Routine Blood Tests

Blood tests, such as CBC, CMP, TFT, and determination of inflammatory markers such as the erythrocyte sedimentation rate (ESR) and level of C-reactive protein (CRP), are often ordered on initial evaluation of patients with suspected IBS. These tests are relatively inexpensive, can be processed rapidly in a local laboratory, and are generally considered noninvasive. However, there is little evidence to suggest that their routine use enhances the diagnosis of organic disease or results in meaningful management changes in patients with suspected IBS. Tolliver and colleagues found that evaluation with CBC, ESR, CMP, TFT, and urinalysis in patients with suspected IBS yielded no alternative diagnoses (other than IBS) to explain the patients’ gastrointestinal symptoms. Sanders and colleagues evaluated 300 patients who fulfilled the Rome II criteria for IBS with a similar battery of blood tests and found 1 patient (0.3%) with an abnormal CBC, 2 (0.7%) with abnormal levels of liver-associated enzymes, and 1 (0.3%) with an elevated ESR. Likewise, Cash and colleagues found no differences in the results of routine blood tests (CBC, CMP, TFT, ESR, CRP) between a cohort of nearly 500 patients with IBS and 500 healthy controls. These results are consistent in that they confirm that the likelihood of identifying organic disease with routine laboratory tests seems to be no greater in patients with IBS (and absent alarm features) than in the general population. However, many of these studies arise from the clinical trial or tertiary care arenas and, therefore, may not be representative of the larger population of patients with suspected IBS who are encountered in community practice.

Testing for Celiac Disease

Patients with IBS often endorse specific food ingestion as a primary trigger of their gastrointestinal symptoms, and many identify gluten-containing foods as problematic, causing a concern of celiac disease in patients with suspected IBS. The symptoms of IBS and celiac disease may be indistinguishable. In a survey of 1032 members of the Celiac Disease Foundation, many respondents endorsed presenting symptoms characteristic of IBS and patients with celiac disease often have been diagnosed with IBS for some time before the formal diagnosis of celiac disease. Celiac disease, a chronic enteropathy resulting from an immune-mediated response to deamidated gliadin, has a prevalence of about 0.7% to 1% in the United States. Common symptoms of celiac disease include abdominal pain (77%), gas or bloating (73%), diarrhea (52%), constipation (7%), and fluctuation between diarrhea and constipation (24%). Untreated celiac disease can result in malnutrition, anemia, osteoporosis, infertility, and an increased risk of gastrointestinal lymphoma. The diagnosis of celiac disease is typically based on the demonstration of elevated levels of antigliadin antibodies (AGAs), antiendomysial antibodies (anti-EMAs), or anti–tissue transglutaminase antibodies, with the last 2 antibody tests preferred because of their higher sensitivity and specificity than AGA. The gold-standard diagnostic test for celiac disease is small intestinal biopsy demonstrating characteristic changes in the intestinal mucosa, including villous atrophy, crypt hyperplasia, and inflammation of the lamina propria. Because of the invasiveness and expense of upper gastrointestinal endoscopy and biopsy, this diagnostic modality is typically reserved for confirmation of abnormal antibody testing or evaluation of treatment effects after the diagnosis has been made.

Multiple case-control studies have demonstrated an increased prevalence of celiac disease among subjects with IBS compared with healthy controls. Sanders and colleagues screened 300 consecutive new patients who met the Rome II IBS criteria and an age-and-gender–matched control group with anti-EMA and antigliadin IgG and IgA antibody tests and used duodenal biopsy to confirm celiac disease in serologically positive subjects. Celiac disease was identified in 14 (4.6%) patients with IBS but in only 2 (0.7%) of the controls (OR, 7.0; P = .004). A more recent prospective US study comparing the prevalence of celiac disease in patients with suspected IBS and in controls found that 7.3% of 492 patients with nonconstipated IBS (diagnosed by Rome II criteria) had elevated celiac disease antibodies. The prevalence of biopsy-confirmed celiac disease in this group was 0.4%, similar to that observed in 458 controls without IBS symptoms ( P = .28) and US population prevalence estimates. In a recent meta-analysis, the prevalence of celiac disease in individuals with suspected IBS was computed to be 4 times greater than in non-IBS controls when data from 14 studies comprising 4202 patients were included.

In its revised statement, the ACG IBS Task Force recommended testing for celiac disease in patients presenting with nonconstipated IBS symptoms, based on studies such as those discussed earlier and analyses that suggest that serologic testing for celiac disease is cost-effective when the pretest probability exceeds 1%. Mein and Ladabaum used decision analytic modeling to evaluate the cost-effectiveness of serologic screening for celiac disease in patients with IBS. They concluded that with a calculated cost per quality adjusted life-year gain of $4600 per case, testing patients with suspected celiac disease is cost-effective. Spiegel and colleagues came to a similar conclusion in their cost-effectiveness study, even after biasing their model against testing by assuming low individual test characteristics and incomplete patient adherence to a gluten-free diet. Given the wide geographic variability in the prevalence of celiac disease among persons with IBS symptoms, it may be necessary to know local prevalence data to understand cost-effectiveness of serologic screening in a specific population.

Colonoscopy

Patients with typical IBS symptoms commonly undergo colonoscopy. For example, community-based surveys indicate that up to 50% of patients with IBS undergo colonoscopy in the course of their diagnostic evaluation. Furthermore, available data suggest that 25% of all colonoscopies performed in the United States are for IBS-related symptoms and 10% of colonoscopies performed on patients younger than 50 years are conducted for the evaluation of IBS symptoms. Despite such broad use of colonoscopy in the evaluation of IBS symptoms, data addressing the actual prevalence of colonic structural abnormalities in patients with IBS are limited.

In a post hoc analysis of data from 2 placebo-controlled IBS trials, Hamm and colleagues examined the yield of flexible sigmoidoscopy in study participants younger than 50 years or colonoscopy/flexible sigmoidoscopy and barium enema in enrollees aged 50 years and older. Of 1452 study participants, 306 (21%) were included in this post hoc analysis. Colonic imaging identified important structural lesions in 4 patients (1.3%; 3 IBD, 1 colonic obstruction). Unfortunately, the results of this post hoc analysis are difficult to interpret, as the proportion of patients who underwent sigmoidoscopy versus colonoscopy was not reported, patients underwent colonic imaging only if they had not undergone such imaging within 2 years of study enrollment, and the proportion of patients with IBS who had undergone previous colonic imaging was not reported.

Tolliver and colleagues evaluated the yield of sigmoidoscopy, barium enema, and/or colonoscopy in an uncontrolled, prospective trial of 196 subjects with suspected IBS. Forty-three colonic structural abnormalities were found in 34 subjects. Most abnormalities found were thought to be incidental and not responsible for the patients’ gastrointestinal symptoms (benign polyps, diverticulosis, hemorrhoids, lipomas, and melanosis coli). Two (1.0%) patients were found to have organic disease (1 IBD, 1 cancer) that could have explained their IBS symptoms. Again, there are several issues that complicate the interpretation of the results from this study, including the inclusion of patients with warning signs (family history of colon cancer and hemoccult positive stool), absence of a control group, and failure to report the percentage of patients with IBS who underwent each type of examination. In an uncontrolled study, Francis and colleagues studied the yield of flexible sigmoidoscopy, barium enema, or colonoscopy in 125 patients who fulfilled the Rome I criteria for IBS. With the exception of diverticulosis that was judged to be an incidental finding, no structural lesions that changed the diagnosis of IBS were identified.

Vanner and colleagues performed a prospective study in 95 patients referred to the general gastroenterology clinics over a 9-month period in 1995–1996 who met the Rome criteria and lacked warning signs. About 91% of patients older than 45 years underwent barium enema or colonoscopy. Aside from 1 patient with rectal bleeding who was found to have ulcerative proctitis, no colonic abnormalities were identified. In another study, investigators reviewed colonoscopy reports of 622 patients with suspected IBS and 642 patients without IBS. All colonoscopies were performed to evaluate new gastrointestinal complaints. Colonoscopy results were normal in only 48.4% of the patients with IBS; however, the lesions identified, such as hemorrhoids (21.1%), polyps (20.3%), diverticulosis (19%), and angiodysplasia (12.1%), were just as frequent in the non-IBS group, and none of these findings were felt to be responsible for IBS symptoms.

In a prospective trial of colonoscopy in patients with suspected IBS, Chey and colleagues performed colonoscopy on 466 patients with suspected IBS (IBS with diarrhea [IBS-D] or IBS with a mixed stool pattern [IBS-M]) and 451 healthy controls. Compared with the control group, the study group was younger and predominantly female. Colonoscopy revealed organic lesions in both groups, with the most common lesions being hemorrhoids, polyps, and diverticulosis. Polyps were more than 2 times as prevalent in the older control group (34.4% vs 14.6%, P ≤.0001) and were more likely to be adenomatous than in patients with suspected IBS (26.1% vs 7.7%, P ≤.0001). Two patients (0.4%) in the IBS group were diagnosed with IBD, and routine colonic mucosal biopsies yielded a diagnosis of microscopic colitis in an additional 7 patients (1.5%) with suspected IBS. The prevalence of microscopic colitis was higher in older patients with IBS-D. A retrospective review of patients with biopsy-proven microscopic colitis found that 56% met the diagnostic criteria for IBS (Rome II) and one-third had previously been diagnosed with IBS. Routine biopsies in the sigmoid colon of patients with IBS older than 40 years undergoing colonoscopy revealed microscopic colitis in more than 5%. These findings provide the rationale for the recommendations of the ACG IBS Task Force to perform colonoscopy in patients with IBS older than 50 years (for colon cancer screening) and those with alarm features. According to the Task Force, when colonoscopy is performed in patients with suspected IBS-D, clinicians should consider performing random mucosal biopsies to rule out microscopic colitis.

Breath Testing for Small Intestinal Bacterial Overgrowth or Carbohydrate Maldigestion

Bloating is one of the most common symptoms associated with IBS, occurring in more than 75% of patients with IBS. The high prevalence of bloating and the increasing recognition of symptom improvement from antibiotic and probiotic therapy for IBS has prompted the consideration of breath testing of patients with IBS to detect small intestinal bacterial overgrowth (SIBO). Similarly, the frequent association of IBS-like symptoms with lactose or fructose ingestion suggests a rationale for the use of breath tests to detect carbohydrate maldigestion. Breath tests are performed by measuring expired hydrogen or methane levels after ingestion of a disaccharide load. Increased gas production resulting from fermentation of carbohydrates by bacteria, either because of an overabundance of fermenting bacteria in the small bowel or because of maldigestion of sugars serving as an additional substrate for these bacteria, is thought to be a possible organic cause for the bloating observed in patients with IBS symptoms.

Whether or not a correlation exists between IBS and SIBO remains controversial. Pimentel and colleagues found that 84% of patients with IBS compared with 20% in a gender-matched control group (OR, 26.2; 95% CI, 4.7–103.9; P <.00001) had abnormal results in the lactulose breath test (LBT) indicative of SIBO. In their randomized, double-blind, placebo-controlled study of neomycin as a therapy for IBS, they found that patients with abnormal baseline LBTs were 10 times more likely to have a significant improvement in IBS symptoms after treatment with neomycin versus placebo. A recent systematic review and meta-analysis of pooled data from 12 case series and case-control studies examined the prevalence of abnormal breath testing in IBS. Challenged by considerable heterogeneity between studies, including the use of 3 different types of breath tests (lactulose, glucose, or sucrose), these investigators concluded that the prevalence of SIBO among a total of nearly 2000 adults meeting the criteria for IBS ranged from 4% to 64%, depending on the type of test used and the criteria defining a positive test result. Even when comparing studies using the same substrate, a broad range of prevalence rates was evident. The ACG Task Force echoed this conclusion when they indicated that there was insufficient evidence to support the routine use of breath tests for SIBO in the evaluation of patients with suspected IBS. Another study used jejunal aspirate and culture, considered to be a more accurate test for SIBO, and found that only 4% of patients with IBS had SIBO (defined as >10 ⁶ colony-forming units/mL), similar to what was observed in non-IBS controls. However, this study was considered supportive of a putative role of small bowel bacteria in the pathophysiology of IBS because it showed significantly higher concentrations of coliform bacteria (albeit at lower thresholds) in the small bowel of patients with IBS compared with those without IBS.

Maldigestion of lactose caused by lactase deficiency is common, with prevalence rates as high as 25% in the United States and 75% in other parts of the world. Rates among certain ethnic populations within the United States can be as high as 90%. In affected individuals, ingestion of significant quantities of lactose can result in a syndrome of diarrhea, abdominal pain, flatulence, and/or bloating. A study conducted by Vernia and colleagues prospectively assessed the prevalence of lactose malabsorption among patients diagnosed with IBS. They compared the results of breath tests in 503 patients with IBS with those in 336 patients who subjectively identified themselves as lactose intolerant and found comparable rates of positive test results between the 2 groups, with 337 patients with IBS (66.9%) and 240 patients with reported lactose intolerance (71.4%) having greater than 20 ppm rise in hydrogen concentration in the expired breath after lactose ingestion. Choi and colleagues identified fructose maldigestion in 33% of patients with Rome II IBS symptoms based on expired hydrogen and methane levels after the ingestion of 25 g of fructose. At 1-year follow-up evaluations, the 46% of patients with fructose maldigestion who adhered to a fructose-restricted diet reported significant improvement in 6 of 9 gastrointestinal symptoms ( P = .02), whereas the fructose-tolerant group reported significant improvement in only 2 symptoms.

Although breath testing may objectively identify those patients who are unable to digest the particular sugar being tested, an optimal breath testing technique and results can be difficult to achieve, and many patients with troublesome gastrointestinal symptoms who have negative breath test results may still benefit from food avoidance or presumed manipulation of their intestinal microbiome with antibiotics or probiotics. Because of this and other concerns about the validity of breath test results in the diagnosis of carbohydrate maldigestion, some have questioned whether it should be used in the evaluation of this condition at all and suggested that patient report and response to a food challenge may be a more appropriate diagnostic tool. Patient report alone is a poor predictor of SIBO or carbohydrate maldigestion. This fact was shown in a randomized, double-blind, cross-over trial of subjects reporting symptoms consistent with severe lactose intolerance in which they underwent blinded hydrogen breath testing and were then randomized to receive a daily 240-mL serving of milk (containing 12 g of lactose) or lactose-hydrolyzed milk (containing no lactose). Subjects were instructed to avoid all other dairy products during the study period and maintain daily dietary records. After 1 week, the subjects were crossed-over to the other milk preparation. Of the 30 self-reported severe lactose-intolerant patients, nearly one-third had a negative result in the hydrogen breath test. Analysis of the symptoms reported during the study period revealed no significant difference in the amount of flatus episodes, abdominal pain, diarrhea, or bloating between the 2 milk preparations. Overall, symptoms experienced were minimal among those with normal and abnormal lactose breath test results. From a practical standpoint, whether food-related gastrointestinal symptoms result from actual enzyme deficiencies or simply food intolerance is of minimal clinical significance, because management is similar with either cause. Consequently, in patients reporting food-related symptoms, either by history or based on correlating symptoms with a food diary, trials of lactose- or fructose-restricted diets are a rational and pragmatic therapeutic approach. Similarly, it was recently reported that rifaximin at a dosage of 1650 mg/d for 14 days resulted in a statistically significant improvement in IBS symptoms and bloating in several large phase 3 trials of this therapy compared with placebo. This approach may also be a realistic empiric therapeutic option in lieu of breath testing because breath testing for SIBO was not performed in patients enrolled in these trials.

Stool Studies

The ACG IBS Task Force does not recommend the routine use of stool studies to evaluate for infectious causes of IBS symptoms in the absence of a relevant travel history or specific alarm features (severe uncontrollable diarrhea, hematochezia, weight loss). Studies by Hamm and colleagues and Tolliver and colleagues demonstrated a low prevalence of positive results of fecal ova and parasite tests in patients with IBS. Only 19 (2%) of Hamm’s cohort of 1452 patients had evidence of intestinal parasites, with nearly half of those being from colonization with Blastocystis hominis , which has been implicated as a possible factor in IBS but is of questionable clinical significance. Stool examinations for parasites in Tolliver’s cohort of 196 patients with IBS had normal findings.

Several neutrophil-derived proteins excreted in the feces, such as lactoferrin and calprotectin, are highly sensitive and specific markers of inflammatory activity in the gut, but their use in differentiating organic from functional disease has not been widely studied. Kane and colleagues measured lactoferrin concentrations in outpatients with Crohn disease, ulcerative colitis, and IBS and in healthy controls. Lactoferrin concentrations were significantly higher in the 2 IBD groups than in the controls or in patients with IBS. An abnormal fecal lactoferrin level was 100% specific in ruling out IBS. Similarly, strong test characteristics were shown for both fecal calprotectin and lactoferrin in a prospective blinded study comparing 64 patients with IBD (all with endoscopically proven active disease) with 30 patients with IBS and 42 healthy controls. Mean stool concentrations of both proteins were more than 10 times higher in the IBD groups than in the IBS group ( P≥ .0001), whereas concentrations were similar in the IBS group and controls. A third study examined the ability of these 2 stool markers, as well as a third marker, polymorphonuclear elastase, to distinguish active IBD from inactive disease and IBS. Concentrations of all of these proteins were significantly higher in active IBD than in quiescent disease or in IBS, but diagnostic accuracy of the 3 assays for differentiating IBS from IBD ranged from 69% to 87%. Because of the high negative predictive value of these tests in ruling out intestinal inflammation, many clinicians use them as a noninvasive screen for IBD in their patients with typical IBS symptoms. Cost-effectiveness studies assessing this strategy are needed before this approach can be advocated.

Tests of Gut Hormones and Other Serologic Biomarkers

Numerous studies over the past several years have attempted to identify serum biomarkers correlating with the presence of IBS, based on many of the 600 different pathways implicated in the pathophysiology of IBS. Many of these tests have focused specifically on the suggestion that the clinical syndrome of IBS may be associated with alterations in the secretion of various gut hormones, neurotransmitters, and inflammatory mediators. Although no single serologic marker has been able to reliably diagnose IBS, the presumed multifactorial nature of IBS has prompted researchers to develop assays to measure plausible biomarkers to clarify the diagnosis of IBS. The article by Spiller elsewhere in this issue discusses this exciting field of study and its implications for the management of IBS.