Comparing radiopaque markers and 13C-labelled breath test in diabetic gastroparesis diagnostics

Abstract

Purpose: Determining gastric emptying is mandatory in the diagnosis of diabetic gastroparesis. Several methods of investigation exist, but none has proven reliable, inexpensive and accessible. In this study, we aimed to compare gastric emptying of radiopaque markers (ROM) and ¹³carbon-labelled gastric emptying breath tests for solids (GEBT). We also aimed to determine any association between gastric emptying and patient-reported symptoms, glycemic control and the patients' age, diabetes duration and occurrence of other late complications. Patients and methods: Forty-five patients (30 women, 15 men) with diabetes mellitus types 1 or 2 (40, 5) and symptoms of gastroparesis were examined with ROM and GEBT. All were interviewed, filled out symptom questionnaires and had HbA1c levels measured. Results: Forty percent of patients had delayed gastric emptying of ROM, while 55% had delayed gastric emptying of GEBT. Correlation between ROM and GEBT was not significant. Compared to GEBT, sensitivity for a positive ROM test was 0.52, while specificity was 0.74. In women, we found a higher specificity of 0.92, sensitivity 0.47. Difference in HbA1c between patients with positive and negative results was of borderline significance for both tests. GEBT (r=0.41, P=0.008) correlated with HbA1c. Patients with any late complications of diabetes had higher gastric retention of ROM (P=0.028), while patients with polyneuropathy (P=0.014) and diabetic wounds (P=0.004) had slower emptying with GEBT. None of the methods identified significant associations between gastric emptying and symptom scores, age or diabetes duration. Conclusions: As a measure of gastric emptying, the ROM test has benefits of being affordable and available. Compared to GEBT, the method has low diagnostic reliability. Before continued use, we recommend additional studies validating the test in diabetes patients.

Keywords: 13carbon-labelled gastric emptying breath tests; diabetes mellitus; gastric emptying; gastroparesis; patient-reported outcomes; radiopaque markers.

Figure 1

Study protocol. Abbreviations: PAGI-SYM, Patient assessment of upper gastrointestinal symptom severity index; GEBT, ¹³Carbon-labelled gastric emptying breath test of solids; ROM, [gastric emptying of] radiopaque markers.

Figure 2

Inclusion flow chart. Abbreviations: ROM, [gastric emptying of] radiopaque markers; GEBT, ¹³Carbon-labelled gastric emptying breath test of solids.

Figure 3

ROC curve for a positive ROM test compared to GEBT. Area under curve for the ROC was 0.63 (95% CI 0.48–0.70). Abbreviations: ROC, receiver operating characteristics; AUC, area under curve; ROM, [gastric emptying of] radiopaque markers; GEBT, ¹³Carbon-labelled gastric emptying breath test of solids.

Figure 4

Inter-rater reliability testing showed a strong correlation of r=0.97, P<0.0001. Abbreviation: ROM, [gastric emptying of] radiopaque markers.

Figure 5

Abdominal radiograph taken after four hours in a patient undergoing ROM testing. As indicated by the red arrow, ROM can be hard to distinguish from the underlying spine, as they have almost identical X-ray opacity. Abbreviation: ROM, [gastric emptying of] radiopaque markers.