The most common cause of small bowel obstruction (SBO) is intra-abdominal adhesions following laparotomy; this accounts for about three fourths of all cases.^1,² Peritoneal adhesions are common after laparotomy and are exacerbated by intra-abdominal infection, the tissue ischemia attending wound closure (e.g., an anastomosis), external beam radiation, and the presence of foreign material, such as sutures. Although SBO can occur any time after laparotomy, the risk is greatest in the first few postoperative years. Lower abdominal or pelvic operations have a higher risk of adhesive SBO than do upper abdominal procedures, such as cholecystectomy.

Several population-based and large case studies have defined the risk of adhesive obstruction after common abdominal operations. For example, the risk of adhesive SBO after appendectomy is about 1% with 30 years of follow-up; the risk is greater for patients with perforated appendicitis (2.76% at 30 years) than nonperforated appendicitis (0.75% at 30 years).³ The incidence of adhesive SBO after gynecologic operations is similar to that of appendectomy except for cesarean delivery, which has a lower risk of subsequent SBO. Using Medicare data, Beck and colleagues found that the incidence of adhesive obstruction after intestinal resection and anastomosis was 14.3% within two years.⁴

About one fourth of patients who present with SBO have an etiology other than intraperitoneal adhesions. Of these, the most common causes are Crohn’s disease (7%) (see Chapter 111), intra-abdominal neoplasms (5%), and abdominal wall hernias (2%).² A comprehensive list of causes of intestinal obstruction is shown in Table 119-1.

Table 119-1 Causes of Intestinal Obstruction

Intrinsic Bowel Lesions

Congenital atresia or stenosis

Inflammatory causes

Diverticulitis

Inflammatory bowel disease (e.g., Crohn’s disease)

Neoplastic strictures

Surgical anastomosis

Extrinsic Bowel Lesions

Abdominal wall hernias (e.g., umbilical) and incisional hernias, as well as inguinal and femoral hernias, are much more common causes of SBO than internal or intra-abdominal hernias (e.g., paraduodenal hernias). Less than 10% of patients presenting to the hospital with SBO have a hernia as the cause of their obstruction,^2,⁵ and up to 30% of patients requiring an operation for SBO have an incarcerated hernia as the etiology of their obstruction.¹

Intestinal obstruction caused by a hernia has a particularly high risk of strangulation, failure to resolve spontaneously, and recurrence when it is not surgically corrected. In one study of 877 patients, three fourths of patients presenting with incarcerated hernias and SBO had ischemic bowel at the time of operation; the bowel was necrotic in 27%.¹ In contrast, only 29% of patients presenting with SBO due to adhesions had a strangulated obstruction, and only 11% had nonviable bowel.¹ The increased risk of obstruction and strangulation is due, at least in part, to the rigid fascial defect through which the herniated intestine passes. Femoral hernias, in particular, pose a high risk of intestinal strangulation. Although SBO from a groin hernia can occur at any age, it is particularly prevalent in the elderly; advanced age, concomitant chronic illnesses, and treatment delay are associated with unfavorable outcomes in this subset of patients presenting with SBO.⁶

Internal hernias may be congenital (e.g., paraduodenal) or acquired (e.g., hernias through mesenteric defects created in the performance of intestinal anastomoses). The 3% incidence of internal herniation of the Roux limb after gastric bypass for weight loss is a particularly important example of an internal hernia, given the frequency with which this procedure is performed today.⁷ Congenital and acquired internal hernias are discussed in greater detail in Chapter 24.

Various authors have reported herniation of a portion of the bowel wall (Richter’s hernia) or a whole segment of the intestine through a laparoscopic trocar site with resultant intestinal obstruction.^8,⁹ The incidence of trocar site hernias is 1% to 2%, and intestinal obstruction is significantly less common.⁸ In these instances, the hernias usually occur at 10-mm port sites positioned at or close to the midline.⁹ Intestinal obstruction after laparoscopic transabdominal preperitoneal herniorrhaphy usually is due to herniation of the bowel through a defect in the peritoneal closure. In these instances, the bowel is tethered by adhesions between the partially peritoneal-covered prosthesis and the intestine, with formation of a kink or a point of torsion.

Neoplasms are a relatively unusual cause of SBO, accounting for less than 10% of cases ^2,¹⁰; adenocarcinoma accounts for more than 50% of instances of colonic obstruction by neolasms. In more than 90% of such neoplastic SBO, the small intestine becomes obstructed by extrinsic compression or local invasion from advanced gastrointestinal or gynecologic malignancies. Advanced colorectal cancer and ovarian adenocarcinoma are the two most common malignancies associated with SBO. Hematogenous metastases from breast cancer, melanoma, or Kaposi’s sarcoma also can involve the small intestine, with subsequent obstruction. Primary neoplasms of the small intestine, of which adenocarcinoma and carcinoid tumors are most common, are the cause of SBO in less than 3% of cases (Chapter 121).

PATHOPHYSIOLOGY

The duration and degree of obstruction and the presence and severity of ischemia determine the local and systemic consequences of intestinal obstruction. The intestinal mucosa is an important and early site of injury in both simple and strangulated intestinal obstruction. Microscopic evidence of epithelial injury occurs within the first four to six hours of simple intestinal obstruction and progresses to focal epithelial necrosis within 8 to 12 hours.¹¹ Strangulated obstruction exacerbates the injury, causing extensive mucosal necrosis and sloughing.

Intestinal obstruction causes the profound accumulation of fluid, swallowed air, and gas within the intestinal lumen proximal to the site of obstruction. Fluid accumulates because of impaired water and electrolyte absorption and also enhanced secretion, which results in the net movement of isotonic fluid from the intravascular space into the intestinal lumen. The accumulation of swallowed air, and to a lesser extent gases generated by bacteria within the obstructed bowel (e.g., hydrogen, carbon dioxide, and methane), contributes to intestinal distention.

Failure of normal intestinal motility causes bacterial overgrowth within the small intestine and loss of the normally increasing concentration gradient of bacteria from the jejunum to the ileum. Disruption of the ecologic balance of the normal enteric microflora is associated with the translocation of bacteria to mesenteric lymph nodes and systemic organs. In a study by Deitch, enteric bacteria, particularly Escherichia coli, were cultured from mesenteric lymph nodes in nearly 60% of patients with simple intestinal obstruction compared with only 4% of controls.¹² These observations are consistent with experimental studies that described the translocation of bacteria into the submucosa within 36 minutes of simple SBO.¹³ Together, these data are consistent with the hypothesis that translocating enteric bacteria contribute to the septic consequences of SBO.

The systemic manifestations of SBO are related to hypovolemia and the inflammatory response incited by bacterial translocation, with or without the presence of ischemic or gangrenous intestine. Hypovolemia primarily results from the loss of fluid into the intestinal lumen, the bowel wall, and the peritoneal cavity. When combined with anorexia and vomiting, a marked reduction in intravascular volume results. Intestinal ischemia markedly exacerbates the loss of intravascular fluid locally into the intestine as well as systemically through a generalized microvascular leak. The generation and activation of proinflammatory mediators, including neutrophils, complement, cytokines, eicosanoids, and oxygen-derived free radicals, also has been linked to remote organ failure and mortality caused by intestinal ischemia and reperfusion injury.

CLINICAL FEATURES

History

Patients with SBO classically present with the acute onset of cramping mid-abdominal pain, vomiting, obstipation, and abdominal distention. The magnitude of symptoms depends upon the degree (complete or partial), site, and duration of the obstruction. Typically, patients describe paroxysms of periumbilical pain occurring at four- to five-minute intervals for proximal obstructions and less frequently for more-distal obstructions. With prolonged obstruction, the cramping pain subsides as the motility in the distended intestine decreases. Patients with proximal intestinal obstruction have profuse vomiting, more-frequent pain, and minimal abdominal distention, whereas patients with distal obstruction present with less-frequent vomiting and much more abdominal distention. The emesis of patients with SBO is often feculent because of the increased bacterial count in the obstructed intestine. Although obstipation is an important symptom of SBO, patients with partial obstruction can continue to pass flatus and stool. Even patients with complete SBO evacuate the bowel contents distal to the obstruction. Continuous severe pain, particularly when localized, strongly suggests the presence of strangulated obstruction. Closed-loop obstructions are associated with the sudden onset of severe unremitting abdominal pain.

Physical Examination

In general, patients with simple SBO appear to be acutely ill, with abdominal distention and systemic evidence of intravascular volume depletion. Auscultation of the abdomen reveals periods of increased bowel sounds separated by intervals of relative quiet. The quality of the bowel sounds usually is described as high-pitched or musical. Borborygmi are pronounced rumbling bowel sounds that correspond with paroxysms of cramping abdominal pain. In the setting of prolonged obstruction, bowel sounds disappear as intestinal motility decreases. As alluded to earlier, the abdomen generally is distended and only minimally tender. Abdominal tenderness with guarding or other evidence of peritonitis suggests strangulation of the obstruction and necessitates urgent laparotomy. Patients with proximal SBO can have minimal abdominal distention if they have been vomiting. Patients with closed-loop obstructions can present with pain out of proportion to the physical findings, much like that of other causes of acute mesenteric ischemia, such as embolus in the superior mesenteric artery (SMA) (see Chapter 114). The presence of a tender mass at the site of an inguinal, femoral, or umbilical hernia strongly suggests that this is the etiology of the obstruction.

LABORATORY FINDINGS

Patients with suspected SBO should have blood drawn for a complete blood cell count (CBC) and serum electrolyte and creatinine concentrations. The CBC in patients with simple intestinal obstruction often reveals a slight leukocytosis as well as evidence of hemoconcentration. Significant neutrophilia and immature white blood cellular forms suggest strangulated obstruction, although the predictive value of this parameter is too low to be useful as a sole determinant of strangulation. Vomiting and the profound loss of fluid and electrolytes from the intravascular space that accompanies SBO cause hemoconcentration, increased blood urea nitrogen and serum creatinine concentration, and abnormalities of serum sodium, potassium, and chloride concentrations. The presence of a metabolic acidosis suggests severe intravascular volume depletion with or without intestinal ischemia.

RADIOLOGIC FINDINGS

Abdominal Plain Films

After history and physical examination, plain abdominal films are an inexpensive and reasonable first step in evaluating patients with suspected intestinal obstruction. Films taken with the patient in supine and upright positions may confirm the diagnosis of intestinal obstruction, localize the obstruction to the small intestine or colon, provide evidence of the degree of obstruction (partial or complete), and, if the upright image includes the diaphragm, detect pneumoperitoneum, thereby suggesting intestinal perforation.

Classically, the abdominal films of patients with SBO demonstrate multiple dilated gas- or fluid-filled loops of small intestine with a decompressed colon (Fig. 119-1). The finding of dilated small bowel loops containing air-fluid levels is insufficient to distinguish SBO from ileus; however, when combined with an absence of colonic gas, the diagnosis of SBO becomes very likely. When dilated small bowel loops are accompanied by colonic distention, ileus or large bowel obstruction become more likely (Fig. 119-2). A gasless abdomen may be seen in patients with a very proximal SBO or those in whom the intestine is filled with fluid (Fig. 119-3).

Figure 119-1. Supine (A) and erect (B) abdominal plain films in two patients with adhesive small bowel obstruction (SBO). A, Air-filled distended small bowel loops (arrows) with collapsed colon (*). Note that the small bowel folds (valvulae conniventes) typically extend completely across the intestinal loops. B, Multiple air-fluid levels in dilated small bowel loops (arrows) in the context of nondistended colon.

Figure 119-2. Supine abdominal plain film in a patient with an ileus. The small intestine (arrows) and the colon (*) are significantly distended.

Figure 119-3. Supine abdominal plain film (A) and coronal computed tomography (CT) image (B) in a patient with recalcitrant vomiting, obstipation, and a remote history of laparotomy for a gunshot wound to the abdomen. In this case, the abdominal plain film demonstrates a gasless abdomen. CT demonstrates proximal jejunal dilatation, with an abrupt caliber change (arrow) and no passage of contrast beyond the obstruction. Proximal small bowel obstruction was relieved by adhesiolysis. D, duodenum; St, stomach.

Lappas and colleagues reviewed 12 radiologic findings associated with SBO and found that the combination of air-fluid levels of different heights in the same bowel loop and a mean air-fluid level diameter of 2.5 cm were most predictive of a high-grade partial or complete SBO.¹⁴ Thompson and coworkers corroborated the predictive value of these types of air-fluid levels and reported their sensitivity for SBO to be 59% to 93%.¹⁵ The limitations of abdominal plain films in determining the presence of intestinal obstruction are well recognized: 20% to 30% of patients with proven SBO have equivocal or normal studies.^16,¹⁷ False-negative plain films are most likely to occur with low-grade, proximal, or closed-loop obstructions, and in such cases further imaging may be diagnostic (see Fig. 119-3).

Computed Tomography

Many studies support the use of abdominal computed tomography (CT) to evaluate patients with suspected intestinal obstruction.^18,¹⁹ Advances in CT hardware and software provide high-resolution reconstructed images in any plane, thus enhancing image resolution and diagnostic confidence.^20,²¹ Overall, CT is 90% to 95% sensitive, 96% specific, and 95% accurate in determining the presence of complete or high-grade SBO, and it provides important information regarding the site of obstruction and etiology in up to 95% of instances.

CT findings of mechanical SBO are listed in Table 119-2 and illustrated in Figures 119-4 to 119-7. The demonstration of dilated, fluid- or gas-filled loops of proximal bowel and collapsed loops of distal bowel supports the diagnosis of intestinal obstruction. A transition point between bowel loops with disparate calibers may be identified (see Fig. 119-4A) and the degree of obstruction estimated by the amount of enteral contrast passing through the obstruction. Tapered bowel at the transition point can form a beak (see Fig. 119-3B), and a thorough search of this area can suggest the cause of obstruction. Although peritoneal adhesions are not usually seen on imaging studies, the presence of a transition point without another identifiable cause strongly favors adhesive obstruction. CT, however, can demonstrate tumors and other nonadhesive causes of bowel obstruction (see Fig. 119-6).

Table 119-2 Computed Tomography Findings in Patients with Small Intestinal Obstruction

Simple Complete Intestinal Obstruction

Proximal bowel dilatation; discrete transition zone, with collapsed distal small bowel and no passage of oral contrast beyond the transition zone

Colon with little gas or fluid

Small bowel feces sign

Closed-loop Obstruction

Bowel Wall Changes

U-shaped, distended, fluid-filled bowel loop

Whirl sign: tightly twisted mesentery around a collapsed bowel segment

Beak sign: fusiform tapering in the longitudinal section of bowel at the site of obstruction

Two adjacent collapsed round, oval, or triangular loops of bowel at the site of obstruction

Mesenteric Changes

Fixed radial distribution of several dilated bowel loops with stretched and thickened mesenteric vessels converging toward the point of obstruction

Strangulated Intestinal Obstruction

Bowel Wall Changes

Bowel wall thickening with increased attenuation on unenhanced images

Target or halo sign: concentric rings of slightly different densities

Pneumatosis intestinalis linearis

Poor enhancement or lack of enhancement of the bowel wall with intravenous contrast

Serrated beak configuration of the obstructed bowel loop

Mesenteric Changes

A spectrum ranging from haziness and blurring of the mesenteric vessels to obliteration of the fatty mesentery and its vessels caused by mesenteric congestion and hemorrhage

Diffuse engorgement of the mesenteric vasculature

Unusual course of the mesenteric vasculature

Other Changes

Large amount of ascites

High-density ascites

Figure 119-4. Axial computed tomography images of simple (A) and closed loop (B) small bowel obstructions. A, The small bowel is dilated proximal to a transition point (arrow) and collapsed distally. Rectal contrast identifies the colon (*). B, Dilated, fluid-filled small bowel loops (*) are radially arranged, and several demonstrate concentric rings of wall thickening and submucosal edema. There is complete loss of mesenteric fat (arrow). Infarcted ileum was resected at surgery.

Figure 119-5. Computed tomography signs in small bowel obstruction. A, Abnormal position of mesenteric vessels: The superior mesenteric artery (*) is anterior to the superior mesenteric vein. B, Whirl sign (arrow). C, Small bowel feces sign (arrow). D, Pneumatosis (arrow) and large-volume ascites. The colon (*) is collapsed (see Table 119-2).

Figure 119-6. Axial computed tomography image (A) and intraoperative photograph (B) of partial small bowel obstruction caused by carcinoid tumor. The small bowel is dilated (*) proximal to the partially calcified primary lesion (arrow), and a metastatic nodal mass is evident in the mesentery (N).

Figure 119-7. Computed tomography image revealing a strangulated obstruction of the small intestine. A long segment of obstructed small bowel wall does not enhance (arrows) after intravenous contrast, indicating poor perfusion. Note the high-density mesenteric fluid (*). Small bowel volvulus and impaired perfusion were found at surgery.

Many classic CT signs have been proposed to distinguish simple from strangulated SBO (see Table 119-2)²²^–²⁸; these include circumferential bowel wall thickening and edema (see Fig. 119-4B), ascites (see Fig. 119-5D), mesenteric engorgement, abnormal vessel course (see Fig. 119-5A), altered enhancement of the bowel wall (see Fig. 119-7), and a bowel configuration suggesting a closed-loop obstruction or volvulus (see Fig. 119-4B). The small bowel feces sign refers to the presence of a mottled admixture of particulate matter and gas within the dilated bowel proximal to a low-grade obstruction or in the setting of intestinal ischemia (see Fig. 119-5C).^24,²⁷ A closed-loop obstruction or small bowel volvulus is suggested by U– or C-shaped dilated bowel loops and a radial distribution of stretched mesenteric vessels converging toward a point of torsion (see Fig. 119-4B).²² The whirl sign²⁵ also suggests intestinal torsion or volvulus and refers to a swirling mass of soft tissue and fat density that occurs as the bowel rotates on its mesentery (see Fig. 119-5B).

Decreased bowel wall enhancement is the most specific finding of intestinal ischemia (see Fig. 119-7), but its sensitivity is relatively low in most series (34% to 56%).^23,^24,²⁸ Porto-mesenteric venous gas, pneumoperitoneum, and pneumatosis intestinalis linearis (see Fig. 119-5D) may be seen very late in the natural history of strangulated obstruction and suggest the presence of extensive necrosis. Despite these many signs, the early diagnosis of strangulated SBO remains a challenge with reported sensitivity, specificity, and accuracy of these various criteria to detect strangulated obstruction ranging from 14% to 95%.²⁴

Detection of low-grade or intermittent bowel obstruction may be extremely difficult using standard radiologic and CT approaches; diagnostic accuracy varies from 48% to 66%.²⁹ In these settings, CT enteroclysis (Fig. 119-8) combines the advantages of active luminal distention with the mural and extraenteric evaluation of cross-sectional imaging; this technique has been shown to raise the accuracy for detecting small bowel diseases to nearly 100%.^30,³¹ Magnetic resonance imaging also has been used to detect SBO and to characterize benign and malignant causes,³² although its greater cost, lower spatial resolution, and lack of incremental diagnostic gain over CT has limited its widespread implementation.³³