Kenneth Nichols PhD, Presenter: Royalties, Syntermed, Inc

Christopher J. Palestro MD, Abstract Co-Author: Research grant, BioMed Valley Discoveries, Inc

Accurate scintigraphic detection of gastroparesis (GP) requires collecting data for 4 hrs. This investigation was undertaken to determine if the length of the gastric emptying (GE) test could be shortened without compromising accuracy.

This was a retrospective study of GE data collected using a standardized protocol at 0, 1, 2, 3 and 4 hrs for 432 pts (111 men & 321 women; mean age 53±19 yrs). GP was defined as >10% gastric retention at 4 hrs (J Nucl Med Technol. 2008;36:44-54). Simultaneous 1-minute anterior and posterior static images were collected with pts upright. Attenuation-corrected conjugate view counts were determined within manually drawn regions over the stomach. Per cent retention values at 1 hr (p1), 2 hrs (p2) & 3 hrs (p3) were compared by ROC analysis to determine optimal thresholds for predicting GP. We also tested recently published (Clin Nucl Med. 2011;36:283-5) criteria (C2t4) that classifies pts as: having GP if GE < 35% at 2 hrs, and normal if GE > 55% at 2hrs; inconclusive if 55%≥GE≥35% at 2 hrs with continued imaging to 4 hrs.

97/432 (23%) pts had GP. At 2 hrs, C2t4 criteria had ROC area of 85±3% for the subgroup with GE < 35% or GE > 55%, and were inconclusive in 17% (75/432) of cases, for which GE values at 2 hrs were 46±7%. Among the remaining 357 (83%) cases, C2t4 failed to detect 28% (18/65) of GP cases. Optimal ROC discrimination thresholds were 71%, 43%, and 28% for p1, p2, and p3, respectively. Using these thresholds to dichotomize methods, and including analysis of C2t4 criteria, resulted in sensitivity of 83%, 85%, 88% and 81%, specificity of 69%, 82%, 95% and 98%, and accuracy of 76%, 83%, 91% and 94% for p1, p2, p3 and C2t4, respectively, for all 432 cases. All sensitivity values were significantly lower (p=0.001) than the 4 hr reference standard sensitivity (100%, by definition). Among the 97 cases of GP, using these criteria to truncate studies at 1, 2 & 3 hrs would have resulted in failure to detect GP in 17%, 15%, 12% & 19% of all cases by p1, p2, p3 & C2t4 criteria, respectively.

Truncating gastric emptying studies at 2 hrs can result in failure to detect 15%-19% of gastroparesis cases.

Gastric emptying studies should be carried out to 4 hrs to maximize test sensitivity.

Nichols, K, Palestro, C, Implications of Truncating Gastric Emptying Tests.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034304.html