Poor reproducibility of breath hydrogen testing: Implications for its application in functional bowel disorders

Abstract

Background: Limited data are available regarding the reproducibility of lactulose and fructose breath testing for clinical application in functional bowel disorders.

Objectives: The purpose of this study was to investigate the reproducibility of lactulose and fructose breath hydrogen testing and assess symptom response to fructose testing.

Methods: Results were analysed from 21 patients with functional bowel disorder with lactulose breath tests and 30 with fructose breath tests who completed another test >2 weeks later. Oro-caecal transit time, hydrogen responses, both qualitatively (positive/negative) and quantitatively (area under the curve (AUC) for hydrogen), were compared between tests. In another 36 patients, data scores for overall abdominal symptoms, abdominal pain, bloating, wind, nausea and fatigue were collected during the fructose test and compared to hydrogen responses.

Results: No correlations were found for lactulose AUC (linear regression, p = 0.58) or transit time (Spearman's p = 0.54) between tests. A significant proportion (30%) lost the presence of fructose malabsorption (p < 0.01). Hydrogen AUC for fructose did not correlate between tests, (r = 0.28, p = 0.17) independent of time between testing (p = 0.82). Whilst patients with fructose malabsorption were more likely to report symptoms than those without (56% vs 17%; p = 0.04), changes in symptom severity were not different (p > 0.05).

Conclusions: Routine use of lactulose and fructose breath tests in functional bowel disorder patients is not supported due to its poor reproducibility and low predictive value for symptom responses.

Keywords: Breath test; diet; fructose; irritable bowel syndrome; lactulose.

Figure 1.

The relationship of breath hydrogen area under the curve (AUC) between two separate testing periods (a) after 15 g lactulose; and (b) after 35 g fructose. Neither analysis showed statistically significant relationships (linear regression, (a) r = 0.13; p = 0.58; and (b) r = 0.28; p = 0.17).

Figure 2.

Comparison of the change in oro-cecal transit time with the change breath hydrogen production (shown as area under the curve (AUC)) after (a) 15 g lactulose or (b) 35 g fructose. A decrease in oro-caecal transit time (OCTT) was significantly correlated with increased hydrogen production (Spearman's r=–0.41; p = 0.09 for lactulose; r=–0.47, p = 0.05 for fructose).

Figure 3.

Changes in symptom scores according to the visual analogue scale for those with and without fructose malabsorption. *This difference was not statistically significant after adjusting for multiple comparisons using Bonferroni's correction, adjusted p-value=0.004. No significant differences were observed for the other symptoms (unpaired t-test).

Figure 4.

Comparison of the prevalence of individuals reporting symptoms across different oro-caecal transit times after consumption of fructose. Comparisons across the groups were computed using chi-square analysis. Where + ve is shown, this indicates positive response or worsening of symptoms; –ve indicates negative response or absence of symptoms.