Diagnostic performance of magnetic resonance cholangiopancreatography (MRCP) versus endoscopic retrograde cholangiopancreatography (ERCP) in the pediatric population: a clinical effectiveness study.

PURPOSE

To determine the diagnostic performance of magnetic resonance cholangiopancreatography (MRCP) for detecting biliary and pancreatic abnormalities in a pediatric population, using endoscopic retrograde cholangiopancreatography (ERCP) as the reference standard.

MATERIALS AND METHODS

Institutional review board approval with a waiver of informed consent was obtained for this retrospective investigation. Records from the Cincinnati Children's Hospital Medical Center, Division of Gastroenterology and Department of Radiology were used to identify patients aged ≤ 18 years who had undergone both ERCP and MRCP within a 4-week interval between January 2013 and May 2017. Biliary and pancreatic duct findings were documented for each modality to determine the diagnostic performance of MRCP (with 95% confidence intervals), using ERCP as the reference standard.

RESULTS

54 patients met inclusion criteria. Mean patient age at time of ERCP was 10.4 ± 4.9 years, and 25 (46%) were male. Mean interval between ERCP and MRCP was 11.2 ± 9.7 days. For detection of any abnormality (n = 99 ERCP findings), MRCP had a sensitivity of 76.8% (67.5-84.0%) and a positive predictive value (PPV) of 81.7% (72.7-88.3%). MRCP was 75.7% (59.9-86.6%) sensitive, with a PPV of 84.9% (69.1-93.4%) for biliary findings (n = 37) and 73.5% (59.7-83.8%) sensitive, with a PPV of 78.3% (64.4-87.7%) for pancreatic findings (n = 49). For pancreatobiliary abnormalities (n = 13), MRCP had a sensitivity of 92.3% (66.7-99.6%) and a PPV of 85.7% (60.1-97.5%).

CONCLUSION

In clinical practice, MRCP is moderately sensitive for biliary and pancreatic abnormalities, with false-negative and false-positive examinations being relatively common.