Diagnosis

Research to characterize the digestive tract microflora has developed mainly from the 1960s onward. The micro-organisms of the digestive tube flora, made up of aerobic and anaerobic bacteria, protozoa and even some viruses, form a complex ecosystem. The terms 'indigenous flora' and 'microbiota' are used as synonyms for normal flora and refer to micro-organisms that can be found in normal individuals. Therefore, in order to establish a diagnosis for SBBO, it is necessary to define the location, the count and the type of bacteria found. In general, studies regarding normal flora have been carried out by analyzing fluid collected by intubation of the Treitz angle region. Dickman et al14 studied the intestinal fluid of 22 healthy individuals, finding 103 bacteria/ml in 16 of them and 104 bacteria/ml in two individuals. In another study, Thadepalli et al15 found up to 104 bacteria/ml in six of 28 individuals. Fagundes-Neto et al16 evaluated the small-bowel microflora in chil dren who were not undernourished or presenting diarrhea, and found similar numbers of bacteria.

Many authors17-27 characterize SBBO based on bacterial count, without taking into consideration the family or genus of these micro-organisms. According to this quantitative principle, various cut-off points can be found in the literature:

  • 1) > 104 bacteria/ml;17-21
  • 2) > 105 bacteria/ml;22-25
  • 3) > 106 bacteria/ml.26,27

Other authors28,29 have used criteria that take into consideration the family as well as the number of the bacteria, characterizing SBBO when >103 coliforms and aerobic or anaerobic entero-bacteria/ml are found in the intestinal fluid.

Finally, there is a group of authors that considers SBBO to be the presence, in any concentration, of aerobic or anaerobic bacteria of the colonic microflora.16,30-33

In spite of the lack of consensus on the interpretation of results, the intestinal fluid culture is considered the gold standard test for the diagnosis of SBBO.6,22,28 However, the invasive character of duodenal intubation and the high cost of aerobic and anaerobic microbiological study have driven research into lower cost, non-invasive methods, of which the hydrogen breath test has shown itself to be the most important.

The basic principle of the hydrogen breath test is the fermentation of carbohydrates by bacteria present in the intestinal lumen, which generates, among other products, hydrogen. After it diffuses through the intestinal mucosa and reaches the lungs through the circulatory system, the hydrogen is eliminated in exhaled air. Normally, fermenting bacteria are found only in the colon: so, when an unabsorbed carbohydrate reaches the colon, it is fermented and the elimination of hydrogen through the breath occurs about 10 min after it reaches the large intestine. Under these circumstances, the carbohydrate must travel the entirety of the small intestine for it to begin to produce hydrogen. This orocecal transit time varies according to the type of carbohydrate and its concentration in the administered solution. The lactulose solution employed in the hydrogen breath test for SBBO diagnosis presents a mean orocecal transit time of approximately 90 min. When SBBO is present, bacterial fermentation occurs prematurely, while still in the small intestine, elevating the concentration of hydrogen in the breath in the samples collected in the 60 min following the ingestion of the carbohydrate. Ideally, when the unabsorbed carbohydrate reaches the colon, it will produce a second peak of hydrogen corresponding to its fermentation by the colonic flora. Thus, after the ingestion of a carbohydrate that is not absorbed, such as lactulose, in individuals with bacterial overgrowth, two peaks in hydrogen production are expected: the first is an early peak, generated by the bacteria of the SBBO, while the second is a late peak, due to the normal production of hydrogen in the colon, as illustrated in Figure 13.1. The other carbohydrate used in the hydrogen breath test is glucose. As glucose is, in ordinary circumstances, completely absorbed by the small intestine, there should be no delayed peak, except if the patient presents intestinal malabsorption of glucose associated with SBBO.19,34-36

Most studies relating the hydrogen breath test to the culture of intestinal fluid have been carried out in adults. However, common criteria were not used for the interpretation of intestinal fluid cultures, which is considered the gold standard for the characterization of SBBO. The main points of these studies are summarized in Table 13.3.22,26,35,37-39

To our knowledge, only six studies have been carried out in children. Davidson et al40 studied nine children aged 2-34 months, and characterized SBBO as the presence of more than 104 bacteria/ml in duodenal fluid. Lactose, sucrose and lactulose were used as substrates. SBBO was deter

Sbbo Glucose Positive Result

0 50 100 150 200 250 300

0 50 100 150 200 250 300

Time (min)

Figure 13.1 Schematic representation of possible results of the hydrogen breath test with lactulose.

Table 13.3 Studies in adults evaluating the diagnostic performance of the hydrogen breath test for the diagnosis of small-bowel bacterial overgrowth

Authors and reference

Number of patients

Probe

Criteria for bacterial overgrowth*

Sensitivity

(%)

King and Toskes26

20

lactulose 10 g

>10ppm

61

41

Kerlin and Wong37

45

glucose 50 g

> 12 ppm

93

78

Corazza et al35

77

lactulose 1 2g

> 1 0 ppm

68

44

Riordan et al38

42

lactulose 1 0 g

> 1 6 ppm

20

75

Riordan et al22

28

lactulose 1 0 g

> 1 0 ppm

16

70

MacMahon et al39

30

glucose 50 g

> 1 0 ppm

75

30

  • Increment of breath hydrogen concentration before colonic peak
  • Increment of breath hydrogen concentration before colonic peak mined by the presence of a double peak. Boissieu et al41 studied cultures of intestinal fluid and the hydrogen breath test after administering glucose to five children. A study carried out by Khin-Maung et al,27 with 19 children aged from 3 to 5 years, considered SBBO to be a concentration of more than 106 bacteria/ml in the enteral fluid. After ingestion of 10g of lactulose, an increase of more than 10ppm in samples collected after 20, 40 and 60 min was considered indicative of SBBO. Only two of the children with positive cultures presented a hydrogen breath test indicative of SBBO. Furthermore, three of the nine children with positive cultures did not present a positive hydrogen breath test indicative of SBBO.

Marcelino42 carried out a study in our institution and observed that only three of 18 unweaned infants, with acute or persistent diarrhea and SBBO, according to an intestinal fluid culture of the small intestine, presented a rise in breath hydrogen of more than 20 ppm after the ingestion of 10g of lactulose. It should be pointed out that almost 80% of the unweaned infants studied were not hydrogen producers and many presented secondary lactose malabsorption.

Guno et al43 studied SBBO in 31 infants with ages ranging from 2 to 24 months suffering from protein-energy malnutrition. The culture of the small-bowel fluid revealed that SBBO was present in ten of the 31 patients. The sensitivity and specificity of the hydrogen breath test using lactulose was 72% and 90%, respectively, while the scores for the hydrogen breath test using glucose were 55% and 90%, respectively. The authors did not include criteria for interpretation of the breath test.

Silva33 compared the results of the culture of aerobic and anaerobic organisms in the intestinal fluid of 31 children, with ages ranging from 6 months to 16 years, with the increase in breath hydrogen after the ingestion of 10g of lactulose. SBBO was characterized by the presence of any bacterium from the colonic flora in the duodenum, and was found in 16 (51.6%) of the 31 children. Given the minimum increase of 10ppm in breath hydrogen concentration in the samples collected up to 60min, the sensitivity of the hydrogen test was 75% and specificity 60%.

Analysis of the available information shows that both in adults and in children, irrespective of the carbohydrate that is used - lactulose or glucose -the occurrence of false-positive and false-negative results in the hydrogen breath test in relation to culture requires cautious application of this test in researching SBBO in a individual patient.

Another utilization of the hydrogen breath test is the analysis of groups of individuals in order to obtain information about the bacterial flora of the

Clinical presentation 207

Negative Carbohydrate Utilization Test

Figure 13.2 Median of the increment of the concentration of hydrogen in the expired air of 16 patients with bacterial overgrowth (unbroken line) and in 15 patients without bacterial overgrowth (dotted line) according to culture of the intestinal fluid. Differences at 15 min, p = 0.98; 30 min, p = 0.90; 45 min, p = 0.58; 60min, p = 0.26; 90 min, p = 0.075; 120 min, p =0.028; 150min, p = 0.031; 180 min, p = 0.007.

Figure 13.2 Median of the increment of the concentration of hydrogen in the expired air of 16 patients with bacterial overgrowth (unbroken line) and in 15 patients without bacterial overgrowth (dotted line) according to culture of the intestinal fluid. Differences at 15 min, p = 0.98; 30 min, p = 0.90; 45 min, p = 0.58; 60min, p = 0.26; 90 min, p = 0.075; 120 min, p =0.028; 150min, p = 0.031; 180 min, p = 0.007.

digestive tract in varying environmental conditions. The results obtained by Silva,33 with two groups of children (15 with SBBO and 16 without SBBO) can, in this light, be reanalyzed differently. Figure 13.2 presents the median values of increases in hydrogen on expiration. It can be seen that the SBBO groups produced a greater amount of hydrogen after 60min, and that there was a statistically significant difference in hydrogen increases at 120, 150 and 180min of the test. The data suggest two possibilities: that individuals with SBBO may not present a double peak of hydrogen; or that the colonic flora of children with SBBO has a greater hydrogen-producing capacity. This is based on the observation that, at the beginning of the test, the difference in hydrogen production in expired breath in SBBO patients was lower than after 60 min of the test. This second aspect was further analyzed using the area under the curve as the basis for calculation. In the fasting sample until 60 min, the 16 patients with SBBO presented a median hydrogen production rate (720ppm/min) not statistically different from the 15 patients without SBBO (923 ppm/min). In turn, from 60 to 180 min this parameter was signifi cantly higher (p = 0.017) in SBBO patients (6943 ppm/min) compared to patients without SBBO (3074 ppm/min).

0 0

Post a comment