Neurogastroenterology & Motility Neurogastroenterol Motil (2015)
doi: 10.1111/nmo.12516
Small intestinal bacterial overgrowth: duodenal aspiration vs glucose breath test A. ERDOGAN ,* S. S. C. RAO ,* D. GULLEY ,* C. JACOBS ,† Y. Y. LEE *,‡
& C. BADGER *
*Section of Gastroenterology and Hepatology, Georgia Regents University, Augusta, GA, USA †Carver College of Medicine, University of Iowa, Iowa City, IA, USA ‡School of Medical Sciences, Universiti Sains Malaysia, Kota Bharu, Kelantan, Malaysia
Key Messages GOALS OF RESEARCH ○
To compare the diagnostic yield of glucose breath test and duodenal aspirate/culture, and analyze the influence of ethnics, age, and, gender on prevalence of SIBO, and assess symptom patterns.
BASIC METHODOLOGY ○
Duodenal aspiration was performed with Liguory catheter during an upper endoscopy. GBT was performed (QuinTron Instrument Company, Inc. Milwaukee, WI, USA) and alveolar gas was analyzed for both H2 and CH4 levels by chromatography.
RESULTS/TAKE HOME MESSAGES There is a higher yield for SIBO with duodenal culture at ≥103 CFU/mL. ○ For glucose breath test, the diagnostic yield is lower but there is adequate agreement with culture and a good specificity. ○ The overall agreement of duodenal culture and GBT is 66%; suggesting neither test is perfect. ○ Other than age and certain symptom patterns (gas), ethnical or gender differences did not appear to influence the prevalence of SIBO. ○
Abstract Background The diagnosis of small intestinal bacterial overgrowth (SIBO) remains challenging. Our aim was to examine the diagnostic yield of duodenal aspiration/culture and glucose breath test (GBT), and
effects of gender, race and demographics on prevalence of SIBO. Methods Patients with unexplained gas, bloating and diarrhea and negative endoscopy, imaging and blood tests were prospectively enrolled in two centers in USA. Randomly, within 1 week each patient underwent both duodenal aspiration/culture and GBT. The diagnostic yield of each test and relationship of symptoms, and effects of ethnicity, age, and gender on prevalence of SIBO were assessed and compared. Key Results Duodenal culture was positive in 62/139 (44.6%) subjects and GBT was positive in 38/139 (27.3%) subjects with an overall diagnostic agreement of 65.5%. The sensitivity, specificity, positive and negative predictive value of GBT was 42%, 84%, 68%, and 64%, respectively. Ethnicity or gender did not influence SIBO, but SIBO positive patients were older (p = 0.0018). Symptom patterns were similar except bloating was more prevalent in GBT positive
Address for Correspondence Satish S.C. Rao, MD, PhD, FRCP (LON) AGAF FACG, Department of Medicine, Section of Gastroenterology and Hepatology, Medical College of Georgia, Georgia Regents University, 1120 15th Street, BBR 2538, Augusta, GA 30912, USA. Tel: 706-721-2238; fax: 706-721-0331; e-mail: [email protected] Portions of this work were presented at: ANMS 2011, St Louis, Missouri, USA. (Neurogastroenterol Motil 2011; 23 Supp s1, 50:16); ACG 2012 Las Vegas, NV, USA (Am J Gastroenterol. 2012; 107 (Supp s1, 329): S141); ANMS 2013, Huntington Beach, California, USA (Neurogastroenterol Motil 2013; 25 (Supp s1, 19): 7). Received: 10 October 2014 Accepted for publication: 22 December 2014
© 2015 John Wiley & Sons Ltd
1
A. Erdogan et al.
Neurogastroenterology and Motility
However, the sensitivity and specificity of breath tests in the diagnosis of SIBO remains unclear because of the heterogeneity of patient populations that have been examined, methodological issues including use of different substrates and different doses of substrate. GBT has been performed with 50 g,12–16 75 g,17 80 g 18 and 100 g glucose,4,19,20 and likewise LBT has been performed with 10 g,18,21,22 16 g,23 and 25 g lactulose.24 Furthermore, LBT may overestimate SIBO (false positive), because the rise in hydrogen (H2) usually coincides with the arrival of substrate in cecum and not necessarily from bacterial fermentation in small bowel.4,25 Many investigators also have analyzed breath samples for H2 alone 13,26 and not methane (CH4). Finally, different investigators have used different cut-offs for defining an abnormal breath test; rise of 10 ppm,18,27 12 ppm,19,28,29 15 ppm,22 and 20 ppm 12,15,30 over baseline. Additionally, whether factors such as geography, race, age, or gender influence the prevalence of SIBO is not known. We tested the hypothesis that GBT provides a similar yield to that of duodenal aspiration/culture for diagnosis of SIBO. Our aim was to examine the diagnostic yield of duodenal aspiration and culture (aerobic and anaerobic) for SIBO, and to compare this with that of GBT in the same subjects. Additionally, we assessed the prevalence of SIBO with both tests in two different geographical regions in USA, each with different ethnic distribution, and examined the relationship of symptoms, age, and gender on the prevalence of SIBO.
and gas in culture positive subjects. Conclusions & Inferences Duodenal aspiration/culture identifies 45% of patients with suspected SIBO. GBT has lower sensitivity but good specificity for detection of SIBO. There were no ethnic or gender differences in the prevalence of SIBO, but patients with SIBO were older. Because GBT is non-invasive, it should be considered first in patients with suspected SIBO. Keywords duodenal culture, ethnics, glucose breath test, sensitivity and specificity, small intestinal bacterial overgrowth.
INTRODUCTION Small intestinal bacterial overgrowth (SIBO) is characterized by the presence of abnormal and high levels of bacteria in the small intestine associated with gastrointestinal (GI) symptoms such as bloating, distention, flatulence, abdominal discomfort, diarrhea, and weight loss.1–3 The anatomical and structural abnormalities that occur in scleroderma or intestinal pseudo-obstruction or those that follow intestinal surgeries that form a blind loop often predispose to SIBO, but it can also occur in the absence of these conditions.2,4–6 Recently, proton pump inhibitor use and small intestinal dysmotility have been shown to be independent risk factors that also predispose to SIBO.2 In clinical practice, the diagnosis of SIBO remains challenging, because there is no widely accepted diagnostic test. Aspiration and culture of small intestinal fluid is generally believed to be the gold standard. However, this is an invasive procedure and requires either upper endoscopy or small intestinal intubation under fluoroscopy. There is also limited information regarding the optimal technique for aspiration and culture, and in clinical practice its diagnostic yield remains unclear. Contamination from oral flora or saliva during oral intubation may cause false positive result and an overestimation of SIBO.7 In contrast, aspiration/culture from the proximal portion of the small bowel may miss bacterial overgrowth in the distal small bowel causing false negative results;8 also, anaerobic culture is not routinely performed in most laboratories.5,8,9 Thus, the utility of aerobic and anaerobic culture in the diagnosis of SIBO is not fully known. There is also a need for less invasive and inexpensive test for the detection of SIBO in routine clinical practice. The glucose breath test (GBT) and the lactulose breath test (LBT) have been advocated for the past few years, because they are safe, easy and can be performed in children and pregnant women.10,11
MATERIALS AND METHODS Patients Adult patients with unexplained GI symptoms such as gas, bloating, diarrhea, and abdominal discomfort and suspected to have SIBO, and who were referred to two specialist motility centers, one in mid-west (University of Iowa) and the second in southern USA (Georgia Regents University) were prospectively enrolled. All patients had normal upper endoscopy, colonoscopy, computerized abdominal tomography scan, and normal hematology, biochemical profiles, tissue transglutaminase antibody, thyroid stimulating hormone, and normal right upper quadrant ultrasound scan. Patients with a history of previous GI surgeries (except for cholecystectomy, hysterectomy, and appendectomy), those with significant co-morbid medical problems (including stroke, COPD (Chronic Obstructive Pulmonary Disease), cancers and disabled) or those who had diabetes mellitus or who were hospitalized or had a history of recent antibiotic use (4 weeks) were excluded from the study. Patients were asked to discontinue the use of proton pump inhibitor, laxatives and drugs that affect intestinal motility (opioids, anticholinergics, antidiarrheals) at least 1 week prior to study, but were allowed to continue with other usual medications. All consecutive subjects underwent both the GBT and
2
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
Statistical analyses
duodenal aspiration/culture within 1 week in a random order. The University of Iowa Hospitals and Clinics and Georgia Regents University Investigation Review Boards approved the study.
Statistical analyses were performed using SPSS for Windows 11.0 (SPSS Inc., Chicago, IL, USA). All data were presented as mean SD unless otherwise stated. Diagnostic agreement (positive %, negative % and overall %) and yield (sensitivity, specificity, positive and negative predictive value) were determined for GBT against the gold standard, duodenal culture. Univariable analysis (t-test for continuous variables and chi-squared or Fisher’s Exact test for categorical variables) and multivariable analysis (multiple logistic regression: backward LR method) were used to determine association between factors including age, gender, ethnics, differences in symptom patterns and with GBT and duodenal culture. A p < 0.05 was considered as statistically significant for above tests.
Duodenal aspiration and culture After an overnight fast, an upper endoscope was passed into the distal duodenum with minimal air insufflation. Next, under aseptic techniques (sterile gloves worn by the endoscopist and nurse assisting the procedure just before aspiration), a 2–3 mm Liguory catheter (COOK Medical, Bloomigton, IN, USA) was advanced through the biopsy channel of the endoscope into third or fourth portions of the duodenum. Using gentle intermittent suction with a 5 mL syringe, 3–5 mL of duodenal juice was aspirated and immediately sent to the microbiology laboratory for aerobic and anaerobic cultures.
RESULTS Glucose breath test Demographics and symptom profiles
One day before the GBT, patients were asked to consume a low carbohydrate diet without vegetables, fiber and dairy products. After an overnight fast, patients were asked to brush their teeth and rinse their mouth with an antiseptic mouthwash, at least 2 h before the test to avoid false positive high basal levels from fermentation of substrate by oral bacteria. Patients were asked to refrain from any exercise, sleep or smoke cigarette during the test to avoid possible effects on the breath H2 levels. On the test day, 75 g of glucose dissolved in 250 mL water was administered, and breath samples were obtained at baseline and at 15 min intervals for a 2 h period. The samples were collected in a bag (QuinTron Instrument Company, Inc.) and alveolar gas was analyzed for both H2 and CH4 levels by chromatography (QuinTron Micro Analyzer, QuinTron Instrument Company, Inc.).
One hundred and thirty-nine consecutive patients were included in this study, of whom 78 (M/F = 17/ 61, mean age = 48.9 1.9 years) were enrolled from Georgia and 61 (M/F = 16/45, mean age = 45.9 1.8 years) from Iowa (p = 0.33). Because there were no geographical differences in the prevalence of SIBO with either GBT or cultures between Georgia and Iowa, the data were pooled for all analyses. The mean age was significantly higher in patients who were GBT positive (54.4 2.4 years) vs GBT negative (45.1 1.5 years), p = 0.002, and also for patients who were culture positive (52.8 2.0 years) vs culture negative (43.5 1.7 years), p = 0.0004. Logistic regression analysis showed that SIBO positive patients were significantly older than SIBO negative patients, both with GBT (OR: 1.04 [95% CI: 1.01–1.07, p = 0.003]) and duodenal culture at ≥103 CFU/mL (OR: 1.04 [95% CI: 1.01–1.06, p = 0.003]) and ≥105 CFU/mL (OR: 1.05 [95% CI: 1.02–1.09, p = 0.002]; Table 1). The dominant symptoms in patients (mean score 6 or higher) were bloating, abdominal pain, fullness, and gas.
Symptom questionnaire All patients filled out a validated questionnaire for the presence/ absence of 9 symptoms using a validated scale31,32 including abdominal pain, belching, bloating, fullness, indigestion (dyspepsia-discomfort centered in upper abdomen), nausea, diarrhea, vomiting and gas. If present, patients were asked to rate the frequency, intensity and duration of each symptom on a 0–3 point Likert scale. Intensity: 0 = no symptom, 1 = mild, 2 = moderate, 3 = severe symptom. Frequency: 0 = none, 1 = 1 episode/week. Duration: 0 = none, 1 = 30 min. The total score could range from 0 to 9 for each symptom. A mean score was calculated for each symptom, and an overall mean total score was calculated for all 9 symptoms.
Table 1 Effects of symptoms and demographic features on prevalence of SIBO Outcome and risk factors*
Data analysis
SIBO based on GBT Older age 1.04 SIBO based on culture ≥103 CFU/mL Older age 1.04 Gas 1.16 5 SIBO based on culture ≥10 CFU/mL Older age 1.05
The culture results were considered as positive for SIBO if one or more organisms (aerobic or anaerobic) were cultured with a colony count of ≥103 CFU/mL.2 We also assessed the prevalence of SIBO at a threshold of ≥105 CFU/mL. Glucose breath test was considered positive for SIBO if the following criteria were met; ≥20 ppm increase above baseline for H2,17 or ≥15 ppm increase above baseline for CH4 or ≥20 ppm increase above baseline for H2 and CH4.2,12,30,33 Additionally, we performed a separate analysis, using an increase of ≥12 ppm H2 values above baseline.19,34
© 2015 John Wiley & Sons Ltd
OR (95% CI)
p-value
(1.01–1.07)
0.003
(1.01–1.06) (1.01–1.33)
0.003 0.03
(1.02–1.09)
0.002
*Only factors in logistic regression (Backward LR method) with significant p-value ( 0.05; Fig. 1A). The mean score for gas was significantly higher for culture positive vs culture negative patients (p = 0.023; Fig. 1B). The mean score for other symptoms was similar (p > 0.05). Multiple logistic regression analysis also found that gas was significantly higher in SIBO positive (duodenal culture) patients at ≥103 CFU/mL (OR: 1.2 [95% CI: 1.01–1.33, p = 0.03]; Table 1). The overall mean total symptom score was similar for patients who tested positive vs negative for GBT (46.5 vs 46.1, p = 0.9) or for duodenal culture (46.7 vs 45.9, p = 0.8). We also compared mean symptom scores and mean total symptom score for culture positive patients at threshold of 103 CFU/mL vs 105 CFU/mL. The symptom scores were not significantly different for any symptom at the thresholds of 103 CFU/mL vs 105 CFU/mL; p > 0.9.
≥105 CFU/mL analysis At this threshold, the duodenal aspiration/culture was positive in 25 (18%) patients. 18 (72%) had positive aerobic cultures and 7 (28%) had both positive aerobic and anaerobic cultures. None of the patients had positive anaerobic cultures alone. Alpha-hemolytic streptococcus, Escherichia coli and Klebsiella species were the most commonly isolated organisms.
Diagnostic yield of GBT Mean baseline H2 and CH4 levels for GBT were 7.2 12.8 ppm and 5.0 13.5 ppm, respectively, and mean peak values were 34.3 76.8 and 8.8 21.3 ppm, respectively; H2 values ranged between 0 to 498 ppm and CH4 values ranged between 0 to 123 ppm. In total, 11 patients had high baseline H2 values (H2 >20 ppm). Of these, eight were positive and three were negative for GBT according to criteria used to define a positive test, and high baseline values were not considered as positive test if it did not meet the diagnostic criteria. A positive GBT was seen in 38 (27.3%) and a negative test in 101 (72.7%) patients. 24
Diagnostic yield of duodenal culture ≥103 CFU/mL analysis The duodenal aspiration/culture yielded a positive result in 62 (44.6%) patients, and 45 (36.3%) samples grew more than one organism. The most common aerobic bacterial organisms were Alpha-hemolytic Streptococcus, Escherichia coli, and
A
*
8
7
Mean symptom scores
6
5
4 Total patients
3
GBT positive 2 GBT negative 1 * p = 0.009
0 Abdom pain
B
Belching
Bloating
Fullness
Indigestion
Nausea
Diarrhea
Vomiting
8
Gas
*
7
Mean symptom score
6 5 4 3 Total patients 2
Culture positive
1
Culture negative * p = 0.023
0 Abdom pain Belching
Bloating
Fullness
Indigestion
Nausea
Diarrhea
Vomiting
Gas
4
Figure 1 Mean symptom scores of (A) all patients and those with a positive and negative GBT and (B) all patients and those with a positive and negative culture (≥103 CFU/mL). GBT, Glucose breath test.
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
sensitivity, specificity, positive and negative predictive values were 41.9%, 84.4%, 68.4%, and 64.4%, respectively. When a ≥12 ppm increase in H2/CH4 from baseline was considered as positive for GBT, the sensitivity, specificity, positive and negative predictive values were 43.6%, 77.9%, 61.4%, 63.2%, respectively. The positive, negative and overall agreements were 43.6%, 77.9% and 62.6%, respectively.
Table 2 Culture results of duodenal aspirates* 3
Culture results (≥10 CFU/mL)
Total, N = 124 (%)
Escherichia coli Klebsiella oxcytoca Klebsiella pneumonia Enterobacter cloacea Enterobacter aerogenes Pseudomonas aeruginosa Neisseria spp. Hemophilus parainfluenzae Hemophilus influenzae Stenotrophomonas maltophilia Bacteriodes vulgatus Velionella spp. Prevotella spp. Alpha-hemolytic streptococcus Non-hemolytic streptococcus Beta-hemolytic streptoccus Staphylococcus aureus Coagulase negative staphylococcus Diphtheroid bacilli Rothia spp. (Stomatococcus) Lactobacillus spp. Clostridium perfringes Bacillus spp.
15 2 7 2 2 1 6 1 1 1 4 1 2 38 6 1 3 6 9 4 8 3 1
(12.1) (1.6) (5.6) (1.6) (1.6) (0.8) (4.8) (0.8) (0.8) (0.8) (3.2) (0.8) (1.6) (30.6) (4.8) (0.8) (2.4) (4.8) (7.3) (3.2) (6.5) (2.4) (0.8)
For ≥105 analysis A positive % agreement for duodenal aspiration/culture and GBT was seen in 16/25 (64%), and negative % agreement in 92/114 (80.7%) subjects with an overall % agreement of 108/139 (77.7%). Sensitivity, specificity, positive, and negative predictive values were 64%, 80.7%, 42.1%, and 91.1%, respectively. When a ≥12 ppm increase from baseline was considered positive for GBT, the sensitivity, specificity, positive, and negative predictive values were 68%, 76.3%, 38.6%, 91.6%, respectively. The positive, negative and overall agreements were 68%, 76.3%, and 74.8%, respectively.
*36.3% of samples grew more than one organism.
Racial differences in prevalence of SIBO
(63.2%) subjects were diagnosed with SIBO on the basis of increase in H2 level alone, seven (18.4%) with CH4 level alone, and 7 (18.4%) with increase in both H2 and CH4 levels. When ≥12 ppm increase was used as the threshold cut-off, the GBT was positive in 44 (31.7%) and negative in 95 (68.3%) patients.
There is a significant difference in the racial demographic population distribution with Whites composing 56% of population in Georgia and 86% in Iowa, and Blacks composing 30% in Georgia but only 3% in Iowa.35 Our study population comprised of 82.1% and 93.4% Whites and 11.5% and 3.3% of Blacks in Georgia and Iowa, respectively. There were fewer Blacks and more Whites than expected in the study population from Georgia (p = 0.016). The prevalence of SIBO was similar in Whites vs Blacks, with a positive GBT in 26.4% vs 18.2% (p = 0.2), and a positive duodenal aspiration/culture in 44.6% vs 36.4% (p = 0.7), respectively (Table 3).
Diagnostic agreement For ≥103 analysis A positive % agreement for duodenal aspiration/culture and GBT was seen in 26/62 (41.9%) subjects, and a negative % agreement in 65/77 (84.4%) subjects. The overall % agreement for duodenal aspiration/culture and GBT was 91/139 (65.5%). The
% of aerobic and anaerobic culture growth (103 CFU/mL)
22.6%
3.2%
Aerobic culture
74.2% Figure 2 Aerobic and anaerobic distribution of positive culture results.
© 2015 John Wiley & Sons Ltd
Anaerobic culture
Aerobic and anaerobic culture
5
A. Erdogan et al.
Neurogastroenterology and Motility
Table 3 Racial distribution of study population for GBT and duodenal aspiration/culture
White African American Hispanic Other
Study population (N = 139)
Positive GBT, N (%)
Negative GBT, N (%)
Positive culture*, N (%)
Negative culture*, N (%)
121 11
32 (26.4) 2 (18.2)
89 (73.6) 9 (81.8)
54 (44.6) 4 (36.4)
67 (55.4) 7 (63.6)
4 3
3 (75) 1 (33.3)
1 (25%) 2 (66.7)
2 (50) 2 (66.7)
2 (50) 1 (33.3)
*103 CFU/mL. GBT, glucose breath test.
Effects of gender on prevalence of SIBO
bacteria from the colon, and its frequent exposure to acid from the stomach, all of which should decrease the risk of SIBO.2 The techniques for aspiration and culture have not been standardized. Some have used endoscopic suction of duodenal juice38,40 with a high chance of cross-contamination as opposed to our technique of placing a sterile catheter into the distal duodenum under aseptic techniques, including use of sterile gloves. In our study, the yield of positive culture was 45% at ≥103 CFU/mL and 18% at ≥105 CFU/mL. A low cut-off of 103 CFU/mL was clinically appropriate for the type of population we studied.2 The ranges for a positive culture reported in the literature have varied from as low as 8% in patients with diarrhea and weight loss40 (>105 CFU/mL aerobic and >104 CFU/ mL anaerobic) to as high as 100% in postsurgical patients with Billroth II.38 Culture was positive in 39.5% of patients with malabsorption,4 66.7% of patients with GI symptoms (20/30),27 65% of patients with cirrhosis.17 Pyleris et al. found the yield of positive culture to be 19.4% at >104 CFU/mL, suggesting that the variability in the yield of culture depends on the threshold used to define a positive culture and the patients’ characteristics.41 The relatively poor symptom correlations with positive cultures in our study suggest that a functional or psychological etiology besides SIBO may also be important in the pathogenesis of unexplained GI symptoms and confirm a previous study 2 that symptoms are poor predictors of SIBO. The diagnostic yield of breath tests for SIBO in the literature has also been variable because of differences in the dose and type of substrate used, the threshold for defining a positive breath test, analysis of breath samples for H2 or a combination of H2 and CH4, and also the patients’ characteristics. In our study, GBT was positive in 27% of patients with unexplained GI symptoms. In IBS patients, GBT was positive in 31%13 and LBT in 34.3–84%.18,42 In a comparative study, GBT (≥12 ppm H2) was positive in 20/75 (26.7%), LBT in 13/71 (18.3%) and culture
Glucose breath test was positive in 26.4% of females and 30.3% of males (OR: 0.8 [95% CI: 0.3–1.9, p = 0.7]). Similarly, 44.3% of females and 45.5% of males were SIBO positive by culture at ≥103 CFU/mL (OR: 1.0 [95% CI: 0.4–2.1, p = 1.0]). There were no gender differences in the prevalence of SIBO (p > 0.05).
DISCUSSION In this study, we assessed and compared the diagnostic yield of GBT with duodenal aspiration and culture in patients with unexplained GI symptoms (gas, bloating and diarrhea) and with negative endoscopy, imaging and blood tests, and the effects of age, gender and race on symptoms and prevalence of SIBO. Glucose breath test was positive in 27% of patients, whereas the duodenal culture was positive in 45% of patients with unexplained GI symptoms. Alpha-hemolytic Streptococcus, Escherichia coli and Klebsiella species were the most common organisms isolated from cultures, and 74% showed a positive aerobic culture, 3% positive anaerobic culture and 23% with both positive aerobic and anaerobic cultures. Thus, the sensitivity and specificity of GBT were 42% and 84%, respectively. Aspiration and culture of small bowel fluid is generally accepted as the gold standard for diagnosis of SIBO, however, there is no established cut-off value for defining the abnormal levels of bacterial growth especially in different parts of the small intestine.2,36–38 The proximal jejunum may contain ≤104 CFU/mL bacteria in healthy subjects.37–39 In one study, 12% of healthy subjects had bacterial counts ranging from 0 to 103 CFU/mL in the proximal jejunum.22 As reported recently, a lower cut-off value of ≥103 CFU/mL may be more clinically relevant for aspirations obtained from the third and fourth portions of the duodenum, given its proximal location, relative protection from translocation of
6
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
(≥105 CFU/mL) in 39.5% of patients with malabsorption.19 In another study, culture (≥103 CFU/mL) was positive in 4/12 (25%) and LBT in 8/15 (53.3%) patients.8 Typically, LBT had higher positive yield compared to GBT largely due to false positive results from cecal rather than small bowel fermentation.18,25,43 Assessment of breath CH4 levels can also affect the yield of GBT. In our study, 18.4% of patients were diagnosed with a positive GBT on the basis of an increase in CH4 values, and this is comparable to yields of 29%12 and 30%30 reported by others.12 The sensitivity and specificity of GBT not only depend on values of H2 and or CH4 cut-offs but also on the culture cut-offs. Not surprisingly, the sensitivity has ranged between 44% and 62.5% and specificity between 36% and 83%.17,19,44,45 In one study, the sensitivity and specificity of GBT (≥12 ppm) were 44% and 80%, respectively using a duodenal culture cut-off value of ≥105 CFU/mL.19 If these criteria were applied to our study, the sensitivity for GBT increased to 68%, whereas the specificity decreased to 76.3%. Unlike our study, Ghoshal el al. did not measure CH4 levels that might have lowered their detection rate of SIBO.19 In another study, Corazza et al. 45 found a sensitivity and specificity of 62% and 83%, respectively for GBT, but they cultured fluid from the proximal jejunum at two different sites and used a bacterial count of ≥106 CFU/mL that might have increased the detection rate for SIBO. In our study, the lower H2 cut-off value (12 ppm) had minimal impact on the sensitivity and positive/negative predictive values, but yielded a lower specificity. Hence, we suggest H2 ≥20 ppm as an appropriate cut-off for diagnosis of SIBO, but this needs further validation. Thus, the specificity for a positive GBT, usually 76–85%, was comparable, although the sensitivity has varied depending on the threshold cut-off values, the site and technique of aspiration, and bacterial concentration (103 vs 105 CFU/mL). The high specificity makes GBT a reliable test especially when positive, whereas the low sensitivity could be due to rapid absorption of glucose and its lack of availability as a substrate in distal small bowel or low concentrations of bacteria,46 or the inability of bacteria to ferment glucose, the solubility of H2 in blood, its transfer and excretion from lungs and lung disease.27 Although there are differences in the ethnic makeup of Georgia and Iowa, we found that the prevalence of SIBO was similar between these regions and between the ethnic groups (Whites and African
© 2015 John Wiley & Sons Ltd
Americans), and between men and women. This lack of gender effect confirms a previous observation47 and suggests that unlike other GI motility disorders, the prevalence of SIBO is not affected by gender. However, we found a higher prevalence of SIBO in older subjects confirming previous studies.30,40 Older subjects are more likely to have risk factors for SIBO including intestinal dysmotility, intestinal surgery, small bowel diverticulosis, achlorhydria, and medication use including proton pump inhibitor.48 The limitations of our study include a potential referral bias because patients were assessed in two tertiary care centers, and may not reflect SIBO prevalence in the community, however, it may not be feasible to do a population-based study because these assessments require breath tests or endoscopy. Aspiration was also performed from the third/fourth portions of duodenum and whether aspiration more distally could have yielded a higher positive rate for culture is unclear. In spite of aseptic techniques, we cannot definitely exclude the potential for contamination from oral bacteria because of performing an endoscopy. However, the sterile technique was consistently used, and by one single operator, and the high prevalence of bacteria in duodenal aspirates confirms the presence of SIBO. Finally, given the sample distribution, a type II error cannot be excluded for some of the demographic observations. In conclusion, GBT has a low sensitivity but high specificity for detection of SIBO when compared to duodenal culture. It is non-invasive and widely available and hence in clinical practice this test may be considered first in a patient with suspected SIBO. However, the overall agreement for duodenal culture and GBT was 66% suggesting that neither test was perfect. Therefore, in selected patients with a high index of suspicion for SIBO, both tests may be required.
ACKNOWLEDGEMENTS We acknowledge the technical and secretarial assistance of m/s Michelle Jackson, Dr. Enrique Coss Adame, and Ms. Helen Smith. Guarantor of article: Satish S.C. Rao, MD, PhD, FACG.
FUNDING None.
CONFLICTS OF INTEREST None.
7
A. Erdogan et al.
Neurogastroenterology and Motility
AUTHOR CONTRIBUTION
data, and drafting of manuscript; DG analysis and interpretation of data, and drafting of manuscript; CJ analysis and interpretation of data, and drafting of manuscript; YYL analysis and interpretation of data, and drafting of manuscript; CB analysis and interpretation of data.
AE study concept and design, data collection, analysis, and interpretation of data, and drafting of manuscript; SSCR study concept and design, data collection, analysis, and interpretation of
REFERENCES 1 Sachdev AH, Pimentel M. Gastrointestinal bacterial overgrowth: pathogenesis and clinical significance. Therapeutic advances in chronic disease 2013; 4: 223–31. 2 Jacobs C, Coss Adame E, Attaluri A, Valestin J, Rao SS. Dysmotility and proton pump inhibitor use are independent risk factors for small intestinal bacterial and/or fungal overgrowth. Aliment Pharmacol Ther 2013; 37: 1103–11. 3 Holt PR. Diarrhea and malabsorption in the elderly. Gastroenterol Clin North Am 2001; 30: 427–44. 4 Ghoshal UC, Ghoshal U, Ayyagari A, Ranjan P, Krishnani N, Misra A, Aggarwal R, Naik S et al. Tropical sprue is associated with contamination of small bowel with aerobic bacteria and reversible prolongation of orocecal transit time. J Gastroenterol Hepatol 2003; 18: 540–7. 5 Simren M, Barbara G, Flint HJ, Spiegel BM, Spiller RC, Vanner S, Verdu EF, Whorwell PJ et al. Intestinal microbiota in functional bowel disorders: a Rome foundation report. Gut 2013; 62: 159–76. 6 Saad RJ, Chey WD. Breath testing for small intestinal bacterial overgrowth: maximizing test accuracy. Clin Gastroenterol Hepatol 2014; 12: 1964–72. 7 Hamilton I, Worsley BW, Cobden I, Cooke EM, Shoesmith JG, Axon AT. Simultaneous culture of saliva and jejunal aspirate in the investigation of small bowel bacterial overgrowth. Gut 1982; 23: 847–53. 8 Kerckhoffs AP, Visser MR, Samsom M, van der Rest ME, de Vogel J, Harmsen W, Akkermans LM. Critical evaluation of diagnosing bacterial overgrowth in the proximal small intestine. J Clin Gastroenterol 2008; 42: 1095–102. 9 Tabaqchali S. The pathophysiological role of small intestinal bacterial flora. Scand J Gastroenterol Suppl 1970; 6: 139–63. 10 Romagnuolo J, Schiller D, Bailey RJ. Using breath tests wisely in a gastroenterology practice: an evidencebased review of indications and
11
12
13
14
15
16
17
18
pitfalls in interpretation. The American journal of gastroenterology 2002; 97: 1113–26. Jones HF, Davidson GP, Brooks DA, Butler RN. Is small-bowel bacterial overgrowth an underdiagnosed disorder in children with gastrointestinal symptoms? J Pediatr Gastroenterol Nutr 2011; 52: 632–4. Majewski M, Reddymasu SC, Sostarich S, Foran P, McCallum RW. Efficacy of rifaximin, a nonabsorbed oral antibiotic, in the treatment of small intestinal bacterial overgrowth. The American journal of the medical sciences 2007; 333: 266–70. Lupascu A, Gabrielli M, Lauritano EC, Scarpellini E, Santoliquido A, Cammarota G, Flore R, Tondi P et al. Hydrogen glucose breath test to detect small intestinal bacterial overgrowth: a prevalence case-control study in irritable bowel syndrome. Aliment Pharmacol Ther 2005; 22: 1157–60. Sabate JM, Jouet P, Harnois F, Mechler C, Msika S, Grossin M, Coffin B. High prevalence of small intestinal bacterial overgrowth in patients with morbid obesity: a contributor to severe hepatic steatosis. Obes Surg 2008; 18: 371–7. Ratuapli SK, Ellington TG, O’Neill MT, Umar SB, Harris LA, Foxx-Orenstein AE, Burdick GE, Dibaise JK et al. Proton pump inhibitor therapy use does not predispose to small intestinal bacterial overgrowth. The American journal of gastroenterology 2012; 107: 730–5. Gabrielli M, D’Angelo G, Di Rienzo T, Scarpellini E, Ojetti V. Diagnosis of small intestinal bacterial overgrowth in the clinical practice. Eur Rev Med Pharmacol Sci 2013; 17 (Suppl. 2): 30–5. Bauer TM, Schwacha H, Steinbruckner B, Brinkmann FE, Ditzen AK, Kist M, Blum HE. Diagnosis of small intestinal bacterial overgrowth in patients with cirrhosis of the liver: poor performance of the glucose breath hydrogen test. J Hepatol 2000; 33: 382–6. Rana SV, Sharma S, Kaur J, Sinha SK, Singh K. Comparison of lactu-
8
19
20
21
22
23
24
25
26
27
lose and glucose breath test for diagnosis of small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Digestion 2012; 85: 243–7. Ghoshal UC, Ghoshal U, Das K, Misra A. Utility of hydrogen breath tests in diagnosis of small intestinal bacterial overgrowth in malabsorption syndrome and its relationship with oro-cecal transit time. Indian J Gastroenterol 2006; 25: 6–10. Pande C, Kumar A, Sarin SK. Smallintestinal bacterial overgrowth in cirrhosis is related to the severity of liver disease. Aliment Pharmacol Ther 2009; 29: 1273–81. Weinstock LB, Klutke CG, Lin HC. Small intestinal bacterial overgrowth in patients with interstitial cystitis and gastrointestinal symptoms. Dig Dis Sci 2008; 53: 1246–51. Posserud I, Stotzer PO, Bjornsson ES, Abrahamsson H, Simren M. Small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Gut 2007; 56: 802–8. Madrid AM, Poniachik J, Quera R, Defilippi C. Small intestinal clustered contractions and bacterial overgrowth: a frequent finding in obese patients. Dig Dis Sci 2011; 56: 155– 60. Rana SV, Sharma S, Sinha SK, Kaur H, Sikander A, Singh K. Incidence of predominant methanogenic flora in irritable bowel syndrome patients and apparently healthy controls from North India. Dig Dis Sci 2009; 54: 132–5. Ghoshal UC. How to interpret hydrogen breath tests. Journal of neurogastroenterology and motility 2011; 17: 312–7. Carrara M, Desideri S, Azzurro M, Bulighin GM, Di Piramo D, Lomonaco L, Adamo S. Small intestine bacterial overgrowth in patients with irritable bowel syndrome. Eur Rev Med Pharmacol Sci 2008; 12: 197–202. Mac Mahon M, Gibbons N, Mullins E, O’Moore RR, Keane CT, Walsh JB, Coakley D. Are hydrogen breath tests valid in the elderly? Gerontology 1996; 42: 40–5.
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
28 Di Stefano M, Malservisi S, Veneto G, Ferrieri A, Corazza GR. Rifaximin versus chlortetracycline in the shortterm treatment of small intestinal bacterial overgrowth. Aliment Pharmacol Ther 2000; 14: 551–6. 29 Sachdeva S, Rawat AK, Reddy RS, Puri AS. Small intestinal bacterial overgrowth (SIBO) in irritable bowel syndrome: frequency and predictors. J Gastroenterol Hepatol 2011; 26(Suppl. 3): 135–8. 30 Reddymasu SC, Sostarich S, McCallum RW. Small intestinal bacterial overgrowth in irritable bowel syndrome: are there any predictors? BMC Gastroenterol 2010; 10: 23. 31 Choi YK, Johlin FC Jr, Summers RW, Jackson M, Rao SS. Fructose intolerance: an under-recognized problem. The American journal of gastroenterology 2003; 98: 1348–53. 32 Choi YK, Kraft N, Zimmerman B, Jackson M, Rao SS. Fructose intolerance in IBS and utility of fructoserestricted diet. J Clin Gastroenterol 2008; 42: 233–8. 33 Bures J, Cyrany J, Kohoutova D, Forstl M, Rejchrt S, Kvetina J, Vorisek V, Kopacova M. Small intestinal bacterial overgrowth syndrome. World J Gastroenterol 2010; 16: 2978–90. 34 Kerlin P, Wong L. Breath hydrogen testing in bacterial overgrowth of the small intestine. Gastroenterology 1988; 95: 982–8. 35 Population Distribution by Race/Ethnicity, 2013. Available at: http:// kff.org/other/state-indicator/distribution-by-raceethnicity/- notes. 36 Quigley EM. Small intestinal bacterial overgrowth: what it is and what it
© 2015 John Wiley & Sons Ltd
37
38
39
40
41
42
is not. Curr Opin Gastroenterol 2014; 30: 141–6. Khoshini R, Dai SC, Lezcano S, Pimentel M. A systematic review of diagnostic tests for small intestinal bacterial overgrowth. Dig Dis Sci 2008; 53: 1443–54. Bardhan PK, Gyr K, Beglinger C, Vogtlin J, Frey R, Vischer W. Diagnosis of bacterial overgrowth after culturing proximal small-bowel aspirate obtained during routine upper gastrointestinal endoscopy. Scand J Gastroenterol 1992; 27: 253–6. Shindo K, Machida M, Miyakawa K, Fukumura M. A syndrome of cirrhosis, achlorhydria, small intestinal bacterial overgrowth, and fat malabsorption. The American journal of gastroenterology 1993; 88: 2084–91. Choung RS, Ruff KC, Malhotra A, Herrick L, Locke GR 3rd, Harmsen WS, Zinsmeister AR, Talley NJ et al. Clinical predictors of small intestinal bacterial overgrowth by duodenal aspirate culture. Aliment Pharmacol Ther 2011; 33: 1059–67. Pyleris E, Giamarellos-Bourboulis EJ, Tzivras D, Koussoulas V, Barbatzas C, Pimentel M. The prevalence of overgrowth by aerobic bacteria in the small intestine by small bowel culture: relationship with irritable bowel syndrome. Dig Dis Sci 2012; 57: 1321–9. Pimentel M, Chow EJ, Lin HC. Normalization of lactulose breath testing correlates with symptom improvement in irritable bowel syndrome. a double-blind, randomized, placebocontrolled study. The American jour-
9
43
44
45
46
47
48
nal of gastroenterology 2003; 98: 412– 9. Bratten JR, Spanier J, Jones MP. Lactulose breath testing does not discriminate patients with irritable bowel syndrome from healthy controls. The American journal of gastroenterology 2008; 103: 958–63. Gasbarrini A, Corazza GR, Gasbarrini G, Montalto M, Di Stefano M, Basilisco G, Parodi A, Usai-Satta P et al. Methodology and indications of H2-breath testing in gastrointestinal diseases: the Rome Consensus Conference. Aliment Pharmacol Ther 2009; 29(Suppl. 1): 1–49. Corazza GR, Menozzi MG, Strocchi A, Rasciti L, Vaira D, Lecchini R, Avanzini P, Chezzi C et al. The diagnosis of small bowel bacterial overgrowth. Reliability of jejunal culture and inadequacy of breath hydrogen testing. Gastroenterology 1990; 98: 302–9. Lin HC. Small intestinal bacterial overgrowth: a framework for understanding irritable bowel syndrome. JAMA 2004; 292: 852–8. Pimentel M, Chow EJ, Lin HC. Eradication of small intestinal bacterial overgrowth reduces symptoms of irritable bowel syndrome. The American journal of gastroenterology 2000; 95: 3503–6. Elphick DA, Chew TS, Higham SE, Bird N, Ahmad A, Sanders DS. Small bowel bacterial overgrowth in symptomatic older people: can it be diagnosed earlier? Gerontology 2005; 51: 396–401.
doi: 10.1111/nmo.12516
Small intestinal bacterial overgrowth: duodenal aspiration vs glucose breath test A. ERDOGAN ,* S. S. C. RAO ,* D. GULLEY ,* C. JACOBS ,† Y. Y. LEE *,‡
& C. BADGER *
*Section of Gastroenterology and Hepatology, Georgia Regents University, Augusta, GA, USA †Carver College of Medicine, University of Iowa, Iowa City, IA, USA ‡School of Medical Sciences, Universiti Sains Malaysia, Kota Bharu, Kelantan, Malaysia
Key Messages GOALS OF RESEARCH ○
To compare the diagnostic yield of glucose breath test and duodenal aspirate/culture, and analyze the influence of ethnics, age, and, gender on prevalence of SIBO, and assess symptom patterns.
BASIC METHODOLOGY ○
Duodenal aspiration was performed with Liguory catheter during an upper endoscopy. GBT was performed (QuinTron Instrument Company, Inc. Milwaukee, WI, USA) and alveolar gas was analyzed for both H2 and CH4 levels by chromatography.
RESULTS/TAKE HOME MESSAGES There is a higher yield for SIBO with duodenal culture at ≥103 CFU/mL. ○ For glucose breath test, the diagnostic yield is lower but there is adequate agreement with culture and a good specificity. ○ The overall agreement of duodenal culture and GBT is 66%; suggesting neither test is perfect. ○ Other than age and certain symptom patterns (gas), ethnical or gender differences did not appear to influence the prevalence of SIBO. ○
Abstract Background The diagnosis of small intestinal bacterial overgrowth (SIBO) remains challenging. Our aim was to examine the diagnostic yield of duodenal aspiration/culture and glucose breath test (GBT), and
effects of gender, race and demographics on prevalence of SIBO. Methods Patients with unexplained gas, bloating and diarrhea and negative endoscopy, imaging and blood tests were prospectively enrolled in two centers in USA. Randomly, within 1 week each patient underwent both duodenal aspiration/culture and GBT. The diagnostic yield of each test and relationship of symptoms, and effects of ethnicity, age, and gender on prevalence of SIBO were assessed and compared. Key Results Duodenal culture was positive in 62/139 (44.6%) subjects and GBT was positive in 38/139 (27.3%) subjects with an overall diagnostic agreement of 65.5%. The sensitivity, specificity, positive and negative predictive value of GBT was 42%, 84%, 68%, and 64%, respectively. Ethnicity or gender did not influence SIBO, but SIBO positive patients were older (p = 0.0018). Symptom patterns were similar except bloating was more prevalent in GBT positive
Address for Correspondence Satish S.C. Rao, MD, PhD, FRCP (LON) AGAF FACG, Department of Medicine, Section of Gastroenterology and Hepatology, Medical College of Georgia, Georgia Regents University, 1120 15th Street, BBR 2538, Augusta, GA 30912, USA. Tel: 706-721-2238; fax: 706-721-0331; e-mail: [email protected] Portions of this work were presented at: ANMS 2011, St Louis, Missouri, USA. (Neurogastroenterol Motil 2011; 23 Supp s1, 50:16); ACG 2012 Las Vegas, NV, USA (Am J Gastroenterol. 2012; 107 (Supp s1, 329): S141); ANMS 2013, Huntington Beach, California, USA (Neurogastroenterol Motil 2013; 25 (Supp s1, 19): 7). Received: 10 October 2014 Accepted for publication: 22 December 2014
© 2015 John Wiley & Sons Ltd
1
A. Erdogan et al.
Neurogastroenterology and Motility
However, the sensitivity and specificity of breath tests in the diagnosis of SIBO remains unclear because of the heterogeneity of patient populations that have been examined, methodological issues including use of different substrates and different doses of substrate. GBT has been performed with 50 g,12–16 75 g,17 80 g 18 and 100 g glucose,4,19,20 and likewise LBT has been performed with 10 g,18,21,22 16 g,23 and 25 g lactulose.24 Furthermore, LBT may overestimate SIBO (false positive), because the rise in hydrogen (H2) usually coincides with the arrival of substrate in cecum and not necessarily from bacterial fermentation in small bowel.4,25 Many investigators also have analyzed breath samples for H2 alone 13,26 and not methane (CH4). Finally, different investigators have used different cut-offs for defining an abnormal breath test; rise of 10 ppm,18,27 12 ppm,19,28,29 15 ppm,22 and 20 ppm 12,15,30 over baseline. Additionally, whether factors such as geography, race, age, or gender influence the prevalence of SIBO is not known. We tested the hypothesis that GBT provides a similar yield to that of duodenal aspiration/culture for diagnosis of SIBO. Our aim was to examine the diagnostic yield of duodenal aspiration and culture (aerobic and anaerobic) for SIBO, and to compare this with that of GBT in the same subjects. Additionally, we assessed the prevalence of SIBO with both tests in two different geographical regions in USA, each with different ethnic distribution, and examined the relationship of symptoms, age, and gender on the prevalence of SIBO.
and gas in culture positive subjects. Conclusions & Inferences Duodenal aspiration/culture identifies 45% of patients with suspected SIBO. GBT has lower sensitivity but good specificity for detection of SIBO. There were no ethnic or gender differences in the prevalence of SIBO, but patients with SIBO were older. Because GBT is non-invasive, it should be considered first in patients with suspected SIBO. Keywords duodenal culture, ethnics, glucose breath test, sensitivity and specificity, small intestinal bacterial overgrowth.
INTRODUCTION Small intestinal bacterial overgrowth (SIBO) is characterized by the presence of abnormal and high levels of bacteria in the small intestine associated with gastrointestinal (GI) symptoms such as bloating, distention, flatulence, abdominal discomfort, diarrhea, and weight loss.1–3 The anatomical and structural abnormalities that occur in scleroderma or intestinal pseudo-obstruction or those that follow intestinal surgeries that form a blind loop often predispose to SIBO, but it can also occur in the absence of these conditions.2,4–6 Recently, proton pump inhibitor use and small intestinal dysmotility have been shown to be independent risk factors that also predispose to SIBO.2 In clinical practice, the diagnosis of SIBO remains challenging, because there is no widely accepted diagnostic test. Aspiration and culture of small intestinal fluid is generally believed to be the gold standard. However, this is an invasive procedure and requires either upper endoscopy or small intestinal intubation under fluoroscopy. There is also limited information regarding the optimal technique for aspiration and culture, and in clinical practice its diagnostic yield remains unclear. Contamination from oral flora or saliva during oral intubation may cause false positive result and an overestimation of SIBO.7 In contrast, aspiration/culture from the proximal portion of the small bowel may miss bacterial overgrowth in the distal small bowel causing false negative results;8 also, anaerobic culture is not routinely performed in most laboratories.5,8,9 Thus, the utility of aerobic and anaerobic culture in the diagnosis of SIBO is not fully known. There is also a need for less invasive and inexpensive test for the detection of SIBO in routine clinical practice. The glucose breath test (GBT) and the lactulose breath test (LBT) have been advocated for the past few years, because they are safe, easy and can be performed in children and pregnant women.10,11
MATERIALS AND METHODS Patients Adult patients with unexplained GI symptoms such as gas, bloating, diarrhea, and abdominal discomfort and suspected to have SIBO, and who were referred to two specialist motility centers, one in mid-west (University of Iowa) and the second in southern USA (Georgia Regents University) were prospectively enrolled. All patients had normal upper endoscopy, colonoscopy, computerized abdominal tomography scan, and normal hematology, biochemical profiles, tissue transglutaminase antibody, thyroid stimulating hormone, and normal right upper quadrant ultrasound scan. Patients with a history of previous GI surgeries (except for cholecystectomy, hysterectomy, and appendectomy), those with significant co-morbid medical problems (including stroke, COPD (Chronic Obstructive Pulmonary Disease), cancers and disabled) or those who had diabetes mellitus or who were hospitalized or had a history of recent antibiotic use (4 weeks) were excluded from the study. Patients were asked to discontinue the use of proton pump inhibitor, laxatives and drugs that affect intestinal motility (opioids, anticholinergics, antidiarrheals) at least 1 week prior to study, but were allowed to continue with other usual medications. All consecutive subjects underwent both the GBT and
2
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
Statistical analyses
duodenal aspiration/culture within 1 week in a random order. The University of Iowa Hospitals and Clinics and Georgia Regents University Investigation Review Boards approved the study.
Statistical analyses were performed using SPSS for Windows 11.0 (SPSS Inc., Chicago, IL, USA). All data were presented as mean SD unless otherwise stated. Diagnostic agreement (positive %, negative % and overall %) and yield (sensitivity, specificity, positive and negative predictive value) were determined for GBT against the gold standard, duodenal culture. Univariable analysis (t-test for continuous variables and chi-squared or Fisher’s Exact test for categorical variables) and multivariable analysis (multiple logistic regression: backward LR method) were used to determine association between factors including age, gender, ethnics, differences in symptom patterns and with GBT and duodenal culture. A p < 0.05 was considered as statistically significant for above tests.
Duodenal aspiration and culture After an overnight fast, an upper endoscope was passed into the distal duodenum with minimal air insufflation. Next, under aseptic techniques (sterile gloves worn by the endoscopist and nurse assisting the procedure just before aspiration), a 2–3 mm Liguory catheter (COOK Medical, Bloomigton, IN, USA) was advanced through the biopsy channel of the endoscope into third or fourth portions of the duodenum. Using gentle intermittent suction with a 5 mL syringe, 3–5 mL of duodenal juice was aspirated and immediately sent to the microbiology laboratory for aerobic and anaerobic cultures.
RESULTS Glucose breath test Demographics and symptom profiles
One day before the GBT, patients were asked to consume a low carbohydrate diet without vegetables, fiber and dairy products. After an overnight fast, patients were asked to brush their teeth and rinse their mouth with an antiseptic mouthwash, at least 2 h before the test to avoid false positive high basal levels from fermentation of substrate by oral bacteria. Patients were asked to refrain from any exercise, sleep or smoke cigarette during the test to avoid possible effects on the breath H2 levels. On the test day, 75 g of glucose dissolved in 250 mL water was administered, and breath samples were obtained at baseline and at 15 min intervals for a 2 h period. The samples were collected in a bag (QuinTron Instrument Company, Inc.) and alveolar gas was analyzed for both H2 and CH4 levels by chromatography (QuinTron Micro Analyzer, QuinTron Instrument Company, Inc.).
One hundred and thirty-nine consecutive patients were included in this study, of whom 78 (M/F = 17/ 61, mean age = 48.9 1.9 years) were enrolled from Georgia and 61 (M/F = 16/45, mean age = 45.9 1.8 years) from Iowa (p = 0.33). Because there were no geographical differences in the prevalence of SIBO with either GBT or cultures between Georgia and Iowa, the data were pooled for all analyses. The mean age was significantly higher in patients who were GBT positive (54.4 2.4 years) vs GBT negative (45.1 1.5 years), p = 0.002, and also for patients who were culture positive (52.8 2.0 years) vs culture negative (43.5 1.7 years), p = 0.0004. Logistic regression analysis showed that SIBO positive patients were significantly older than SIBO negative patients, both with GBT (OR: 1.04 [95% CI: 1.01–1.07, p = 0.003]) and duodenal culture at ≥103 CFU/mL (OR: 1.04 [95% CI: 1.01–1.06, p = 0.003]) and ≥105 CFU/mL (OR: 1.05 [95% CI: 1.02–1.09, p = 0.002]; Table 1). The dominant symptoms in patients (mean score 6 or higher) were bloating, abdominal pain, fullness, and gas.
Symptom questionnaire All patients filled out a validated questionnaire for the presence/ absence of 9 symptoms using a validated scale31,32 including abdominal pain, belching, bloating, fullness, indigestion (dyspepsia-discomfort centered in upper abdomen), nausea, diarrhea, vomiting and gas. If present, patients were asked to rate the frequency, intensity and duration of each symptom on a 0–3 point Likert scale. Intensity: 0 = no symptom, 1 = mild, 2 = moderate, 3 = severe symptom. Frequency: 0 = none, 1 = 1 episode/week. Duration: 0 = none, 1 = 30 min. The total score could range from 0 to 9 for each symptom. A mean score was calculated for each symptom, and an overall mean total score was calculated for all 9 symptoms.
Table 1 Effects of symptoms and demographic features on prevalence of SIBO Outcome and risk factors*
Data analysis
SIBO based on GBT Older age 1.04 SIBO based on culture ≥103 CFU/mL Older age 1.04 Gas 1.16 5 SIBO based on culture ≥10 CFU/mL Older age 1.05
The culture results were considered as positive for SIBO if one or more organisms (aerobic or anaerobic) were cultured with a colony count of ≥103 CFU/mL.2 We also assessed the prevalence of SIBO at a threshold of ≥105 CFU/mL. Glucose breath test was considered positive for SIBO if the following criteria were met; ≥20 ppm increase above baseline for H2,17 or ≥15 ppm increase above baseline for CH4 or ≥20 ppm increase above baseline for H2 and CH4.2,12,30,33 Additionally, we performed a separate analysis, using an increase of ≥12 ppm H2 values above baseline.19,34
© 2015 John Wiley & Sons Ltd
OR (95% CI)
p-value
(1.01–1.07)
0.003
(1.01–1.06) (1.01–1.33)
0.003 0.03
(1.02–1.09)
0.002
*Only factors in logistic regression (Backward LR method) with significant p-value ( 0.05; Fig. 1A). The mean score for gas was significantly higher for culture positive vs culture negative patients (p = 0.023; Fig. 1B). The mean score for other symptoms was similar (p > 0.05). Multiple logistic regression analysis also found that gas was significantly higher in SIBO positive (duodenal culture) patients at ≥103 CFU/mL (OR: 1.2 [95% CI: 1.01–1.33, p = 0.03]; Table 1). The overall mean total symptom score was similar for patients who tested positive vs negative for GBT (46.5 vs 46.1, p = 0.9) or for duodenal culture (46.7 vs 45.9, p = 0.8). We also compared mean symptom scores and mean total symptom score for culture positive patients at threshold of 103 CFU/mL vs 105 CFU/mL. The symptom scores were not significantly different for any symptom at the thresholds of 103 CFU/mL vs 105 CFU/mL; p > 0.9.
≥105 CFU/mL analysis At this threshold, the duodenal aspiration/culture was positive in 25 (18%) patients. 18 (72%) had positive aerobic cultures and 7 (28%) had both positive aerobic and anaerobic cultures. None of the patients had positive anaerobic cultures alone. Alpha-hemolytic streptococcus, Escherichia coli and Klebsiella species were the most commonly isolated organisms.
Diagnostic yield of GBT Mean baseline H2 and CH4 levels for GBT were 7.2 12.8 ppm and 5.0 13.5 ppm, respectively, and mean peak values were 34.3 76.8 and 8.8 21.3 ppm, respectively; H2 values ranged between 0 to 498 ppm and CH4 values ranged between 0 to 123 ppm. In total, 11 patients had high baseline H2 values (H2 >20 ppm). Of these, eight were positive and three were negative for GBT according to criteria used to define a positive test, and high baseline values were not considered as positive test if it did not meet the diagnostic criteria. A positive GBT was seen in 38 (27.3%) and a negative test in 101 (72.7%) patients. 24
Diagnostic yield of duodenal culture ≥103 CFU/mL analysis The duodenal aspiration/culture yielded a positive result in 62 (44.6%) patients, and 45 (36.3%) samples grew more than one organism. The most common aerobic bacterial organisms were Alpha-hemolytic Streptococcus, Escherichia coli, and
A
*
8
7
Mean symptom scores
6
5
4 Total patients
3
GBT positive 2 GBT negative 1 * p = 0.009
0 Abdom pain
B
Belching
Bloating
Fullness
Indigestion
Nausea
Diarrhea
Vomiting
8
Gas
*
7
Mean symptom score
6 5 4 3 Total patients 2
Culture positive
1
Culture negative * p = 0.023
0 Abdom pain Belching
Bloating
Fullness
Indigestion
Nausea
Diarrhea
Vomiting
Gas
4
Figure 1 Mean symptom scores of (A) all patients and those with a positive and negative GBT and (B) all patients and those with a positive and negative culture (≥103 CFU/mL). GBT, Glucose breath test.
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
sensitivity, specificity, positive and negative predictive values were 41.9%, 84.4%, 68.4%, and 64.4%, respectively. When a ≥12 ppm increase in H2/CH4 from baseline was considered as positive for GBT, the sensitivity, specificity, positive and negative predictive values were 43.6%, 77.9%, 61.4%, 63.2%, respectively. The positive, negative and overall agreements were 43.6%, 77.9% and 62.6%, respectively.
Table 2 Culture results of duodenal aspirates* 3
Culture results (≥10 CFU/mL)
Total, N = 124 (%)
Escherichia coli Klebsiella oxcytoca Klebsiella pneumonia Enterobacter cloacea Enterobacter aerogenes Pseudomonas aeruginosa Neisseria spp. Hemophilus parainfluenzae Hemophilus influenzae Stenotrophomonas maltophilia Bacteriodes vulgatus Velionella spp. Prevotella spp. Alpha-hemolytic streptococcus Non-hemolytic streptococcus Beta-hemolytic streptoccus Staphylococcus aureus Coagulase negative staphylococcus Diphtheroid bacilli Rothia spp. (Stomatococcus) Lactobacillus spp. Clostridium perfringes Bacillus spp.
15 2 7 2 2 1 6 1 1 1 4 1 2 38 6 1 3 6 9 4 8 3 1
(12.1) (1.6) (5.6) (1.6) (1.6) (0.8) (4.8) (0.8) (0.8) (0.8) (3.2) (0.8) (1.6) (30.6) (4.8) (0.8) (2.4) (4.8) (7.3) (3.2) (6.5) (2.4) (0.8)
For ≥105 analysis A positive % agreement for duodenal aspiration/culture and GBT was seen in 16/25 (64%), and negative % agreement in 92/114 (80.7%) subjects with an overall % agreement of 108/139 (77.7%). Sensitivity, specificity, positive, and negative predictive values were 64%, 80.7%, 42.1%, and 91.1%, respectively. When a ≥12 ppm increase from baseline was considered positive for GBT, the sensitivity, specificity, positive, and negative predictive values were 68%, 76.3%, 38.6%, 91.6%, respectively. The positive, negative and overall agreements were 68%, 76.3%, and 74.8%, respectively.
*36.3% of samples grew more than one organism.
Racial differences in prevalence of SIBO
(63.2%) subjects were diagnosed with SIBO on the basis of increase in H2 level alone, seven (18.4%) with CH4 level alone, and 7 (18.4%) with increase in both H2 and CH4 levels. When ≥12 ppm increase was used as the threshold cut-off, the GBT was positive in 44 (31.7%) and negative in 95 (68.3%) patients.
There is a significant difference in the racial demographic population distribution with Whites composing 56% of population in Georgia and 86% in Iowa, and Blacks composing 30% in Georgia but only 3% in Iowa.35 Our study population comprised of 82.1% and 93.4% Whites and 11.5% and 3.3% of Blacks in Georgia and Iowa, respectively. There were fewer Blacks and more Whites than expected in the study population from Georgia (p = 0.016). The prevalence of SIBO was similar in Whites vs Blacks, with a positive GBT in 26.4% vs 18.2% (p = 0.2), and a positive duodenal aspiration/culture in 44.6% vs 36.4% (p = 0.7), respectively (Table 3).
Diagnostic agreement For ≥103 analysis A positive % agreement for duodenal aspiration/culture and GBT was seen in 26/62 (41.9%) subjects, and a negative % agreement in 65/77 (84.4%) subjects. The overall % agreement for duodenal aspiration/culture and GBT was 91/139 (65.5%). The
% of aerobic and anaerobic culture growth (103 CFU/mL)
22.6%
3.2%
Aerobic culture
74.2% Figure 2 Aerobic and anaerobic distribution of positive culture results.
© 2015 John Wiley & Sons Ltd
Anaerobic culture
Aerobic and anaerobic culture
5
A. Erdogan et al.
Neurogastroenterology and Motility
Table 3 Racial distribution of study population for GBT and duodenal aspiration/culture
White African American Hispanic Other
Study population (N = 139)
Positive GBT, N (%)
Negative GBT, N (%)
Positive culture*, N (%)
Negative culture*, N (%)
121 11
32 (26.4) 2 (18.2)
89 (73.6) 9 (81.8)
54 (44.6) 4 (36.4)
67 (55.4) 7 (63.6)
4 3
3 (75) 1 (33.3)
1 (25%) 2 (66.7)
2 (50) 2 (66.7)
2 (50) 1 (33.3)
*103 CFU/mL. GBT, glucose breath test.
Effects of gender on prevalence of SIBO
bacteria from the colon, and its frequent exposure to acid from the stomach, all of which should decrease the risk of SIBO.2 The techniques for aspiration and culture have not been standardized. Some have used endoscopic suction of duodenal juice38,40 with a high chance of cross-contamination as opposed to our technique of placing a sterile catheter into the distal duodenum under aseptic techniques, including use of sterile gloves. In our study, the yield of positive culture was 45% at ≥103 CFU/mL and 18% at ≥105 CFU/mL. A low cut-off of 103 CFU/mL was clinically appropriate for the type of population we studied.2 The ranges for a positive culture reported in the literature have varied from as low as 8% in patients with diarrhea and weight loss40 (>105 CFU/mL aerobic and >104 CFU/ mL anaerobic) to as high as 100% in postsurgical patients with Billroth II.38 Culture was positive in 39.5% of patients with malabsorption,4 66.7% of patients with GI symptoms (20/30),27 65% of patients with cirrhosis.17 Pyleris et al. found the yield of positive culture to be 19.4% at >104 CFU/mL, suggesting that the variability in the yield of culture depends on the threshold used to define a positive culture and the patients’ characteristics.41 The relatively poor symptom correlations with positive cultures in our study suggest that a functional or psychological etiology besides SIBO may also be important in the pathogenesis of unexplained GI symptoms and confirm a previous study 2 that symptoms are poor predictors of SIBO. The diagnostic yield of breath tests for SIBO in the literature has also been variable because of differences in the dose and type of substrate used, the threshold for defining a positive breath test, analysis of breath samples for H2 or a combination of H2 and CH4, and also the patients’ characteristics. In our study, GBT was positive in 27% of patients with unexplained GI symptoms. In IBS patients, GBT was positive in 31%13 and LBT in 34.3–84%.18,42 In a comparative study, GBT (≥12 ppm H2) was positive in 20/75 (26.7%), LBT in 13/71 (18.3%) and culture
Glucose breath test was positive in 26.4% of females and 30.3% of males (OR: 0.8 [95% CI: 0.3–1.9, p = 0.7]). Similarly, 44.3% of females and 45.5% of males were SIBO positive by culture at ≥103 CFU/mL (OR: 1.0 [95% CI: 0.4–2.1, p = 1.0]). There were no gender differences in the prevalence of SIBO (p > 0.05).
DISCUSSION In this study, we assessed and compared the diagnostic yield of GBT with duodenal aspiration and culture in patients with unexplained GI symptoms (gas, bloating and diarrhea) and with negative endoscopy, imaging and blood tests, and the effects of age, gender and race on symptoms and prevalence of SIBO. Glucose breath test was positive in 27% of patients, whereas the duodenal culture was positive in 45% of patients with unexplained GI symptoms. Alpha-hemolytic Streptococcus, Escherichia coli and Klebsiella species were the most common organisms isolated from cultures, and 74% showed a positive aerobic culture, 3% positive anaerobic culture and 23% with both positive aerobic and anaerobic cultures. Thus, the sensitivity and specificity of GBT were 42% and 84%, respectively. Aspiration and culture of small bowel fluid is generally accepted as the gold standard for diagnosis of SIBO, however, there is no established cut-off value for defining the abnormal levels of bacterial growth especially in different parts of the small intestine.2,36–38 The proximal jejunum may contain ≤104 CFU/mL bacteria in healthy subjects.37–39 In one study, 12% of healthy subjects had bacterial counts ranging from 0 to 103 CFU/mL in the proximal jejunum.22 As reported recently, a lower cut-off value of ≥103 CFU/mL may be more clinically relevant for aspirations obtained from the third and fourth portions of the duodenum, given its proximal location, relative protection from translocation of
6
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
(≥105 CFU/mL) in 39.5% of patients with malabsorption.19 In another study, culture (≥103 CFU/mL) was positive in 4/12 (25%) and LBT in 8/15 (53.3%) patients.8 Typically, LBT had higher positive yield compared to GBT largely due to false positive results from cecal rather than small bowel fermentation.18,25,43 Assessment of breath CH4 levels can also affect the yield of GBT. In our study, 18.4% of patients were diagnosed with a positive GBT on the basis of an increase in CH4 values, and this is comparable to yields of 29%12 and 30%30 reported by others.12 The sensitivity and specificity of GBT not only depend on values of H2 and or CH4 cut-offs but also on the culture cut-offs. Not surprisingly, the sensitivity has ranged between 44% and 62.5% and specificity between 36% and 83%.17,19,44,45 In one study, the sensitivity and specificity of GBT (≥12 ppm) were 44% and 80%, respectively using a duodenal culture cut-off value of ≥105 CFU/mL.19 If these criteria were applied to our study, the sensitivity for GBT increased to 68%, whereas the specificity decreased to 76.3%. Unlike our study, Ghoshal el al. did not measure CH4 levels that might have lowered their detection rate of SIBO.19 In another study, Corazza et al. 45 found a sensitivity and specificity of 62% and 83%, respectively for GBT, but they cultured fluid from the proximal jejunum at two different sites and used a bacterial count of ≥106 CFU/mL that might have increased the detection rate for SIBO. In our study, the lower H2 cut-off value (12 ppm) had minimal impact on the sensitivity and positive/negative predictive values, but yielded a lower specificity. Hence, we suggest H2 ≥20 ppm as an appropriate cut-off for diagnosis of SIBO, but this needs further validation. Thus, the specificity for a positive GBT, usually 76–85%, was comparable, although the sensitivity has varied depending on the threshold cut-off values, the site and technique of aspiration, and bacterial concentration (103 vs 105 CFU/mL). The high specificity makes GBT a reliable test especially when positive, whereas the low sensitivity could be due to rapid absorption of glucose and its lack of availability as a substrate in distal small bowel or low concentrations of bacteria,46 or the inability of bacteria to ferment glucose, the solubility of H2 in blood, its transfer and excretion from lungs and lung disease.27 Although there are differences in the ethnic makeup of Georgia and Iowa, we found that the prevalence of SIBO was similar between these regions and between the ethnic groups (Whites and African
© 2015 John Wiley & Sons Ltd
Americans), and between men and women. This lack of gender effect confirms a previous observation47 and suggests that unlike other GI motility disorders, the prevalence of SIBO is not affected by gender. However, we found a higher prevalence of SIBO in older subjects confirming previous studies.30,40 Older subjects are more likely to have risk factors for SIBO including intestinal dysmotility, intestinal surgery, small bowel diverticulosis, achlorhydria, and medication use including proton pump inhibitor.48 The limitations of our study include a potential referral bias because patients were assessed in two tertiary care centers, and may not reflect SIBO prevalence in the community, however, it may not be feasible to do a population-based study because these assessments require breath tests or endoscopy. Aspiration was also performed from the third/fourth portions of duodenum and whether aspiration more distally could have yielded a higher positive rate for culture is unclear. In spite of aseptic techniques, we cannot definitely exclude the potential for contamination from oral bacteria because of performing an endoscopy. However, the sterile technique was consistently used, and by one single operator, and the high prevalence of bacteria in duodenal aspirates confirms the presence of SIBO. Finally, given the sample distribution, a type II error cannot be excluded for some of the demographic observations. In conclusion, GBT has a low sensitivity but high specificity for detection of SIBO when compared to duodenal culture. It is non-invasive and widely available and hence in clinical practice this test may be considered first in a patient with suspected SIBO. However, the overall agreement for duodenal culture and GBT was 66% suggesting that neither test was perfect. Therefore, in selected patients with a high index of suspicion for SIBO, both tests may be required.
ACKNOWLEDGEMENTS We acknowledge the technical and secretarial assistance of m/s Michelle Jackson, Dr. Enrique Coss Adame, and Ms. Helen Smith. Guarantor of article: Satish S.C. Rao, MD, PhD, FACG.
FUNDING None.
CONFLICTS OF INTEREST None.
7
A. Erdogan et al.
Neurogastroenterology and Motility
AUTHOR CONTRIBUTION
data, and drafting of manuscript; DG analysis and interpretation of data, and drafting of manuscript; CJ analysis and interpretation of data, and drafting of manuscript; YYL analysis and interpretation of data, and drafting of manuscript; CB analysis and interpretation of data.
AE study concept and design, data collection, analysis, and interpretation of data, and drafting of manuscript; SSCR study concept and design, data collection, analysis, and interpretation of
REFERENCES 1 Sachdev AH, Pimentel M. Gastrointestinal bacterial overgrowth: pathogenesis and clinical significance. Therapeutic advances in chronic disease 2013; 4: 223–31. 2 Jacobs C, Coss Adame E, Attaluri A, Valestin J, Rao SS. Dysmotility and proton pump inhibitor use are independent risk factors for small intestinal bacterial and/or fungal overgrowth. Aliment Pharmacol Ther 2013; 37: 1103–11. 3 Holt PR. Diarrhea and malabsorption in the elderly. Gastroenterol Clin North Am 2001; 30: 427–44. 4 Ghoshal UC, Ghoshal U, Ayyagari A, Ranjan P, Krishnani N, Misra A, Aggarwal R, Naik S et al. Tropical sprue is associated with contamination of small bowel with aerobic bacteria and reversible prolongation of orocecal transit time. J Gastroenterol Hepatol 2003; 18: 540–7. 5 Simren M, Barbara G, Flint HJ, Spiegel BM, Spiller RC, Vanner S, Verdu EF, Whorwell PJ et al. Intestinal microbiota in functional bowel disorders: a Rome foundation report. Gut 2013; 62: 159–76. 6 Saad RJ, Chey WD. Breath testing for small intestinal bacterial overgrowth: maximizing test accuracy. Clin Gastroenterol Hepatol 2014; 12: 1964–72. 7 Hamilton I, Worsley BW, Cobden I, Cooke EM, Shoesmith JG, Axon AT. Simultaneous culture of saliva and jejunal aspirate in the investigation of small bowel bacterial overgrowth. Gut 1982; 23: 847–53. 8 Kerckhoffs AP, Visser MR, Samsom M, van der Rest ME, de Vogel J, Harmsen W, Akkermans LM. Critical evaluation of diagnosing bacterial overgrowth in the proximal small intestine. J Clin Gastroenterol 2008; 42: 1095–102. 9 Tabaqchali S. The pathophysiological role of small intestinal bacterial flora. Scand J Gastroenterol Suppl 1970; 6: 139–63. 10 Romagnuolo J, Schiller D, Bailey RJ. Using breath tests wisely in a gastroenterology practice: an evidencebased review of indications and
11
12
13
14
15
16
17
18
pitfalls in interpretation. The American journal of gastroenterology 2002; 97: 1113–26. Jones HF, Davidson GP, Brooks DA, Butler RN. Is small-bowel bacterial overgrowth an underdiagnosed disorder in children with gastrointestinal symptoms? J Pediatr Gastroenterol Nutr 2011; 52: 632–4. Majewski M, Reddymasu SC, Sostarich S, Foran P, McCallum RW. Efficacy of rifaximin, a nonabsorbed oral antibiotic, in the treatment of small intestinal bacterial overgrowth. The American journal of the medical sciences 2007; 333: 266–70. Lupascu A, Gabrielli M, Lauritano EC, Scarpellini E, Santoliquido A, Cammarota G, Flore R, Tondi P et al. Hydrogen glucose breath test to detect small intestinal bacterial overgrowth: a prevalence case-control study in irritable bowel syndrome. Aliment Pharmacol Ther 2005; 22: 1157–60. Sabate JM, Jouet P, Harnois F, Mechler C, Msika S, Grossin M, Coffin B. High prevalence of small intestinal bacterial overgrowth in patients with morbid obesity: a contributor to severe hepatic steatosis. Obes Surg 2008; 18: 371–7. Ratuapli SK, Ellington TG, O’Neill MT, Umar SB, Harris LA, Foxx-Orenstein AE, Burdick GE, Dibaise JK et al. Proton pump inhibitor therapy use does not predispose to small intestinal bacterial overgrowth. The American journal of gastroenterology 2012; 107: 730–5. Gabrielli M, D’Angelo G, Di Rienzo T, Scarpellini E, Ojetti V. Diagnosis of small intestinal bacterial overgrowth in the clinical practice. Eur Rev Med Pharmacol Sci 2013; 17 (Suppl. 2): 30–5. Bauer TM, Schwacha H, Steinbruckner B, Brinkmann FE, Ditzen AK, Kist M, Blum HE. Diagnosis of small intestinal bacterial overgrowth in patients with cirrhosis of the liver: poor performance of the glucose breath hydrogen test. J Hepatol 2000; 33: 382–6. Rana SV, Sharma S, Kaur J, Sinha SK, Singh K. Comparison of lactu-
8
19
20
21
22
23
24
25
26
27
lose and glucose breath test for diagnosis of small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Digestion 2012; 85: 243–7. Ghoshal UC, Ghoshal U, Das K, Misra A. Utility of hydrogen breath tests in diagnosis of small intestinal bacterial overgrowth in malabsorption syndrome and its relationship with oro-cecal transit time. Indian J Gastroenterol 2006; 25: 6–10. Pande C, Kumar A, Sarin SK. Smallintestinal bacterial overgrowth in cirrhosis is related to the severity of liver disease. Aliment Pharmacol Ther 2009; 29: 1273–81. Weinstock LB, Klutke CG, Lin HC. Small intestinal bacterial overgrowth in patients with interstitial cystitis and gastrointestinal symptoms. Dig Dis Sci 2008; 53: 1246–51. Posserud I, Stotzer PO, Bjornsson ES, Abrahamsson H, Simren M. Small intestinal bacterial overgrowth in patients with irritable bowel syndrome. Gut 2007; 56: 802–8. Madrid AM, Poniachik J, Quera R, Defilippi C. Small intestinal clustered contractions and bacterial overgrowth: a frequent finding in obese patients. Dig Dis Sci 2011; 56: 155– 60. Rana SV, Sharma S, Sinha SK, Kaur H, Sikander A, Singh K. Incidence of predominant methanogenic flora in irritable bowel syndrome patients and apparently healthy controls from North India. Dig Dis Sci 2009; 54: 132–5. Ghoshal UC. How to interpret hydrogen breath tests. Journal of neurogastroenterology and motility 2011; 17: 312–7. Carrara M, Desideri S, Azzurro M, Bulighin GM, Di Piramo D, Lomonaco L, Adamo S. Small intestine bacterial overgrowth in patients with irritable bowel syndrome. Eur Rev Med Pharmacol Sci 2008; 12: 197–202. Mac Mahon M, Gibbons N, Mullins E, O’Moore RR, Keane CT, Walsh JB, Coakley D. Are hydrogen breath tests valid in the elderly? Gerontology 1996; 42: 40–5.
© 2015 John Wiley & Sons Ltd
SIBO: Duodenal culture vs breath test
28 Di Stefano M, Malservisi S, Veneto G, Ferrieri A, Corazza GR. Rifaximin versus chlortetracycline in the shortterm treatment of small intestinal bacterial overgrowth. Aliment Pharmacol Ther 2000; 14: 551–6. 29 Sachdeva S, Rawat AK, Reddy RS, Puri AS. Small intestinal bacterial overgrowth (SIBO) in irritable bowel syndrome: frequency and predictors. J Gastroenterol Hepatol 2011; 26(Suppl. 3): 135–8. 30 Reddymasu SC, Sostarich S, McCallum RW. Small intestinal bacterial overgrowth in irritable bowel syndrome: are there any predictors? BMC Gastroenterol 2010; 10: 23. 31 Choi YK, Johlin FC Jr, Summers RW, Jackson M, Rao SS. Fructose intolerance: an under-recognized problem. The American journal of gastroenterology 2003; 98: 1348–53. 32 Choi YK, Kraft N, Zimmerman B, Jackson M, Rao SS. Fructose intolerance in IBS and utility of fructoserestricted diet. J Clin Gastroenterol 2008; 42: 233–8. 33 Bures J, Cyrany J, Kohoutova D, Forstl M, Rejchrt S, Kvetina J, Vorisek V, Kopacova M. Small intestinal bacterial overgrowth syndrome. World J Gastroenterol 2010; 16: 2978–90. 34 Kerlin P, Wong L. Breath hydrogen testing in bacterial overgrowth of the small intestine. Gastroenterology 1988; 95: 982–8. 35 Population Distribution by Race/Ethnicity, 2013. Available at: http:// kff.org/other/state-indicator/distribution-by-raceethnicity/- notes. 36 Quigley EM. Small intestinal bacterial overgrowth: what it is and what it
© 2015 John Wiley & Sons Ltd
37
38
39
40
41
42
is not. Curr Opin Gastroenterol 2014; 30: 141–6. Khoshini R, Dai SC, Lezcano S, Pimentel M. A systematic review of diagnostic tests for small intestinal bacterial overgrowth. Dig Dis Sci 2008; 53: 1443–54. Bardhan PK, Gyr K, Beglinger C, Vogtlin J, Frey R, Vischer W. Diagnosis of bacterial overgrowth after culturing proximal small-bowel aspirate obtained during routine upper gastrointestinal endoscopy. Scand J Gastroenterol 1992; 27: 253–6. Shindo K, Machida M, Miyakawa K, Fukumura M. A syndrome of cirrhosis, achlorhydria, small intestinal bacterial overgrowth, and fat malabsorption. The American journal of gastroenterology 1993; 88: 2084–91. Choung RS, Ruff KC, Malhotra A, Herrick L, Locke GR 3rd, Harmsen WS, Zinsmeister AR, Talley NJ et al. Clinical predictors of small intestinal bacterial overgrowth by duodenal aspirate culture. Aliment Pharmacol Ther 2011; 33: 1059–67. Pyleris E, Giamarellos-Bourboulis EJ, Tzivras D, Koussoulas V, Barbatzas C, Pimentel M. The prevalence of overgrowth by aerobic bacteria in the small intestine by small bowel culture: relationship with irritable bowel syndrome. Dig Dis Sci 2012; 57: 1321–9. Pimentel M, Chow EJ, Lin HC. Normalization of lactulose breath testing correlates with symptom improvement in irritable bowel syndrome. a double-blind, randomized, placebocontrolled study. The American jour-
9
43
44
45
46
47
48
nal of gastroenterology 2003; 98: 412– 9. Bratten JR, Spanier J, Jones MP. Lactulose breath testing does not discriminate patients with irritable bowel syndrome from healthy controls. The American journal of gastroenterology 2008; 103: 958–63. Gasbarrini A, Corazza GR, Gasbarrini G, Montalto M, Di Stefano M, Basilisco G, Parodi A, Usai-Satta P et al. Methodology and indications of H2-breath testing in gastrointestinal diseases: the Rome Consensus Conference. Aliment Pharmacol Ther 2009; 29(Suppl. 1): 1–49. Corazza GR, Menozzi MG, Strocchi A, Rasciti L, Vaira D, Lecchini R, Avanzini P, Chezzi C et al. The diagnosis of small bowel bacterial overgrowth. Reliability of jejunal culture and inadequacy of breath hydrogen testing. Gastroenterology 1990; 98: 302–9. Lin HC. Small intestinal bacterial overgrowth: a framework for understanding irritable bowel syndrome. JAMA 2004; 292: 852–8. Pimentel M, Chow EJ, Lin HC. Eradication of small intestinal bacterial overgrowth reduces symptoms of irritable bowel syndrome. The American journal of gastroenterology 2000; 95: 3503–6. Elphick DA, Chew TS, Higham SE, Bird N, Ahmad A, Sanders DS. Small bowel bacterial overgrowth in symptomatic older people: can it be diagnosed earlier? Gerontology 2005; 51: 396–401.
