Lactulose Breath Testing Does Not Discriminate Patients With Irritable Bowel Syndrome From Healthy Controls

Jason R. Bratten, B.S.; Jennifer Spanier, D.O.; Michael P. Jones, M.D.


Am J Gastroenterol. 2008;103(4):958-963. 

In This Article

Abstract and Introduction

Introduction: Recent reports suggest that abnormalities of lactulose breath testing (LBT) are common in patients with irritable bowel syndrome (IBS), although the criteria for abnormal studies are poorly validated, and controlled comparisons are limited. The goal of this study was to determine the prevalence of abnormal LBT using the previously published criteria in both IBS patients and healthy controls, as well as to determine the prevalence and symptom association with methane (CH4) and hydrogen (H2) productions during LBT.
Methods: Consecutive LBT from patients meeting Rome II criteria for IBS and healthy control subjects were examined. Patients listed their most bothersome digestive symptom at the start of the test. LBT was performed using 10 g of lactulose mixed in 240 mL of water, and breath samples collected every 20 min for a 180-min period. Both breath H2 and CH4 were measured. LBT was considered positive if it met any of the previously published criteria: (a) breath H2 of >20 parts per million (ppm), (b) increase in breath H2 in <90 min, (c) dual H2 peaks (12-ppm increase over baseline with a decrease of ≥5 ppm before 2nd peak), and (d) breath CH4 of >1 ppm.
Results: In total, 224 patients with IBS and 40 controls were studied. Twenty percent of IBS patients were CH4(+) compared with 15% of controls. CH4(+) IBS patients were significantly more likely than CH4(-) IBS patients to have constipation, and significantly less likely to have diarrhea; however, the association did not hold for symptoms of bloating or pain. Patients and controls did not differ significantly with respect to the frequency of a positive study defined by increase in breath H2 in <90 min (121 per 180 vs 26 per 40, P = 0.79), increase in breath H2 of >20 ppm (92 per 180 vs 24 per 40, P = 0.31), or dual peaks (25 per 180 vs 9 per 40, P = 0.17).
Conclusions: The majority of patients with IBS and healthy subjects meet criteria for an "abnormal" LBT using previously published test criteria, and groups are not discriminated using this diagnostic method. Similarly, while CH4 production was associated with constipation among IBS patients, the prevalence of CH4-positive subjects did not significantly differ between IBS patients and controls. The utility of LBT, in its current form as a diagnostic tool in IBS requires critical reappraisal.

Irritable bowel syndrome (IBS) is a chronic gastrointestinal disorder affecting approximately 11-14% of the population.[1,2] The pathophysiology of the disorder remains unknown, and is likely multifactorial. A variety of possible mechanisms have been proposed, including alterations in intestinal motility and sensation, alterations in brain-gut regulatory pathways, postinfectious or postinflammatory changes in digestive neuroimmune function, and alterations in intestinal microflora.[3]

Much recent attention has focused on the role of luminal microflora in intestinal function and dysfunction.[4,5,6] Within IBS specifically, tremendous attention has been focused upon abnormalities in lactulose breath testing (LBT) as a presumptive surrogate marker for small intestinal bacterial overgrowth (SIBO).[7,8,9] Reports from a single center cite abnormal LBT in 78-84% of IBS patients meeting Rome II criteria.[10,11] Subsequent studies report symptom improvement associated with normalization of LBT using either an uncontrolled 14-day elemental diet,[9] a variety of antibiotics studied in an uncontrolled fashion,[10] or a placebo-controlled trial of neomycin.[11] Two recent studies have demonstrated symptom improvement following treatment with the nonabsorbable antibiotic rifaximin.[12,13] Symptom improvement was limited to symptoms of bloating and flatulence. Bowel pattern was not significantly affected. Only Sharara et al. reported LBT results that symptom improvement was associated with decreased breath hydrogen (H2) production.

While the role of enteric microflora as either a pathogen or therapeutic ally in IBS is an exciting and potentially important area, the construct of an abnormal LBT as synonymous with SIBO, and, indeed, the clinical utility of LBT in IBS, requires critical appraisal. The basis for this statement derives from several facts. First, SIBO has never been established in any study reporting abnormalities of LBT. The difficulties of proving SIBO in general, and distal SIBO in particular, are acknowledged. However, in methodological comparisons of patients with culture-proven SIBO, LBT fares poorly compared with either glucose H2 breath testing or the 14C-xylose test.[14,15,16,17] In particular, the concordance of LBT with the 14C-xylose test, generally regarded as the most robust noninvasive diagnostic measure, has been quite poor.[14,16,17]

Perhaps more importantly, there are limitations with the criteria for an abnormal LBT in patients with IBS. Pimentel et al. have published varying criteria for normal LBT.[7,8,10,11] The reported criteria for an abnormal LBT are any sustained increase in breath H2 concentration in less than 90 min after the ingestion of lactulose, or an increase in breath H2 concentration of more than 20 parts per million (ppm) at any time during the study. The citations supporting these criteria in the work of Pimentel et al. are the original work of Bond et al. on LBT as a measure of intestinal transit time,[18] a study comparing 14C-glycocholate and LBT in patients with suspected proximal SIBO and controls,[19] and a letter in a pediatric nutrition journal on LBT as a measure of orocecal transit time in hospitalized infants.[20] The final proposed criterion is the presence of two distinct breath H2 peaks, but the validity of this measure is questionable as the pattern may be seen with bacterial fermentation in the right and left colons.[15,19]

As the criteria used to define normality with LBT have been variously drawn from studies using the measure to assess intestinal transit time as well as bacterial overgrowth, it would seem critical to validate these criteria in both healthy subjects and patients with IBS. Unfortunately, this does not seem to be the case. Pimentel et al. have reported a group of 15 controls with an abnormal LBT based upon rapidity or magnitude of breath H2 rise in 20%.[7,11] Dual peaks were seen in 13% of controls. Walters and Vanner reported that, in a series of 39 patients with IBS and 20 controls, no differences existed between the two groups when applying the various criteria used by Pimentel et al.[16]

Because of these methodological shortcomings, we sought to evaluate the utility of LBT by determining the prevalence of abnormal breath tests using the previously published criteria in patients with IBS and healthy controls. The aim of the proposed study is to test the null hypothesis that the proportion of positive LBT is identical between controls and IBS patients, and alternatively to estimate the difference between the two populations.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.