Refinement of the positive predictive value of gallbladder nonvisualization after morphine administration for acute...

The authors previously reported two major patterns in the time-activity curve of the common hepatic bile duct (BD) after morphine administration in patients with gallbladder nonvisualization. The first pattern consists of a gradual increase in BD activity (of variable duration) occurring during a simultaneous decrease in liver parenchymal activity (BD increase), representing the physiologic effects of morphine administration. The second pattern consists of a continuous decrease in BD activity that parallels the activity in the liver parenchyma (BD decrease), representing lower or no physiologic effects of morphine administration. The authors hypothesize that gallbladder nonvisualization associated with a continuous decrease in BD activity after morphine administration will have a lower positive predictive value (PPV) for acute cholecystitis than gallbladder nonvisualization associated with an increase in BD activity.

Thirty-six patients who had morphine-augmented cholescintigraphy were divided into two groups: 19 with BD increase after morphine administration and 17 with BD decrease.

Of the 36 patients, 22 had acute cholecystitis. The positive predictive value (PPV) of gallbladder nonvisualization was 61%. All of the remaining 14 had chronic cholecystitis. Of 19 patients with BD increase, 15 had acute cholecystitis (PPV = 79%), whereas only 7 of 17 patients with BD increase (PPV = 41 %) had acute cholecystitis (P = 0.023 by the one-tailed and 0.038 by the two-tailed Fisher exact tests).

Gallbladder nonvisualization after morphine administration with the pattern of BD decrease is not as reliable (intermediate probability in this series) for the diagnosis of acute cholecystitis as is nonvisualization of the gallbladder in patients with a pattern of BD increase (high probability).