The purpose of this study was to compare the diagnostic accuracy of the non-invasive diagnostic methods and rectal suction biopsy for the detection of Hirschsprung disease (HD).

We reviewed diagnostic methods and results retrospectively in patients who underwent anorectal manometry, barium enema and rectal suction biopsy for the diagnosis of HD at Asan Medical Center from January 2000 to December 2012.

There were 97 patients (59 neonates and 38 infants) in the study period. The overall accuracy of anorectal manometry for the diagnosis of HD was 71.1% and its sensitivity was 51.4% (48.1% in neonate and 62.5% in infant, respectively) and its overall specificity was 82.3% (81.3% in neonate and 83.3% in infant, respectively). The Overall accuracy of barium enema was 66.0% (72.8% in neonate and 55.3% in infant, respectively) and specificity of barium enema was 53.2% (56.3% in neonate and 50.0% in infant, respectively). These results were lower than those of anorectal manometry. The overall sensitivity of barium enema was 88.6% (92.6% in neonate and 75.0% in infant, respectively) and it was higher than the sensitivity of anorectal manometry. Histological studies confirmed HD in 35 patients, in one of whom the suction biopsy showed negative finding.

Accuracy of non-invasive methods for diagnosis of HD in our study is lower than those in previous study, so we need to improve the quality of diagnostic tools in our hospital. We conclude that the rectal suction biopsy is the most accurate test for diagnosing HD, so the biopsy to confirm the diagnosis of the HD is very important.

To investigate the diagnostic accuracy and applicability of barium enema (BE) and rectal suction biopsy with acetyl cholinesterase (AChE) histochemistry in the diagnosis of neonatal Hirschsprung's disease (HD), we retrospectively reviewed the findings of BE and AChE staining in 96 neonates with suspected HD during a 10-year period from January 1991 to December 2000. Sixty-nine cases of HD (58 males and 11 females) and 27 cases of non-HD are included in this study. In regard to BE, HD was based on definite transitional zone, suspicious HD on reversed rectosigmoid index (RSI <1), and non-HD on normal RSI (RSI>1). The histochemical criterion used for the diagnosis of HD was that of Chow et al (1977), i.e., the presence of many coarse discrete cholinergic nerve fibers in the muscularis mucosae and in the immediately subjacent submucosa regardless of infiltration of cholinergic nerve fibers in the lamina propria. Of 66 neonates with HD who underwent BE, transitional zone was identified in 33 cases (50%) and reversed RSI in 19 cases (21%), microcolon in 4 cases and normal finding in 10 cases (15%) while of 27 neonates with non-HD, there was normal finding in 16 cases and reversed RSI in 9 cases (41%). Thus diagnostic accuracy based on transitional zone was 64%. The positive predictive value of reversed RSI for the diagnosis of HD was 68%. Of 42 neonates with HD who underwent AChE histochemistry, there were 41 AChE-positive reactions and one AChE-negative reaction in a neonate with total colonic aganglionosis, while of 27 cases of non-HD, there were one equivocal AChE-positive reaction and 26 AChE-negative reactions. Thus AChE histochemical study showed a 97% diagnostic accuracy with a 98% sensitivity and a 96% specificity. In conclusion, we believe that BE is valuable as a first diagnostic step since about 80% of neonates with HD show significant radiologic findings such as a transitional zone or reversed RSI. AChE histochemical study was a more reliable diagnostic tool showing a 97% diagnostic accuracy, and is particularly valuable in neonates with HD who showed reversed RSI or normal RSI on BE. A diagnostic algorithm in neonates with suspicious HD was proposed.particularly valuable in neonates with HD who showed reversed RSI or normal RSI on BE. A diagnostic algorithm in neonates with suspicious HD was proposed.