SEARCH

SEARCH BY CITATION

Kerner C, Carey K, Mills AM, et al. Use of abdominopelvic computed tomography in emergency departments and rates of urgent diagnoses in Crohn's disease. Clin Gastroenterol Hepatol 2012; 10:52–57.

In their retrospective study of 648 adults with Crohn's disease (CD) between 2001–2009, Kerner et al1 evaluated abdomen/pelvis computed tomography (APCT) utilization trends and, as an outcome of interest, identified CT findings warranting urgent care in CD patients. Using ICD-9 codes for CD, they reviewed CT findings of patients with CD with gastrointestinal chief complaints who presented to the emergency department (ED) of two hospitals providing both community care to local residents and tertiary level care within the University of Pennsylvania Health System during 2001–2003 and 2007–2009. Overall, 69.1% of ED encounters among CD patients with gastrointestinal complaints involved the use of APCT; 74.3% of ED encounters with a triage complaint of abdominal pain involved the use of APCT; 77.5% of ED encounters resulting in admissions involved the use of APCT; and 52% of ED encounters resulting in discharge involved APCT use.

There was an increase in ED encounters for CD from 140 in 2001 to 246 in 2009 at the two medical centers. CT utilization trends in patients with CD paralleled those seen in the general population. There was a statistically significant increase in APCT utilization from 47.1% in 2001 to 77.5% in 2009 (P = 0.005). When limited to patients with abdominal pain or to first-time ED visitors, this trend was still observed. Despite increasing APCT utilization, admission rates from 2001–2009 remained stable. Only in first-time ED visitors was there an increase in admission rates over time (P = 0.002).

The authors also reviewed the rates of positive findings on CT scan, including new or worsening perforation, obstruction, or abscess (POA). They also collected the rates of other non-CD-related urgent findings on CT such as cholecystitis, appendicitis, diverticulitis, ischemia or vascular emergencies, pyelonephritis, pancreatitis, new neoplasms, complicated urolithiasis, or gynecologic emergencies (POANCD). Of all the 652 APCT scans, 29% revealed a POA and 34.9% revealed POANCD. There was no significant difference in these findings when comparing 2001–2003 findings to 2007–2009 findings with 30% POAs detected among 169 scans in 2001–2003 and 28.6% POAs detected among 482 CT scans in 2007–2009. (P = 0.92) Interestingly, despite stable CT findings and stable admission rates, there was a statistically significant increase in the number of patients taking biologic agents for CD from 2001–2003 to 2007–2009. (P = 0.003)

DISCUSSION

  1. Top of page
  2. DISCUSSION
  3. REFERENCES

The use of CT scans in the United States has increased dramatically from ≈3 million annually in 1980 to ≈70 million in 2007.2, 3 A recent study showed that CT use during ED visits alone increased 330% from 3.2% of encounters in 1992 to 13.9% in 2007.4 Rates of growth were highest for abdominal pain and flank pain, two complaints with which CD patients often present. As CD can affect any part of the gastrointestinal tract with complications that extend beyond the luminal surface and clinical symptoms that do not always reflect disease activity, physicians will often reach for any tool that can increase diagnostic certainty. Consequently, as technological advances have increased the speed and accuracy of CT scans and financial reimbursements have improved, CT scans have become pervasive.4 The results are often actionable, with CT scans having been shown to change management in 28% of CD patients.5

However, repeat exposure to ionizing radiation can come at a cost and CT scans should be used prudently. While the radiation dose of one CT scan is not large, cumulative radiation is thought to potentially increase average lifetime cancer mortality. An estimate of the average lifetime cancer mortality risk associated with a 25 mSv dose is about 1 in 1000 or 0.1%.6–8 One retrospective 5-year study of 392 adult CD patients found that 7% of CD patients were exposed to high levels of radiation (>50 mSv/5 years) and that 75% of radiation exposure came from CT scans.9 In the pediatric CD population, one study estimated that 67% of CD children would exceed 50 mSv of radiation exposure by 35 years of age.10 The obvious question is whether findings on CT scan are important enough to justify repeated use. Kerner et al focused this question on the ED, where many CT scans for CD patients are performed even before admission, addressing the important question of whether findings from ED CT scans are significant enough to impact triage decisions and ultimately medical management.1

The authors rightly point out that if there were an increase in unnecessary CT scans over time, one would expect the proportion of normal APCTs to increase with a corresponding decrease over time of urgent APCT findings. However, this was not the case. The fact that roughly one-third of all APCT scans in CD patients revealed an urgent condition requiring specific medical or surgical therapy, a rate that remained unchanged from 2001 to 2009 despite significantly more patients on biologic therapy in 2007–2009, confirms that complication rates in this patient population have been and continue to remain high, requiring close monitoring by physicians. The authors offer several potential explanations for the high POA and POANCD rate, including higher-sensitivity CT scanners allowing for higher detection of POA and POANCD despite a lower threshold to order APCT in the later years, and potentially greater degree of disease severity among the ED patients over time. However, in a separate analysis limited to first-time ED visitors, similar rates of POA and POANCD were detected between the early and later years, decreasing the likelihood that the stable rates of POA and POANCD were due to an increase in the number of CD patients with more severe disease in their later years.

These data point to the fact that imaging modalities will likely remain an important diagnostic modality to determine urgent disease complications in the ED. While data for magnetic resonance (MR) enterography show that it is a promising and radiation-free imaging modality, it is not easily accessible or widely available at this time and is unlikely to supplant CT scans in the ER in the foreseeable future.11, 12 Therefore, it is important that we determine the characteristics that best determine which patients are likely to have POA and POANCD. It would have been interesting to see what percentage of patients without POA or POANCD were admitted despite negative imaging. In this population, was the CT scan necessary? Could the ED APCT scan have waited for in-hospital MR enterography, thus limiting exposure of the patient to ionizing radiation? Did the CT scan play a significant role in determining discharge from the ED in patients without POA and POANCD? Could another test have supplanted the CT scan in the low-risk patients? As clinicians, how accurate are we in our assessment of disease activity? Do we seek imaging modalities to confirm our initial suspicions or do we truly require these imaging modalities to discover hidden pathology? Were hospital stays shortened or was medical/surgical management significantly expedited due to the imaging performed in the ED? Was there a close dialog between the gastroenterologist and the ER doctor to determine whether a CT scan should have been performed and was the patient asked about when his/her last imaging exam took place? In future studies, these questions should be asked to determine what circumstances warrant early CT scan use in CD patients.

REFERENCES

  1. Top of page
  2. DISCUSSION
  3. REFERENCES