891

Downloaded from www.ajronline.org by 36.82.86.104 on 11/30/13 from IP address 36.82.86.104. Copyright ARRS. For personal use only; all rights reserved

Distinction Between Postoperative Ileus and Mechanical Small-Bowel Obstruction: Value of CT Compared with
Clinical and Other Radiographic
,

Findings
r-

David H. Frager1 Jeanne W. Baer Allen Rothpearl Peter A. Bossart

OBJECTIVE.
bowel obstruction,

The expeditious

diagnosis

of complete

and partial mechanical

small-

as opposed to paralytic ileus, during the immediate postoperative period may be difficult on the basis of clinical and plain film radiographic findings. For this reason, we prospectively evaluated the use of CT in this setting and compared it with the clinical and plain film evaluations as well as with various contrast examinations. SUBJECTS AND METHODS. Thirty-six postoperative patients with signs and symptoms

of paralytic ileus or mechanical

plain abdominal radiographs.

small-bowel
Based

obstruction

were examined

clinically and had

the surgeon

on the findings

of these

examinations,

assigned patients to one of the following categories: (1) paralytic ileus, (2) Indeterminate, (3) partial mechanical obstruction, or (4) complete mechanical obstruction. CT scans were
obtained within 24 hr of the initial diagnostic studies, and patients were then recategonized according to the above classification solely based on CT findings. Initial examination results were then compared with the CT results. In addition, the results of contrast studies, namely, enteroclysis and barium enema, performed after CT small-bowel series, were evaluated. The gold standard for diagnosis was laparotomy in 20 patients, clinical course and follow-up in 13 patients, and clinical course and contrast studies in the other three patients. RESULTS. CT was effective (sensitivity and specificity, 100%) in distinguishing between postoperative ileus and complete mechanical small-bowel obstruction. The
combined clinical and plain film findings were often confusing and nondlagnostic

(sensitivity,

19%).

CT

was

also

valuable

in diagnosing

and

distinguishing

partial

mechanical small-bowel obstruction from paralytic ileus. Contrast studies (enteroclysis) in four patients with partial mechanical small-bowel obstruction were useful In
grading

the degree and severity of the obstruction. CONCLUSION. Our results suggest that in the immediate

postoperative

period, evaluating

CT

is the method
tinguishing

of choice

for diagnosing
ileus.

mechanical
studies

small-bowel
are useful

obstruction
in

and dis-

it from

paralytic

Contrast

further

partial
AJR

mechanical
1995;164:891-894

small-bowel

obstruction.

Distinguishing between paralytic ileus and mechanical small-bowel obstruction (SBO) in the postoperative period is critical yet extremely difficult and confusing [1]. CT has been advocated as a useful tool for making this distinction [2-6]. This study addresses the usefulness of CT in diagnosing early postoperative mechanical SBO
and distinguishing The it from value paralytic of contrast ileus compared in this to the traditional setting is also clinical addressed. and plain film evaluation.
Received October26, July 7, 1 994; 1994. accepted after revision

studies

Subjects

and

Methods
patients services and who distension, clinically. in nine, in two, The (15 males, had signs nausea, 21 females; and symptoms and vomiting, laparotomies small-bowel in one, and 3-102 years old; mean, ileus bowel colorectal in two, of adhesions 62 years) within from SBO 10 of paralytic diminished were for lysis or mechanical penistalsis) resections appendectomy in one. Abdomi-

Presented at the annual meeting of the American Roentgen Ray Society, New Orleans, LA, April 1994.
1All authors: Columbia University College of Physicians and Surgeons, Department of Radiology, St. Lukes-Roosevelt Hospital Center, 1111 Amsterdam Ave., New York, NY 10025. Address correspondence to D. H. Frager.

Thirty-six
one of three (i.e., abdominal days) two, were external

consecutive surgical pain evaluated trauma hernia

days of laparotomy
abdominal

(33 patients

between

3 and 10 days of laparotomy,

in two, laparotomy resection

three patients

within

1-3
in 17, in

initial

hysterectomy exploratory

0361-803X195/1 644-891 American Roentgen Ray Society

nal radiographs

were obtained

from patients

in the supine

and erect positions

and interpreted

by

892

FRAGER

ET AL.

AJA:164,

April

1995

Downloaded from www.ajronline.org by 36.82.86.104 on 11/30/13 from IP address 36.82.86.104. Copyright ARRS. For personal use only; all rights reserved

an attending staff radiologist or by a senior radiology resident who knew the diagnosis of mechanical SBO was being considered. The surgeon then assigned the patients to one of four categories based on plain film interpretations in conjunction with the clinical and laboratory findings. The categories were defined as follows: 1 no obstruction-reflex ileus, 2 indeterminate-mechanical obstruction oould not be ruled out, 3 partial mechanical obstruction, and 4 complete obstruction. CT scans were obtained within 24 hr of the initial evaluation in all cases. Oral contrast material (500 ml of Gastrografmn, 2% iodine;
= = = =

oral contrast

material

beyond

the transition

zone. The colon

is usually

collapsed and contains at most minimal fluid or gas. Paralytic ileus exhibits the opposite corresponding findings. The proximal and distal
small intestine are dilated with no transition tended with gas, fluid, or oral contrast zone, and the colon is dismaterial when scans are

delayed. Partial mechanical SBO falls between these two entities. The bowel distal to the obstruction is not completely collapsed, the transition zone is less distinct, and the colon contains moderate amounts of

gas and fluid. Nine patients had contrast studies after the CT examination (small-bowel series in three patients, enteroclysis in four patients, and contrast enema in two patients). Contrast studies were performed to evaluate CT diagnoses of partial mechanical SBO or because the surgeons refused to rely solely on the CT results. Of the 36 patients studied, 16 had complete obstruction proved at surgery, 15 had partial obstruction, of whom four underwent surgery, and five patients had paralytic ileus (Fig. 1). The causes of obstruction in the 20 patients who underwent a second laparotomy were adhesions in 14 patients (Fig. 2); abscess in three patients; and hematoma, intussusception (Fig. 3), and ischemic stricture in one patient each. Criteria for reoperation included CT findings of complete mechanical SBO or strangula-

Squibb,

Princeton,

NJ) was given 30 mm to 2 hr before

scanning

in all

patients. Contrast material (Renografin 60 or Isovue 300; Squibb) was given IV as a 50 ml bolus (rate, 1 .5-2.0 mI/sec) followed by 50-70 ml at 0.5-1.0 mVsec in 24 ofthe 36 patients. In addition, four patients had delayed scans 12-24 hr later. CT scans were interpreted by a staff
radiologist or senior residents and categorized blindly on the basis of

the CT findings into the same groupings used by the surgeon. CT criteria for distinguishing between paralytic ileus and mechanical SBO have already been described [2]. Complete mechanical obstruction appears as proximal bowel dilatation (>2.5 cm in diameter), a discrete transition zone with distal collapsed small bowel, and no passage of

Fig. 1.-Postoperative year-old man 1 0 days

after

paralytic subtotal

ileus in 80colectomy

and Ileoproctostomy.
A, Erect abdominal bowel dilatation and radiograph shows smallair-fluid levels. Diagnosis

was indeterminate. B, CT scan performed to evaluate bowel distention and fever shows Ileoproctostomy (arrow at
staple rather line) with dilated small-bowel obstruction. management and rectum

and no transition
than mechanical

indicative

of paralytic

ileus

ered with conservative

Patient recov1 week later.

.,

Fig. 2.-Postoperative mechanical small-bowel obstruction caused by adhesions In 45-year-old woman 7 days after partial colectomy. A, Supine abdominal radiograph shows dilated loop of jejunum. Erect view (not shown) had one air-fluid level. Although obstruction was consideration based on abdominal radlographs, surgeon was not convInced on clinIcal grounds. B, CT scan shows proximal Jejunal dilatation wIth abrupt transition (arrow) and distally collapsed small bowel consistent with complete mechanical small-bowel obstruction-proved at repeat laparotomy. Note that, although unusual, the most common cause of mechanical small-bowel obstruction In Immediate postoperatIve period Is adhesive bands [13].

Fig. 3.-Postoperative

mechanical

small-bowel

obstruction duetoidlopathic Intussusception In 75year-old woman 3 days after sigmold resection. CT scan shows manlced small-boweldilatation with collapsed loop of ileum (intussusceptum)contained In another more distal loop (Intussusclplens) (arrow); proved at laparotomy. Abdominal radiograph demonstrated dilated small bowel and no colonlc gas consistentwlth mechanical small-bowel obstruction.

AJR:164,

April

1995

CT

OF

ILEUS

AND

SMALL-BOWEL

OBSTRUCTION

893

tion, lack of clinical response within 24-48 hr, and inability

attempts over a 3-7-day period

to nasogastric
to feed
despite initial

suctioning
after
response

and IV fluids
two or three
to nasogastric

in this

series)

(Fig.

1) or when

it is filled

with

fluid.

In addition,

the patient

suctioning.

The results of the clinical, radiographic, and CT studies compared with the gold standards of the laparotomy findings

were in 20

Downloaded from www.ajronline.org by 36.82.86.104 on 11/30/13 from IP address 36.82.86.104. Copyright ARRS. For personal use only; all rights reserved

patients, contrast

clinical studies

follow-up in three

in 13 patients, and clinical follow-up and patients. The resulting groups, based on

the final diagnoses of no obstruction (including indeterminate), partial obstruction, and complete obstruction, were analyzed with the McNemar test for paired samples [7] to determine whether CT provided any clear advantage over the traditional diagnostic approach
in diagnosing mechanical SBO. In addition, the role of contrast stud-

ies was evaluated.

Results CT readily enables the diagnosis of complete mechanical SBO even in the setting of the immediate postoperative period, where paralytic ileus is so common (Table 1). CT shows this distinction effectively (100% sensitivity, 100% specificity) in contradistinction to an evaluation that relies on the combined efforts of clinical and plain film parameters (1 9% sensitivity, 1 00% specificity). CT in this series was effective (100% sensitivity) in diagnosing and distinguishing partial mechanical SBO. The gold standard for diagnosing partial mechanical SBO from paralytic ileus, however, was laparotomy in only four of 15 patients. Contrast studies provided additional valuable information in four of the nine patients (44%) studied. In these four patients, the contrast examination (entenoclysis) charactenized the severity of partial mechanical SBO better than the plain film or CT studies did. CT findings suggested strangulation in all four patients with proved strangulation at surgery.
Discussion

the clinical signs of bowel obstruction, such as abdominal pain distension and obstipation, are all considered normal in the immediate postoperative period [1 10-13] and represent the expected paralytic ileus. This paralytic ileus is commonly presentfor the first 3 days after surgery [12]. Also, the surgeon may be reluctant to consider the complication of mechanical SBO soon after surgery. Thus, in our series, complete mechanical SBO was definitively diagnosed and immediately distinguished from paralytic ileus without CT in only 1 9% of the patients. CT, on the other hand, readily makes the diagnosis and the distinction. Although this study shows that traditional clinical and plain film evaluation is poor (Table 1), the truth is more complicated. Virtually all negative diagnoses were actually indeterminate (Table 1). In other words, mechanical obstruction remained a significant diagnostic consideration, but there was not enough evidence to indicate the need for reoperation. Delaying sungery and managing the patient conservatively in the immediate postoperative period can be argued to be correct because strangulation is rare [13]. This study does not support the latter contention, because four (25%) patients undergoing surgery for complete mechanical SBO had strangulation (Fig. 4). CT allows confident diagnosis of complete mechanical SBO and CT findings can suggest strangulation [14]. In the case of partial obstruction, the need for an immediate definitive diagnosis is less urgent. The same holds true for patients with paralytic ileus. These patients should be managed conservatively, because in this setting, strangula,

tion

is indeed

rare

and

the

majority

of these

patients

will

Previous studies [2-6] have reported the value of CT in diagnosing complete mechanical obstruction of the small bowel. Nonetheless, the traditional clinical and plain film evaluations suffice in approximately 50-80% of cases [3, 8]. The immediate postoperative period, however, is different. The plain films in this setting are difficult to interpret [8-101 because bowel distention and paralytic ileus are so common. Furthermore, the gas-filled colon is difficult to visualize, particulanly when a colectomy has been done (1 7 of 36 patients

recover without surgery [13]. This study seems to purport that CT can readily distinguish between partial mechanical SBO and paralytic ileus, but the evidence is somewhat shaky. Of the 1 5 patients with partial mechanical SBO, only four had surgical proof. The other 11 had only CT evidence and clinical follow-up. CT signs of partial SBO are the least reliable [5, 6] and certainly overlap with paralytic ileus. Some authors contend that paralytic ileus and partial mechanical SBO are part of a single disease spectrum [5]. In any event, for the vast majority of immediate postoperative partial mechanical SBOs and paralytic ileus, conservative management suffices and the patient recovers uneventfully. In those patients with paralytic ileus or partial mechanical SBO who do not recover as expected,

TABLE

1 : Clinical/Plain

Film Evaluation

Versus 5) Specificity

CT Partial Truebstructionb False(n

=

No Ob structiona
Evaluation TrueFalse-

(n

15)
Sensitivity

Complete
True-

Obstructionc
False-

16)
Sensitivity

Negative Clinical/ plainfilm CT

5

Positive
0

(%)
100

Positive
2

Negative
13(10)

(%)
13d

Positive

Negative 13(11)

(%)
19e 100e

100

15

100d

16

Note-Numbers in parentheses represent indeterminate diagnoses. aproved by clinical course and CT. bproved in four patients by lapanotomy, in two by CT contrast studies CAll proved at surgery.
d<

and clinical

course,

and in nine by CT and clinical

course.

.ooi.
.01.

ep<

894

FRAGER

ET AL.

AJR:164,

April

1995

Downloaded from www.ajronline.org by 36.82.86.104 on 11/30/13 from IP address 36.82.86.104. Copyright ARRS. For personal use only; all rights reserved

Fig. 4.-Marked

partial

postoperative

mechanical

small-bowel

obstruction

in 20-year-old

man 7 days

after laparotomy for a stab wound. Plain films showed dilated large and small bowel with air fluid levels. A, CT scan shows contrast material In colon and narrowing of lleum at site of an adhesion to anterior
abdominal wall (arrow), Indicative of partial mechanical small-bowel obstruction. B, Small-bowel series via long tube (modified enteroclysis) shows little barium past stenosls (arrow), Indicating marked partial mechanical small-bowel obstruction. Loop distal to stenosis only partially col-

lapsed. Surgery performed based on this study after patient showed no Improvement adhesive band from abdominal Incision to Ileum was lysed.

clinically.

Dense

Fig. 5.-Postoperative mechanical smallbowelobstruction with strangulation in 50-yearold man 10 days after partial colectomy. Plain film findings were nonspecific with a few dilated loops of small bowel noted. CT scan following IV contrast administration shows dilated and collapsed small bowel and dilated thickened Ileum with hyperdense wall Indicative of small-bowel obstruction with strangulation. Pathologic examination of surgical specimen

showed venous occlusive Infarction.

contrast studies can play an important role. Contrast studies, especially entenoclysis in this situation, can determine whether there is partial obstruction and, more important, its severity (Fig. 5). Entenoclysis is an excellent technique for evaluating all types of mechanical SBO, and radiologists have specifically advocated its use in the immediate postoperative period [15]. The major drawback of this technique is its reliance on barium. If no obstruction is seen on enteroclysis and CT scanning is requested to determine the cause of the reflex ileus, the CT study cannot be done for several days. A more sensible approach, therefore, would be to do the CT scanning first. CT scanning can be used to determine whether there is an obstruction, and if there is none, to detect underlying abscess hematomas, leaks, colitides, and so on. Furthermore, in the case of partial obstruction, a delayed CT or abdominal film obtained 12-24 hr later can show whether contrast material has reached the colon. This finding does not absolutely preclude surgery but is of some value in excluding complete obstruction [16].
This study by no means with early provides the final mechanical word in evaluatSBO. Even ing patients postoperative

REFERENCES
1 . Sykes PA, Schofield
1974:61:594-600

PF. Early postoperative

small-bowel

obstruction.

BrJ Surg

2. Frager D, Medwid SW, Baer JW, Mollinelli B, Friedman M. CT of smallbowel obstruction: value in establishing the diagnosis and determining the degree and cause. AJR1994:162:37-41 3. Balthazar EJ, Cho KC, Medwid SW, Birnbaum BA, Noz ME. Bowel obstruction: evaluation with CT. Radiology 1991:180:313-318 4. Fukuya T, Hawes DR, Lu CC, Chang PJ, Barloon TJ. CT diagnosis of small-bowel obstruction: efficacy in 60 patients. AJR 1992:158:765-769 5. Gazelle GS, Goldberg MA, Wittenberg J, Halpern EF, Pinkney L, Mueller PR. Efficacy of CT in distinguishing small-bowel obstruction from other causes of small-bowel dilatation. AJR 1994;162:43-47 6. Maglinte DD, Gage SN, Harmon BH, et al. Obstruction of the small intestine: accuracy and role of CT in diagnosis. Radiology 1993:188:61-64 7. McNeman Q. Note on the sampling error of the difference between comelated proportions or percentages. Psychometrika 1947:12:153-157 8. Baker SR. The abdominal plain film. Norwalk, CT: Appleton & Lange, 1990:155-242 9. Samuel E, Duncan JG, Philip T, Sumerling MD. Radiology ofthe postoperative abdomen. Clin Radiol 1 963; 14:133-148 10. Quatromoni JC, Rosoff L, Halls JM, Yellin AE. Early postoperative smallbowel obstruction. Ann Surg 1980:191:72-74 11 . Coletti L, Bossart PA. Intestinal obstruction during the eariy postoperative period. Ann Surg 1964:88:774-778 12. Frykberg ER, Phillips JW. Obstruction of the small-bowel in the eariy postoperative period. South Med J 1989;82:169-173 13. Pickelman J, Lee RM. The management of patients with suspected early postoperative small-bowel obstruction. Ann Surg 1989:210:216-219 14. Balthazan EJ, Bimnbaum BA, Megibow AJ, Gordon RB, Whelan CA, Hulnick DH. Closed loop and strangulating intestinal obstruction: CT signs. Radiology 1992:185:769-775 15. Dehn TCB, Nolan DJ. Enteroclysis in the diagnosis of intestinal obstruction in the early postoperative period. Gastrointest Radioll989;14:15-21 16. Zer M, Kaznelson D, Feigenberg Z, Dintsman M. The value of Gastrografin in the differential diagnosis of paralytic ileus versus mechanical obstruction: a critical review and report of two cases. Dis Colon Rectum 1977:20:573-579

though statistical significance was achieved, the sample size was small. Interpretation of CT scans is difficult in these cases, and accurate results come with experience. Nonetheless, we believe that CT should be used early (occasionally even before plain films) to evaluate possible mechanical SBO or the cause of paralytic ileus in the immediate postoperative period. We believe that such an approach, by establishing a fast and accurate diagnosis, might reduce the current morbidity, mortality, and resulting monetary costs in these circumstances.