Acute colonic pseudo-obstruction, also called Ogilvie's syndrome, refers to marked dilation of the colon in the absence of mechanical obstruction. It generally develops in hospitalized patients over a period of days, and up to 95 percent of affected patients have an associated medical or surgical condition, (1,2) such as trauma, recent surgery, or serious infection.

The chief criterion for the diagnosis is the diameter of the colon on abdominal radiographs. However, there is no consensus regarding the minimal diameter required for the diagnosis. Perhaps the most commonly used value is 9 cm, based on a frequently cited study from 1956, in which 19 surgically treated patients who had cecal perforation or "impending" cecal perforation due to colonic obstruction all had cecal diameters of at least 9 cm and only 3 of 100 control patients had such cecal diameters after cecal distention during a barium enema. (3) The applicability of these data to acute colonic pseudo-obstruction is questionable.

The most clinically meaningful diagnostic criterion for acute colonic pseudo-obstruction should be the threshold diameter above which there is a risk of colonic perforation. In a review of 400 cases, perforation or ischemia was not seen when the diameter of the cecum was less than 12 cm. (1) Other studies have also suggested that perforation is uncommon unless the diameter of the cecum is at least 12 cm. (4) However, there is a broad overlap in cecal diameters between patients in whom acute colonic pseudo-obstruction resolves and those in whom perforation occurs. Thus, once the threshold diameter is reached in an individual patient, the actual extent of dilation does not appear to matter. Some have suggested that the duration of dilation may be a more important risk factor. (5,6)

Acute colonic pseudo-obstruction can lead to colonic perforation and death. In a 1997 review of published studies, Rex reported that the risk of perforation was approximately 3 percent. (2) However, since this figure is largely based on retrospective case series, its generalizability is unclear and it may represent an overestimate. At best, we can conclude that perforation does occur in patients with acute colonic pseudo-obstruction, but that it is uncommon. Furthermore, although perforation appears to increase the risk of death, patients with acute colonic pseudo-obstruction may die of their underlying conditions, even when the pseudo-obstruction resolves without complications.

The initial management of acute colonic pseudo-obstruction is conservative: the underlying cause is treated if possible, metabolic disturbances are corrected, and medications that may decrease colonic motility (e.g., narcotics, anticholinergic agents, and calcium-channel antagonists) are stopped. Nasogastric suction, rectal tubes, and frequent changes in the patient's position are often used. If symptoms persist or worsen and if the colonic diameter increases or remains above a certain level (e.g., 12 cm), colonoscopy is generally performed. Colonoscopic decompression reduces the diameter of the cecum on abdominal radiographs in about 70 percent of patients. (2) However, the condition will recur in 40 percent of these patients, requiring repeated colonoscopy. (2) The risk of recurrence may be decreased by the placement of a drainage tube into the right side of the colon at the time of initial colonoscopy. (7,8) Bedside colonoscopy of an unprepared bowel is technically difficult and not without risk: a number of cases of perforation have been reported in this setting. (2)

Surgery is generally recommended for patients with persistent or worsening acute colonic pseudo-obstruction despite colonoscopic decompression. However, surgery also carries a risk in patients with serious concurrent illnesses, even in the absence of perforation: the mortality rate was 26 percent in a review of 125 surgically treated patients who were found to have viable bowel at operation. (1) Thus, in the absence of randomized trials, it is uncertain whether the benefit of colonoscopic or surgical therapy outweighs the risks in these patients. Some have questioned the need for early endoscopic or surgical treatment. (9)

In this issue of the Journal, Ponec et al. (10) present the results of a randomized, double-blind, controlled trial of neostigmine for patients with acute colonic pseudo-obstruction. Ten of 11 patients who were treated with intravenous neostigmine had prompt passage of flatus or stool, with reduced abdominal distention (median time to response, four minutes), as compared with none of the 10 patients who received placebo injections. Significant decreases were also seen in abdominal circumference and colonic diameters on radiographs. All the patients had had no response to at least 24 hours of conservative treatment. Two of the 10 patients with an initial response had a recurrence and underwent colonoscopy, surgery, or both. All seven patients in the placebo group who were given open-label neostigmine also had an immediate clinical response, and none had a recurrence. Symptomatic bradycardia requiring atropine developed in two patients; other side effects included abdominal pain, excessive salivation, and vomiting. Two of the 18 patients who received neostigmine died of causes unrelated to acute colonic pseudo-obstruction or its treatment, reinforcing the fact that such patients often die of their underlying illness.

The response to neostigmine, which increases cholinergic activity, may shed light on the cause of acute colonic pseudo-obstruction. In 1948, Ogilvie suggested that sympathetic activity of the colon was interrupted, allowing unopposed sacral parasympathetic innervation. () More recently, it has been proposed that the condition is due to sympathetic overactivity, parasympathetic suppression, or both. Hutchinson and Griffiths (12) studied sequential treatment with guanethidine (an adrenergic inhibitor) and neostigmine and found that improvement occurred only after neostigmine was given. Two subsequent uncontrolled studies reported that intravenous neostigmine was effective in over 80 percent of patients. (13,14) These studies support the theory that acute colonic pseudo-obstruction is due to decreased parasympathetic activity.

How should we integrate the findings of Ponec et al. and others into clinical practice? Acute colonic pseudo-obstruction should be diagnosed only when symptoms and signs of abdominal distention are present and when marked dilation of the cecum or right colon is seen radiographically without evidence of distal obstruction. Although a diameter of 9 cm may be used as a threshold for diagnosis, 12 cm may be a more appropriate measure in terms of concern about perforation. Conservative treatment should still be used initially. If the condition persists or worsens after 24 hours of conservative measures and if there are no contraindications, such as bradycardia, neostigmine should be given. The most common potentially serious side effect is bradycardia. Therefore, patients should be monitored and remain supine before and for some period after the infusion.

Because colonic perforation is uncommon and the risk of death is greatly influenced by the underlying illness, it seems unlikely that any trial could be large enough to address adequately the effects of neostigmine on these important clinical outcomes. Nonetheless, the findings of Ponec et al. suggest an important role for neostigmine, which may speed the resolution of acute colonic pseudo-obstruction and reduce the need for colonoscopy and surgery.

Loren Laine, M.D.
University of Southern California School of Medicine
Los Angeles, CA 90033

Background. Acute colonic pseudo-obstruction -- that is, massive dilation of the colon without mechanical obstruction -- may develop after surgery or severe illness. Although it may resolve with conservative therapy, colonoscopic decompression is sometimes needed to prevent ischemia and perforation of the bowel. Uncontrolled studies have suggested that neostigmine may be an effective treatment.

Methods. We studied 21 patients with acute colonic pseudo-obstruction. All had abdominal distention and radiographic evidence of colonic dilation, with a cecal diameter of at least 10 cm, and had had no response to at least 24 hours of conservative treatment. We randomly assigned 11 to receive 2.0 mg of neostigmine intravenously and 10 to receive intravenous saline. A physician who was unaware of the patients' treatment assignments recorded clinical response (defined as prompt evacuation of flatus or stool and a reduction in abdominal distention), abdominal circumference, and measurements of the colon on radiographs. Patients who had no response to the initial injection were eligible to receive open-label neostigmine three hours later.

Results. Ten of the 11 patients who received neostigmine had prompt colonic decompression, as compared with none of the 10 patients who received placebo (P<0.001). The median time to response was 4 minutes (range, 3 to 30). Seven patients in the placebo group and the one patient in the neostigmine group without an initial response received open-label neostigmine; all had colonic decompression. Two patients who had an initial response to neostigmine required colonoscopic decompression for recurrence of colonic distention; one eventually underwent subtotal colectomy. Side effects of neostigmine included abdominal pain, excess salivation, and vomiting. Symptomatic bradycardia developed in two patients and was treated with atropine.

Conclusions. In patients with acute colonic pseudo-obstruction who have not had a response to conservative therapy, treatment with neostigmine rapidly decompresses the colon. (N Engl J Med 1999;341:137-41.)

Acute colonic pseudo-obstruction consists of massive dilation of the colon in the absence of mechanical obstruction. This severe form of adynamic ileus, also known as Ogilvie's syndrome, (1) develops in hospitalized patients and is associated with a wide variety of medical and surgical conditions. (2,3) The approximate risk of spontaneous perforation is 3 percent, with an attendant mortality rate of 50 percent. (4) Most cases respond to conservative management. (5) Although its value is unproved, colonoscopic decompression is often performed to prevent ischemia and perforation of the bowel in patients who have no response to conservative management. Colonoscopy in such patients is technically difficult, is not always successful, and can be accompanied by complications including perforation. Colonic distention may recur in up to 40 percent of patients despite initial decompression. (4)

The results of three uncontrolled studies suggest that the intravenous administration of neostigmine, an acetylcholinesterase inhibitor, produces rapid colonic decompression in patients with acute colonic pseudo-obstruction. (6,7,8) This pharmacologic approach is based on the theory that acute colonic pseudo-obstruction results from ineffectual colonic motility caused by excessive sympathetic stimulation, parasympathetic dysfunction, or both. (1,9,10,11) Therefore, we conducted a prospective, double-blind, placebo-controlled trial of neostigmine as a treatment for acute colonic pseudo-obstruction.

Patients

Patients with acute colonic pseudo-obstruction who were 18 years of age or older were recruited for the study between August 1995 and November 1997 from inpatient medical and surgical wards of hospitals affiliated with the University of Washington. Acute colonic pseudo-obstruction was defined as marked colonic distention in the absence of mechanical obstruction. To be eligible for the study, patients had to have a cecal diameter of at least 10 cm on plain radiographs. Mechanical obstruction was ruled out by the finding of air throughout all colonic segments including the rectosigmoid on plain abdominal radiographs. When air was not demonstrable in the rectosigmoid colon, mechanical obstruction was ruled out by radiographic contrast enemas. Patients were enrolled in the study if colonic distention, documented by clinical examination and abdominal radiographs, failed to improve after 24 hours of conservative management that included administering nothing by mouth, nasogastric suction, and intravenous fluid and electrolyte replacement. Any drugs that could adversely affect colonic motility, specifically narcotics and anticholinergic agents, were discontinued when possible. One patient, who was subsequently randomly assigned to the placebo group, was enrolled after only 18 hours of conservative therapy, when the consulting gastroenterologist determined that urgent decompression was warranted.

Exclusion criteria included a base-line heart rate of less than 60 beats per minute or systolic blood pressure of less than 90 mm Hg; signs of bowel perforation, with peritoneal signs on physical examination or free air on radiographs; active bronchospasm requiring medication; treatment with prokinetic drugs such as cisapride or metoclopramide in the 24 hours before evaluation; a history of colon cancer or partial colonic resection; active gastrointestinal bleeding; pregnancy; or a serum creatinine concentration of more than 3 mg per deciliter (265 ?mol per liter).

The human-subjects committee of the University of Washington and its affiliated hospitals approved the study protocol. All patients provided written informed consent.

Study Design

Patients were randomly assigned to receive 2.0 mg of neostigmine intravenously over a period of three to five minutes or identical-appearing saline placebo. The injections were given by a physician who was unaware of the patients' treatment assignments. All patients were monitored by electrocardiography; atropine was available at the bedside, and 1.0 mg was given intravenously as needed for symptomatic bradycardia. Patients were instructed to remain supine for at least 60 minutes after the injection. Vital signs were recorded immediately before the injection and five minutes and three hours afterward.

The physician administering the infusion monitored the clinical response for 30 minutes after the injection. The maximal abdominal circumference and the diameter of the cecum, ascending colon, and transverse colon on plain radiographs were measured before and three hours after the injection by an investigator who was unaware of the patients' treatment assignments.

Three hours after the infusion, patients who did not have a reduction in colonic distention on both clinical examination and radiographs were eligible to receive open-label neostigmine (2.0 mg intravenously) administered by a physician who was unaware of the identity of the study drug. The three-hour period was chosen because of the short half-life of neostigmine. Three hours later, abdominal circumference, colonic diameters, and clinical response were again measured. Patients were monitored for adverse effects during the initial treatment and during open-label treatment and were then followed for the remainder of their hospitalization. The treatment assignments were not revealed to the investigators, treating physicians, or patients until the last patient had been discharged from the hospital.

Assessment of Outcomes

The outcomes assessed included an immediate clinical response to the study drug, changes in abdominal girth and colonic diameters on abdominal radiographs three hours after treatment, and the need for colonoscopic decompression or surgery during hospitalization. An immediate clinical response was defined as the passage of flatus or stool with a reduction in abdominal distention on physical examination within 30 minutes after the injection. Treatment was considered to have failed if open-label neostigmine, colonoscopic or surgical intervention, or both were required because of the recurrence or persistence of colonic distention.

Statistical Analysis

On the basis of prior reports, we estimated that 10 patients would be required in each group for the study to have the power at an alpha level of 0.05, with a beta error of 0.2, to detect a significant difference between groups, assuming a response rate of 80 percent in the neostigmine group and 30 percent in the placebo group. We used Fisher's exact test to compare the frequency of clinical responses and treatment failures in the two groups. (12)

Eleven patients were randomly assigned to receive neostigmine, and 10 to receive saline placebo. All patients had acute abdominal distention. Fifteen patients (71 percent) reported abdominal pain at base line. The two groups were similar with regard to age, sex, duration and degree of colonic distention, use of narcotics and anticholinergic medications, history of recent surgical procedures, and severity of illness (Table 1).

In two patients, the underlying medical diagnosis was spinal cord injury with paralysis, and one patient each had bleeding peptic ulcer, cerebrovascular accident, pneumonia with respiratory failure, liver failure due to primary sclerosing cholangitis, sepsis, wrist fracture with urinary tract infection, graft-versus-host disease after bone marrow transplantation, renal transplantation, and metastatic insulinoma. The underlying surgical diagnoses included total knee replacement in two patients, coronary-artery bypass grafting in two patients, and total hip replacement, amputation of a leg after a burn, prostatectomy, lumbar laminectomy, exploratory laparotomy after a gunshot wound, and open reduction of multiple fractures, with internal fixation, in one patient each.

After treatment, there was prompt evacuation of flatus or stool with a reduction in abdominal distention on physical examination in 10 patients in the neostigmine group (91 percent) and none in the placebo group (P<0.001) (Table 2). The median time to response was 4 minutes (range, 3 to 30). There were also significant reductions in abdominal circumference and colonic diameters in the neostigmine group as compared with the placebo group (Table 2).

Treatment was considered to have failed in three patients who received neostigmine (27 percent) and eight who received placebo (80 percent, P=0.04). One of the three patients in the neostigmine group had no immediate clinical response to initial treatment but did have a response to open-label therapy, with no recurrence of dilation. The other two patients required colonoscopic decompression for recurrence of colonic distention. One of these two also received atropine for symptomatic bradycardia and underwent colonoscopic decompression three hours after receiving neostigmine. The other patient, who had metastatic insulinoma, had a clinical response to a second dose of neostigmine given 24 hours after the initial injection. However, distention recurred, leading to colonoscopic decompression. After a third recurrence of colonic dilation, the patient underwent subtotal colectomy.

Three of the 10 patients who were assigned to the placebo group did not receive open-label therapy. At the discretion of their attending physicians, two were treated with conservative measures alone, and colonic distention gradually resolved over the next 48 to 72 hours. One patient did not receive open-label treatment, because of the development of renal failure. This patient underwent two unsuccessful attempts at colonoscopic decompression and was found at laparotomy to have ischemic colonic necrosis that required bowel resection.

The results of open-label treatment with neostigmine are given in Table 3. Of the eight patients who received open-label therapy, one had previously received neostigmine and seven had received placebo. All eight patients had an immediate clinical response, and none had a recurrence of dilation or required colonoscopic or surgical decompression. Of the 18 patients who received neostigmine, either initially or during open-label treatment, 17 (94 percent) had an immediate clinical response and 2 (11 percent) had recurrent colonic dilation.

The most frequent adverse effect of neostigmine treatment was abdominal pain, which occurred in 13 patients; it was described as mild cramping by 9 and as moderate-to-severe cramping by 4. In all patients, the abdominal pain was transient and had no sequelae. Eight patients had excessive salivation, and two vomited. Symptomatic bradycardia requiring atropine occurred in two patients: one patient had a syncopal episode while walking to the bathroom 30 minutes after receiving neostigmine (a violation of the protocol), and the other felt lightheaded within minutes after the infusion. None of the patients in the placebo group had an adverse effect.

Two patients (both of whom were in the neostigmine group) died, but in neither patient was death related to acute colonic pseudo-obstruction or its treatment. One patient, who had metastatic insulinoma, underwent hepatic-artery chemoembolization and died of multiorgan failure 16 days after receiving neostigmine. The other patient, who had end-stage liver disease from primary sclerosing cholangitis, underwent liver transplantation 16 days after receiving neostigmine and died of invasive aspergillosis 17 days after transplantation.

Although the pathophysiology of acute colonic pseudo-obstruction is not fully understood, it is thought to result from an imbalance in the regulation of colonic motor activity by the autonomic nervous system. In 1948, Ogilvie described two patients with colonic pseudo-obstruction resulting from malignant infiltration of the celiac plexus and attributed the syndrome to sympathetic deprivation. (1) There is now a better understanding of the autonomic innervation of the colon. The parasympathetic nervous system increases contractility, whereas the sympathetic nerves decrease motility. (11) Multiple pharmacologic and metabolic factors, as well as spinal and retroperitoneal trauma, can alter the autonomic regulation of colonic function, leading to excessive parasympathetic suppression, sympathetic stimulation, or both. This imbalance results in colonic atony or pseudo-obstruction and forms the rationale for pharmacologic manipulation of the autonomic innervation of the colon in patients with this condition.

We found that neostigmine decompressed the colon in patients with acute colonic pseudo-obstruction who had not responded to conservative therapy. Colonic distention recurred infrequently. Even though the elimination half-life of neostigmine is short, most patients had sustained improvement. This result may reflect resolution of the underlying problem with continued use of conservative measures. Our results confirm those of uncontrolled studies.

An example of the response to neostigmine is shown in Figure 1.

(6,7,8) However, our study was too small to evaluate the effect of neostigmine treatment on the risk of colonic perforation and mortality.

In most patients with acute colonic pseudo-obstruction, conservative management will result in the resolution of colonic distention within three days. (5) Decisions about the need for medical therapy, colonoscopy, or surgery should be individualized and should be based on the patient's clinical status. The risk of colonic perforation is reportedly increased when the cecal diameter exceeds 12 cm and when distention has been present for more than six days. (2,13)

Colonoscopy is successful in about 70 percent of patients with acute colonic pseudo-obstruction, as determined by a reduction in radiographically measured cecal diameter. (4,14) The recurrence rate of approximately 40 percent may be decreased by placement of a decompression tube at the time of the procedure. (4) In this setting, colonoscopy is a difficult procedure and has a morbidity rate of 3 percent and a mortality rate of 1 percent. (9) Surgical intervention is associated with high morbidity and mortality rates and should be reserved for patients with signs of ischemia or perforation or those in whom colonoscopic decompression fails. (2)

Although the majority of side effects in our study were minor, the use of parasympathomimetic agents such as neostigmine is not without risk. Patients with underlying bradyarrhythmias or those receiving (beta)-adrenergic antagonists may be more susceptible to neostigmine-induced bradycardia. Similarly, neostigmine increases airway secretions and bronchial reactivity, which may exacerbate active bronchospasm. Concomitant treatment with neostigmine and the anticholinergic agent glycopyrrolate has been reported to diminish the central cholinergic effects of neostigmine without reducing the increases in colonic motility. (15) Thus, the combination of neostigmine and glycopyrrolate merits further study in patients with colonic pseudo-obstruction.

The elimination half-life of neostigmine after intravenous infusion averages 80 minutes. (16) Neostigmine is hydrolyzed by plasma cholinesterase and is metabolized by microsomal liver enzymes. Renal excretion accounts for the clearance of 50 percent of the drug, and the serum half-life is prolonged in patients with renal dysfunction. (16) Therefore, patients with renal impairment may have an increased or prolonged vagomimetic response after the administration of neostigmine. (17)

A 2-mg ampule of neostigmine for parenteral use costs $3. The cost of neostigmine to the patient after storage and handling fees are included is approximately $15 -- substantially less than the cost of colonoscopy. In addition, colonoscopic decompression can be a technically demanding and unpleasant procedure that involves a considerable amount of effort and time on the part of physicians. The procedure is often done at the bedside, and preparation of the colon may not be possible.

In summary, we found that treatment with neostigmine was an effective way to decompress the colon in patients with acute colonic pseudo-obstruction. The use of neostigmine should be considered before colonoscopy is performed in patients with acute colonic pseudo-obstruction who have not had a response to conservative management.

Supported by a grant from the American Society for Gastrointestinal Endoscopy.

We are indebted to Dr. Jason Dominitz for his contribution to the statistical analyses.

From the Division of Gastroenterology, University of Washington Medical Center, Seattle. Address reprint requests to Dr. Kimmey at the Division of Gastroenterology, Box 356424, University of Washington, 1959 NE Pacific St., Seattle, WA 98195.

1. Ogilvie H. Large-intestine colic due to sympathetic deprivation: a new clinical syndrome. BMJ 1948;2:671-3.
2. Vanek VW, Al-Salti M. Acute pseudo-obstruction of the colon (Ogilvie's syndrome): an analysis of 400 cases. Dis Colon Rectum 1986;29:203-10.
3. Nanni G, Garbini A, Luchetti P, Nanni G, Ronconi P, Castagneto M. Ogilvie's syndrome (acute colonic pseudo-obstruction): review of the literature (October 1948 to March 1980) and report of four additional cases. Dis Colon Rectum 1982;25:157-66.
4. Rex DK. Colonoscopy and acute colonic pseudo-obstruction. Gastrointest Endosc Clin North Am 1997;7:499-508.
5. Sloyer AF, Panella VS, Demas BE, et al. Ogilvie's syndrome: successful management without colonoscopy. Dig Dis Sci 1988;33:1391-6.
6. Hutchinson R, Griffiths C. Acute colonic pseudo-obstruction: a pharmacological approach. Ann R Coll Surg Engl 1992;74:364-7.
7. Stephenson BM, Morgan AR, Salaman JR, Wheeler MH. Ogilvie's syndrome: a new approach to an old problem. Dis Colon Rectum 1995;38:424-7.
8. Turegano-Fuentes F, Munoz-Jimenez F, Del Valle-Hernandez E, et al. Early resolution of Ogilvie's syndrome with intravenous neostigmine: a simple, effective treatment. Dis Colon Rectum 1997;40:1353-7.
9. Vantrappen G. Acute colonic pseudo-obstruction. Lancet 1993;341:152-3.
10. Dorudi S, Berry AR, Kettlewell MG. Acute colonic pseudo-obstruction. Br J Surg 1992;79:99-103.
11. Spira IA, Rodrigues R, Wolff WI. Pseudo-obstruction of the colon. Am J Gastroenterol 1976;65:397-408.
12. Swinscow TDV. Statistics at square one. 2nd ed. London: British Medical Association, 1977.
13. Johnson CD, Rice RP, Kelvin FM, Foster WL, Williford ME. The radiographic evaluation of gross cecal distention: emphasis on cecal ileus. AJR Am J Roentgenol 1985;145:1211-7.
14. Jetmore AB, Timmcke AE, Gathright JB Jr, Hicks TC, Ray JE, Baker JW. Ogilvie's syndrome: colonoscopic decompression and analysis of predisposing factors. Dis Colon Rectum 1992;35:1135-42.
15. Child CS. Prevention of neostigmine-induced colonic activity: a comparison of atropine and glycopyrronium. Anaesthesia 1984;38:1083-5.
16. Cronnelly R, Stanski DR, Miller RD, Sheiner LB, Sohn YJ. Renal function and the pharmacokinetics of neostigmine in anesthetized man. Anesthesiology 1979;51:222-6.
17. Webb MD. Type I second-degree AV block after neostigmine administration in a child with renal failure. Anesth Prog 1995;42:21-2.

The New England Journal of Medicine -- November 18, 1999 -- Vol. 341, No. 21

To the Editor:

Ponec et al. (July 15 issue) (1) reported that treatment with neostigmine rapidly decompresses the colon in patients with acute colonic pseudo-obstruction who have not had a response to conservative therapy. Symptomatic bradycardia, however, developed in two patients.

Because bradycardia is a well-recognized and important complication of neostigmine therapy, use of neostigmine for reversal of neuromuscular blockade in the operating room is always accompanied by administration of an antimuscarinic anticholinergic agent such as atropine or glycopyrrolate. Although the authors recognized that administration of glycopyrrolate has not been shown to decrease colonic motility, they did not administer it prophylactically. Vital signs were monitored before the injection of neostigmine and 5 and 30 minutes after injection. Since vital signs were not continuously monitored, asymptomatic bradycardia might not have been detected, and thus the true incidence of important bradycardia might have been underestimated. Moreover, even if bradycardia is treated, the effects of neostigmine may outlast those of glycopyrrolate or atropine.

We recommend that patients who are given neostigmine for colonic pseudo-obstruction also receive either atropine or glycopyrrolate prophylactically and that they be monitored continuously by electrocardiography and blood-pressure measurement for one hour after neostigmine administration.

Monica S. Vavilala, M.D.
Arthur M. Lam, M.D.
University of Washington School of Medicine
Seattle, WA 98104

References

1. Ponec RJ, Saunders MD, Kimmey MB. Neostigmine for the treatment of acute colonic pseudo-obstruction. N Engl J Med 1999;341:137-41.

We have been using neostigmine to treat patients with acute colonic pseudo-obstruction for more than six years and have found it to be extremely effective and safe. (1) However, we are concerned that some patients in the study by Ponec et al. -- specifically, those with air in the rectosigmoid colon on plain abdominal radiography -- did not receive radiographic contrast enemas to rule out a mechanical obstruction. In our experience and that of others, (2) the presence of air at the presumed rectosigmoid junction in association with dilatation of the proximal colon can be misleading and can wrongly imply the absence of an obstructing lesion. Such a false sense of reassurance can lead to incorrect diagnosis and treatment of a patient with a potent colonic stimulant such as neostigmine. (3) This can have severe adverse consequences.

We believe that a water-soluble contrast enema should be used for all patients with possible acute colonic pseudo-obstruction when pharmacologic treatment with neostigmine is considered. Also, over the past four years, we have used a combination of glycopyrrolate and neostigmine (Robinul-Neostigmine, Wyeth Laboratories) (500 ?g and 2.5 mg, respectively) to good effect in seven consecutive patients in the intensive care unit who had acute colonic pseudo-obstruction. (4) None had even a transient bradycardia.

Faisal Abbasakoor, F.R.C.S.
Alison Evans, M.B., B.Ch.
Brian M. Stephenson, F.R.C.S.
Royal Gwent Hospital
Newport, NSW NP20 2UB, Australia

References

1. Stephenson BM, Morgan AR, Drake N, Salaman JR, Wheeler MH. Parasympathomimetic decompression of acute colonic pseudo-obstruction. Lancet 1993;342:1181-2.
2. Stewart J, Finan PJ, Courtney DF, Brennan TG. Does a water soluble contrast enema assist in the management of acute large bowel obstruction: a prospective study of 117 cases. Br J Surg 1984;71:799-801.
3. Trevisani G, Hyman N, Church J. Neostigmine: a new treatment for Ogilvie's syndrome. Dis Colon Rectum 1998;41:A29. abstract.
4. Stephenson BM, Morgan AR, Salaman JR, Wheeler MH. Ogilvie's syndrome: a new approach to an old problem. Dis Colon Rectum 1995;38:424-7.

I was surprised by the enrollment of an obviously very sick patient in the study by Ponec et al. The authors stated that "one patient, who was subsequently randomly assigned to the placebo group, was enrolled after only 18 hours of conservative therapy, when the consulting gastroenterologist determined that urgent decompression was warranted." I wonder how they justify their disregard of the patient's consultant, who obviously assessed the situation as urgent. What was the outcome for this patient?

Claus A. Pierach, M.D.
Abbott Northwestern Hospital
Minneapolis, MN 55407

In 1948, Ogilvie described two cases of large-intestine colic due to sympathetic deprivation associated with abdominal carcinoma. (1) No massive distention was found. To refer to massive distention of the colon without mechanical obstruction as "Ogilvie's syndrome" is not appropriate.

David Nicholson, M.D.
117 Colonial Ct.
Little Rock, AR 72205-4221

References

1. Ogilvie H. Large-intestine colic due to sympathetic deprivation: a new clinical syndrome. BMJ 1948;2:671-3.

To the Editor:

We appreciate the comments of Abbasakoor and colleagues and acknowledge that the report by Stephenson et al. (1) was one of the reasons we performed our controlled study. We also are concerned about the use of neostigmine in the presence of mechanical obstruction and agree that contrast radiography should be used to rule out obstruction if plain abdominal radiographs do not have the classic appearance of acute colonic pseudo-obstruction. We cannot agree, however, that contrast radiography must be performed in all cases, since in the right clinical setting, plain abdominal radiographs with classic findings are seldom incorrect.

Vavilala and Lam discuss the importance of monitoring patients when they are given neostigmine. We agree with this and monitor all our patients continuously with portable electrocardiographic and automated blood-pressure equipment for 30 minutes after infusion of neostigmine. Our protocol called for atropine for patients who had symptomatic bradycardia. We are very interested in the potential application of glycopyrrolate as a means of avoiding this adverse effect of neostigmine, as suggested by both Abbasakoor et al. and Vavilala and Lam, and look forward to the results of a controlled trial to prove its efficacy.

Pierach raises an important point with regard to the use of neostigmine. Patients whose cardiovascular or respiratory condition is unstable should probably not receive neostigmine. Thus, we adhered to the strict inclusion and exclusion criteria listed in the Methods section of our article. With regard to the specific patient mentioned, he was offered entry into the study early because the consulting gastroenterologist did not believe that he should wait for an additional 6 hours to allow the 24 hours of conservative therapy otherwise required before study entry. According to our protocol, patients who did not have decompression three hours after infusion of the study drug were offered open-label neostigmine, as was this particular patient, who had a response to open-label treatment. The protocol was designed in this manner specifically to avoid the ethical dilemma described by Pierach.

Finally, as Nicholson points out, it may be a misnomer to use the term "Ogilvie's syndrome" to refer to acute colonic pseudo-obstruction. In his original report of two patients with widespread cancer, Ogilvie called attention to the importance of a balanced autonomic nervous system for maintenance of colonic function. His patients, however, actually presented with chronic symptoms. Ironically, today we would probably not diagnose acute colonic pseudo-obstruction in his two patients.

Robert J. Ponec, M.D.
Michael D. Saunders, M.D.
Michael B. Kimmey, M.D.
University of Washington Medical Center
Seattle, WA 98195

References

1. Stephenson BM, Morgan AR, Drake N, Salaman JR, Wheeler MH. Parasympathomimetic decompression of acute colonic pseudo-obstruction. Lancet 1993;342:1181-2