|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Neutropenic Enterocolitis in Patients With Acute Leukemia: Prognostic Significance of Bowel Wall Thickening Detected by UltrasonographyFrom the Departments of Biotecnologie Cellulari ed Ematologia and Experimental Medicine and Pathology, and Surgical Pathology IX, Policlinico Umberto I, University "La Sapienza," Rome, Italy. Address reprint requests to Claudio Cartoni, MD, Department of "Biotecnologie Cellulari ed Ematologia," Università "La Sapienza," Via Benevento 6, 00161 Rome, Italy; email: cartoni{at}bce.med.uniroma1.it
PURPOSE: Neutropenic enterocolitis (NE) is a severe complication of intensive chemotherapy and is barely identifiable by clinical signs alone. Ultrasonography (US) supports the diagnosis of NE by showing pathologic thickening of the bowel wall. The aim of this study was to evaluate the prognostic value of the degree of mural thickening evaluated by US in patients with clinically suspected NE. PATIENTS AND METHODS: Neutropenic patients with fever, diarrhea, and abdominal pain after intensive chemotherapy for hematologic malignancies were studied with abdominal US. We evaluated the degree of bowel wall thickening detected by US and its correlation with the duration of the clinical syndrome as well as NE-related mortality.
RESULTS: Eighty-eight (6%) of 1,450 consecutive patients treated for leukemia had clinical signs of NE. In 44 (50%) of 88 patients, US revealed pathologic wall thickening (mean ± SD, 10.2 ± 2.9 mm; range, 6 to 18). The mean duration of symptoms was significantly longer in this group (7.9 days) than among patients without mural thickening (3.8 days, P < .0001), and the NE-related mortality rate was higher (29.5% v 0%, P < .001). Patients with bowel wall thickness of more than 10 mm had a significantly higher mortality rate (60%) than did those with bowel wall thickness CONCLUSION: Symptomatic patients with sonographically detected bowel wall thickening have a poor prognosis compared with patients without this finding. In addition, mural thickness of more than 10 mm is associated with poorer outcome among patients with NE.
NEUTROPENIC enterocolitis (NE) is a severe complication of chemotherapeutic regimens administered in the treatment of hematologic malignancies such as acute leukemias or lymphomas.1 The acute inflammatory disease may involve cecum, colon, and the terminal part of the ileum. Neutropenia, infections, and drug-induced alterations of the bowel mucosal surface have been suggested as etiologic factors in NE.2,3 The diagnosis of NE may be delayed, because the presenting clinical features (fever, abdominal pain, and diarrhea) are not specific and may suggest other abdominal diseases.4,5 To support the clinical diagnosis of NE, imaging techniques have therefore been used. Abdominal radiography does not reveal specific abnormalities, whereas computed tomography (CT) of the abdomen yields characteristic findings such as thickening of the bowel wall.6-8 This thickening is the main macroscopic finding in patients with NE and reflects the pathology of the disease: transmural inflammation of the bowel with mucosal ulceration, wall edema, necrosis, telangiectasia, and formation of intramural hematomas.2 However, CT cannot be easily performed and repeated (eg, at the bedside) in severely ill patients in intensive care or transplantation units. Ultrasonography (US) has been used successfully in the study of bowel loop abnormalities occurring in intestinal tract diseases such as Crohn’s disease or nonspecific ulcerative colitis.9,10 This low-cost imaging technique permits analysis of the intestinal wall under particular conditions during an episode of NE, such as gross, diffuse mural thickening or fluid-filled bowel. Although the detection of bowel wall thickening has been repeatedly used to make a definite diagnosis of NE in neutropenic patients, the prognostic impact of the degree of thickening of the bowel wall in leukemic patients affected by NE has not been previously established.11,12 Therefore, we used abdominal US to study patients with hematologic malignancies who developed clinical signs suggestive of NE (fever, abdominal pain, and diarrhea) during the neutropenic phase after intensive chemotherapy. The aims of this study were to estimate how many patients sharing the same symptoms at the onset of the abdominal complication showed thickened bowel wall on ultrasound scans and to evaluate the prognostic significance of the degree of bowel thickening with regard to the severity of the abdominal syndrome, by assessing the mortality rate and the duration of symptoms.
We reviewed the clinical records of 1,720 consecutive patients with acute leukemia, chronic myeloid leukemia in blast crisis, or non-Hodgkin’s lymphoma treated at our institution with intensive chemotherapy between May 1995 and December 1998. During this period, in accordance with a standardized imaging protocol, US of the abdomen was performed in neutropenic patients (absolute neutrophil count < 500/mm3) if they had concurrent symptoms of NE such as fever (axillary temperature > 38°C), diarrhea (> four stools daily), and abdominal pain. The sonographic scans and medical records of this selected group of patients with abdominal complications were therefore reviewed. US of the lower and upper abdomen was performed in fasting patients without preparation, using 3.75-MHz convex and 5-MHz linear electronically focused probes (Toshiba, Tokyo, Japan). A standard scanning technique (including graded compression) was used. We measured the intestinal wall thickness (outer wall to luminal surface). A thickness of less than 5 mm was considered normal (US-negative for a diagnosis of NE) ( Fig 1), and a thickness of more than 5 mm was considered abnormal (US-positive for a diagnosis of NE).13-15 The morphology and the five layers of ileum and colon were evaluated when possible, using the description by Wang et al.15
We monitored patients by performing multiple blood and feces cultures. In addition, results of surveillance nasopharynx, oropharynx, and external-genital cultures were also available. Treatment for clinically suspected episodes of NE consisted of empiric antibiotic therapy (usually a beta-lactam agent plus amikacin, with the addition of teicoplanin if fever persisted, followed by empiric antifungal agents in selected cases) and bowel rest. Albumin, packed RBCs, and platelets were infused when appropriate. One patient underwent left hemicolectomy after the neutrophil and platelet counts returned to normal values.
Pathologic Studies
Statistical Evaluation Differences between the two groups of patients (US-positive and US-negative) with regard to duration of symptoms (from occurrence of the three main symptoms [fever, abdominal pain, and diarrhea] to disappearance of all of these symptoms), the time from initiation of chemotherapy to onset of symptoms, duration of neutropenia, and the mean number of antibiotics administered were analyzed using the nonparametric Wilcoxon rank sum two-tailed test and parametric analysis of variance.
We used the
The 1,720 consecutive patients treated during the study period included 1,392 patients with acute leukemia, 270 with non-Hodgkin’s lymphoma, and 58 with chronic myeloid leukemia in blast crisis. The clinical syndrome characterized by fever, diarrhea, and abdominal pain occurred in 88 patients during the neutropenic phase. Sixty-four patients (72.8%) had acute myeloid leukemia, 21 (23.8%) had acute lymphoid leukemia, and three (3.4%) had chronic myeloid leukemia in blast crisis (Table 1). No symptoms of NE were noted in patients with non-Hodgkin’s lymphoma. Thus, the syndrome occurred in 88 (6%) of 1,450 courses of treatment for leukemia. Intensive chemotherapies in accordance with protocols for the underlying diseases preceded the onset of symptoms. All protocols included administration of drugs at daily doses considered to be toxic for the gastrointestinal mucosa, drugs such as cytarabine (standard doses, 100 to 200 mg/m2; high doses, 500 to 3,000 mg/m2), etoposide (100 to 200 mg/m2), idarubicin (10 to 12 mg/m2), mitoxantrone (10 mg/m2), daunorubicin (60 mg/m2), doxorubicin (50 mg/m2), and methotrexate (500 to 3,000 mg/m2). Chemotherapy was administered to induce a first complete remission in 67 (76%) of 88 cases and a second complete remission in 22 relapsed patients (24%).
US of the abdomen was performed in all 88 cases, with a mean number of two examinations performed per episode (range, one to four). US revealed abnormal intestinal wall thickening of the large and/or small intestine in 44 (50%) of 88 cases. In this group, mean bowel wall thickness was 10.2 mm (SD 2.9; range, 6 to 18); mean bowel wall thickness was 3.6 mm (SD, 1.36; range, 2 to 5) in the US-negative group. Mural thickening appeared in different patterns, such as complete effacement of the lumen with no discernible layers in the wall or reduced numbers of layers along with resolution in a heterogeneous zone of medium echogenicity ( Fig 2). In some cases, US detected mural hyperechoic septa floating into the lumen; these septa corresponded to the tags of necrotic mucosa, as seen at autopsy ( Fig 3).
The outcome of the abdominal complication significantly differed according to intestinal mural thickness ( Table 2). Complete resolution of fever, diarrhea, and abdominal symptoms through medical treatment occurred in 31 (70.5%) of 44 patients in the US-positive group, with a mean duration of symptoms of 7.9 days (SD, 2.34; range, 5 to 15); complete resolution occurred in 44 (100%) of 44 patients without bowel wall thickening, with a mean duration of symptoms of 3.8 days (SD, 1.54; range, 2 to 7; P < .0001). Also, the intestinal complication–related mortality rate was significantly higher in the US-positive group; such complications accounted for the deaths of 13 (29.5%) of 44 patients in that group (CI, 15.8 to 43.2), compared with no deaths in the US-negative group (P < .001).
US also showed fluid-filled bowel in 29 patients (66%) in the US-positive group and 20 patients (47%) in the US-negative group (P < .001), hepatomegaly in 36 cases (84%) versus 29 (66%; P < .001), and ascites in 13 cases (29.5%) versus zero (P < .001). The outcome of the intestinal complication was not affected by the presence of any of these sonographic abnormalities.
Moreover, the degree of mural thickening was predictive of mortality in US-positive patients ( Fig 4). Death occurred in 12 (60%) of 20 patients with bowel wall thickness of more than 10 mm, compared with one (4.2%) of 24 patients with bowel wall thickness
No significant differences were observed between the US-positive and US-negative groups of patients with regard to the duration of neutropenia before the onset of symptoms (US-positive group: mean, 12 days; range, 1 to 34; US-negative group: mean, 10 days; range, 0 to 39), the interval from the start of chemotherapy to the onset of symptoms of NE (US-positive group: mean, 12.3 days; range, 4 to 34; US-negative group: mean, 14 days; range, 6 to 41), the length of time with neutrophils less than 500/mm3 (US-positive group: mean, 23 days; range, 6 to 35; US-negative group: mean, 25 days; range, 5 to 47) or neutrophils less than 100/mm3 (US-positive group: 18 days, range, 4 to 29; US-negative group: mean, 16 days; range, 1 to 33), the number of antibiotics administered at the time the abdominal symptoms occurred (US-positive group: mean, three; range, one to four; US-negative group: mean, three; range, one to five), or antifungal prophylaxis (US-positive group: 38 patients; US-negative group: 41 patients). In all 31 US-positive patients who recovered from NE, sonographic follow-up demonstrated progressive reduction of intestinal mural thickening along with disappearance of abdominal symptoms.
Pathologic Findings
NE is a life-threatening complication in patients treated for hematologic malignancies and has been more frequently seen in these patients in recent years because of the increasing use of intensive chemotherapy regimens.17 In a retrospective analysis of cases over 30 years at St Jude Hospital, Sloas et al18 diagnosed (by means of US or CT) typhlitis, a localized type of enterocolitis involving only the cecum, in 24 of 6,911 patients treated for cancer. The overall incidence was 0.35%, and the incidence was significantly higher (2.1%) among those treated for acute leukemia. In addition, these investigators noted an increase in cases of typhlitis during the last 5 years of the study period, when patients with acute myeloid leukemia were treated with high doses of cytarabine and etoposide. Our study confirms these observations; we found a 6% incidence of NE between 1995 and 1998 among 1,450 patients with acute leukemia or chronic myeloid leukemia in blast crisis treated with aggressive chemotherapy. By revealing bowel wall thickening, US helped establish a diagnosis of NE and excluded other clinical entities with similar clinical presentations, such as acute appendicitis, acute cholecystitis, intussusception, ileus due to vincristine toxicity, or acute pancreatitis. When clinical signs suggestive of NE are noted, an imaging test capable of detecting the associated bowel alterations must be performed before a definitive diagnosis can be made.19 At the University of Texas M.D. Anderson Cancer Center, Gomez et al20 studied patients who had a clinical picture suggestive of NE. These authors noted that detection of bowel wall thickening allowed them to make a definitive diagnosis of NE. Patients with this imaging finding, detected by US or CT, had diarrhea for a significantly longer period and had a higher frequency of bloody stools.
In our study, the mean duration of symptoms was significantly longer among patients with sonographically detected bowel wall thickening (7.9 days) than among patients without mural thickening (3.8 days), and the NE-related mortality rate was higher (29.5% v 0%). Moreover, the degree of bowel wall thickening significantly correlated with the outcome of patients with NE. In particular, 60% of patients with bowel wall thickness of more than 10 mm died from this complication, compared with only 4.2% of those with mural thickness The accuracy of US in terms of demonstrating bowel wall thickening was confirmed in our study by the pathologic findings obtained in six cases. The slight overestimation by US in assessing the amount of thickening might in some cases be ascribed to the inability of this technique to discriminate between damaged and edematous mucosa and overlying necrotic and fibrinous debris. US was also found to be useful for monitoring the clinical course of NE, by showing the progressive reduction of bowel wall thickening in responding patients. In conclusion, in neutropenic patients who have completed intensive chemotherapy, the sonographic finding of thickened bowel wall, in association with a clinical syndrome characterized by fever, diarrhea, and abdominal pain, confirms the clinical diagnosis of NE. In our study, such patients had a higher mortality rate than did the US-negative group. In addition, patients with thicker bowel wall (> 10 mm) require intensive supportive treatment consisting of bowel rest with total parenteral nutrition and antibiotic and antimycotic therapy, because they have a very poor prognosis.
Supported in part by grants from the Italian Association Against Leukemias of Rome.
1. Alt B, Glass NR, Sollinger H: Neutropenic enterocolitis in adults: Review of the literature and assessment of surgical intervention. Am J Surg 149: 405-408, 1985[Medline] 2. Slavin RE, Dias MA, Saral R: Cytosine arabinoside induced gastrointestinal toxic alterations in sequential chemotherapeutic protocols: A clinical-pathologic study of 33 patients. Cancer 42: 1747-1759, 1978[Medline] 3. Girmenia C, Micozzi A, Cartoni C, et al: Detection of Candida mannoproteinemia in patients with neutropenic enterocolitis. Eur J Clin Microbiol Infect Dis 18: 55-58, 1999[Medline] 4. Vlasveld LT, Zwann FE, Fibbe WE, et al: Neutropenic enterocolitis following treatment with cytosine arabinoside-containing regimens for hematological malignancies: A potentiating role for amsacrine. Ann Hematol 62: 129-134, 1991[Medline] 5. Micozzi A, Cartoni C, Monaco M, et al: High incidence of infectious gastro-intestinal complications observed in patients with acute myeloid leukemia receiving intensive chemotherapy for first induction of remission. Support Care Cancer 4: 294-297, 1996[Medline]
6.
Tjon A, Tham RT, Vlasveld LT, et al: Gastrointestinal complications of cytosine-arabinoside chemotherapy: Findings on plain abdominal radiographs. AJR Am J Roentgenol 154: 95-98, 1990 7. Wall SD, Jones B: Gastrointestinal tract in the immunocompromised host: Opportunistic infections and other complications. Radiology 85: 327-335, 1992
8.
Fishman EK, Kavuru M, Jones B, et al: Pseudomembranous colitis: CT evaluation of 26 cases. Radiology 180: 57-60, 1991 9. Kaftori JK, Pery M, Kleinhaus U, et al: Ultrasonography in Crohn’s disease. Gastrointest Radiol 9: 137-142, 1984[Medline]
10.
Morgan CL, Trought WS, Oddson TA, et al: Ultrasound patterns of disorders affecting the gastrointestinal tract. Radiology 135: 129-135, 1980 11. Gootenberg JE, Abbondanzo SL: Rapid diagnosis of neutropenic enterocolitis (typhlitis) by ultrasonography. Am J Pediatr Hematol Oncol 9: 222-227, 1987[Medline] 12. Patel U, Leonidas JC, Furie D: Sonographic detection of necrotizing enterocolitis in infancy. J Ultrasound Med 9: 673-675, 1990[Medline]
13.
Fleischer AC, Muhletaler CA, James AE Jr: Sonographic assessment of the bowel wall. AJR Am J Roentgenol 136: 887-891, 1981 14. Bolondi L, Ferrentino M, Trevisani F, et al: Sonographic appearance of pseudomembranous colitis. J Ultrasound Med 4: 489-492, 1985[Medline]
15.
Teefey SA, Montana MA, Goldfogel GA, et al: Sonographic diagnosis of neutropenic typhlitis. AJR Am J Roentgenol 149: 731-733, 1987
16.
Wang KY, Kimmey MB, Nyberg DA, et al: Colorectal neoplasms: Accuracy of US in demonstrating the depth of invasion. Radiology 165: 827-829, 1987 17. Woodrow G, Greenberg BR: Successful medical management of neutropenic enterocolitis. Cancer 51: 1551-1555, 1983[Medline] 18. Sloas MM, Flynn PM, Kaste SC, et al: Typhlitis in children with cancer: A 30-year experience. Clin Infect Dis 17: 484-490, 1993[Medline]
19.
Siegel MJ, Friedland JA, Hildebolt CF: Bowel wall thickening in children: Differentiation with US. Radiology 203: 631-635, 1997 20. Gomez L, Martino P, Rolston KV: Neutropenic enterocolitis: Spectrum of the disease and comparison of definite and possible cases. Clin Infect Dis 27: 695-699, 1998[Medline] Submitted April 24, 2000; accepted September 13, 2000. This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
10 mm (4.2%, P < .001). 
2 test and/or the Fisher’s exact test to evaluate differences in sonographic findings (ascites, bowel loop distention, and hepatomegaly), antifungal prophylaxis, cases of symptom resolution, and the NE-related death rate. SE and 95% confidence intervals (CIs) were calculated. P < .05 was considered significant, and tests were two-tailed. 
