Medical students are taught that 90% of all diagnoses are made through careful assessment of the patients’ symptoms. Clinicians now rely heavily on techniques such as endoscopy or radiology before making a definitive diagnosis of organic disease. Most gastroenterologists would require endoscopic confirmation before labeling a patient as having peptic ulcer disease and would make a diagnosis of Crohn disease based on small bowel radiology or colonoscopy. However, the most common causes of symptoms of the gastrointestinal tract are functional. It is important that clinicians obtain a thorough history so that the disorder of the patient can be accurately defined.
Medical students are taught that 90% of all diagnoses are made through the careful assessment of the patients’ symptoms. This methodology was probably true in the past, but clinicians now rely heavily on techniques such as endoscopy or radiology before making a definitive diagnosis of organic disease. Indeed, most gastroenterologists would require endoscopic confirmation before labeling a patient as having peptic ulcer disease and would make a diagnosis of Crohn disease based on small bowel radiology or colonoscopy. However, the most common causes of symptoms of the gastrointestinal (GI) tract are functional, and most GI investigations are normal. It is therefore important that clinicians obtain a thorough history so that the disorder of the patient can be accurately defined. Gastroenterologists seem to be aware of this method, because office visits are longer than that for primary care providers, particularly for those patients diagnosed with irritable bowel syndrome (IBS). However, community gastroenterologists and primary care providers still think that IBS is a diagnosis of exclusion, whereas IBS experts are more likely to make a positive diagnosis. This scenario suggests that experts are more confident on the accuracy of symptoms in diagnosing IBS.
The rigorous evaluation of symptoms that identify patients with IBS started with the criteria identified by Manning and colleagues. Researchers then developed more complex models that relied on symptoms and simple investigations to diagnose IBS. From an academic perspective, the field was still in disarray, with researchers recruiting subjects using a variety of definitions of IBS. This scenario made it difficult to compare studies because it was unclear as to what type of patient was being evaluated. Progress came with the Rome Foundation, which brought together a group of experts to reach a consensus on the appropriate definition of IBS. There have been 3 iterations of Rome Foundation’s symptom-based definitions of IBS and these have been predominantly used in research studies evaluating functional GI disorders; they have also been used to help manage IBS in clinical practice. Although the Rome criteria have undoubtedly brought order to functional GI research, the question remains as to their accuracy in diagnosis and whether they are any better than the previous symptom-based criteria.
Accuracy of symptom-based criteria for IBS
The authors have updated their previous systematic review addressing this issue searching MEDLINE (1950–October 2010) and EMBASE (1908–October 2010) using a predefined search strategy. No new studies were eligible for the review, but a further systematic review that also evaluated the accuracy of symptom-based criteria in IBS has been published subsequently. This review identified 15 more studies than that of the authors’, so it is important to establish the differences between the 2 systematic reviews. It has been shown that there can be errors in systematic reviews of IBS that can change the interpretation of the data. It is therefore important that the researchers account for any discrepancy between these 2 systematic reviews. The systematic review of the authors required imaging of the colon with colonoscopy, barium enema, or computed tomographic colography to exclude organic disease. One article included in the systematic review by Jellema and colleagues investigated patients with flexible sigmoidoscopy only and was therefore excluded from the review. A study was excluded that just compared Rome I with Rome II criteria without a gold standard for diagnosing IBS that was included by Jellema and colleagues. One study included by Jemella and colleagues was excluded by the authors’ review because data to provide sensitivity and specificity could not be extracted. The authors excluded 3 papers that evaluated scoring models that separated organic from any functional GI disease but were not specifically designed to identify patients with IBS. Also, the review excluded 9 studies that enrolled patients with both upper and lower GI symptoms (the 2 types of symptoms could not be separated in the review). This exclusion is because the main question is whether symptom-based criteria can identify patients with IBS in subjects with lower GI symptoms, not any GI symptom (eg, it was anticipated that the criteria could easily distinguish a patient with gastroesophageal reflux disease from a patient with IBS).
The 10 studies assessing 2355 patients with lower GI symptoms included in the original review were reanalyzed to evaluate the diagnostic utility of symptom-based criteria in identifying patients with IBS. The study characteristics are described in Table 1 . Individual symptoms perform poorly in identifying patients with IBS ( Table 2 ), so it is important to consider symptom complexes when making the diagnosis. Studies investigated the accuracy of Manning criteria ( Box 1 ), Rome I criteria (see Box 1 ), and statistical models ( Box 2 ). There were 4 studies that evaluated 574 patients reporting on the accuracy of Manning criteria. Sensitivity varied between 66% and 90%, with specificity between 56% and 87% ( Fig. 1 ). One study involving 602 participants evaluated the Rome I criteria and reported a sensitivity of 71% and specificity of 85%. The Kruis model was evaluated in 4 studies in 1171 patients, and sensitivities ranged between 56% and 83%, with specificities between 65% and 97% (see Fig. 1 ). The accuracy of other statistical models was investigated by 4 studies (1 study evaluated a different scoring system of the Kruis model) in 863 patients. Sensitivity varied between 76% and 91%, with specificity between 53% and 100% (see Fig. 1 ).
| Refs. | Country | Total Number | Proportion IBS (%) | Method of Assessment | Type of Patient | Setting | Assessors Blinded? |
|---|---|---|---|---|---|---|---|
| England | 65 | 49 | Symptom-based criteria | Referred to outpatient clinic with lower GI symptoms; unclear who referred | Gastroenterology and surgical clinic in a single hospital | Yes | |
| Germany | 317 | 34 | Statistical model | Referred to outpatient clinic with lower GI symptoms by external physician | Internal medicine clinic in a single hospital | Yes | |
| Korea | 74 | 78 | Symptom-based criteria | Referred to outpatient clinic with lower GI symptoms; unclear who referred | Internal medicine clinic in a single hospital | Yes | |
| India | 88 | 74 | Symptom-based criteria | Referred to outpatient clinic with lower GI symptoms; unclear who referred | Gastroenterology clinic in a single hospital | Unclear | |
| Turkey | 347 | 48 | Symptom-based criteria and statistical model | Referred to outpatient clinic with lower GI symptoms; unclear who referred | Gastroenterology clinic and internal medicine clinic in 2 hospitals | Yes | |
| Italy | 253 | 21 | Statistical model | Referred to outpatient clinic with lower GI symptoms by PCP | Gastroenterology clinic in a single hospital | Yes | |
| Italy | 254 | 60 | Statistical model | Consulted PCP or referred to outpatient clinic with lower GI symptoms | 14 PCPs and a gastroenterology clinic in a single hospital | Yes | |
| India | 75 | 73 | Statistical model | Referred to outpatient clinic with lower GI symptoms, unclear who referred | Gastroenterology clinic in a single hospital | Yes | |
| Australia | 280 | 76 | Statistical model | Referred to outpatient clinic with lower GI symptoms by PCP primarily and also by surgeons and internists | Gastroenterology clinic in a single hospital | Yes | |
| England | 602 | 56 | Symptom-based criteria | Referred to outpatient clinic with lower GI symptoms by PCP | Gastroenterology clinic in a single hospital | Yes |
| Symptom Item | Refs. | Sensitivity (95% CI) | Specificity (95% CI) | Positive Likelihood Ratio (95% CI) | Negative Likelihood Ratio (95% CI) |
|---|---|---|---|---|---|
| Lower abdominal pain | 0. 97 (0.84–1.0) 0.96 (0.91–0.99) 0.87 (0.74–0.94) 0.80 (0.74–0.85) | 0.09 (0.02–0.24) 0.45 (0.39–0.52) 0.36 (0.29–0.43) 0.33 (0.22–0.46) | 1.1 (0.92–1.3) 1.8 (1.6–2.0) 1.4 (1.1–1.6) 1.2 (1.0–1.5) | 0.34 (0.05–2.3) 0.08 (0.03–0.20) 0.38 (0.18–0.73) 0.60 (0.40–0.94) | |
| Mucus per rectum | 0.47 (0.29–0.65) 0.19 (0.10–0.31) 0.78 (0.67–0.88) 0.36 (0.29–0.42) | 0.79 (0.61–0.91) 0.81 (0.54–0.96) 0.35 (0.1–0.57) 0.65 (0.52–0.76) | 2.2 (1.1–4.7) 1.0 (0.36–3.2) 1.2 (0.92–1.8) 1.0 (0.71–1.5) | 0.67 (0.45–0.96) 1.0 (0.80–1.4) 0.62 (0.31–1.3) 0.99 (0.82–1.2) | |
| Incomplete evacuation | 0.59 (0.41–0.76) 0.78 (0.65–0.87) 0.85 (0.74–0.92) 0.72 (0.66–0.78) | 0.67 (0.48–0.82) 0.31 (0.11–0.59) 0.35 (0.16–0.57) 0.42 (0.30–0.55) | 1.8 (1.0–3.2) 1.1 (0.85–1.8) 1.3 (1.0–1.9) 1.3 (1.0–1.6) | 0.61 (0.37–0.97) 0.72 (0.33–1.8) 0.44 (0.21–0.99) 0.65 (0.46–0.94) | |
| Looser stools at onset of pain | 0.81 (0.63–0.93) 0.59 (0.45–0.71) 0.48 (0.35–0.60) 0.50 (0.43–0.57) | 0.73 (0.54–0.88) 0.63 (0.35–0.85) 0.87 (0.66–0.97) 0.70 (0.57–0.80) | 3.0 (1.7–5.8) 1.6 (0.89–3.3) 3.7 (1.4–11) 1.7 (1.2–2.5) | 0.26 (0.12–0.52) 0.66 (0.42–1.2) 0.60 (0.45–0.82) 0.72 (0.59–0.90) | |
| More frequent stools at onset of pain | 0.74 (0.55–0.88) 0.57 (0.43–0.70) 0.35 (0.24–0.48) 0.51 (0.44–0.58) | 0.70 (0.51–0.85) 0.69 (0.41–0.89) 0.91 (0.72–0.99) 0.62 (0.49–0.74) | 2.5 (1.5–4.6) 1.8 (0.96–4.1) 4.1 (1.3–15) 1.3 (0.98–1.9) | 0.37 (0.19–0.67) 0.63 (0.41–1.0) 0.71 (0.56–0.92) 0.79 (0.63–1.0) | |
| Pain relieved by defecation | 0.71 (0.52–0.86) 0.59 (0.45–0.71) 0.66 (0.53–0.77) 0.55 (0.48–0.62) | 0.70 (0.51–0.85) 0.69 (0.41–0.89) 0.57 (0.34–0.77) 0.67 (0.54–0.78) | 2.4 (1.4–4.4) 1.9 (1.0–4.2) 1.5 (0.99–2.6) 1.7 (1.2–2.4) | 0.41 (0.22–0.72) 0.60 (0.39–1.0) 0.60 (0.37–1.0) 0.67 (0.54–0.86) | |
| Patient reported visible abdominal distension | 0.59 (0.41–0.76) 0.40 (0.27–0.53) 0.22 (0.12–0.33) | 0.79 (0.61–0.91) 0.63 (0.35–0.85) 0.87 (0.66–0.97) | 2.8 (1.4–5.8) 1.1 (0.57–2.3) 1.7 (0.59–5.1) | 0.52 (0.32–0.78) 0.97 (0.67–1.6) 0.90 (0.75–1.2) |
Manning criteria
Symptom duration not defined
Number of symptoms required for diagnosis also not defined but more than 3 generally used
- •
Abdominal pain relieved by defecation
- •
More frequent stools with the onset of abdominal pain
- •
Looser stools with the onset of abdominal pain
- •
Visible abdominal distension reported by patient
- •
Incomplete evacuation
- •
Passage of mucus per rectum
- •
Rome I criteria
Symptoms present for at least 3 months. Patient must experience abdominal pain or discomfort relieved with defecation or associated with a change in stool frequency or consistency. In addition, there should be at least 2 of the following features for 25% of days or more:
- •
Altered stool frequency
- •
Altered stool form
- •
Altered stool passage
- •
Bloating or distension
- •
Passage of mucus per rectum
- •
Rome II criteria
Symptoms present for at least 12 weeks in 1 year (need not be consecutive). Patient must experience abdominal pain with at least 2 of the following features:
- •
Relieved by defecation
- •
Onset of pain associated with change in stool frequency
- •
Onset of pain associated with change in stool form
- •
Rome III criteria
Symptoms present for at least 3 days per month in the last 3 months (with symptom onset at least 6 months previously) with at least 2 of the following features:
- •
Pain improved with defecation
- •
Onset of pain associated with change in stool frequency
- •
Onset of pain associated with change in stool form
- •
Kruis and colleagues
Score greater than 44 defined as IBS
| Item | Score | |
|---|---|---|
| 1 | At least 1 of the following: abdominal pain, flatulence, change in bowel habit | +34 |
| 2 | Symptoms for more than 2 years | +16 |
| 3 | Abdominal pain described as burning, cutting, very strong, terrible, feeling of pressure, dull, boring, or not so bad a | +23 |
| 4 | Alternating constipation and diarrhea | +14 |
| 5 | History or physical pathognomonic for any diagnosis other than IBS | −47 |
| 6 | ESR>20 mm/2 h | −13 |
| 7 | Leucocytosis>10,000/cm 3 | −50 |
| 8 | Hemoglobin low (<12 g% female, <14 g% male) | −98 |
| 9 | History of blood in stool | −98 |
a Scored only if 1 statement in the first line or more than 2 statements in total endorsed.
Bellentani and colleagues
Score less than 0 defined as IBS
| Item | Score | |
|---|---|---|
| 1 | Visible distension of the abdomen | −39 |
| 2 | First-degree relatives have colitis | −35 |
| 3 | Have a feeling of distension | −34 |
| 4 | Suffer from flatulence | −33 |
| 5 | Suffer from irregularities of bowel movements | −26 |
| 6 | ESR>17 mm/h | +134 |
| 7 | History of blood in the stool | +112 |
| 8 | Age>45 years | +95 |
| 9 | Leucocytosis>10,000/cm 3 | +85 |
| 10 | Fever between 37°C and 38°C | +74 |
| 11 | History of cancer in first-degree relatives | +33 |
Mazumdar and colleagues
Score less than 0 defined as IBS
| Item | Score | ||
|---|---|---|---|
| Present | Absent | ||
| 1 | Abdominal pain | +3 | −9 |
| 2 | Early morning abdominal pain | +15 | −2 |
| 3 | Postprandial abdominal pain | −5 | +3 |
| 4 | Poorly localized pain | +14 | −30 |
| 5 | Food aggravating pain | +17 | −11 |
| 6 | Pain relieved after flatus or defacation | +2 | −10 |
| 7 | Nocturnal diarrhea | −24 | +4 |
| 8 | Alternating constipation/diarrhea | +16 | −5 |
| 9 | Repeated attempts to pass stool | +14 | −9 |
| 10 | Straining at defecation | +11 | −5 |
| 11 | Feeling of incomplete evacuation | +11 | −9 |
| 12 | Food precipitating bowel movement | +12 | −10 |
| 13 | Stress factor | +12 | −5 |
| 14 | Excess mucus in stool | +10 | −30 |
| 15 | Blood in stool | −4 | +6 |
| 16 | Blood uniformly mixed with stool | −37 | +21 |
| 17 | Blood after stool | +30 | −18 |
| 18 | Gas bloat/belching | +25 | −4 |
| 19 | Borborygmi | +8 | −4 |
Related posts:
Inflammation and Microflora
Centrally Acting Therapies for Irritable Bowel Syndrome
Emerging Pharmacological Therapies for the Irritable Bowel Syndrome
Inflammation and Microflora
The Role of Diagnostic Testing in Irritable Bowel Syndrome
Emerging Pharmacological Therapies for the Irritable Bowel Syndrome
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
