You are here
The impact of symptoms, irritable bowel syndrome pattern and diagnostic investigations on the diagnostic delay of Crohn's disease: A prospective study
Digestive and Liver Disease, Volume 47, Issue 8, August 2015, Pages 646 - 651
We investigated symptoms and tests performed prior to a formal diagnosis of Crohn's disease and the reasons for diagnostic delay.
Consecutive patients recently diagnosed with Crohn's disease were enrolled between October 2012 and November 2013. Clinical data, symptoms including Rome III criteria at onset and at diagnosis, location and disease phenotype were recorded. Faecal calprotectin, radiological and endoscopic examinations performed prior to diagnosis were analysed. Diagnostic delay, stratified into tertiles and median time, was analysed using parametric and nonparametric tests.
83 patients (49.4% males, median age 31 years) were enrolled. The median diagnostic delay was 8 (0–324) months. Twenty-six patients did not consult a general practitioner until diagnosis (31.3%), 18 presented to the emergency department (21.7%) and 8 directly to a gastroenterologist (9.6%). Diagnostic delay was not associated with specific symptoms. However, patients with bloating at presentation had a longer delay compared to those who did not (median, 6.1 vs. 16.8 months, respectively; p = 0.016).
Nineteen patients underwent incomplete ileocolonoscopies (22.9%) and 7 had no biopsies (8.4%), with a consequent diagnostic delay (median, 24 and 24 vs. 6 months, respectively; p = 0.025 and p = 0.008).
Diagnostic delay for Crohn's disease is significantly associated with incomplete ileocolonoscopies, but not with symptoms, except bloating at presentation.
Keywords: Calprotectin, Crohn's disease, Diagnostic delay, Ileocolonoscopy, Irritable bowel syndrome, Symptoms.
A definitive diagnosis of Crohn's disease (CD) is often a difficult process and frequently associated with a variable time delay. Several studies have shown that the time interval between the initial onset of symptoms secondary to CD and formal diagnosis can vary from 5 months to 11 years , , , , , and . Interestingly, some of these studies have highlighted that >25% of patients were diagnosed beyond 24 months from the initial presentation.
This diagnostic delay is known to translate into adverse outcomes for CD patients, such as an increased risk of bowel stricture formation and related intestinal surgery  . Thus, the importance of avoiding any delay in the diagnosis of CD has been highlighted, but identification of specific risk factors for this delay needs to be explored  . In this regard, the evaluation of presenting symptoms as risk factors has not been adequately performed. Irritable bowel syndrome (IBS)-like symptoms as presenting complaints in inflammatory bowel disease (IBD) patients have been studied, but their association as a risk factor for diagnostic delay remains controversial. Burgmann et al. have illustrated that IBS-like symptoms are likely to cause a delay in the diagnosis of IBD, but other studies have not verified this  and . The symptoms are the initial presenting complaint of patients, based on which clinicians decide their index of suspicion for IBD diagnosis.
To determine whether specific symptoms are risk factors for a delay in diagnosis, we assessed the characteristics and behaviour of symptoms at initial presentation and at formal diagnosis, as well as the correlation between the delay in diagnosis and diagnostic investigations. These were then analysed to identify whether specific risk factors can be identified to be associated with a diagnostic delay for Crohn's disease.
2. Patients and methods
A total of 83 consecutive subjects recently (<6 months) diagnosed with CD, between October 2012 and November 2013, were enrolled in the study. All patients provided informed consent and agreed to the collection of demographic and clinical data via a specific questionnaire. Patients with indeterminate colitis and those in whom the diagnosis changed (from ulcerative colitis (UC) to CD or vice versa) were excluded from the study.
The protocol was approved by the institutional review board of the L. Sacco University Hospital in Milan.
For each patient, the following data were obtained: sex, patients’ age and time of symptom onset, date of definitive diagnosis and site and behaviour of CD at diagnosis (according to the Montreal classification). The time duration between onset of symptoms and diagnosis was further subdivided into three groups: time elapsed between onset of symptoms and first consultation with the general practitioner (GP), time elapsed between GP consultation and the gastroenterologist consultation and, finally, the time elapsed between gastroenterologist consultation and final diagnosis. In patients who did not consult a GP and who presented directly to either the emergency department (ED) or the specialist, the time elapsed between this consultation and final diagnosis was assessed.
Symptoms included the number of bowel movements, rectal bleeding, abdominal pain rated on a Likert scale (0 = no symptom; 1 = minimal symptoms, it can be easily ignored without effort; 2 = mild symptoms, it can be ignored with effort; 3 = moderate symptoms, it cannot be ignored but does not influence my daily activities; 4 = moderately severe symptoms, it cannot be ignored and occasionally limits my daily activities; 5 = severe symptoms, it cannot be ignored and often limits my concentration on daily activities; and 6 = very severe symptoms, it cannot be ignored and markedly limits my daily activities and often requires rest), fever, fatigue, bloating, vomiting and weight loss (>3 kg in the last 3 months). These symptoms were assessed both at initial onset and at formal diagnosis. Consultations that occurred in the ED were also collected. Symptoms consistent with the Rome III criteria for IBS (IBS-D: diarrhoea-predominant; IBS-C: constipation-predominant; and IBS-M: mixed), at the onset of symptoms and at formal diagnosis, were analysed  .
We then collected information on the diagnostic examinations performed prior to a formal diagnosis of CD including assessment of faecal calprotectin, intestinal ultrasound, magnetic resonance imaging (MRI), computed tomography (CT) and ileocolonoscopy. For completion, inconclusive endoscopic examinations were recorded; these were defined as colonoscopies without ileoscopies or ileocolonoscopies without biopsies. Furthermore, we analysed the diagnostic delay in patients who underwent faecal calprotectin assessment, CT scan, MRI or intestinal ultrasound prior to the diagnostic endoscopies, irrespective of their result.
2.2. Statistical analysis
Quantitative data are presented as either mean ± standard deviation (for parametric data) or median and range plus interquartile range (IQR, for nonparametric data).
Diagnostic delay was stratified into three groups according to the tertiles of its distribution, and the differences in quantitative data distribution groups were compared with the analysis of variance (parametric data) and the Kruskal–Wallis test (nonparametric data). The Bonferroni correction for multiple testing was used for one-by-one differences amongst groups. The differences in the frequencies for categorical data were compared using the χ2 test.
Moreover, the diagnostic delay according to each clinical variable was compared using Student's t-test or the Mann–Whitney U test. A p-value of <0.05 was considered to be statistically significant.
The demographic and clinical features of the patients are shown in Table 1 .
|Male, n (%)||41 (49.4)|
|Age at first symptoms (years), mean ± SD||35.2 ± 14.5|
|Median, IQR, range||(31; 24-45; 13-74)|
|Age at diagnosis (years), mean ± SD||37.2 ± 15.3|
|Median, IQR, range||(34; 25-47; 14-74)|
|Age at diagnosis, n (%)|
|<40 years old||48 (57.8)|
|>40 years old||35 (42.2)|
|Age at enrolment (years), mean ± SD||38.5 ± 15.3|
|Median, IQR, range||(35; 26-48; 14-76)|
|Disease location, n (%)|
|Upper gastrointestinal tract||7 (8.4)|
|Disease duration since first symptoms (months), mean ± SD||24.5 ± 49.7|
|Median, IQR, range||(8; 3-27; 0-324)|
|Disease behaviour at diagnosis, n (%)|
3.1. Delay and consultations
The median diagnostic delay was 8 months (IRQ, range: 3–27, 0–324).
Twenty-six patients did not consult their GP until after formal diagnosis, 18 of whom visited the ED for acute symptoms and eight of whom were directly examined by a gastroenterologist. Fifty-seven patients consulted their GP: four were directly investigated by colonoscopy with a subsequent definite diagnosis of CD, whilst 53 were referred to a gastroenterologist. The time intervals between the onset of symptoms and diagnostic delays in these patients, as well as those who did not consult their GP, are shown in Table 2 .
|Time (months)||Patients with GP consultation (n = 57)||Patients without GP consultation (n = 26)||p value|
|From onset of symptoms to GP consultation, Mean ± SD||2.66 ± 3.95||–|
|Median; IQR, range||1.0; 0.5–2, 0.5–24|
|From GP consultation to specialist referral, Mean ± SD||10.54 ± 18.80||–|
|Median; IQR, range||3.0; 1.5–8, 0.5–80|
|From emergency department or specialist referral to diagnosis, Mean ± SD||14.52 ± 38.99||4.11 ± 5.44|
|Median; IQR, range||3.0; 1–10, 0–272||1.0; 0–5, 0–16||0.181|
|From onset of symptom to diagnosis, Mean ± SD||32.89 ± 57.92||6.24 ± 8.73|
|Median; IQR, range||14.2; 5–38.5, 1–324||1.5; 1–7.5, 0–30||<0.001|
|Emergency room visits; n (%)||26 (45.6)||18 (69.2)||0.078|
GP, general practitioner; SD, standard deviation; IQR, inter-quartile range.
Patients without a prior GP consultation, compared to those who had, showed a significantly shorter diagnostic delay (median, IQR 1.5, 1–7.5 vs. 14, 5–38.5 months; p < 0.0001) and a higher rate of consultation in the ED (69.2% vs. 45.6%; p = 0.078). Both groups of patients exhibited a comparable prevalence of acute and alarming symptoms at first presentation and at formal diagnosis, but a lower prevalence of diarrhoea (53.8% vs. 80.7%, p = 0.023) at symptom onset (Table S1). However, patients who initially consulted their GP had a short interval between onset of symptoms and initial GP consultation (median; IQR, range: 1, 0.5–2, 0.5–24 months); the time intervals between the GP and gastroenterology consultations and between this latter consultation and final diagnosis were comparable ( Table 2 ).
3.2. Delay and symptoms
The most frequent symptoms reported at onset were diarrhoea (72.3%) with an average of 4.1 bowel movements a day, abdominal pain (69.9%), bloating (33.7%), fatigue (26.5%), weight loss (27.7%) and fever (19.3%). IBS symptoms, as per the Rome III criteria, were reported by 36 patients (43.4%).
Symptoms remained unchanged from the time of onset to formal diagnosis in only 9 patients (10.8%), while 38.5% of patients showed a change in ≥2 out of the 10 initial presenting symptoms – mainly weight loss and fatigue ( Table 3 ). Diarrhoea and abdominal pain worsened in 25.5% and 37.3% of patients, respectively. IBS-like symptoms remained unchanged until formal diagnosis in 20 of 34 patients (58.8%), and these symptoms disappeared in 14 patients (41.2%); three patients developed new-onset IBS symptoms.
|Symptoms||n, %||Diagnostic delay (no of patients)||p value|
|0–4 months (27)||5–18 months (28)||20–324 months (28)|
|At onset||60 (72.3%)||16 (59.2)||21 (75.0)||23 (82.1)||0.153|
|At diagnosis||55 (66.3%)||15 (55.5)||19 (67.8)||21 (75.0)||0.305|
|N. of bowel movements|
|1–2||5 (18.3)||4 (14.3)||4 (14.3)|
|3–6||7 (25.9)||15 (53.6)||18 (64.1)|
|>7||4 (14.8)||2 (7.1)||1 (3.6)||0.177|
|1–2||6 (40.0)||5 (26.3)||2 (9.5)|
|3–6||7 (46.7)||13 (68.4)||12 (57.2)|
|>7||2 (13.3)||1 (5.3)||7 (33.3)||0.060|
|At onset||58 (69.9%)||17 (63.0)||21 (75.0)||20 (71.4)||0.608|
|At diagnosis||50 (60.2%)||11 (40.7)||17 (60.7)||22 (79.6)||0.169|
|Severity of pain|
|0–1||10 (37.0)||13 (46.4)||11 (39.3)|
|2–3||10 (37.0)||11 (39.3)||16 (57.1)|
|4–6||7 (25.9)||4 (14.3)||1 (3.6)||0.159|
|0–1||12 (44.4)||12 (42.9)||8 (28.6)|
|2–3||8 (29.7)||10 (35.7)||11 (39.3)|
|4–6||7 (25.9)||6 (21.4)||9 (32.1)||0.720|
|At onset||16 (19.3%)||7 (25.9)||5 (17.8)||4 (14.3)||0.535|
|At diagnosis||25 (30.2%)||12 (45.5)||5 (17.8)||8 (28.6)||0.097|
|At onset||22 (26.5%)||7 (25.9)||10 (35.7)||5 (17.9)||0.317|
|At diagnosis||37 (44.5%) #||11 (40.7)||13 (46.4)||13 (46.4)||0.888|
|At onset||23 (27.7%)||9 (33.3)||6 (21.4)||8 (28.6)||0.610|
|At diagnosis||37 (44.6%) §||13 (48.1)||13 (46.4)||11 (39.3)||0.746|
|At onset||11 (13.2%)||6 (22.2)||3 (10.7)||2 (7.1)||0.228|
|At diagnosis||17 (20.5%)||5 (18.5)||6 (21.4)||6 (21.4)||0.954|
|At onset||14 (18.1%)||6 (22.2)||5 (17.9)||3 (10.7)||0.515|
|At diagnosis||15 (16.9%)||5 (18.5)||3 (10.7)||7 (25.0)||0.380|
|At onset||28 (33.7%)||4 (14.8)||11 (39.3)||13 (46.4)||0.035|
|At diagnosis||32 (38.5%)||7 (25.9)||11 (39.3)||14 (50.0)||0.185|
|At onset||34 (41.0%)||8 (29.6)||13 (46.4)||13 (46.4)||0.345|
|IBS-D||5 (62.5)||11 (84.6)||8 (61.5)|
|IBS non D||3 (37.5)||2 (15.4)||5 (38.5)||0.368|
|At diagnosis||23 (27.7%)||4 (14.8)||11 (39.3)||8 (28.6)||0.127|
|IBS-D||4 (100)||9 (81.8)||5 (62.5)|
|IBS non D||0 (0)||2 (18.2)||3 (37.5)||0.307|
# p: 0.023 between onset and at diagnosis.
§ p: 0.036 between onset and at diagnosis.
IBS, irritable bowel syndrome.
Patients complaining of bloating at the onset were shown to have a significantly greater diagnostic delay ( Table 3 ). No other specific symptoms were significantly associated with a delay in diagnosis.
The median (IQR) diagnostic delay in patients with and without diarrhoea (9.5 (3.25–34.75) vs. 5 (1–16) months; p = 0.177), abdominal pain (median, IQR: 9, 3–26 vs. 7, 2–29 months; p = 0.796), fever (median, IQR: 6, 3–22 vs. 9, 3–28 months; p = 0.255) and IBS symptoms (median, IQR: 7.5, 4.75–41.25 vs. 9, 1.5–24 months; p = 0.442) did not significantly differ. However, patients presenting with rectal bleeding had a shorted diagnostic delay (median, IQR: 3, 1–6 vs. 9.5, 3–27.75 months; p = 0.055), and patients complaining of bloating had a significantly longer delay compared with those who did not (median, IQR: 16.5, 6–45.25 vs. 6, 2–24 months; p = 0.016).
Diagnostic delay showed a significant correlation with age at diagnosis and stricturing CD phenotype ( Table 4 ).
|Diagnostic delay||p value|
|0–4 months||5–18 months||19–324 months|
|Mean ± SD (Median)||1.8 ± 1.1 (1.0)||9.2 ± 4 (8.1)||61.9 ± 72.6 (38)|
|Number of patients||27||28||28|
|Male n (%)||14 (50)||15 (53.6)||12 (42.8)||0.691|
|Age at diagnosis (years), Mean ± SD||34.6 ± 11.4||33.5 ± 16.9||43.4 ± 15.5|
|(Median; range)||(33; 15–57)||(26.4; 19–63)||(43.8; 18–74)||0.030|
|Disease location, n (%)|
|Ileal||11 (40.8)||11 (39.3)||19 (67.8)|
|Colic||3 (11.1)||6 (21.4)||1 (3.6)|
|Ileocolic||13 (48.1)||11 (39.3)||8 (28.6)||0.09|
|Upper gastrointestinal tract||1 (14.3)||2 (28.6)||4 (57.1)||0.223|
|Disease behaviour at diagnosis, n (%)|
|Inflammatory||20 (74)||20 (71.4)||17 (60.7)|
|Stenosing||2 (7.4)||7 (25.0)||10 (35.7)|
|Fistulizing||5 (18.6)||1 (3.6)||1 (3.6)||0.038|
SD, standard deviation.
3.3. Delay and previous investigations
Prior to diagnosis, 45 patients were subject to one or more investigations, namely MRI (11 patients), CT (20 patients) intestinal ultrasound (IUS; 22 patients), faecal calprotectin (19 patients) or video capsule endoscopy (3 patients).
In 38 patients, the diagnosis was made solely with endoscopy ( Table 5 ). However, 26 patients were deemed to have had inconclusive endoscopic tests prior to diagnosis: no ileoscopy was performed in 19 patients (22.9%) or no colonic and/or ileal biopsies were taken in 7 patients (8.4%). Both of these situations were associated with a significant diagnostic delay ( Table 5 ). In particular, compared with patients who had undergone a complete ileocolonoscopy with biopsies (median (IQR) delay of 6 (2–15.25) months), patients without biopsies at ileocolonoscopy had a median time delay of 24 (15–28) months (p = 0.008) and a median (IQR) delay of 24 (5–49) months (p = 0.025) in incomplete ileocolonoscopy.
|Diagnostic examination (n, %)||Diagnostic delay||p value|
|0–4 Months (27)||5–18 Months (28)||20–324 Months (28)|
|Magnetic resonance imaging||3 (11.1)||4 (14.3)||4 (14.3)||0.923|
|Computed tomography||7 (25.9)||3 (10.7)||10 (35.7)||0.088|
|Intestinal ultrasound||4 (14.8)||10 (35.7)||8 (28.6)||0.204|
|Faecal calprotectin||3 (11.1)||9 (32.1)||7 (25.0)||0.169|
|Only ileocolonoscopy||17 (60.7)||11 (39.3)||10 (35.7)||0.089|
|Ileocolonoscopy without biopsies||0 (0)||2 (7.1)||5 (17.9)||0.056|
|Incomplete colonoscopy or without ileoscopy||4 (14.8)||4 (14.3)||11 (39.3)||0.025|
No significant time delay was found to be associated with other diagnostic investigations.
Crohn's disease is frequently diagnosed with a variable diagnostic time delay, the reasons for which, however, have not been fully investigated thus far. This study prospectively investigated the reasons for such diagnostic delays in order to identify and rectify them in patient clinical care.
We found that incomplete performance of endoscopic investigations, rather than features and interpretation of symptoms, was a major reason for diagnostic delay.
With respect to symptoms, the presence of abdominal bloating resulted in a statistically significant delay, whilst this was not the case for other IBS symptoms meeting Rome III criteria.
The median delay from symptom onset to definitive diagnosis in all enrolled patients was 8 months. This finding, in view of the fact that it emerges from a prospective study carried out on recently diagnosed (October 2012 to November 2013) incident cases, is in keeping with the remaining literature where the mean duration of symptoms prior to formal diagnosis varied between 6 and 9 months, with a significant proportion having a delay >24 months  and .
In addition, we found a correlation between diagnostic delay and age at diagnosis, with the delay increasing with age. There is conflicting evidence in the literature on this matter. In agreement with our results, Pimentel et al.  observed that the duration of the prodromal period in CD is directly correlated to the age; other studies have also confirmed this finding  and . It has been postulated that this correlation between age at diagnosis and delay in formal diagnosis may result from a different pathogenesis of CD in the young and in the elderly or due to a higher frequency of medical errors in the latter age group. On the contrary, others  have shown that age below 40 years is an independent risk factor for delay due to a more aggressive approach and the use of diagnostic investigations in the elderly to exclude organic pathology. However, this issue remains to be fully answered, and it requires further in-depth investigation.
We demonstrated that pure ileal localization is associated, although not significantly, with the diagnostic delay. This has been well described in the literature  , and it is attributed to a greater difficulty in obtaining the diagnosis with endoscopic investigation alongside a higher likelihood of a more subtle disease at presentation. In fact, we found that localization of CD to the colon led to the development of warning symptoms such as rectal bleeding, and this helped speed up the diagnostic process (median delay 3 vs. 9.6 months; p = 0.055).
We did not find a significant correlation between any other specific symptom at onset (diarrhoea, abdominal pain, fever, fatigue, weight loss and vomiting) and diagnostic delay. This may be explained by the considerable variability in symptoms (only 10.7% of patients had symptoms that remained unchanged from onset to diagnosis), their relative high frequency and their non-specific nature , , and . This is the first study to investigate the association between specific symptoms and CD diagnostic delay.
Our study did not support the hypothesis that IBS symptoms at presentation or at diagnosis adversely affects the time duration to formal diagnosis of CD. It has been stated that patients with likely or possible pre-existing IBS were more likely to experience a longer, although not statistically significant, diagnostic delay of IBD  . Moreover, Pimentel et al.  asserted in their work that the duration of the prodromal period of the disease in CD patients with no IBS symptoms is indeed reduced, although not statistically significant, compared to those CD patients with IBS symptoms; by contrast, Barratt et al. showed that the duration of the prodromal CD symptoms is increased in patients with prodromal IBS symptoms at onset  . Further investigations are required to clarify this aspect. However, despite the absence of a specific correlation between IBS symptoms and diagnostic delay of CD, we found a significant increase in the diagnostic delay in patients complaining of abdominal bloating at presentation (6.1 vs. 16.8 months; p = 0.016). The presence of this symptom, typically related to functional disorders, delays the diagnosis of CD.
Our study showed that diagnostic delay of CD is also related to specific medical care consultations. In particular, the delay was shorter in those patients who had not initially consulted their GP. In this cohort, the median delay was only 1.5 months compared to 14.2 months for patients examined by the GP (p < 0.001). This does not seem to be attributable to the presence of acute or alarming symptoms at presentation. It may suggest that on the one hand there is a lack of attention and awareness on IBD in general practice, and on the other hand the different CD phenotypes may play an important role, for example, some patients experience a (long) prodromal asymptomatic period followed by an acute onset, which would primarily require evaluation in the ED (e.g., for intestinal obstruction). This has been supported by another study  illustrating that a late diagnosis of CD was caused by a delay in consultation between the patient and GP as well as the referral delay between the GP and gastroenterologist. Increased awareness of GPs regarding IBD may reduce the delay in diagnosis of CD by encouraging the necessary investigations and referral promptly  . Moreover, part of the diagnostic delay could be attributed to the burdened health-care system, in particular for patients referred by the GPs, who may wait up to 2 months for a gastroenterologist consultation, colonoscopy or MRI. For most of these patients, in particular for those with severe symptoms, visiting the emergency unit, even in the absence of a true acute symptom, can reduce the wait for clinical or radiographic assessment.
One of the most important findings of our study is the strong impact that incomplete endoscopic examinations may have on the diagnostic delay of CD. In particular, we found that incomplete endoscopic evaluation (see above for definitions) results in a significant delay with a median duration of 24 (no ileoscopy) and 24 months (no biopsies) compared to 6 months for patients with a complete ileocolonoscopy with biopsies. No study has previously assessed the impact of an incomplete examination on diagnostic delay of CD or its association with diagnostic delay of CD.
However, this is not surprising; colonoscopy is an invasive procedure and thus not easily or quickly repeated even in the presence of an incomplete examination. To date, there is no scientific evidence that reports on the impact of the diagnostic instrumental technique on the delay in diagnosis of CD. However, a considerably high rate (up to 50%) of incomplete colonoscopies has been reported in other wide retrospective series , , and . In this regard, a cost-effective analysis assessed that incomplete colonoscopy is the major detrimental factor affecting costs of the diagnostic strategy in patients with CD  . Moreover, incorrect use of double-contrast barium enema and enteroclysis has been shown to contribute to a delay in the diagnosis of CD  .
The limitations of this study are that its population is representative of patients referred to a tertiary centre, and that a recall bias for symptom onset cannot be excluded. In conclusion, our study has shown that the characteristics of the symptoms, with the exception of abdominal bloating, and their evolution are not associated with a delay in the diagnosis of CD. By contrast, incorrect or incomplete endoscopy significantly affects diagnostic delay.
Conflict of interest
-  S.R. Vavricka, S.M. Spigaglia, G. Rogler, et al. Systematic evaluation of risk factors for diagnostic delay in inflammatory bowel disease. Inflammatory Bowel Diseases. 2012;18:496-505 Crossref
-  S.M. Barratt, J.S. Leeds, K. Robinson, et al. Prodromal irritable bowel syndrome may be responsible for delays in diagnosis in patients presenting with unrecognized Crohn's disease and celiac disease, but not ulcerative colitis. Digestive Diseases and Sciences. 2011;56:3270-3275 Crossref
-  G. Maconi, S. Ardizzone, C. Cucino, et al. Pre-illness changes in dietary habits and diet as a risk factor for inflammatory bowel disease: a case–control study. World Journal of Gastroenterology. 2010;16:4297-4304 Crossref
-  T. Burgmann, I. Clara, L. Graff, et al. The Manitoba Inflammatory Bowel Disease Cohort Study: prolonged symptoms before diagnosis – how much is irritable bowel syndrome?. Clinical Gastroenterology and Hepatology. 2006;4:614-620 Crossref
-  M. Pimentel, M. Chang, E.J. Chow, et al. Identification of a prodromal period in Crohn's disease but not ulcerative colitis. American Journal of Gastroenterology. 2000;5:3458-3462 Crossref
-  M.J.L. Romberg-Camps, M.A.M. Hesselink-van de Kruijs, L.J. Schouten, et al. Inflammatory Bowel Disease in South Limburg (the Netherlands) 1991–2002: incidence, diagnostic delay, and seasonal variations in onset of symptoms. Journal of Crohn's and Colitis. 2009;3:115-124 Crossref
-  A.M. Schoepfer, M.A. Dehlavi, N. Fournier, et al. Diagnostic delay in Crohn's disease is associated with a complicated disease course and increased operation rate. American Journal of Gastroenterology. 2013;108:1744-1753 Crossref
-  Rome III Diagnostic Criteria for Functional Gastrointestinal Disorders. Available from www.romecriteria.org .
-  D.M. Foxworthy, J.A. Wilson. Crohn's disease in the elderly. Prolonged delay in diagnosis. Journal of the American Geriatrics Society. 1985;33:492-495
-  M.J. Wagtmans, H.W. Verspaget, C.B. Lamer. Crohn's disease in the elderly: a comparison with young adults. Journal of Clinical Gastroenterology. 1998;27:129-133 Crossref
-  APDW2004 Chinese IBD Working Group. A retrospective analysis of 515 cases of Crohn's disease hospitalization in China: nationwide study from 1990 to 2003. Journal of Gastroenterology and Hepatology. 2006;21:1009-1015 Crossref
-  B.E. Sands. Crohn's disease. M. Feldman, L.S. Friedman, L.J. Brandt (Eds.) Sleisenger and Fordtran's gastrointestinal and liver disease 8th ed. (Saunders, Philadelphia, 2006) 2459-2498
-  J. Belaiche, E. Luis, G. D’Haens, et al. Acute lower gastrointestinal bleeding in Crohn's disease: characteristics of a unique series of 34 patients. Belgian IBD Res Group. American Journal of Gastroenterology. 1999;94:2177-2181 Crossref
-  French Federation of Digestive Oncology.
-  J.S. Brown, W.G. Humphreys, T.G. Parks. Some clinical aspects of Crohn's disease in Northern Ireland: an aid to earlier diagnosis?. Journal of the Royal College of General Practitioners. 1988;38:549-551
-  I.F. Yusoff, D.G. Ormonde, N.E. Hoffman. Routine colonic mucosal biopsy and ileoscopy increases diagnostic yield in patients undergoing colonoscopy for diarrhea. Journal of Gastroenterology and Hepatology. 2002;17:276-280 Crossref
-  H.S. Wijewantha, A.P. de Silva, M.A. Niriella, et al. Usefulness of routine terminal ileoscopy and biopsy during colonoscopy in a tropical setting: a retrospective record-based study. Gastroenterology Research and Practice. 2014;2014:343849 10.1155/2014/343849
-  R. Makkar, R. Lopez, B. Shen. Clinical utility of retrograde terminal ileum intubation in the evaluation of chronic non-bloody diarrhea. Journal of Digestive Diseases. 2013;14:536-542 Crossref
-  B.G. Levesque, L.E. Cipriano, S.L. Chang, et al. Cost effectiveness of alternative imaging strategies for the diagnosis of small-bowel Crohn's disease. Clinical Gastroenterology and Hepatology. 2010;8:261-267
-  R.K. Gedgaudas, F.M. Kelvin, W.M. Thompson, et al. The value of the preoperative barium-enema examination in the assessment of pelvic masses. Radiology. 1983;146:609-613
a Gastroenterology Unit, Department of Biomedical and Clinical Sciences, L. Sacco University Hospital, Milan, Italy
b Department of Gastroenterology, The Alfred Hospital and Monash University, Melbourne, VIC, Australia
⁎ Corresponding author at: Gastroenterology Unit, Department of Biomedical and Clinical Sciences, ‘L. Sacco’ University Hospital, Via G.B. Grassi, 74, 20157 Milan, Italy. Tel.: +39 02 39043164; fax: +39 02 39042232.
© 2015 Editrice Gastroenterologica Italiana S.r.l., Published by Elsevier B.V.