ranging 9-223 days). Costs for home care were 1/6 that of hospital care. Australian authors report another positive experience of the role of venting PG in control of symptoms due to obstructive carcinomatosis infiveadvanced cancer patients [38]. In another study [39], five patients with abdominal cancer were submitted to PG, obtaining a decompression of the gastrointestinal tract without nasogastric tubes. All patients went home using PG. The insertion of a long intraluminal decompressive Baker tube across the abdominal wall into the proximal
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
20
jejunum was performed on 12 patients with peritoneal carcinomatosis [40]. Venting of the small intestine distally to the inoperable obstruction permitted the continuous decompression of fluids and gas, and helped patients to control the symptoms. The survival rate of these patients was quite different: three patients died 3, 6, and 14 days after their operations. One patient died five weeks and one two months after operation. Both of the latter were able to eat and drink without vomiting. The remaining patients lived for a period of 3.5 to 36 months. PG may be a valuable alternative to the few with refractory symptoms who do not respond adequately to pharmacological measures. When obstructive symptoms cannot be controlled by drugs, PG is believed to be a more effective and acceptable alternative to the prolonged use of nasogastric tube. Feeding and hydration
When relief of symptoms is obtained through palliative treatment, patients are able to drink water and other fluids and to eat small amounts of their favorite foods by mouth. The exact role of TPN in the management of these patients is controversial, also from the viewpoint of surgeons [6, 24]. No data is available to date on the survival rates or quality of life of advanced cancer patients treated with TPN. We do not administer TPN In our clinical practice, but rather fluids by intravenous infusion (if a central venous catheter was previously used), or by hypodermoclysis [41], only for control of symptoms due to dehydration.
Conclusion The treatment of bowel obstruction via a surgical vs. conservative approach is still an open and widely debated issue in clinical practice for advanced and terminal cancer patients. A patient is a suitable candidate for surgery when he or she has a life expectancy of not less than two months. Yet who can predict how long a patient will live? An accurate study of the prognostic indicators of survival in advanced cancer patients is needed to provide a basis on which doctors, together with patients, can make correct therapeutic decisions. In the management of bowel obstruction, further research is needed: 1. to evaluate the efficacy, tolerability and costs of pharmacological therapy vs. percutaneous gastrostomy, as well as the quality of life of the patients treated in these different ways; 2. to evaluate how many patients with bowel obstruction actually require intravenous or subcutaneous hydration to relieve thirst or other symptoms due to dehydration and electrolyte imbalance.
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
References 1. Addison NV. Pseudo-obstruction of the large bowel. J Royal SocMed 1983; 76: 252-5. 2. Schuffler MD, Baird HW, Fleming CR. Intestinal pseudoobstruction as the presenting manifestation of small-cell carcinoma of the lung. Ann intMed 1983; 98: 129-34. 3. Tunca JC, Buchler DA, Mack EA et al. The management of ovarian-cancer-caused bowel obstruction. Gynecologic Oncology 1981; 12: 186-92. 4. Gallick HL, Weaver DW, Sachs RJ et al. Intestinal Obstruction in cancer patients. An assessment of risk factors and outcome. The American Surgeon 1986; 52: 434-7. 5. Baines M, Oliver DJ, Carter RL. Medical management of intestinal obstruction in patients with advanced malignant disease: A clinical and pathological study. Lancet 1985; 2: 990-3. 6. Castaldo TW, Petrilli ES, Ballon SC et al. Intestinal operations in patients with ovarian carcinoma. Am J Obstet Gynecol 1981; 139: 80-4. 7. Lund B, Hansen M, Lundvall F et al. Intestinal obstruction in patients with advanced carcinoma of the ovaries treated with combination chemotherapy. Surgery, Gynecology & Obstretics 1989; 169:213-8. 8. Steiner N. Controle des symptomes en soins palliatifs: Pileus terminal. Medecine & Hygiene 1991; 49:1182-92. 9. Beattie GJ, Leonard R, Smyth JF. Bowel obstruction in ovarian carcinoma: A retrospective study and review of the literature. Palliative Medicine 1989; 3: 275-80. 10. Solomon HJ. Bowel complications in the management of ovarian cancer. The Australian and New Zeland Journal of Obstetrics and Gynaecology 1983; 23: 65-8. 11. Kyllonen LEJ. Obstruction and perforation complicating colorectal carcinoma. Acta Chir Scand 1987; 153: 607-14. 12. Soo KC, Davidson T, Parker M et al. Intestinal Obstruction in patients with Gynaecological Malignancies. Annals Academy of Medicine 1988; 17(1): 72-5. 13. Phillips RKS, Hittinger R, Fry JS et al. Malignant large bowel obstruction. Br J Surg 1985; 72: 296-302. 14. Yuhasz M, Laufer I, Sutton G et al. Radiography of the small bowel in patients with gynecologic malignancies. Am J Radiology 1985; 144:303-7. 15. Ziter FMH. Radiologic diagnosis: Small bowel. In Welch JP (ed): Bowel Obstruction. Philadelphia: WB Saunders 1990; 96-108. 16. Markowitz SK. Radiologic diagnosis: Colon. In Welch JP (ed): Bowel Obstruction. Philadelphia: WB Saunders 1990; 108-21. 17. Ventafridda V, Ripamonti C, Caraceni A et al. The management of inoperable gastrointestinal obstruction in terminal cancer patients. Tumori 1990; 76: 389-93. 18. Welch JP. Bowel obstruction. Philadelphia: WB. Saunders 1990. 19. Ketcham AS, Hoye RC, Pilch YH et al. Delayed intestinal obstruction following treatment for cancer. Cancer 1970; 25: 406-10. 20. Osteen RT, Guyton S, Steele G et al. Malignant intestinal obstruction. Surgery 1980; 87 (6> 611 -5. 21. Spears H, Petrelli NJ, Herrera L et al. Treatment of bowel obstruction after operation for colorectal carcinoma. Am J Surg 1988; 155:383-6. 22. Turnbull ADM, Guerra J, Starners HF. Results of surgery for obstructing carcinomatosis of gastrointestinal, pancreatic, or biliary origin. J Clin Oncol 1989; 7: 381-6. 23. Piver MS, Barlow JJ, Lele SB et al. Survival after ovarian cancer induced intestinal obstruction. Gynecologic Oncology 1982; 13:44-9. 24. Krebs HB, Goplerud DR. Surgical management of bowel obstruction in advanced ovarian carcinoma. Obstet Gynecol 1983; 61: 327-30. 25. Annest LS, Jolly PC. The results of surgical treatment of bowel obstruction caused by peritoneal carcinomatosis. The American Surgeon 1979; 45: 718-21.
21 26. Rubin SC, Hoskins WJ, Benjamin I et al. Palliative surgery for intestinal obstruction in advanced ovarian cancer. Gynecologic Oncology 1989; 34: 16-9. 27. Taylor RH. Laparotomy for obstruction with recurrent tumor. BrJSurgl985;72:327. 28. Aabo K, Pedersen H, Bach F et al. Surgical management of intestinal obstruction in the late course of malignant disease. Acta Chir Scand 1984; 150: 173-6. 29. Glass RL, LeDuc RJ. Small Intestinal Obstruction from Peritoneal Carcinomatosis. Am J Sur 1973; 125: 316-7. 30. Clarke-Pearson DL, Chin NO, DeLong ER et al. Surgical Management of Intestinal Obstruction in Ovarian Cancer. I. Clinical features, postoperative complication^, and survival. Gynecologic Oncology 1987; 26: 11-8. 31. Isbister WH, Elder P, Symons L. Non-operative management of malignant intestinal obstruction. J R Coll Surg Edinb 1990; 35:369-72. 32. De Conno F, Caraceni A, Zecca E et al. Continuous subcutaneous infusion of hyoscine butylbromide reduces secretions in patients with gastrointestinal obstruction. J Pain Sympt Manage 1991; 6:484-6. 33. Ripamonti C, Bruera E. Rectal, buccal, and sublingual narcotics for the management of cancer pain. J Palliative Care 1991; 7(1): 30-5. 34. George J, Crawford D, Lewis T et al. Percutaneous endoscopic gastrostomy: a two year experience. Med J Aust 1990; 152: 17-9. 35. MacDonald N. Palliative Cancer Care: Pain Relief and Management of other symptoms WHO, 1991. 36. Malone JM, Koonce T, Larson DM et al. Palliation of small bowel obstructon by percutaneous gastrostomy in patients with progressive ovarian carcinoma. Obstetrics and Gynecology 1986; 68:431-3. 37. Gemlo B, Rayner AA, Lewis B. Home support of patients with
38.
39. 40. 41.
42.
43.
44.
45.
end-stage malignant bowel obstruction using hydration and venting gastrostomy. Am J Surgery 1986; 152: 100-4. Ashby MA, Game PA, Devitt P et al. Percutaneous gastrostomy as a venting procedure in palliative care. Palliative Medicine 1991; 5: 147-50. Stellato TA, Shenk RR. Gastrointestinal emergencies in the oncology patient. Semin Oncol 1989; 16 (6): 521-31. McCarthy D. Strategy for intestinal obstruction in peritoneal carcinomatosis. Archives Surgery 1986; 121: 1081-2. Bruera E, Legris MA, Kuehn N et al. Hypodermoclysis for the administration of fluids and narcotic analgesics in patients with advanced cancer. J Pain and Sympt Manage 1990; 5: 218-20. Walsh HPJ, Schofield PF. Is laparotomy for small bowel obstruction justified in patients with previously treated malignancy? Br J Surg 1984; 71: 933-5. Aranha GV, Folk FA, Greenlee HB. Surgical palliation of small bowel obstruction due to metastatic carcinoma. The American Surgeon 1981; 47: 99-102. Chan A, Woodruff RK. Intestinal obstruction in patients with widespread intra-abdominal malignancy. J Pain Sympt Manage 1992 (in press). Baines M. The pathophysiology and management of malignant intestinal obstruction. In: Oxford Text Book of Palliative Medicine, 1993 (in press).
Received 26 March 1992; accepted 27 May 1992.
Correspondence to: Carla Ripamonti M.D. Pain Therapy and Palliative Care Division National Cancer Institute via Venezian, 1 20133, Milan, Italy
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
Special article Management of bowel obstruction in advanced and terminal cancer patients C. Ripamonti,1 F. De Conno,1 V. Ventafridda,1 B. Rossi2 & M. J. Baines3 'Pain Therapy and Palliative Care Division, National Cancer Institute, 2Surgical Division, Desio Hospital, Milan, Italy; 3Sl. Christopher's Hospice, Sydcnham, London, U.K. Summary Background: Bowel obstruction is a common and distressing outcome in patients with abdominal or pelvic cancer. Patients and methods: Patients may develop bowel obstruction at any time in their clinical history, with an incidence ranging from 5.5% to 42% in ovarian carcinoma and from 10% to 28.4% in colorectal cancer. The causes of the obstruction may be benign postoperative adhesions, a focal malignant or benign deposit, relapse or diffuse carcinomatosis. The symptoms which are almost always present are intestinal colic (reported in 72%-76% of patients), abdominal pain due to distension, hepatomegaly or tumor masses (in 92% of patients) and vomiting (68%-10O%) of cases.
Introduction
Bowel obstruction is a common clinical complication in patients with abdominal or pelvic cancers such as colonic, ovarian and gastric. Moreover, extra-abdominal cancers, such as lung, breast, and melanoma can spread to the abdomen, causing secondary bowel obstruction. Bowel obstruction can be caused by intestinal muscle paralysis (paralytic ileus) or occlusion of the lumen (mechanical ileus), or both, leading to the blocking of faeces and gas through the intestinal passage. The obstruction can be partial or complete, single or multiple, due to benign causes (adhesions, post-irradiation bowel damage, inflammatory bowel disease, hernia) or malignant causes (previous or new tumor, recurrence, carcinomatosis). Pseudo-obstruction is caused by tumor infiltration of the mesentery or bowel muscle or, rarely, involvement of the coeliac plexus. It may also occur as a paraneoplastic neuropathy in patients with lung cancer [1,2]. Incidence
Cancer patients may develop bowel obstruction at any time in their clinical history, more quickly at the advanced stage [3]. The duration between initial tumor diagnosis and development of bowel obstruction is significantly longer between intra-abdominal (mean 22.4 Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
Conclusion: While surgery must remain the primary treatment for malignant obstruction, it is now recognised that there is a group of patients with advanced disease or poor general condition who are unfit for surgery and require alternative management to relieve distressing symptoms. A number of treatment options are now available for the patient with advanced cancer who develops intestinal obstruction. In this review of the literature, the indications for surgery will be examined, the use of nasogastric tube and percutaneous gastrostomy evaluated and the place of drugs for symptom control described.
Key words: bowel obstruction, advanced and terminal cancer patients
months) and extra-abdominal (mean 57.5 months) tumors [4]. In patients with ovarian cancer, the incidence of obstruction ranges from 5.5% to 42% and from 10% to 28.4% in patients with colorectal malignancies [3, 5-13]. Gastrointestinal obstruction occurs in about 3% of advanced cancer patients who are receiveing hospice treatment [5]. Symptoms
The diagnosis of bowel obstruction is based on clinical findings and radiological examination [14-16]. In advanced and terminal cancer patients, the onset of obstruction is rarely an acute event. Symptoms gradually worsen until they become continuous, and their presence and intensity depends on the level involved. Vomiting develops early and in large amounts in gastric, duodenal and small bowel obstruction, and develops later in large bowel obstruction. The incidence of vomiting is 68% and 100% according to two different studies [5, 17]. Vomiting can be intermittent or continuous. Pain is due to distension and abdominal colic in small and large bowel obstruction. It can also be due to the tumor masses present. The incidence of intestinal colic is about 75% [5, 17], while continuous abdominal pain is present in more than 90% of the patients [5, 17]. In cases of complete obstruction, constipation is present for faeces and flatus. Sometimes overflow diarrhoea results from bacterial liquefaction of the
16
shows the rate of bowel obstruction due to benign causes. In patients with a history of colorectal carcinoma developing into intestinal obstruction, it is more probable that a benign cause will be found [21]. Not all patients are suitable candidates for surgery. The rate of inoperable patients ranges from 6% to 50% according to different authors [3, 7, 12, 22-27). The most frequent causes reported in the literature are: 1) extensive rumor; 2) multiple partial obstructions; 3) surgical correction impossible. Benefit from surgery is defined as at least 60 days of survival after operation [6]; however, this definition does not take into account the well-being of the patient, the presence or absence of symptoms, or post-operative complications, nor how long the patients had to remain in the hospital. In Lund's study [7], for example, 56% of the patients operated on survived 60 days after the operation, but 43% of them manifested intermittent symptoms of incomplete and complete intestinal obstruction until death. Surgery does not always solve the problem of controlling obstruction and symptoms in advanced cancer patients, and, in a large percentage of cases, it can lead to further complications. Table 2 shows the rate of operative mortality (defined as death from any cause within 30 days of the operation), the morbidity and the length of survival after surgery for bowel obstruction in cancer patients. The most frequent complications are: wound infections and/or dehiscence, sepsis, enterocutaneous fistula, further obstruction, peritoneal abscess, anastomosis dehiscence, gastrointestinal bleeding, pulmonary embolus, deep venous thrombosis. Several authors have pointed out the prognostic criteria needed to help doctors in selecting patients who are likely to benefit from surgical intervention. Poor prognostic factors are:
faecal material blocked in the colon or rectum. Abdominal distension, visible peristalsis, intermittent borborygmi and anorexia are commonly manifested in patients with bowel obstruction. Diagnosis of constipation and simple paralytic ileus should be ruled out. Treatment Surgery
Different surgical techniques appropriate to the type and site of obstruction are available [18]. In advanced cancer patients, guidelines for conservative vs. surgical treatment are still conflicting. According to Ketcham [19] curative or palliative surgical intervention is necessary even when multiple episodes are experienced. According to Osteen [20], if obstruction due to recurrent cancer has been established, the chances of relieving a second obstruction are poor. The decision is further complicated by the fact that the causes of the obstruction may be benign. Table 1 Table I. Rate of bowel obstruction due to benign causes in advanced cancer patients. Author
Soo|12] Clarke-Pearson [30] Tunca [3| Gallick [4] Osteen [20] Spears (21] Aabo [28] Annest [25]
patients
Primary cancer
Benign causes
64 49 127 50 66 62 41 34
Gynaecologic Ovary Ovary Various Vanous Colorectal Vanous Various
34.0
No.
6.1 9.4
26.0 31.8 48.0 12.0 3.0
Table 2. Complications and length of survival after surgery. Author
No. pts
Primary cancer
30-Day mortality (%)
Other operative complications (%)
Survival months
Soo [12[ Lund [7] Rubin |26| Castaldo |6] Clarke-Pearson [30| Krebs [24] Tunca [3| Piver [231 Beattie [91 Walsh [42] Aranha [43] Aranha [43] Osteen |20] Aabo [28] Chan [44] Annest [25] Turnbull [22]
64 25 43 23 49 98 90 60 11 36 40 26 32 41 10 34 89
Gynaecol. Ovary Ovary Ovary Ovary Ovary Ovary Ovary Ovary Various Various Various Various Various Various Various Abdominal
11 32 9 13 14 12 14
15.5 32.0 11.5
2.5 median 2.0 median 6.8 mean 1 7% 1 year 4.5 median 3.1 median 7.0 mean 2.5 median 7.0 mean 11 median 7.0 mean 4.5 mean 3 median 4.5 median 2 median 4.0 mean 4.5 mean
* not reported.
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
16.5 9
43 49 12 a
31 9
19
27.5 46
22.5 15.0 a
—•
•
24.4 40 18 13
80 44 44
17
1. intestinal motility problems due to diffuse intraperitoneal carcinomatosis (4, 24, 28, 29] 2. patient age over 65 years in association with severe nutritional deprivation [24] 3. ascites [4,29] requiring frequent paracentesis [24] 4. previous radiotherapy of the abdomen or pelvis [6,24] 5. combination chemotherapy [24] 6. palpable intra-abdominal masses [3, 4, 24, 29] and liver involvement, or distant metastases, pleural effusion or pulmonary metastases 7. multiple partial bowel obstruction with prolonged passage time on x-ray examination [3] 8. obstruction of either the small or small and large intestines has an even higher complication rate than patients with only obstruction of the large intestine [6, 28, 30] 9. end-stage disease and poor performance status [22]
Table 3. Nausea and vomiting: pharmacological therapy. 1. Hyoscine butylbromide 60-120 mg/day (SC) [32] It is possible to increase the dosage until xerostomia and drowsiness are tolerable for the patient. 2. Haloperidol 5-15 mg/day SC 117] It can be associated with Hyoscine butylbromide in the same syringe. 3. Metoclopramide 60-240 mg/day SC |311 It was not used in Baines' study |5] as it may increase colic. 4. Hyoscine hydrobromide 0.8-2 mg/day SC |45) 5. Cyclizine 100-150 mg/day SC or 50 mg 8 hourly rectally It can be added to prochlorperazine suppositories and to haloperidol SC. Crystallisation may occur at higher dosages or in association with other drugs |45|. 6. Methotrimeprazine 50-150 mg/day SC |45| It is very effective but it causes sedation and SC irritation. 7. Prochlorperazine 25 mg 8 hourly rectally 8. Chlorpromazine 50-100 mg 8 hourly rectally or SC |8, 45] SC — subcutaneously. Table 4. Colicky Pain: pharmacological therapy.
According to Rubin [26], the clinical variables analyzed (i.e., patient age, interval after cancer diagnosis, previous RT, mean number of prior laparotomies, use of perioperative total parenteral nutrition (TPN), or site of intestinal obstruction) were not correlated with operability or survival following definitive surgery. Since surgical palliation in advanced cancer patients is a complex issue, the decision to proceed with surgery must be carefully evaluated for each individual patient. The decision should be made by the doctors together with the patient and family members. Surgery has to be considered for patients with a life expectancy > than two months [3, 22, 24]. Nasogastric suction and intravenous fluids This is a useful treatment for decompressing the stomach and/or intestine, and for correcting fluid and electrolyte imbalance before surgery or while a decision is being made. Prolonged nasogastric suction and intravenous fluids for symptomatic treatment of inoperable patients is not suggested. This treatment can create discomfort in patients who are generally already distressed by previous anticancer and surgical therapies. Such methods are regarded by the patient as barriers between him/her and family members, as they often require hospitalization. These patients need to find an alternative therapy for symptom management. Pharmacological treatment The few published studies in which symptom control is successfully achieved by administration of analgesic, anticholinergic and antiemetic drugs, without the use of decompressive tube or intravenous fluids, come from hospice [5, 31] and palliative care unit experiences [8, 17, 32]. Table 3 shows the drugs used to control nausea and Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
1. 2. 3. 4. 5.
Hyoscine butylbromide starting with 40 up to 380 mg/day |17) Hyoscine hydrobromide 0.8-2.0 mg/day SC |5] Hyoscine hydrobromide 0.3—0.6 mg sublingually prn |5] Loperamide 2 mg four times daily |5] Morphine SC starting with 2.5 mg/h increasing the dose until relief is achieved [31] 6. Transdermal scopolamine 1.5-3 mg every 3 days [8] 7. Coeliac plexus block with alcohol
Gastrokinetic antiemetics like metoclopramide and domperidone and stimulant laxatives should be discontinued.
vomiting and Table 4 shows the treatment of colicky pain. To control continuous abdominal pain, strong opioids such as morphine, hydromorphone or diamorphine have to be administered subcutaneously using the equianalgesic dose switching from oral to parenteral route and increasing the dose until symptom control. In patients with symptoms due to bowel obstruction, the route of drug administration must be personalized. As vomiting is a very frequent symptom, the oral route is not always possible. Although rectal and sublingual routes are very safe and useful in a home setting [33], only a few drugs are available for these routes of administration. Sometimes the patients have a central venous catheter previously inserted which can also be used to administer drugs. Continuous subcutaneous infusion of drugs through a portable syringe pump allows for the parenteral administration of different combinations of drugs, minimal discomfort for the patient and ease of use in a home setting. Some years ago, Baines et al. [5], showed the possibility of controlling nausea, vomiting and pain by means of a pharmacological approach. Drugs were administered either by mouth, by continuous subcutaneous infusion, or by the rectal route. In 68% of the patients, intestinal colic was completely controlled
18
using smooth muscle relaxants, such as scopolamine Inoperable patients were treated with a pharmacologi(hyoscine), atropine and loperamide. Continuous ab- cal combination of morphine hydrochloride to reduce dominal pain due to distension, tumor mass or hepato- pain (initial dosage of 0.5 mg/kg), hyoscine butylbromegaly was treated using strong opioids, namely dia- mide to control colicky abdominal pain and to reduce morphine, morphine and oxycodone, and it was re- gastrointestinal secretions (initial dosage of 1 mg/kg) lieved in 89% of the patients. Nausea and vomiting and haloperidol as an antiemetic (initial dosage of 0.05 were the most difficult symptoms to manage; only 13% mg/kg). of the patients were rendered free of this symptom. The dosage and combination of drugs were changed Prochlorperazine, chlorpromazine and haloperidol by the hour or day by day until the symptoms were were effective antiemetics while metoclopramide was under control. The drugs were administered by continnot found to be effective and in fact worsened colicky uous subcutaneous infusion via a syringe pump or pain. intravenously only when a central venous catheter had The average survival rate of inoperable patients previously been inserted. treated with drugs was 3.7 months after the onset of Placement of the nasogastric tube was considered obstruction. Seven patients (18%) lived longer than 7 only in patients in whom control of the symptomatolmonths. The survival rates of this series of patients ogy with drugs proved ineffective, whereas intravenous were greater than the survival rates reported by other fluids were administered when patients complained of authors using a similar approach [17, 34]. This differ- thirst, despite the fact that drinking water or ice cubes ence is probably due to the fact that some of Baines' had been provided. series of patients had subacute occlusions as opposed The period of treatment of symptoms lasted 2 to 50 to other patient populations in which complete intes- days with a mean value of 13.4 days and a SD = 12.1. tinal occlusion was reported. Seventeen patients were treated at home and 5 in the Isbister et al. [31] treated 24 patients with advanced hospital. In both cases, treatment continued until death. abdominal malignancy and previous operative or Table 5 shows the pharmacological combination of radiological evidence of intestinal obstruction without drugs, dosages and routes of administration for sympoperation. Morphine was administered to control back- tom control in all 22 patients under study. ground pain and intestinal colic (mean dose of 9.2 mg/h) and metoclopramide to control vomiting (mean Table 5. Drug associations, dosages and routes of administration. dose of 6.9 mg/h). A useful starting dose of morphine was found to be 2.5 mg/h, increasing by 2.5 mg/h until No. Morphine Scopolamine Halopendol Route of butylbromide (mg/day) administration relief was obtained. The starting dose of metoclopra- PTS (mg/day) Mean + SD Mean + SD (mg/day) mide was 5 mg/h increased by 5 mg/h until control was Mean + SD (range) (range) achieved. A combination of drugs was administered (range) through continuous subcutaneous infusion via a syringe 47.7 + 21.0 103.3 + 88 5.8 + 3.6 SC driver. The patients were encouraged to take free fluids 9 (20-80) (40-320) (2-15) and a low-fibre diet while treated at home. The average survival rate after the onset of complete obstruction 5 38+ 16.4 144+135 SC — (20-60) (60-380) was 29.2 days. No patients needed to use nasogastric tubes to control symptoms. 4 47.5+ 15.0 — 8.0 + 3.5 SC (30-60) (5-12) Ventafridda et al. [17] carried out a prospective study to assess vomiting and pain control in terminal 2 80 5.0 SC — (40-120) cancer patients with inoperable gastrointestinal obstruction, using a pharmacological symptomatic treat- 2 IV 45 130 — (30-60) (60-200) ment. After diagnosis of obstruction had been established by the examination of clinical signs, symptoms and radiographic results, the opinion of the referring From Ventfridda et al. [ 17|. surgeon was requested. Twenty-two patients (16 women and 6 men) were judged as inoperable, (age An analysis of the patients' self-descriptive records range, 40 to 80 years; mean ± SD, 57.9 ± 10.6). showed that the pain score, after 2 days of treatment, had decreased by more than 80% as compared to the initial value (p < 0.001). A further decrease was reContraindications for surgery: corded 2 days before the patients' deaths (p < 0.05). - previous laparatomy for obstruction at Fifteen of 22 self-descriptive records reported vomwhich no procedure was possible 8 patients iting. At baseline time, 2/3 of the patients had had 4 or - previous palliative surgery 3 patients more vomiting episodes daily. After 2 days of treat- Performance Status < 30 4 patients ment, the symptom completely disappeared in 8 pa- multiple sites of obstruction 4 patients tients. In 4 patients it was reduced to a single episode - surgery refused 3 patients daily. For all patients, symptom control was maintained until death. Three patients with gastric, pancreatic and Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
19
hepatic carcinoma had nasogastric tube placements because after 2 days of treatment the drug therapy failed to reduce the number of vomiting episodes. Dry mouth showed an upward trend throughout this observation period (p < 0.05), but was successfully treated by administration of liquids by mouth or ice cubes to suck. Drowsiness was also on the rise between TO to two days before death (p < 0.001). Only one of 16 patients who reported being thirsty required intravenous hydration. Thanks to the pharmacological approach, it was possible to avert placement of a nasogastric tube in 19 of 22 patients treated. In 3 patients who presented gastric, pancreatic and hepatic carcinoma, nasogastric tubes were inserted because antiemetic drug therapy had proven ineffective. A common feature of these cases was the presence of upper bowel obstruction, in contrast with the other patients. Thus, the results of the study showed that, in cases of inoperable lower abdomen obstructions, placement of a nasogastric tube to control vomiting and pain until death is not necessary, whereas it is required in instances of upper abdominal obstructions. Infusion therapy was not necessary to control dry mouth, but was needed in one case to relieve thirst. Most patients were able to remain at home until death. De Conno et al. [32] described the antiemetic role of scopolamine butylbromide administered subcutaneously by means of a syringe pump in 3 women with bowel obstructions due to advanced ovarian cancer. All patients were judged inoperable because of multiple sites of obstruction and were treated with total parenteral nutrition through a central venous catheter (3000 cc/ day) and with nasogastric drainage. During daily administration of 120, 120 and 80 mg of scopolamine butylbromide, the quantities of gastrointestinal fluids drained through the tube were significantly reduced. Moreover, during the first week of treatment, patients reported a good relief of colicky pain thanks to administration of scopolamine butylbromide alone. Later, morphine hydrochloride was added to the other drugs to control continuous abdominal pain. In all patients, it was possible to remove the tubes after the first week of treatment with the anticholinergic drug. Therapy continued until the death of the patients, having lasted 55, 172 and 54 days. Dry mouth was the most frequently reported side effect of therapy, but patients tolerated it by sucking ice cubes and drinking small sips of water. Several authors recommend the use of corticosteroids because they can reduce peritumoral inflammatory oedema, thus improving intestinal passage. No controlled clinical trials have been carried out and administration routes and dosing of these drugs have as yet not been standardized. MacDonald et al. [35] suggest either dexamethasone (8 mg/day) or prednisone (25 mg every 12 h for five days), thereby gradually reducing the dose when response is positive. When there
was no response to dexamethasone with increased doses (5 to 10 mg a day subcutaneously), Steiner et al. [8] administered a single dose intravenously of 20-50 mg, then 100 mg. Percutaneous gastrostomy
Percutaneous gastrostomy (PG) is a technique whereby a tube is introduced into the stomach through the abdominal wall without requirement of an operative procedure or general anesthesia [34, 36]. For this reason, PG can be performed safely in advanced cancer patients, as a venting procedure in patients suffering from nausea and vomiting due to bowel obstruction. In addition, PG can serve to provide enteral nutrition if the patient obtains partial relief from intestinal obstruction. PG is superior to both nasogastric suction and operative gastrostomy for palliation of small bowel obstruction in terminal ovarian cancer [36]. In Malone's study, the most severe complication occurred in one of the ten women submitted to PG after gastric secretions spilled out of a small tube, causing autodigestion of the abdomen wall. This technique requires the patient's hospitalization for an average of 4.6 days. After PG, the average length of survival was 35 days (ranging 26-56 days) for seven patients, while two patients were still alive after five months. Gemlo et al. [37] reported the experience of 27 inoperable terminally-ill patients with abdominal cancer and partial or complete bowel obstruction. A central venous catheter was placed in all patients for administration of 2—3 liters of standard electrolyte solution and drugs (opioids, antiemetics, antispasmotics) when oral and rectal routes of administration were not viable. In 13 patients, a venting gastrostomy (9 percutaneous and 4 operative) connected to passive drainage, was performed to control intractable vomiting due to complete obstruction. All patients were cared for at home by nurses after an average stay in the hospital of 11.2 days. During this period the patients and their families underwent extensive training in the use of these catheters, infusion pumps and gastrostomy tubes. There were no complications with gastrostomies. Vomiting was under control and, thanks to intravenous infusion, few patients complained of hunger and none complained of thirst. The average length of survival in these patients was 64 days (ranging 9-223 days). Costs for home care were 1/6 that of hospital care. Australian authors report another positive experience of the role of venting PG in control of symptoms due to obstructive carcinomatosis infiveadvanced cancer patients [38]. In another study [39], five patients with abdominal cancer were submitted to PG, obtaining a decompression of the gastrointestinal tract without nasogastric tubes. All patients went home using PG. The insertion of a long intraluminal decompressive Baker tube across the abdominal wall into the proximal
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
20
jejunum was performed on 12 patients with peritoneal carcinomatosis [40]. Venting of the small intestine distally to the inoperable obstruction permitted the continuous decompression of fluids and gas, and helped patients to control the symptoms. The survival rate of these patients was quite different: three patients died 3, 6, and 14 days after their operations. One patient died five weeks and one two months after operation. Both of the latter were able to eat and drink without vomiting. The remaining patients lived for a period of 3.5 to 36 months. PG may be a valuable alternative to the few with refractory symptoms who do not respond adequately to pharmacological measures. When obstructive symptoms cannot be controlled by drugs, PG is believed to be a more effective and acceptable alternative to the prolonged use of nasogastric tube. Feeding and hydration
When relief of symptoms is obtained through palliative treatment, patients are able to drink water and other fluids and to eat small amounts of their favorite foods by mouth. The exact role of TPN in the management of these patients is controversial, also from the viewpoint of surgeons [6, 24]. No data is available to date on the survival rates or quality of life of advanced cancer patients treated with TPN. We do not administer TPN In our clinical practice, but rather fluids by intravenous infusion (if a central venous catheter was previously used), or by hypodermoclysis [41], only for control of symptoms due to dehydration.
Conclusion The treatment of bowel obstruction via a surgical vs. conservative approach is still an open and widely debated issue in clinical practice for advanced and terminal cancer patients. A patient is a suitable candidate for surgery when he or she has a life expectancy of not less than two months. Yet who can predict how long a patient will live? An accurate study of the prognostic indicators of survival in advanced cancer patients is needed to provide a basis on which doctors, together with patients, can make correct therapeutic decisions. In the management of bowel obstruction, further research is needed: 1. to evaluate the efficacy, tolerability and costs of pharmacological therapy vs. percutaneous gastrostomy, as well as the quality of life of the patients treated in these different ways; 2. to evaluate how many patients with bowel obstruction actually require intravenous or subcutaneous hydration to relieve thirst or other symptoms due to dehydration and electrolyte imbalance.
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018
References 1. Addison NV. Pseudo-obstruction of the large bowel. J Royal SocMed 1983; 76: 252-5. 2. Schuffler MD, Baird HW, Fleming CR. Intestinal pseudoobstruction as the presenting manifestation of small-cell carcinoma of the lung. Ann intMed 1983; 98: 129-34. 3. Tunca JC, Buchler DA, Mack EA et al. The management of ovarian-cancer-caused bowel obstruction. Gynecologic Oncology 1981; 12: 186-92. 4. Gallick HL, Weaver DW, Sachs RJ et al. Intestinal Obstruction in cancer patients. An assessment of risk factors and outcome. The American Surgeon 1986; 52: 434-7. 5. Baines M, Oliver DJ, Carter RL. Medical management of intestinal obstruction in patients with advanced malignant disease: A clinical and pathological study. Lancet 1985; 2: 990-3. 6. Castaldo TW, Petrilli ES, Ballon SC et al. Intestinal operations in patients with ovarian carcinoma. Am J Obstet Gynecol 1981; 139: 80-4. 7. Lund B, Hansen M, Lundvall F et al. Intestinal obstruction in patients with advanced carcinoma of the ovaries treated with combination chemotherapy. Surgery, Gynecology & Obstretics 1989; 169:213-8. 8. Steiner N. Controle des symptomes en soins palliatifs: Pileus terminal. Medecine & Hygiene 1991; 49:1182-92. 9. Beattie GJ, Leonard R, Smyth JF. Bowel obstruction in ovarian carcinoma: A retrospective study and review of the literature. Palliative Medicine 1989; 3: 275-80. 10. Solomon HJ. Bowel complications in the management of ovarian cancer. The Australian and New Zeland Journal of Obstetrics and Gynaecology 1983; 23: 65-8. 11. Kyllonen LEJ. Obstruction and perforation complicating colorectal carcinoma. Acta Chir Scand 1987; 153: 607-14. 12. Soo KC, Davidson T, Parker M et al. Intestinal Obstruction in patients with Gynaecological Malignancies. Annals Academy of Medicine 1988; 17(1): 72-5. 13. Phillips RKS, Hittinger R, Fry JS et al. Malignant large bowel obstruction. Br J Surg 1985; 72: 296-302. 14. Yuhasz M, Laufer I, Sutton G et al. Radiography of the small bowel in patients with gynecologic malignancies. Am J Radiology 1985; 144:303-7. 15. Ziter FMH. Radiologic diagnosis: Small bowel. In Welch JP (ed): Bowel Obstruction. Philadelphia: WB Saunders 1990; 96-108. 16. Markowitz SK. Radiologic diagnosis: Colon. In Welch JP (ed): Bowel Obstruction. Philadelphia: WB Saunders 1990; 108-21. 17. Ventafridda V, Ripamonti C, Caraceni A et al. The management of inoperable gastrointestinal obstruction in terminal cancer patients. Tumori 1990; 76: 389-93. 18. Welch JP. Bowel obstruction. Philadelphia: WB. Saunders 1990. 19. Ketcham AS, Hoye RC, Pilch YH et al. Delayed intestinal obstruction following treatment for cancer. Cancer 1970; 25: 406-10. 20. Osteen RT, Guyton S, Steele G et al. Malignant intestinal obstruction. Surgery 1980; 87 (6> 611 -5. 21. Spears H, Petrelli NJ, Herrera L et al. Treatment of bowel obstruction after operation for colorectal carcinoma. Am J Surg 1988; 155:383-6. 22. Turnbull ADM, Guerra J, Starners HF. Results of surgery for obstructing carcinomatosis of gastrointestinal, pancreatic, or biliary origin. J Clin Oncol 1989; 7: 381-6. 23. Piver MS, Barlow JJ, Lele SB et al. Survival after ovarian cancer induced intestinal obstruction. Gynecologic Oncology 1982; 13:44-9. 24. Krebs HB, Goplerud DR. Surgical management of bowel obstruction in advanced ovarian carcinoma. Obstet Gynecol 1983; 61: 327-30. 25. Annest LS, Jolly PC. The results of surgical treatment of bowel obstruction caused by peritoneal carcinomatosis. The American Surgeon 1979; 45: 718-21.
21 26. Rubin SC, Hoskins WJ, Benjamin I et al. Palliative surgery for intestinal obstruction in advanced ovarian cancer. Gynecologic Oncology 1989; 34: 16-9. 27. Taylor RH. Laparotomy for obstruction with recurrent tumor. BrJSurgl985;72:327. 28. Aabo K, Pedersen H, Bach F et al. Surgical management of intestinal obstruction in the late course of malignant disease. Acta Chir Scand 1984; 150: 173-6. 29. Glass RL, LeDuc RJ. Small Intestinal Obstruction from Peritoneal Carcinomatosis. Am J Sur 1973; 125: 316-7. 30. Clarke-Pearson DL, Chin NO, DeLong ER et al. Surgical Management of Intestinal Obstruction in Ovarian Cancer. I. Clinical features, postoperative complication^, and survival. Gynecologic Oncology 1987; 26: 11-8. 31. Isbister WH, Elder P, Symons L. Non-operative management of malignant intestinal obstruction. J R Coll Surg Edinb 1990; 35:369-72. 32. De Conno F, Caraceni A, Zecca E et al. Continuous subcutaneous infusion of hyoscine butylbromide reduces secretions in patients with gastrointestinal obstruction. J Pain Sympt Manage 1991; 6:484-6. 33. Ripamonti C, Bruera E. Rectal, buccal, and sublingual narcotics for the management of cancer pain. J Palliative Care 1991; 7(1): 30-5. 34. George J, Crawford D, Lewis T et al. Percutaneous endoscopic gastrostomy: a two year experience. Med J Aust 1990; 152: 17-9. 35. MacDonald N. Palliative Cancer Care: Pain Relief and Management of other symptoms WHO, 1991. 36. Malone JM, Koonce T, Larson DM et al. Palliation of small bowel obstructon by percutaneous gastrostomy in patients with progressive ovarian carcinoma. Obstetrics and Gynecology 1986; 68:431-3. 37. Gemlo B, Rayner AA, Lewis B. Home support of patients with
38.
39. 40. 41.
42.
43.
44.
45.
end-stage malignant bowel obstruction using hydration and venting gastrostomy. Am J Surgery 1986; 152: 100-4. Ashby MA, Game PA, Devitt P et al. Percutaneous gastrostomy as a venting procedure in palliative care. Palliative Medicine 1991; 5: 147-50. Stellato TA, Shenk RR. Gastrointestinal emergencies in the oncology patient. Semin Oncol 1989; 16 (6): 521-31. McCarthy D. Strategy for intestinal obstruction in peritoneal carcinomatosis. Archives Surgery 1986; 121: 1081-2. Bruera E, Legris MA, Kuehn N et al. Hypodermoclysis for the administration of fluids and narcotic analgesics in patients with advanced cancer. J Pain and Sympt Manage 1990; 5: 218-20. Walsh HPJ, Schofield PF. Is laparotomy for small bowel obstruction justified in patients with previously treated malignancy? Br J Surg 1984; 71: 933-5. Aranha GV, Folk FA, Greenlee HB. Surgical palliation of small bowel obstruction due to metastatic carcinoma. The American Surgeon 1981; 47: 99-102. Chan A, Woodruff RK. Intestinal obstruction in patients with widespread intra-abdominal malignancy. J Pain Sympt Manage 1992 (in press). Baines M. The pathophysiology and management of malignant intestinal obstruction. In: Oxford Text Book of Palliative Medicine, 1993 (in press).
Received 26 March 1992; accepted 27 May 1992.
Correspondence to: Carla Ripamonti M.D. Pain Therapy and Palliative Care Division National Cancer Institute via Venezian, 1 20133, Milan, Italy
Downloaded from https://academic.oup.com/annonc/article-abstract/4/1/15/205974 by University of California Santa Barbara/Davidson Library user on 25 March 2018