3 True and false large bowel obstruction K. C. R. FARMER R. K. S. PHILLIPS Large bowel obstruction is a common cause of emergency admission to hospital and still results in significant morbidity and mortality. Patients are often elderly and unwell; their survival may be optimized by an appreciation of the aetiology and pathophysiology of large bowel obstruction, combined with a working knowledge of alternative treatments. Experience is important when a variety of options are available. Large bowel obstruction may be defined as being a cessation in the passage of intraluminal contents in the colon and/or rectum thereby producing the clinical features of abdominal distension and constipation. True large bowel obstruction implies the presence of a mechanical cause such as colorectal cancer. The term false large bowel obstruction or colonic pseudoobstruction (Dudley et al, 1958) refers to an interruption in colorectal peristalsis in the absence of mechanical obstruction. AETIOLOGY

OF LARGE

BOWEL

OBSTRUCTION

True large bowel obstruction

Malignant primary tumours are responsible for the great majority of mechanical large bowel obstructions (Table 1). For example, in a series of Table

1. Aetiology

of true large

bowel

obstruction.

Tumour-intrinsic or extrinsic to the bowel wall v01vu1us Diverticulitis Hernia Inflammatory bowel disease Obturation-foreign body, faeces, gall stone Intussusception Endometriosis Iatrogenic: Postoperative adhesions Sepsis Radiation stricture Anastomotic stricture Baillidre’s Clinical Gastroenter-ologyVol. 5, No. 3, September 1991 ISBN&702@1543-1

563 Copyright 0 1991, by Bailhere Tindall All rights of reproduction in any form reserved

564

K.

C.

R. FARMER

AND

R.

K.

S. PHILLIPS

127 cases of complete large bowel obstruction reported by Buechter et al (1988), 90% were due to malignant disease. The incidence of colonic volvulus varies geographically. In developed countries approximately 5% of large bowel obstructions are caused by colonic volvulus (McEntee et al, 1987), whereas in Iran, where diet may be contributory, 42% of large bowel obstructions are caused by colonic volvulus (Ballantyne, 1982). Sigmoid volvulus is the commonest form of vo1vu1us, particularly when that part of the colon is redundant and mobile and the mesentery is narrow. Rarer causes of obstruction include diverticular disease, hernia, inflammatory bowel disease, obturation by faeces, foreign body or gall stones, colonic intussusception, endometriosis, and previous radiotherapy (Bevan, 1982). Large bowel endometriosis may be confused with carcinoma endoscopically and radiologically; the definitive diagnosis is often made after the bowel has been resected (Graham and Mazier, 1988). There is a rising incidence of bowel disease following radiotherapy for gynaecological malignancies. Allen-Mersh et al (1986) reported a 20-year review from the Royal London Hospital and found a 1.5% incidence of rectal stricture, intracavity irradiation being a major risk factor. Colonic pseudo-obstruction In 1948 Ogilvie reported several cases of non-mechanical large bowel obstruction, which he attributed to malignant infiltration of the mesentery and coeliac axis. Later, Dudley et al (1958) coined the term ‘colonic pseudo-obstruction’ to encompass a variety of causes and this is now the preferred term. The pathogenesis of acute colonic pseudo-obstruction remains unclear, although it is known to complicate several conditions, including major systemicillness, trauma, immobility, anticholinergicor narcoticmedications, electrolyte disturbances and endocrine dysfunction (Table 2). Disturbance of the autonomic supply may be responsible. The parasympathetic nerve supply to the colon comes mainly from the vagus nerve, while the distal colon receives additional parasympathetic innervation from S2 to S4. Spira and his colleagues (1976) have speculated that an imbalance between proximal and distal parasympathetic innervation or between parasympathetic and sympathetic innervation could give rise to a functional large bowel obstruction through segmental or generalized loss of motor activity. Chronic colonic pseudo-obstruction (chronic megacolon) is a separate entity (Table 2) and may be due to smooth muscle or neurological disease, including Hirschsprung’s disease, collagen diseases, Chagas’ disease, familial visceral myopathy and familial visceral neuropathy. PATHOPHYSIOLOGY Large bowel obstruction leads to clinically important changes in intestinal motility, intraluminal colonic environment and colonic blood flow.

TRUE

AND

FALSE

LARGE

BOWEL

565

OBSTRUCTION

Table 2. Aetiology of false large bowel obstruction. Acute

colonic

pseudo-obstruction

Intra-abdominal malignancy (Ogilvie’s syndrome) Major systemic illness Electrolyte disturbance Extraperitoneal inflammationihaemorrhage Immobility Pharmacological agents: Opiates Antidepressants Tranquillizers Anticholinergics Acute endocrine dysfunction Chronic

colonic

pseudo-obstruction

Smooth muscle diseases: Visceral myopathy Collagen diseases Muscular dystrophies Metabolic disturbances (e.g. myxoedema, hypoparathyroidism, Neurological diseases: Visceral neuropathy Hirschsprung’s disease Parkinson’s disease Diabetes mellitus Pharmacological agents Chagas’ disease

porphyria)

Colonic motility

Animal models of left-sided colonic obstruction have shown reduced motility in the ileum and right colon and increased contractions just proximal to the site of obstruction (Coxon et al, 1985). Mass actions lasting up to 40s are thought to be responsible for the clinical features of colic, visible motility and tinkling bowel sounds. They may occur 5 days or more before obstruction is total and differ markedly from normal peristaltic waves as they do not progress along the colon in an orderly fashion. Fraser et al (1980) have postulated that mass actions may even play a part in milking tumour cells and/or bacteria into the lymphatic or vascular circulation, thereby accounting for the poorer short- and long-term prognosis in malignant large bowel obstruction. Progressive distension eventually results in general hypomotility; this persists after relief of the obstruction for a period related to its duration. Intraluminal

environment

The intraluminal environment of the large bowel consists of gas and faeces. Faeces is composed of solid matter, fluid and electrolytes. Obstruction affects them all. Gas causes much of the distension in large bowel obstruction. The adequacy of the ileocaecal valve determines whether the distension is

566

K.

C.

R. FARMER

AND

R.

K.

S. PHILLIPS

confined to the colon or extends proximally to affect the small bowel. Most of the gas is nitrogen derived from swallowed air; a lesser proportion is carbon dioxide, hydrogen and methane produced by bacterial fermentation (Chappuis and Cohn, 1987)) and these last two gases carry a risk of explosion when using diathermy either endoscopically or at operation. Nearly 40% of the dry weight of faeces consists of bacteria. Anaerobic species predominate, particularly in the static environment of a large bowel obstruction (Sykes et al, 1976). There is an active secretion of fluids and electrolytes into the bowel lumen in large bowel obstruction, the mechanism of which is poorly understood. It may relate to anaerobic bacterial overgrowth as it does not occur to the same extent in germ-free animals. Intravenously administered fluids increase this active secretion (Chappuis and Cohn, 1987). Colonic blood flow

Local blood flow changes in the obstructed colon have recently been examined in detail in animal models by Coxon et al (1984) and Papanicolaou et al (1989). Using a technique of radiolabelled microsphere injection, Papanicolaou and colleagues (1989) demonstrated a two-fold increase in blood flow to the distended colon and increased flow to the ileum, whereas flow to other organs was unchanged. This elevated blood flow may result in a degree of high output failure in elderly patients after decompression (following resorption of large quantities of fluid), which may account for some of their high morbidity and mortality. However, this good blood supply should facilitate anastomotic healing if a primary resection with anastomosis has been performed. Systemic effects

Active sequestration of fluid and electrolytes into the bowel lumen may deplete the intravascular volume, although this is less common in large bowel obstruction than in small bowel obstruction. This can be compounded by further fluid shifts shortly after intestinal decompression and, combined with the increased colonic blood flow, may lead to postoperative haemodynamic instability. Abdominal distension will exacerbate any pre-existing respiratory condition and may influence the indication for and urgency of surgical decompression. Bacterial and endotoxin translocation into the portal circulation and peritoneal cavity occur in the presence of obstructed and distended bowel (Sykes et al, 1976) and may contribute to postoperative sepsis. CLINICAL

PRESENTATION

AND INVESTIGATIONS

Signs and symptoms

Large bowel obstruction causes abdominal pain, distension and constipation. Vomiting is unusual and a late sign.

TRUE

AND

FALSE

LARGE

BOWEL

567

OBSTRUCTION

The pain is colicky in nature, and is infraumbilical when the hindgut is the source and periumbilical from the right half of the colon, which is a midgut structure. Colonic mass actions occur up to 15-20 min apart and last up to 60 s. They correspond to the pain and tinkling bowel sounds. Abdominal distension is usually marked once the obstruction has become fully established. To begin with, stools become infrequent and flatus continues to be passed but eventually absolute constipation sets in, where neither faeces nor flatus are passed (even after an enema). The patient may give a clue as to the likely cause of obstruction, change of bowel habit and altered blood rectally or a strong family history being suggestive of cancer, recurrent previous episodes that have settled nonoperatively suggesting vo1vu1us, and so on. Abdominal examination will reveal a distended abdomen. Tenderness, particularly over the caecum, is an important sign as it markedly increases the urgency of surgical decompression. Examination may also give clues as to the aetiology of the obstruction, such as evidence of metastatic disease or a strangulated hernia. Rectal examination usually reveals an empty rectum below an obstruction. An obstructing rectal cancer is rare, accounting for only 6% of cases in one series (Phillips et al, 1985). A computer analysis was performed on the spectra of bowel sounds in 231 patients with mechanical small and large bowel obstruction (Yoshino et al, 1990). There were three different types of sound relating to lower, peak and upper frequencies detected by the analysis. These could not be separately identified by auscultation, but could be distinguished from the pattern found in five healthy controls. Increasingly severe clinical obstruction was associated with spread of the higher frequency range and shift of the peak frequency. The severest obstructions showed a loss of high frequency sounds above 900 Hz, perhaps because of intestinal decompensation. Sigmoidoscopy is essential, but is best left until after plain X-rays have been taken so that air insufflation of the rectum and distal colon does not confuse the X-ray appearance. Investigations

Investigation will reveal the current status of the patient and should allow colonic pseudo-obstruction, which can often be managed without recourse to laparotomy, to be distinguished from a mechanical obstruction. Table

3. Investigations

in large

bowel

obstruction.

Initial: Urinalysis Urea and electrolytes Glucose Haemoglobin, white cell count, platelet count Group and save serum Electrocardiogram Plain X-rays--erect and supine abdomen, chest Rigid sigmoidoscopyiflexible endoscopy Contrast enema

568

K.

C. R. FARMER

AND

R. K. S. PHILLIPS

The suggested minimum investigations are listed in Table 3. Hypokalaemia, for example, is not uncommon (Welch and Donaldson, 1974), and a full blood count will disclose anaemia which may need correction and might indicate an underlying malignancy. Plain X-rays of the abdomen and chest usually show gaseous distension of the large bowel with fluid levels and often a cut-off point at the site of obstruction (Figure 1). Volvulus of the sigmoid or caecum may have typical appearances (Figure 2). If the ileocaecal valve remains competent then distension is limited to the large bowel and the caecum can reach massive proportions with a risk of perforation. A caecal diameter of greater than 12 cm carries a significant risk of perforation; however it would appear from a review of 46 patients with caecal distension (Johnson et al, 1985) that the risk of perforation is more related to the duration of distension than to the absolute caecal diameter, the greatest risk being in patients with obstructions lasting longer than 4 days. Gas in the distended large bowel and the rectum raises the possibility of a

Figure 1. Plain erect mid-sigmoid colon.

X-ray

showing

large

bowel

distension,

fluid

levels

and cut-off

point

in

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

569

570

K.

C. R.

FARMER

AND

R.

K.

S. PHILLIPS

colonic pseudo-obstruction, but plain abdominal X-rays can be quite misleading. It is for this reason that contrast enemas are now considered important in all but the most urgent of cases, i.e. where there is impending or actual caecal rupture and laparotomy would have to be performed whatever the cause. Stewart and his colleagues (1984) prospectively examined 117 patients undergoing a single contrast water-soluble enema for suspected large bowel obstruction to determine its bearing on management. In 95 evaluable cases plain X-ray suggested a mechanical obstruction, which was confirmed in 60 patients by contrast enema. However, in 3.5 cases a mechanical cause was ruled out by free flow of contrast into the caecum. On the other hand, of the 17 patients diagnosed as having colonic pseudoobstruction on plain film, although the diagnosis was confirmed in 15 of them, there were two patients who actually had obstructing colonic cancers. An alternative to a contrast enema is flexible endoscopic examination, which affords both diagnostic and therapeutic capabilities (see below). RESUSCITATION

It should be appreciated that expert cardiovascular fluid resuscitation is important in the elderly after intestinal decompression because peroperative third space sequestration of water and electrolytes in the dilated intestine might be exacerbated by high blood flow through the colon which is no longer counterbalanced by high intraluminal colonic pressure. Unstable postoperative patients should be admitted to the intensive care unit for active monitoring, possibly including the use of a Swan-Ganz catheter. Preoperative optimization should also be considered. These measures may well be an important means of reducing mortality and morbidity in large bowel obstruction, as cardiopulmonary complications have been reported in up to 50% of cases (Phillips et al, 1985). A nasogastric tube is rarely indicated in large bowel obstruction, but care of the airway during anaesthetic induction, including the use of firm cricoid pressure to occlude the oesophagus, is important if inhalation of vomit is to be avoided. A urinary catheter facilitates proper fluid management. Antibiotics are indicated in most cases. A short perioperative prophylactic course is recommended unless there is a clear focus of sepsis (e.g. perforation) when more prolonged therapy is appropriate. A cephalosporin or an aminoglycoside combined with metronidazole should cover most pathogenic organisms. DEFINITIVE Colorectal

MANAGEMENT

ACCORDING

TO AETIOLOGY

cancer

Although colorectal cancer is the commonest cause of mechanical large bowel obstruction, only 7-29% of large bowel cancers will present in this way (Ohman, 1982). It is not known precisely what precipitates the actual acute obstruction, but possible contributing factors include the site of the

TRUE

AND

FALSE

LARGE

BOWEL

571

OBSTRUCTION

tumour, fibrosis, inflammatory impaction of solid faeces.

oedema, fatigue of intestinal

muscle and

Site of obstruction Table 4 shows the pattern of distribution of obstructing tumours as reported in several large series. A comparison of the risk of obstruction for a given site was undertaken by Phillips et al (1985) and the results are shown in Table 5. The chance of obstruction for the right and left colon was similar (approximately 23%), but a cancer at the splenic flexure carried a 50% chance of obstruction. Perhaps the acute angle of the colon at this site is a contributing factor. Table

4. Sites of malignant

Reference

n

Welch and Donaldson, Fielding et al, 1979 Phillips et al, 1985 Buechter et al, 1988

1974

Total/Mean

Table

5. Risk

Non-obstruction Obstruction Risk

(n)

of obstruction

Mortality

colon

obstruction. Spfenic

flexure

Left

colon

30% 44% 34% 23%

12% 18% 11% 14%

58% 39% 51% 63%

1101

33%

14%

53%

at each tumour Splenic flexure

857 242

(n)

Right

bowel

124 137 713 127

of obstruction Right colon

large

84 80

22%

49%

site (Phillips

et al, 1985).

Left colon

Rectum and rectosigmoid

784 236

1881 128

(e.g.

Other multiple) 267 27

6%

23%

9%

and long-term survival

Obstructing large bowel cancer carries a disappointingly high morbidity and operative mortality and poor long-term survival when compared with nonobstructed cases. The series of Kaufman et al (1989), for example, revealed an operative mortality for obstructed cases that was three times that for elective patients (Table 6). The two major causes of in-hospital mortality are abdominal sepsis and cardiopulmonary complications. Table

6. Mortality

and survival

of obstructed

versus

Obstructed

Ohman, 1982 Phillips et al, 1985 Kaufman et al, 1989 * Curative

resections

8%* 23% 22% only.

large 5-year (curative

Mortality Reference

non-obstructed

Non-obstructed 7%* 11% 7%

Obstructed 31% 40% 53%

bowel

obstruction.

survival resections) Non-obstructed 50% 63% 76%

572

K.

C. R.

FARMER

AND

R. K.

S. PHILLIPS

Operative morbidity and mortality are in part occasioned by the age and general condition of the presenting patient. Nonetheless, the choice of management and the skill and experience of the surgeon play an important role in minimizing the adverse outcomes. However, the poor long-term survival is less easily explained. Large studies agree that obstructing carcinomas are more advanced and aggressive than non-obstructing tumours (Serpell et al, 1989), but not by such an extent to account adequately for the poor survival. The difference in survival between obstructed and non-obstructed cases occurs in the first 18 months after surgery and thereafter the survival curves run parallel to each other (Figure 3). This suggests that clinical factors, which influence the short and medium term, are more important than pathological factors, which would be expected to be still exerting a difference after 18 months. 1

---

0.9 76 .z ir a z .Z> .3 a z P P I% 2‘G

0.8

-

0.7

--.

-_-_

0.6

___

--, ---

0.5

“‘:--

--_

0.4

_-.

0.3

z 3

Not obstructed Obstructed

---

--

--_ .--.-em

-

---.--.

I-4

---m--m-.

------Wm.

0.2

wI

0.1 0

Not obstr. At risk I (Jbstr, Figure 3. Comparison carcinoma. Modified

Staged or primary

6

12

18

24 Time

30 36 (months)

42

48

54

60

3870

2518

1936

1379

686

340

713

349

240

153

90

44

of survival from Phillips

of obstructed and non-obstructed et al (1985), with permission.

cases

of large

bowel

resection

In 1974 Fielding and Wells suggested that whereas primary resection carried a higher in-hospital mortality than did staged resection, the longer term outcome in survivors was better after primary resection. They speculated that this was because of a period of accelerated tumour growth after operation with colostomy in the staged resection group and once again stimulated the controversy between the two treatment policies.

TRUE

AND

FALSE

LARGE

BOWEL

573

OBSTRUCTION

Table 7. Outcome according to operative approach. Operative mortality Reference Fielding and Wells, 1974 Umpleby and Williamson, 1984 Phillips et al, 1985 Buechter et al, 1988 Serpell et al, 1989

Hospital stay (days)

5-year survival (curative procedures)

II

SR

PR

SR

PR

SR

PR

90 124 713 127 148

4% 20% 22% 28% 7%

18% 17% 19% 25% 6%

49

26

40

20

25% 17% 31% 40% 46%

50% 34% 29% 39% 59%

SR = staged resection; PR = primary resection.

Later authors (Table 7) have failed to confirm the higher in-hospital mortality of primary resection, perhaps because in practice more senior surgeons tend to be involved when this policy is carried out. In addition, when considering the mortality after a staged resection patients who died after one of the earlier stages of the policy but where the surgical intention had been to perform all three stages should also be included. A staged resection requires that the often elderly patient survives three operations, whereas a primary resection with on-table colonic irrigation and primary anastomosis requires that the patient survives just one operation. The claim that longer-term survival is enhanced by primary resection (Serpell et al, 1989) was not substantiated in the report from the Large Bowel Cancer Project (Phillips et al, 1985). Clearly when faced with a particularly high-risk patient a defunctioning stoma may be all that is possible, and this can be done under local anaesthetic if need be (Gutman et al, 1989). Seniority of the surgeon

The seniority of the surgeon does appear to make a significant difference to the outcome if a primary resection is undertaken (Table 8). The reason may simply be that a more senior surgeon commands a more senior team in general and tends to operate in daylight hours if possible. Table

8. Operative mortality according to seniority of surgeon. Operative mortality Staged resection

Reference Fielding et al, 1979 Phillips et al, 1985

Surgeon-in-training 32% 22%

Primary resection

Consultant 22% 21%

Surgeon-in-training 30% 24%

Consultant 10% 13%

Timing of surgery

Despite two-thirds of large bowel obstructions being closed loop in nature by virtue of a competent ileocaecal valve there is seldom the need to rush to theatre. The timing of surgery rests on a balance between needing time to prepare the patient properly from a cardiorespiratory viewpoint and the

574

K.

C. R.

FARMER

AND

R.

K.

S. PHILLIPS

desire to avoid undue delay. A careful deliberate approach is called for and in most cases surgery can be undertaken in daylight hours. Exceptions to this strategy are: (a) when there is unequivocal evidence of peritonitis; (b) when X-ray reveals a pneumoperitoneum or perforation; and (c) if the patient is suffering from severe respiratory distress due to abdominal distension. These situations call for surgery to be conducted as soon as possible. Operative details

It is wise to warn patients of the possibility of a stoma and to mark potential sites preoperatively. With the patient anaesthetized in the lithotomy-Trendelenberg (LloydDavies) position, the abdomen is opened using a midline incision. Before performing a full laparotomy the distended colon should be decompressed using a 21-gauge needle attached to suction tubing. The needle is passed obliquely through a taenia in the transverse colon and the point kept within the gas bubble. The colon will collapse rapidly with this manoeuvre and the entry point can be made safe with a suture if necessary. Exploratory laparotomy will now reveal the full extent of the tumour in terms of local and distant spread. cancers. There is general acceptance amongst colorectal surgeons that lesions proximal to the splenic flexure can be managed with an immediate right hemicolectomy and ileocolic anastomosis. This affords a satisfactory clearance of the cancer and an anastomosis that is safe in experienced hands even in the presence of local sepsis. In 1966 Hughes introduced the concept of extending this type of resection in the form of a subtotal colectomy for lesions distal to the splenic flexure and there are a number of supporters of this approach. Table 9 lists the results of several recent series. Operative mortality and morbidity appear to be low and hospital stay is not prolonged. Other advantages are that it is a single stage procedure which avoids a stoma and obviates the risks of an anastomosis using proximal unprepared bowel. It also simplifies long-term follow-up for metachronous tumours. The major disadvantages are the magnitude of the

Right-sided

Table9.Resultsofsubtotalcolectomyforobstructionwithileosigmoidorileorectalanastomosis. Mean age (years)

n

Deutsch et al, 1983 Morgan et al, 1985 Halevy et al, 1989 Wilson and Gollock, 1989 Stephenson et al, 1990

14 16 22 18

71 69 67 71

1 2 1 2

0 0 1 0

18 NS 26 15

3-4 l-3 l-3 l-3

31

73

1

0

17

l-4

101

70

7 (7%)

1 (1%)

19

l-3

NS = not stated.

Leak

Daily bowel frequency

Reference

Total/Mean

Mortality

Hospital stay (mean in days)

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

575

procedure, the increase in postoperative bowel frequency, and the risk of elderly people with poor anal sphincter tone developing faecal soiling or incontinence. There are, however, theoretical reasons for recommending a more extensive resection when a carcinoma is situated in the region of the splenic flexure based on known variations in colonic vascular anatomy (Griffiths, 1956), since in about 6% of patients the left colic artery is absent and supply to the splenic flexure comes from an enlarged middle colic artery, and in about 22% of cases the middle colic artery is absent and an enlarged right colic artery is involved. As adequate cancer surgery entails removal of the primary tumour with its lymphatic drainage, and the lymphatic drainage follows the arterial blood supply which here is unpredictable, the surgeon should set out to remove the right colic, middle colic and left colic arteries when dealing with splenic flexure carcinoma, whether simple or obstructed (Aldridge et al, 1986). This means that mobilization for an extended right hemicolectomy should be performed for lesions up to and including the splenic flexure and proximal descending colon. Either a caeco-descending/ sigmoid or ileo-descending/sigmoid anastomosis can be constructed, the latter being preferable in obstruction because the caecum may have been damaged by gross distension. Left-sided curzcers. The traditional management of an obstructing carcinoma of the left colon is a three stage approach (decompressive colostomy, resection of tumour and delayed closure of colostomy) (Welch and Donaldson, 1974; Irvin and Greaney, 1977). With greater understanding of the pathophysiology, better resuscitation and safer anaesthetics, primary resection of obstructing lesions distal to the splenic flexure has become more acceptable (Valerio and Jones, 1978; Fielding et al, 1979; Phillips et al, 1985; Huddy et al, 1988). Following resection, an end colostomy and mucous fistula or a Hartmann’s procedure have been popular because of the anticipated risks of an immediate colocolic anastomosis using distended unprepared proximal colon. A major contributing factor to anastomotic breakdown is faecal loading (Irvin and Goligher, 1973), although there are some who would challenge this (Irving and Scrimgeour, 1987). A significant advance in the management of malignant large bowel obstruction has been the recent reintroduction of techniques for intraoperative mechanical cleansing of the proximal colon. Dudley et al (1980) have described a technique of antegrade on-table lavage and, although others have developed individual variations (White and Macfie, 1985; Vigder et al, 1985; Munro et al, 1987; Hardy et al, 1989), the underlying principle remains the same and is becoming more widely accepted as a means of facilitating an immediate colocolic anastomosis for selected patients with left-sided obstructions. Experimental evidence also lends support to this approach. Not only did Smith et al (1983) demonstrate that a clean colon was associated with a lower incidence of anastomotic breakdown, Foster and colleagues (1986) showed that intraoperative lavage favoured increased collagen synthesis at the anastomotic site.

576

K.

C.

R. FARMER

AND

R. K.

S. PHILLIPS

The essential steps are as follows. Having evacuated colonic gas as described above, the large bowel is fully mobilized and the appropriate resection completed. A 20 French Foley catheter is then inserted either into the caecum through the appendix stump or the base of the caecum if the appendix is not present, or transileally through the ileocaecal valve. A purse-string suture prevents leakage and the catheter is secured by inflating the balloon. Anaesthetic scavenger tubing is inserted into the colon close to the chosen point of anastomosis and firmly secured by tapes. The other end is tied into a large plastic bag (Figure 4). The colon is flushed with 4 1of warm (37°C) Hartmann’s solution while gently massaging the faeces along the bowel. When the effluent is clear the remaining solution in the colon is syphoned out by placing the infusion bag on the floor. The caecostomy should not be maintained as it can be a source of added morbidity (Pollock et al, 1987). A peritoneal lavage seems advisable following this procedure. Experience has shown that once the technique is mastered the majority of patients with left-sided obstructions are suitable (Koruth et al, 1985). One

----,l ,-7.

I

i:-.=il-

I

?

Figure 4. Illustration catheter; B = double pressures.

of on-table irrigation apparatus. linen tapes to secure anaesthetic

A = purse-string suture around Foley tubing; C= 19-gauge needle to equalize

TRUE

AND

FALSE

LARGE

BOWEL

577

OBSTRUCTION

needs to exercise some judgement, however, in the selection of patients. Situations where it would be wise to be cautious include advanced peritonitis, concern about bowel viability, immunosuppression, poor general condition and cardiovascular instability. The results of five series reporting the clinical outcome in selected patients of on-table lavage, primary resection and immediate anastomosis are found in Table 10 and indicate that the procedure can be safely carried out with minimal morbidity and mortality, without a stoma and with a brief hospital stay. Table

Reference

10.

Clinical outcome of on-table lavage.

n

Radcliffe and Dudley, 1983 50 Koruth et al, 1985 47 Thomson and Carter, 1986 126* Pollock et al, 1987 41 Konishi et al, 1988 25$ Total/Mean 289

Hospital stay 13 13 17 12 NS 14

Wound infection

Clinical leak

Hospital mortality

6% 2% 3% 5% 0% 3.2%

4% 8% 5% O%T 4% 4.2%

2% 9% 4% 17% 4% 7.2%

Added operation time NS 39 min 30-45 min 60 min 30 min 42 min

NS = not stated. * 20 were acutely obstructed. i 9.8% with radiological leak. $ 4 with sigmoid volvulus.

It should be emphasized that this aggressive approach to left-sided obstructions requires experience, patience and a certain degree of enthusiasm if one is to perform the procedure without local contamination and maintain a minimum operative mortality. If there is any doubt, deferral of the anastomosis or a covering stoma is recommended. Endoscopic

management

The flexible endoscope has an established place in the diagnosis and treatment of a variety of elective colorectal conditions. Its use in acute situations, however, is less well defined but is gaining favour. Flexible endoscopy in large bowel obstruction not only enables a diagnosis to be made but also offers the possibility of converting an acutely obstructed case into an elective non-obstructed case, with a concomitant improvement in operative mortality and morbidity and the possibility of avoiding a stoma. Three methods to achieve this aim have been described: tube decompression, laser recanalization and balloon dilatation. Lelcuk et al (1986) reported three cases of acute obstruction at the level of the sigmoid colon that were successfully decompressed by a flatus tube passed over an endoscopically directed guide-wire through a pinhole lumen. The authors stressed that this technique is not suitable for obstructions more proximal than the sigmoid colon because of difficulty in directing the decompression tube above that level. In the past few years lasers have been used endoscopically to recanalize advanced tumours of the oesophagus and major airways. More recently the

578

K.

C. R.

FARMER

AND

R.

K.

S. PHILLIPS

neodymium-yttrium aluminium garnet (Nd : YAG) laser has been shown to be effective in alleviating malignant obstruction in the large bowel (Bown et al, 1986; Kiefhaber et al, 1986). The greatest risk is perforation, particularly with lesions above the peritoneal reflection. Low et al (1989) reported on 26 obstructing tumours (18 rectal, eight colonic) and demonstrated that, in experienced hands, use of the laser is a safe and effective means of relieving obstruction. An alternative means of endoscopic decompression using balloon dilatation was recently described by Stone and Bloom (1989). A 54 or 60 French balloon catheter was used in three cases of acute sigmoid or rectal obstruction and in two patients was combined with Nd : YAG laser therapy. The authors recommend using a dual channel endoscope because this enables optimal suction capability. Experience with these endoscopic modalities is limited and for the present should be confined to institutions which are in a position to refine the techniques, determine the indications and risks, and analyse the outcome. Palliative measures

Occasionally the surgeon is confronted with a patient with extensive malignant disease and resection is inappropriate. Apart from endoscopic manoeuvres, ileocolic or colocolic bypass may be performed, or sometimes just a simple defunctioning stoma. In general these palliative manoeuvres are worthwhile as they give the patient relief from the distressing symptoms of obstruction. Colonic volvulus

Up to 80% of colonic volvulus involves the sigmoid colon, the caecum is next most commonly involved and rarely the transverse colon will undergo torsion (Ballantyne, 1990). Sigmoid volvulus in developed countries most often occurs in elderly, institutionalized patients, and predisposing factors include a redundant mobile colon and an accompanying narrow mesentery. It is much more common in parts of the Middle East and Africa, where dietary differences are probably important, and it is also common in people who live at high altitude (e.g. Bolivia), probably because the low atmospheric pressure leads to gas expansion and also because many peasants chew coca leaves which are constipating. In Brazil most cases of sigmoid volvulus are due to Chagas’ disease. If the diagnosis has not been made from the plain X-rays, then a watersoluble contrast enema will usually demonstrate the site of the volvulus. In the absence of signs of strangulation or perforation, the management can be divided into two phases: decompression and definitive treatment. Sigmoid volvulus

Traditionally the rigid sigmoidoscope and flatus tube (Bruusgaard, 1947) have been used to decompress a sigmoid volvulus, and this is successful in up

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

579

to 80% of cases. More recently the flexible endoscope has improved this rate to 90% or more (Arigbabu et al, 1985). Although non-operative decompression relieves the vo1vu1us, it does not prevent recurrence, which occurs in over 40% of cases (Shepard, 1968). Most surgeons therefore recommend that decompression should be followed by a definitive sigmoid resection during the same hospital admission in those who are fit. Obviously if decompression fails or if there are signs of strangulation or perforation then emergency surgery is indicated. Primary resection and anastomosis as described above can be performed but in the worst cases with gangrenous bowel or gross peritonitis a Paul-Mikulicz resection with double-barrel colostomy or a Hartmann’s operation should be considered. The expected mortality of a patient with sigmoid volvulus with viable bowel is approximately 12%. If the bowel is non-viable the mortality rises to over 50% (Ballantyne, 1990). Caecal volvulus

There have been sporadic reports of colonoscopic decompression of caecal volvulus (Anderson et al, 1978); however the success rate is low in most series and the risk of perforation is significant in unskilled hands (Friedman et al, 1989). Accordingly the current choice of definitive management following urgent operative detorsion lies between caecopexy, formation of a tube caecostomy and right hemicolectomy. Caecopexy has fallen into disrepute because of an unacceptable rate of recurrent volvulus (Ballantyne, 1990). Proponents of tube caecostomy claim that it is a simple and quick procedure and that recurrence is rare (Todd and Forde, 1979). However, caecostomies are not without mortality and can be associated with severe local sepsis. In addition, they can be difficult to manage and may need to be closed formally. Consequently there is much to recommend resection in most circumstances, and certainly when the bowel is gangrenous. Diverticular

disease

Diverticular disease is responsible for a small proportion of mechanical large bowel obstructions. The obstruction may be due to progressive fibrosis or secondary to pericolic sepsis. It can often be difficult for a radiologist to distinguish diverticular from malignant obstruction on a contrast enema and it is not unusual for even experienced surgeons to find it impossible to be certain of the pathology at laparotomy. Nevertheless, the principles that have been discussed in relation to malignant large bowel obstruction apply equally well to diverticular obstruction, and immediate resection is to be recommended, together with drainage of sepsis if present. With advanced local sepsis, on-table lavage and immediate anastomosis is best avoided. Hernia

Occasionally a segment of colon becomes obstructed in either an external or

580

K.

C. R.

FARMER

AND

R. K.

S. PHILLIPS

internal hernia. At operation if the colon is viable then all that is required is for it to be returned to the abdominal cavity and the hernia repaired. If resection of non-viable bowel is necessary then a formal laparotomy is advised. Obturation

The large bowel may become obstructed by impaction of an intraluminal object or substance (e.g. a gall stone, barium, faeces or a foreign body). Gall stones and other foreign bodies can be delivered via a small colotomy made proximal to the site of obstruction. Other pliable substances can usually be massaged along the colon with or without the aid of on-table lavage. A lesion at the site of obstruction should be excluded and the length of the intestine palpated proximally for any further sources of potential obstruction in order to prevent an early recurrence. Radiation

stricture

Radiation stricture presents a difficult problem. The patient’s past history lends a clue to the diagnosis and the possibility of recurrent malignancy should not be overlooked. The situation may be compounded by the presence of a fistula (Allen-Mersh et al, 1986). Problems with tissue healing because of a poor blood supply strongly influence the surgical options. A defunctioning stoma may be a wise choice. If resection is contemplated at least one of the bowel ends to be anastomosed should come from outside the radiotherapy field (Galland and Spencer, 1987). With a rectal stricture after radiotherapy the splenic flexure is a suitable choice, often anastomosed endo-anally to the dentate line after residual mucosal proctectomy has been performed (Parks et al, 1978). Colonic pseudo-obstruction

A colonic pseudo-obstruction may be suspected clinically and from the appearance of the plain abdominal X-rays. Even so, it is imperative that a mechanical obstruction be excluded by performing a contrast enema or colonoscopy. There are three therapeutic alternatives in this condition: medical, endoscopic or surgical treatment. Medical treatment

Provided there are no signs of perforation and the caecal diameter is less than 12 cm, initial conservative management is appropriate. This consists of correcting fluid and electrolyte imbalance, maintaining the patient nil-bymouth, and twice daily enemas and/or passage of a proctoscope. Incriminating medications (especially opiates) should be stopped or the drug regimen at least rationalized. Associated medical conditions should be treated vigorously. Daily X-rays to monitor caecal diameter may be of value in

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

581

judging the effectiveness of the treatment. Caecal perforation is associated with a mortality of up to 50% (Strode1 and Brothers, 1989) and every effort must be made to prevent this catastrophic complication. Endoscopic

treatment

Controversy exists over the value of colonoscopic decompression in this condition. Sloyer and colleagues (1988) reported satisfactory resolution in 23 out of 24 patients treated along traditional medical lines without colonoscopy. Only one patient required a caecostomy for failure of resolution. The mean time to resolution in the remainder was 1.6 days and 90% had resolved by 4 days. The authors felt that colonoscopy would not have hastened resolution in these patients and pointed to the potential risks of endoscopy in distended thin-walled bowel. There were no cases of caecal perforation in this series. Kukora and Dent (1977) were the first to describe colonoscopic decompression in six patients with acute colonic pseudo-obstruction, and there are now several other reports (Bode et al, 1984; Strode1 and Brothers, 1989). Colonoscopy in these cases can be a tedious affair and must be performed by an experienced endoscopist. Several preparatory enemas are advisable to clear viscous liquid stool. Frequent irrigation may be necessary to maintain visibility as the instrument is advanced. Once the right colon has been reached a long fenestrated wide-bore nasogastric tube can be railroaded into position with the aid of a length of monofilament nylon previously sited in the biopsy channel of the endoscope. Residual gas and liquid are aspirated during withdrawal of the instrument. The long intestinal tube needs to be regularly irrigated to maintain patency. Gosche et al (1989) have reviewed the cumulated results of this technique from nine series with a total of 169 patients, many of whom would have otherwise undergone surgery. The initial success rate was 84%, with a 25% recurrence rate. Definitive treatment was achieved in 85% overall at the expense of a 2% mortality. Endoscopic decompression therefore appears to be a worthwhile endeavour for committed endoscopists in patients who do not settle rapidly with initial medical measures and in whom emergency surgery is not required. Surgery Surgery is indicated in acute colonic pseudo-obstruction when there are signs of impending or actual colonic perforation, and in the presence of an increasing caecal diameter (greater than 12cm) despite medical or endoscopic intervention. ‘Blind’ tube caecostomy can be performed under general or local anaesthesia and is recommended when there is no evidence of perforation and the sole aim of the operation is to deal with the dilated colon. A large Foley catheter is ideally suited to the procedure. Soreide et al (1977) reported a 12.5% mortality for tube caecostomy in a collected series of 47 patients over a 30-year period.

582

K.

C. R.

FARMER

AND

R. K.

S. PHILLIPS

In the event of perforation laparotomy is advised. If the caecal hole is small and the surrounding bowel wall is healthy then it may be fashioned into a caecostomy. Occasionally perforations are multiple and resection is required.

SUMMARY

Acute large bowel obstruction can be the result of mechanical causes (such as colorectal cancer) or motility disturbances, the latter being termed colonic pseudo-obstruction. Whatever the aetiology, the pathophysiology of large bowel obstruction has clinical significance. Changes in motility augmented by increased colonic blood flow may play a role in dissemination of tumour cells and/or bacteria. Intravascular fluid depletion, especially shortly after intestinal decompression, has important haemodynamic implications. The diagnosis is often confirmed on plain abdominal X-ray, but watersoluble contrast studies are important in distinguishing a mechanical obstruction (which almost always requires an operation) from a pseudoobstruction (which can usually be managed without surgery). Mortality and morbidity may be reduced by optimization of the patient’s condition both before and after the operation using intensive care facilities and by careful timing of surgery. The surgical management of malignant large bowel obstruction is best directed by a senior surgeon. For tumours up to and including the splenic flexure, an extended right hemicolectomy is advisable since it offers adequate removal of the tumour and allows an immediate safe ileocolic anastomosis. More distal tumours should be resected if possible, and there is much to recommend on-table irrigation and immediate anastomosis, although a colostomy with a mucous fistula or Hartmann’s procedure still have a place. Endoscopic diagnosis and decompression enables definitive surgery to be undertaken electively and several techniques are being evaluated. Non-operative reduction of sigmoid volvulus by rigid or flexible endoscopy is achieved with high success rates, but is not recommended for caecal volvulus. Resection is usually necessary in both to prevent recurrence. Mortality of colonic volvulus is closely related to bowel viability. Uncomplicated colonic pseudo-obstruction may be managed medically or by endoscopic decompression. It often occurs in association with systemic medical conditions, which need to be treated vigorously. Surgery is indicated if there are signs of impending or frank perforation, or if nonoperative measures fail. Acknowledgements Mr Farmer Jill Maybee

is supported by the St Mark’s for photographic assistance.

Research

Foundation.

The authors

wish to thankMiss

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

583

REFERENCES Aldridge MC, Phillips RKS, Hittinger R, Fry JS & Fielding LP (1986) Influence of tumour site on presentation, management and subsequent outcome in large bowel cancer. British Journal of Surgery 73: 663-670. Allen-Mersh TG, Wilson EJ, Hope-Stone HF & Mann CV (1986) Has the incidence of radiation-induced bowel damage following treatment of uterine carcinoma changed in the last 20 years? Journal of the Royal Society of Medicine 79: 387-390. Anderson MJ, Okike N & Spencer RJ (1978) The colonoscope in cecal volvulus: report of three cases. Diseases of the Colon and Rectum 21: 71-74. Arigbabu AO, Badejo OA & Akinola DO (1985) Colonoscopy in the emergency treatment of colonic volvulus. Diseases of the Colon and Rectum 28: 795-798. Ballantyne GH (1982) Review of sigmoid volvulus: clinical patterns and pathogenesis. Diseases of the Colon and Rectum 25: 823-830. Ballantyne GH (1990) Volvulus of the colon. In Fazio VW (ed.) Current Therapy in Colon and Rectal Surgerv. DIJ 254-265. TorontoiPhiladelohia: B. C. Dicker. Bevan PG (1982) Acute intestinal obstruction in the’ adult. British Journal of Hospital Medicine 28:258-265. Bode WE, Beart RW Jr, Spencer RJ, Culp CE, Wolff BG & Taylor BM (1984) Colonoscopic decompression for acute pseudoobstruction of the colon (Ogilvie’s syndrome). American Journal of Surgery 147: 243-245. Bown SG, Barr H, Matthewson K et al (1986) Endoscopic treatment of inoperable colorectal cancers with the Nd : YAG laser. British Journal of Surgery 73: 949-952. Bruusgaard C (1947) Volvulus of the sigmoid colon and its treatment. Surgery 22: 466-478. Buechter KJ, Boustany C, Cailliouette R & Cohn I(l988) Surgical management of the acutely obstructed colon: a review of 127 cases. American Journal of Surgery 156: 163-168. Chappuis CW & Cohn I Jr (1987) Pathophysiological effects on intraluminal contents. In Fielding LP & Welch J (eds) Intestinal Obstruction, pp 32-40. Edinburgh: Churchill Livingstone. Coxon JE, Dixon C & Taylor I (1984) Changes in intestinal blood flow during the development of chronic large bowel obstruction. British Journal of Surgery 71: 795-798. Coxon JE, Dixon C & Taylor I (1985) Changes in colonic motility during the development of chronic large bowel obstruction. British Journal of Surgery 72: 690-693. Deutsch AA, Zelikovski A, Sternberg A & Reiss R (1983) One-stage subtotal colectomy with anastomosis for obstructing carcinoma of the left colon. Diseases of the Colon and Rectum 26: 227-230. Dudley HAF, Sinclair IS, McLaren IF; McNair TJ & Newsam JE (1958) Intestinal pseudoobstruction. Journal of the Royal College of Surgeons of Edinburgh 3: 206-217. Dudley HAF, Radcliffe AG & McGeehan D (1980) Intraoperative irrigation of the colon to permit primary anastomosis. British Journal of Surgery 67: 80-81. Fielding LP & Wells BW (1974) Survival after primary and staged resection for large bowel obstruction caused by cancer. British Journal of Surgery 61: 16-19. Fielding LP, Stewart-Brown S & Blesovsky L (1979) Large-bowel obstruction caused by cancer: a prospective study. British Medical Journal 2: 515-517. Foster ME, Johnson CD, Billings PJ, Davies PW & Leaper DJ (1986) Intraoperative antegrade lavage and anastomotic healing in acute colonic obstruction. Diseases of the Colon and Rectum 29: 255-259. Fraser ID, Condon RE, Schulte WJ, DeCosse JJ & Cowles VE (1980) Intestinal motility changes in experimental large bowel obstruction. Surgery 87: 677-682. Friedman JM, Odland MD & Bubrick MP (1989) Experience with colonic volvulus. Diseases of the Colon and Rectum 32: 409-416. Galland RB & Spencer J (1987) Natural history and surgical management of radiation enteritis. British Journal of Surgery’ 74: 742-747. Gosche JR, Sharoe JN & Larson GM (1989) Colonoscooic decomoression for oseudoI obstruction 0; the colon. American Surgeon 55: 111-115: Graham B & Mazier WP (1988) Diagnosis and management of endometriosis of the colon and rectum. Diseases of the Colon and Rectum 31: 952-956. u

,/II

584

K.

C.

R. FARMER

AND

R. K.

Griffiths JD (1956) Surgical anatomy of the blood supply of the distal colon. Annuls College

of Surgeons

of England

S. PHILLIPS

ofthe

Royal

19: 241-256.

Gutman M, Kaplan 0, Skornick V, Greif F, Kahn P & Rozin RR (1989) Proximal colostomy: still an effective emergency measure in obstructing carcinoma of the large bowel. Journal of Surgical

Oncology

41: 21C-212.

Halevy A, Levi J & Orda R (1989) Emergency subtotal colectomy. Annals of Surgery 210: 22@223. Hardy TG, Aguilar PS & Stewart WRC (1989) Complete obstruction of the sigmoid colon treated by primary resection and anastomosis-an improved technique (preliminary report). Diseases of the Colon and Rectum 32: 528-532. Huddy SP, Shorthouse AJ & Marks CG (1988) The surgical treatment of intestinal obstruction due to left sided carcinoma of the colon. Annals of the Royal College of Surgeons of England

IO: N-42.

Hughes ESR (1966) Mortality of acute large bowel obstruction. British Journal of Surgery 53: 593-594. Irvin TT & Goligher JC (1973) Aetiology of disruption of intestinal anastomoses. British Journal

of Surgery

60: 461-464.

Irvin TT & Greaney MG (1977) The treatment of colonic cancer presenting with intestinal obstruction. British Journal of Surgery 64: 741-744. Irving AD & Scrimgeour D (1987) Mechanical bowel preparation for colonic resection and anastomosis. British Journal of Surgery 74: 580-581. Johnson CD, Rice RP, Kelvin FM, Foster WL & Williford ME (1985) The radiologic evaluation of gross cecal distension: emphasis on cecal ileus. American Journal of Roentgenology 145: 1211-1217. Kaufman Z, Eltch E & Dinbar A (1989) Completely obstructive colorectal cancer. Journal of Surgical

Oncology

41: 230-235.

Kiefhaber P, Kiefhaber K & Huber F (1986) Preoperative neodymium-YAG laser treatment of obstructive colon cancer. Endoscopy 18: 44-46. Konishi F, Muto T, Kanazawz K & Morioka Y (1988) Intraoperative irrigation and primary resection for obstructing lesions of the left colon. International Journal of Colorectal Disease

3: 204-206.

Koruth NM, Krukowski ZH, Youngson GG et al (1985) Intra-operative colonic irrigation in the management of left sided large bowel emergencies. British Journal of Surgery 72: 708-711. Kukora JS & Dent TL (1977) Colonoscopic decompression of massive nonobstructive cecal dilation. Archives of Surgery 112: 512-517. Lelcuk S, Klausner JM, Merhav A, Ratan J, Skornick Y & Rozin RR (1986) Endoscopic decompression of acute colonic obstruction. Annals of Surgery 203: 292-294. Low DE, Kozarek RA, Ball TJ & Patterson DJ (1989) Colorectal neodymium-YAG photoablative therapy. Archives of Surgery 124: 684-688. McEntee G, Pender D, Mulvin D et al (1987) Current spectrum of intestinal obstruction. British

Journal

of Surgery

74: 976-980.

Morgan WP, Jenkins N, Lewis P & Aubrey DA (1985) Management of obstructing carcinoma of the left colon by extended right hemicolectomy. American Journal of Surgery 149: 327-329. Munro A, Steele RJC & Logie JRC (1987) Technique for intraoperative colonic irrigation. British Journal of Surgery 74: 1039-1040. Ogilvie H (1948) Large-intestine colic due to sympathetic deprivation. A new clinical syndrome. British Medical Journal ii: 671-673. Ohman U (1982) Prognosis in patients with obstructing colorectal carcinoma. American Journal

of Surgery

143: 742-747.

Papanicolaou G, Ahn YK, Nikas DJ & Fielding LP (1989) Effect of large-bowel obstruction on colonic blood flow: an experimental study. Diseases of the Colon and Rectum 32: 673-679. Parks AG, Allen CLO, Frank JD et al (1978) A method of treating post-irradiation rectovaginal fistula. British Journal of Surgery 65: 417-421. Phillips RKS, Hittinger R, Fry JS & Fielding LP (1985) Malignant large bowel obstruction. British

Journal

of Surgery

72: 296-302.

Pollock AV, Playforth MJ &Evans M (1987) Peroperative lavage of the obstructed left colon to allow safe primary anastomosis. Diseases of the Colon and Rectum 30: 171-173.

TRUE

AND

FALSE

LARGE

BOWEL

OBSTRUCTION

585

Radcliffe AG & Dudley HAF (1983) Intraoperative antegrade irrigation of the large intestine. Surgery, Gynecology and Obstetrics 156: 721-723. Serpell JW, McDermott FT, Katrivessis H & Hughes ESR (1989) Obstructing carcinomas of the colon. British Journal of Surgery 76: 965-969. Shepard JJ (1968) Treatment of volvulus of the sigmoid colon: a review of 425 cases. British Medical Journal i: 280-283. Sloyer AF, Panella VS, Demas BE et al (1988) Ogilvie’s syndrome: successful management without colonoscopy. Digestive Diseases and Sciences 33: 1392-1396. Smith SRG, Connolly JC & Gilmore OJA (1983) The effect of faecal loading on colonic anastomotic healing. British Journal of Surgery 70: 49-50. Soreide 0, Bjerkest T & Fossdal JE (1977) Pseudo-obstruction of the colon (Ogilvie’s syndrome). A genuine clinical condition? Diseases of the Colon and Rectum 20: 487-491. Soira IA. Rodriaues R & Wolff WI (1976) Pseudo-obstruction of the colon. American Journal L of Gastroen;erology 65: 397-408. ’ Stephenson BM, Shandall AA, Farouk R & Griffith G (1980) Malignant left-sided large bowel obstruction managed by subtotal/total colectomy. British Journal of Surgery 77: 10981102. Stewart J, Finan PJ, Courtney DF & Brennan TG (1984) Does a water soluble contrast enema assist in the management of acute large bowel obstruction: a prospective study of 117 cases. British Journal of Surgery 71: 799-801. Stone JM & Bloom RJ (1989) Transendoscopic balloon dilatation of complete colonic obstruction. An adjunct in the treatment of colorectal cancer: report of three cases. Diseases of the Colon and Rectum 32: 429-431. Strode1 WE & Brothers T (1989) Colonoscopic decompression of pseudo-obstruction and volvulus. Surgical Clinics of North America 69: 1327-1335. Sykes PA, Boulte; KH & Schoheld PF (1976) The microflora of the obstructed bowel. British Journal of Surgery 63: 721-725. Thomson WHF & Carter SStC (1986) On-table lavage to achieve safe restorative rectal and emergency left colonic resection without covering colostomy. British Journal of Surgery 73: 61-63. Todd GJ & Forde KA (1979) Volvulus of the caecum: choice of operation. American Journal of Surgery 138: 632-634. Umpleby HC & Williamson RCN (1984) S urvival in acute obstructing colorectal carcinoma. Diseases of the Colon and Rectum 27: 299-304. Valerio D &Jones PF (1978) Immediate resection in the treatment of large bowel emergencies. British Journal of Surgery 65: 712-716. Vigder L, Tzur N, Huber M, Mahagna M & Amir I (1985) Management of obstructive carcinoma of the left colon: comparative study of staged and primary resection. Archives of Surgery 120: 825-828. Welch JP &Donaldson GA (1974) Management of severe obstruction of the large bowel due to malignant disease. American Journal of Surgery 127: 492-499. White CM & Macfie J (1985) Immediate colectomy and primary anastomosis for acute obstruction due to carcinoma of the left colon and rectum. Diseases of the Colon and Rectum 28: 155-157. Wilson RG & Gollock JM (1989) Obstructing carcinoma of the left colon managed by subtotal colectomy. Journal of the Royal College of Surgeons of Edinburgh 34: 25-26. Yoshino H, Abe Y, Yoshiro T & Ohsato K (1990) Clinical application of spectral analysis of bowel sounds in intestinal obstruction. Diseases of the Colon and Rectum 33: 753-757.