Introduction: Many diseases causing abdominal symptoms can occur in patients with leukemia such as neutropenic enterocolitis, leukemic infiltration of GI tract, and infectious enterocolitis. However, many imaging finding of such diseases can have overlapping findings. The purpose of this study was to evaluate clinical data and CT features of acute abdominal pain in patients with acute leukemia to find out the causes of acute abdominal pain in our institution. Also to find out any CT feature that can be used as a prognostic factor, we analyzed the CT findings seen in these patients and sought any correlation between each CT feature and mortality
Material and Methods: This retrospective study was approved by our institutional review board which waived the requirement to obtain informed consent. Between January 2008 and January 2013, 107 CT examinations were performed in 76 patients with acute leukemia. Presence of neutropenia and fever (mortality and were investigated. The patients were divided into two groups, patients with mortality within one month after CT examination were considered as mortality group (n=21) and those without mortality within one month were considered as non-mortality group (n=69). Two reviewers evaluated the images for bowel related changes, ancillary findings, and other organ related findings. Differences between the mortality group and the non-mortality group with regard to small bowel wall thickness and large bowel wall thickness were analyzed using independent t-test. Comparison of proportions was used to analyze differences of other clinical and CT findings between two groups.
Results: Numerous diagnoses were made to the enrolled patients complaining of abdominal pain based on clinical and CT findings. The most common clinical diagnosis was “no evident cause of acute abdomen” (n=23, 25.56%).The second most common clinical diagnosis was infectious enterocolitis (n=13, 14.44%). Other common diagnoses were neutropenic enterocolitis (n=12, 13.33%), acute pancreatitis (n=7, 7.78%), and ischemic enterocolitis (n=5, 5.56%). Presence of neutropenia and fever were not different between two groups. The mean small bowel wall thickness was significantly thicker in the mortality group than the non-mortality group (p=0.0062). Mesenteric stranding was seen in 11 cases (52.38%) in the mortality group and 18 cases (26.09%) in the non-mortality group which showed statistical difference (p=0.0465). Small bowel thickness cut off value of 3 mm has sensitivity and specificity of 52.38% and 79.71% respectively to be included in the mortality group. With the presence of mesenteric stranding, sensitivity and specificity to be included in mortality group, was 52.38% and 73.91%.
Conclusion: In cases when exact diagnosis of a patient may not be properly diagnosed using clinical and imaging studies, small bowel wall thickness can be measure and used as a prognostic factor that has high negative predictive value. This can provide clinicians relieving information in managing leukemic patients with abdominal pain if the patient does not have small bowel wall thickening. When a patient presents with bowel wall thickening, this information may give suggestions to clinicians in determining more intensive therapeutic approach.