Scandinavian Journal of Gastroenterology. 2014; 49: 1084–1090
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
Capsule endoscopic findings correlate with fecal calprotectin and C-reactive protein in patients with suspected small-bowel Crohn’s disease
CHARLOTTE M. HÖÖG1,2, LARS-ÅKE BARK1, OLLE BROSTRÖM1 & URBAN SJÖQVIST1 1
Department for Clinical Science and Education, Karolinska Institutet, Södersjukhuset, Section for Gastroenterology and Hepatology, Stockholm, Sweden, and 2Centre for Digestive Diseases, Karolinska University Hospital, Huddinge, Sweden
Abstract Objective. Capsule endoscopy (CE) is a sensitive method for detecting inflammatory lesions in the small bowel. Such lesions may be due to Crohn’s disease but also to other causes and a histological diagnosis may be difficult to achieve in the small bowel. The aim of the study was to find a possible correlation between capsule endoscopic findings, biochemical parameters, and symptoms in patients with suspected or known small-bowel Crohns disease. Materials and methods. Thirty patients with inflammatory lesions in the small bowel diagnosed by CE were included. CE findings of inflammation were graded using the Lewis score. C-reactive protein (CRP) and fecal calprotectin were used as biochemical parameters. Symptoms were graded using the Harvey–Bradshaw index. The patients were followed up after 9 months with a second CE, CRP, fecal calprotectin, and Harvey–Bradshaw index. Results. There was a significant persistent correlation between endoscopic inflammation and fecal calprotectin (p = 0.003 at inclusion and p < 0.001 at follow-up). CRP was correlated to endoscopic inflammation at inclusion (p = 0.006), but not at follow-up. Symptoms were not correlated with endoscopic inflammation. Conclusion. Inflammatory lesions in the small bowel diagnosed by CE in patients with suspected Crohns disease are correlated to fecal calprotectin and CRP, but not to symptoms.
Key Words: capsule endoscopy, Crohns disease, fecal calprotectin, Lewis score
Introduction Capsule endoscopy (CE) has become a common and useful tool for examining the small bowel since its introduction in 2000 [1]. It is the recommended firstline examination tool for investigating a suspected bleeding source [2], and a second-line method for diagnosing and evaluating small-bowel Crohn’s disease (CD) [3]. CD is an inflammatory bowel disease that can occur in all parts of the gastrointestinal tract and typically causes inflammation and ulcers in the mucosa [4]. The small bowel is involved at diagnosis in 45% of patients [5]. CD is a clinical diagnosis that integrates the patient’s history and physical findings with data from imaging and laboratory studies, including
histopathology, and it should be neither confirmed nor excluded on the basis of any one variable or result [4]. CE is a useful tool for examining the small-bowel mucosa, with a higher sensitivity for inflammatory lesions than magnetic resonance imaging (MRI) or computed tomography [6], though there is a small risk of capsule retention in patients with known CD [7]. CE also demonstrates a relatively low specificity [6,8] and an inability to confirm the nature of detected lesions histologically. The introduction of deep enteroscopy via device-assisted enteroscopy (DAE) [9–11] has led to improvements, but DAE is timeand resource consuming [8] and does not always succeed in reaching the deeper part of the small bowel [11]. Moreover, the results of biopsy analysis can be
Correspondence: Charlotte M Höög, MD PhD, Centre for Digestive Diseases, Karolinska University Hospital, K63, 14186 Huddinge, Sweden. E-mail: [email protected]
(Received 10 March 2014; revised 24 April 2014; accepted 28 April 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.920915
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CE findings correlate to FC and CRP inconclusive because of sampling errors, or because of the unpredictable, relapsing–remitting course of the disease [12]. It is therefore of importance to know more about the significance of CE findings suspicious of CD and how they correlate with biochemical parameters as well as subjective symptoms. This study aimed to investigate the correlations between CE findings of inflammation in the small bowel in patients with suspected or known CD, biochemical parameters predictive of CD, and patient symptoms. We also sought to determine if any such correlations persisted over time and during different levels of disease activity.
Methods This prospective study recruited 30 patients with inflammatory lesions in the small bowel, diagnosed by CE, from the Karolinska University Hospital and South Hospital, both in Stockholm, Sweden. Both hospitals are tertiary referral centers providing CE examinations for 1.5 million inhabitants, and performing around 150 and 200 CE examinations per year, respectively. The capsule systems used were PillCamSB2 (Given Imaging, Yoqneam, Israel) at South Hospital, and MiroCam (Intromedic, Seoul, Korea) at Karolinska University Hospital. Patient preparation included a liquid diet the day before CE and a total fast from midnight. After swallowing the capsule, the patients were allowed clear liquid food after 2 h and an ordinary diet after 4 h. All CE examinations at both participating centers since the technique’s introduction in Stockholm 2003 have been performed by only three gastroenterologists, giving them extensive experience of interpreting the results of CEs. The patients had been referred for CE due to a suspicion of small-bowel CD (n = 26) or suspicion of small-bowel activity in patients with known CD (n = 4). Symptoms were mainly diarrhea and abdominal pain but to be accepted for CE in the first place the patients must fulfill the criteria for suspected CD proposed by the International Conference of Capsule Endoscopy [13]. According to these criteria, smallbowel CD should be suspected in patients with at least one gastrointestinal symptom (diarrhea, abdominal pain, weight loss) + one extra intestinal sign/symptom and/or positive biochemical inflammatory marker and/or positive abdominal imaging findings. Clinical features are presented in Table I. Previous work-up contained at a minimum gastroscopy with duodenal biopsies and recent colonoscopy with biopsies that
1085
ruled out ongoing colonic inflammation. Bacterial fecal tests were negative. Small-bowel inflammation was graded by two of the authors using the Lewis score (LS), which has been developed [14] and used [15] for grading inflammatory mucosal lesions detected by CE. The LS gives points for mucosal edema, ulcers, and strictures. Only the most affected tertile of the small bowel is counted. Ulcers are defined as mucosal breaks with white or yellow bases surrounded by red or pink collars. A score of 790 represents moderate to severe inflammation. The inclusion criteria were inflammatory lesions with an LS ‡135 diagnosed by CE, known or suspected CD, and recent colonoscopy that ruled out active inflammatory bowel disease in the colon. The exclusion criteria were use of non-steroidal antiinflammatory agents, active inflammation in another part of the gastrointestinal tract, and known or suspected small-bowel stricture. Patients’ symptoms were graded using the Harvey– Bradshaw index (HBI), which is a validated symptombased scale for grading CD activity [16]. C-reactive protein (CRP) is known to correlate well with inflammatory activity in the bowel in CD [17] and was measured as a blood marker of inflammation. Fecal calprotectin (FC) is a protein released from the granulocytes in the bowel mucosa and the levels correlate with inflammatory activity in CD [18,19]. FC was used as a fecal marker of inflammation and measured using an enzyme-linked immunosorbent assay test (Bühlmann Laboratories, Switzerland). The patients were examined and asked to fill in a questionnaire about their symptoms and medication use. The questions were based on the HBI. Patients were also requested to provide a blood sample for CRP analysis and a stool test for FC. Patients referred for CE by a gastroenterologist were followed up for treatment and/or further investigations as required. Patients referred from primary care were given an appointment with a gastroenterologist at the clinic. Decisions about treatment and/or further investigations, usually DAE or a second ileocolonoscopy for new biopsies, were based on the clinical picture and were not standardized for this study. Treatment included standard drugs used for CD in clinical practice: corticosteroids, 5-aminosalicylic acid, thiopurines, and tumor necrosis factor-a blockers. Watchful waiting, rather than active treatment, was chosen in patients with mild disease activity and uncertain diagnosis. Data about treatment are shown in Table I. All patients, irrespective of their chosen therapy, were followed up after approximately 9 months by
1086
C. M. Höög et al.
Table I. Clinical data, results of CEs, medication during the study period, and final diagnosis.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
Before inclusion
At inclusion
Study period
Follow-up
Symptom at capsule endoscopy referral
LS 0
CRP 0
FC 0
PAD
Treatment for CD
LS 1
CRP 1
FC 1
Final diagnosis
D+P+E D+P+E D+P+E B+P D+P+E D+P+E P+E D+P+E D+P D+P+E D+P P+E D+P D+E D D+P+E P+E D+P+E D+P+E D+P D+P D+P D+P D+E D+P D+P+E D+P B+E D+P B+P
412 3120 3593 143 1968 233 515 675 4128 225 1690 450 303 675 393 458 233 768 225 562 412 790 1690 143 1068 361 450 1350 1690 2764
1 7 13 1 4 1 53 13 15 6 15 1 3 4 1 4 1 39 1 12 3 3 2 6 12 13 1 21 8 3
121 520 524 144 900 117 17 231 1508 115 725 98 50 151 68 67 61 214 1324 124 1200 930 130 423 1145 90 67 689 1598
CD CD CD CD CD CD CD CD CD CD CD CD CD -
Budesonide Budesonide Thiopurin 5-ASA Anti-TNF Budesonide 5-ASA Thiopurin Anti-TNF Thiopurin Steroids Anti-TNF Thiopurin 5-ASA Anti-TNF Steroids 5-ASA Steroids Budesonide Steroids Budesonide Anti-TNF Thiopurin Budesonide Budesonide
337 3664 233 233 0 0 0 900 768 0 225 135 0 225 135 0 1368 225 562 143 562 1217 225 450 112 143 135 1630 225
2 2 9 2 1 1 5 1 14 4 4 1 1 4 1 6 2 147 1 4 4 2 7 6 11 7 1 3 6 2
159 848 480 156 327 15 15 15 160 488 15 40 15 72 15 20 15 271 89 153 15 1073 460 396 444 42 40 191
CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD
Abbreviations: 0 = at inclusion; 1 = at follow-up; D = diarrhea; B = gastrointestinal bleeding; P = abdominal pain; E = extraintestinal symptoms; LS = Lewis score; CRP = C-reactive protein; FC = fecal calprotectin; CD = Crohns disease; PAD = pathologic anatomic diagnosis; 5-ASA = 5-aminosalicylic acid; Anti-TNF = tumor necrosis factor blockers.
repeat CE, and assessments of HBI, CRP, and FC. Patients diagnosed with a narrow stricture at the first CE underwent MRI. Figure 1 shows a flowchart of the study. The sample size of 30 was based on the assumption of a correlation >0.49 with a power of 80%. Spearman’s rank sum test was used to calculate the correlations between LS, CRP, HBI, and FC. This test was chosen because HBI is based partly on nominal data and because of the presence of skewed data with outliers. Means were compared between patients with and without biopsy-verified diagnoses using Mann–Whitney U-tests. The t-test was not applied because of skewed data. The two-tailed significance level was set at 0.05 for all analyses. All calculations were made using SPSS Statistics 20 software (IBM, Somers, NY, USA). The study was approved by the Swedish Committee of Ethics: Etikprövningsnämnden Stockholm (no. 2011-446-31/4).
Results A total of 30 patients were included in this study. Their median age was 37 years (range 19–77) and 12 patients were female. The distribution of smallbowel lesion was extensive in 13 (43%), upper tertile in 1 (3%) and lower tertile in 16 (53%) of the patients. All patients completed the whole study except for one woman who did not receive a second CE because of several narrow strictures detected at the first CE. MRI was performed instead in this patient and showed normal findings despite the known strictures. Medians, means, and ranges for LS, HBI, and CRP and FC levels at inclusion (0) and at follow-up (1) are shown in Table II. LS was correlated with FC both at inclusion (p = 0.003, r = 0.54) and follow-up (p < 0.001, r = 0.83) (Figures 2 and 3). LS was correlated with CRP at inclusion (p = 0.006, r = 0.49) (Figure 4), but not at
CE findings correlate to FC and CRP
follow-up (p = 0.11, r = 0.30). There was no correlation between LS and HBI either at inclusion (p = 0.211, r = 0.24) or follow-up (p = 0.98, r = 0.04). CD was confirmed in 13 out of 30 patients by histopathological diagnosis (Table I). The lesions in the other 17 patients were not possible to reach, not remaining at the time of endoscopy, or biopsy resulted in an inconclusive histological diagnosis. LS and FC levels were significantly higher among patients with a histological diagnosis of CD (p = 0.026 and p = 0.012, respectively) compared with patients without a conclusive histopathological diagnosis. Final diagnosis of small-bowel Crohns disease had been set in 21 of the 30 study patients (including those 4 patients with an earlier diagnosis of CD) at a follow-up 1 year after the closure of the study (Table I). Diagnosis was based on a combination of clinical, endoscopical, histopathological, and biochemical results.
Clinical suspicion of CD in the small bowel
Referral for CE Lewis score > 135 Inclusion in study
Decision of further investigations and/or treatment is made by the patients clinician No treatmentwhatchful waiting
Follow up: CE, CRP, FC, HBI
Discussion Figure 1. Flowchart of the study.
This study demonstrated a correlation between inflammation detected by CE and inflammatory parameters in the blood and feces in patients with Table II. Medians, means, and ranges for Lewis score, Harvey–Bradshaw index, and C-reactive protein and fecal calprotectin at inclusion (0) and at follow-up (1). LS 0 N Mean Median Range
30 1050 538 135–3985
HBI 0
CRP 0
30 5 4 0–12
30 9 4 1–52
FC 0
LS 1
29 461 151 15–1581
29 478 225 0–3664
HBI 1
CRP 1
FC 1
30 4 4 0-13
30 9 4 1–146
28 215 121 15–1058
Abbreviations: LS = Lewis score; HBI = Harvey–Bradshaw index; CRP = C-reactive protein; FC = fecal calprotectin.
R2 Linear = 0.348
2000
1500 Calprotectin
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CRP, FC, HBI
Treatment
1087
1000
500
0 0
1000
2000 3000 Lewis score
4000
5000
Figure 2. Correlation between Lewis score and calprotectin at inclusion.
1088
C. M. Höög et al. R2 Linear = 0.413
1200
Calprotectin
1000 800 600 400
0 0
1000
2000 Lewis score
3000
4000
Figure 3. Correlation between Lewis score and calprotectin at follow-up.
R2 Linear = 0.011
60 50
40 CRP
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
200
30
20 10 0 0
1000
2000 3000 Lewis score
4000
5000
Figure 4. Correlation between Lewis score and C-reactive protein at inclusion.
suspected CD, but no correlation with gastrointestinal symptoms. The study patients had been admitted for CE because of gastrointestinal symptom. Previous examinations, including colonoscopy, had been negative and use of non-steroidal anti-inflammatory drugs had been excluded, thus raising the suspicion of smallbowel CD. CE showed inflammatory lesions in the small bowel, which in combination with clinical symptoms were the inclusion criteria. Biochemical parameters were tested and the patients were followed up with new examinations after 9 months. A histological diagnosis of CD was achieved in approximately half of the patients.
A lack of positive biopsy findings in the small bowel is not uncommon in clinical practice, due to several reasons (mentioned in the Results section), and the diagnosis must therefore be based on other parameters or remain unclear. Also, inflammatory lesions are known to be present in up to 13% of healthy individuals [20,21]. It is thus vital to learn more of the clinical implications of inflammatory CE findings. In this study, patients with biopsy-confirmed CD had higher LS and FC levels, possibly reflecting the fact that more lesions indicated a higher possibility of true CD. These findings also suggest that the combination of a high LS and high FC value implicates CD and strengthens the indication for moving further in the
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CE findings correlate to FC and CRP diagnostics, for example, by performing DAE with biopsies for histology, or moving on directly to medical treatment. HBI was not significantly correlated with endoscopic inflammation. This may have been because HBI did not fully reflect the symptoms of strict small-bowel CD, or because some of the patients may have suffered from irritable bowel syndrome, which also commonly coexists with CD [22]. The results are in line with a recent study where no correlation between the endoscopic and clinical picture was found [23]. CD usually follows a relapsing-remitting course, and the endoscopic picture may therefore vary over time to reflect this, or in line with the effects of medication. To confirm the reliability of the data, repeat CE, HBI, CRP, and FC measurements were performed after 9 months follow-up. Correlations between LS and FC but not to CRP remained at follow-up, indicating an association between CE inflammation and FC persisting over time and disease activity. This result supports that in the treatment of patients with small-bowel CD, tests with FC alone will facilitate the evaluation of disease activity. At 1 year follow-up after the closure of the study, 21 of 30 patients had been diagnosed with smallbowel CD. CD is a clinical diagnosis that should integrate all findings [4] and it is our opinion that in clinical practice it is sometimes necessary to follow the course of the disease and the responses to medical treatment before a conclusive diagnosis are made. The rigorous patient follow-up and low drop-out rate represented strengths of this study. The fact that not all patients were biopsied for histopathological diagnosis at the affected part of the small bowel was a limitation. The study do not give an answer to how to interpret all inflammatory lesions shown by CE but may contribute to knowledge of the relations between the shown small-bowel lesions and biochemical parameters in patients with symptoms of an inflammatory bowel disease. In conclusion this study demonstrated a correlation between the severity of inflammation in the small bowel detected by CE and biochemical inflammatory parameters (FC and CRP) in patients with a clinical picture suggestive of small-bowel CD. Correlation to FC remained at follow-up. No correlation between CE findings and symptoms was found any time. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
1089
References [1] Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature 2000;405:417. [2] Fisher L, Lee Krinsky M, Anderson MA, Appalaneni V, Banerjee S, Ben-Menachem T, et al. The role of endoscopy in the management of obscure GI bleeding. Gastrointest Endosc 2010;72:471–9. [3] Bourreille A, Ignjatovic A, Aabakken L, Loftus EV, Eliakim R, Pennazio M, et al. Role of small-bowel endoscopy in the management of patients with inflammatory bowel disease: an international OMED-ECCO consensus. Endoscopy 2009;41:618–37. [4] Baumgart DC, Sandborn WJ. Crohn’s disease. Lancet 2012; 380:1590–605. [5] Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005;19:5–36. [6] Jensen MD, Nathan T, Rafaelsen SR, Kjeldsen J. Diagnostic accuracy of capsule endoscopy for small bowel Crohn’s disease is superior to that of MR enterography or CT enterography. Clin Gastroenterol Hepatol 2011;9:124–9. [7] Höög CM, Bark L, Arkani J, Gorsetman J, Broström O, Sjöqvist U. Capsule retentions and incomplete capsule endoscopy examinations: an analysis of 2300 examinations. Gastroenterol Res Pract 2012;2012:518718. [8] Rondonotti E, Marmo R, Petracchini M, de Franchis R, Pennazio M. The American Society for Gastrointestinal Endoscopy (ASGE) diagnostic algorithm for obscure gastrointestinal bleeding: Eight burning questions from everyday clinical practice. Dig Liver Dis 2013;45:179–85. [9] Yamamoto H, Sugano K. A new method of enteroscopy– the double-balloon method. Can J Gastroenterol 2003;17: 4.273–4. [10] Morgan D, Upchurch B, Draganov P, Binmoeller KF, Haluszka O, Jonnalagadda S, et al. Spiral enteroscopy: prospective U.S. multicenter study in patients with small-bowel disorders. Gastrointest Endosc 2010;72:992–8. [11] May A, Färber M, Aschmoneit I, Pohl J, Manner H, Lotterer E, et al. Prospective multicenter trial comparing push-and-pull enteroscopy with the single- and doubleballoon techniques in patients with small-bowel disorders. Am J Gastroenterol 2010;105:575–81. [12] Girelli CM, Porta P, Malacrida V, Barzaghi F, Rocca F. Clinical outcome of patients examined by capsule endoscopy for suspected small bowel Crohn’s disease. Dig Liver Dis 2007;39:148–54. [13] Mergener K, Ponchon T, Gralnek I, Pennazio M, Gay G, Selby W, et al. Literature review and recommendations for clinical application of small-bowel capsule endoscopy, based on a panel discussion by international experts. Consensus statements for small-bowel capsule endoscopy, 2006/2007. Endoscopy 2007;39:895–909. [14] Gralnek IM, Defranchis R, Seidman E, Leighton JA, Legnani P, Lewis BS. Development of a capsule endoscopy scoring index for small bowel mucosal inflammatory change. Aliment Pharmacol Ther 2008;27:146–54. [15] Rosa B, Moreira MJ, Rebelo A, Cotter J. Lewis Score: a useful clinical tool for patients with suspected Crohn’s Disease submitted to capsule endoscopy. J Crohns Colitis 2012;6:692–7.
1090
C. M. Höög et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
[16] Best WR. Predicting the Crohn’s disease activity index from the Harvey-Bradshaw Index. Inflamm Bowel Dis 2006;12: 304–10. [17] Vermeire S, Van Assche G, Rutgeerts P. C-reactive protein as a marker for inflammatory bowel disease. Inflamm Bowel Dis 2004;10:661–5. [18] Gaya DR, Lyon TD, Duncan A, Neilly JB, Han S, Howell J, et al. Faecal calprotectin in the assessment of Crohn’s disease activity. QJM 2005;98:435–41. [19] Jensen MD, Kjeldsen J, Nathan T. Fecal calprotectin is equally sensitive in Crohn’s disease affecting the small bowel and colon. Scand J Gastroenterol 2011;46:694–700. [20] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Zlotnick S, Fort JG, et al. Video capsule endoscopy to prospectively assess small bowel injury with celecoxib, naproxen plus
omeprazole, and placebo. Clin Gastroenterol Hepatol 2005;3:133–41. [21] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Aisenberg J, Bhadra P, et al. Small bowel mucosal injury is reduced in healthy subjects treated with celecoxib compared with ibuprofen plus omeprazole, as assessed by video capsule endoscopy. Aliment Pharmacol Ther 2007;25:1211–22. [22] Halpin SJ, Ford AC. Prevalence of symptoms meeting criteria for irritable bowel syndrome in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2012;107:1474–82. [23] Yang L, Ge ZZ, Gao YJ, Li XB, Dai J, Zhang Y, et al. Assessment of capsule endoscopy scoring index, clinical disease activity, and C-reactive protein in small bowel Crohn’s disease. J Gastroenterol Hepatol 2013;28:829–33.
ORIGINAL ARTICLE
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
Capsule endoscopic findings correlate with fecal calprotectin and C-reactive protein in patients with suspected small-bowel Crohn’s disease
CHARLOTTE M. HÖÖG1,2, LARS-ÅKE BARK1, OLLE BROSTRÖM1 & URBAN SJÖQVIST1 1
Department for Clinical Science and Education, Karolinska Institutet, Södersjukhuset, Section for Gastroenterology and Hepatology, Stockholm, Sweden, and 2Centre for Digestive Diseases, Karolinska University Hospital, Huddinge, Sweden
Abstract Objective. Capsule endoscopy (CE) is a sensitive method for detecting inflammatory lesions in the small bowel. Such lesions may be due to Crohn’s disease but also to other causes and a histological diagnosis may be difficult to achieve in the small bowel. The aim of the study was to find a possible correlation between capsule endoscopic findings, biochemical parameters, and symptoms in patients with suspected or known small-bowel Crohns disease. Materials and methods. Thirty patients with inflammatory lesions in the small bowel diagnosed by CE were included. CE findings of inflammation were graded using the Lewis score. C-reactive protein (CRP) and fecal calprotectin were used as biochemical parameters. Symptoms were graded using the Harvey–Bradshaw index. The patients were followed up after 9 months with a second CE, CRP, fecal calprotectin, and Harvey–Bradshaw index. Results. There was a significant persistent correlation between endoscopic inflammation and fecal calprotectin (p = 0.003 at inclusion and p < 0.001 at follow-up). CRP was correlated to endoscopic inflammation at inclusion (p = 0.006), but not at follow-up. Symptoms were not correlated with endoscopic inflammation. Conclusion. Inflammatory lesions in the small bowel diagnosed by CE in patients with suspected Crohns disease are correlated to fecal calprotectin and CRP, but not to symptoms.
Key Words: capsule endoscopy, Crohns disease, fecal calprotectin, Lewis score
Introduction Capsule endoscopy (CE) has become a common and useful tool for examining the small bowel since its introduction in 2000 [1]. It is the recommended firstline examination tool for investigating a suspected bleeding source [2], and a second-line method for diagnosing and evaluating small-bowel Crohn’s disease (CD) [3]. CD is an inflammatory bowel disease that can occur in all parts of the gastrointestinal tract and typically causes inflammation and ulcers in the mucosa [4]. The small bowel is involved at diagnosis in 45% of patients [5]. CD is a clinical diagnosis that integrates the patient’s history and physical findings with data from imaging and laboratory studies, including
histopathology, and it should be neither confirmed nor excluded on the basis of any one variable or result [4]. CE is a useful tool for examining the small-bowel mucosa, with a higher sensitivity for inflammatory lesions than magnetic resonance imaging (MRI) or computed tomography [6], though there is a small risk of capsule retention in patients with known CD [7]. CE also demonstrates a relatively low specificity [6,8] and an inability to confirm the nature of detected lesions histologically. The introduction of deep enteroscopy via device-assisted enteroscopy (DAE) [9–11] has led to improvements, but DAE is timeand resource consuming [8] and does not always succeed in reaching the deeper part of the small bowel [11]. Moreover, the results of biopsy analysis can be
Correspondence: Charlotte M Höög, MD PhD, Centre for Digestive Diseases, Karolinska University Hospital, K63, 14186 Huddinge, Sweden. E-mail: [email protected]
(Received 10 March 2014; revised 24 April 2014; accepted 28 April 2014) ISSN 0036-5521 print/ISSN 1502-7708 online 2014 Informa Healthcare DOI: 10.3109/00365521.2014.920915
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CE findings correlate to FC and CRP inconclusive because of sampling errors, or because of the unpredictable, relapsing–remitting course of the disease [12]. It is therefore of importance to know more about the significance of CE findings suspicious of CD and how they correlate with biochemical parameters as well as subjective symptoms. This study aimed to investigate the correlations between CE findings of inflammation in the small bowel in patients with suspected or known CD, biochemical parameters predictive of CD, and patient symptoms. We also sought to determine if any such correlations persisted over time and during different levels of disease activity.
Methods This prospective study recruited 30 patients with inflammatory lesions in the small bowel, diagnosed by CE, from the Karolinska University Hospital and South Hospital, both in Stockholm, Sweden. Both hospitals are tertiary referral centers providing CE examinations for 1.5 million inhabitants, and performing around 150 and 200 CE examinations per year, respectively. The capsule systems used were PillCamSB2 (Given Imaging, Yoqneam, Israel) at South Hospital, and MiroCam (Intromedic, Seoul, Korea) at Karolinska University Hospital. Patient preparation included a liquid diet the day before CE and a total fast from midnight. After swallowing the capsule, the patients were allowed clear liquid food after 2 h and an ordinary diet after 4 h. All CE examinations at both participating centers since the technique’s introduction in Stockholm 2003 have been performed by only three gastroenterologists, giving them extensive experience of interpreting the results of CEs. The patients had been referred for CE due to a suspicion of small-bowel CD (n = 26) or suspicion of small-bowel activity in patients with known CD (n = 4). Symptoms were mainly diarrhea and abdominal pain but to be accepted for CE in the first place the patients must fulfill the criteria for suspected CD proposed by the International Conference of Capsule Endoscopy [13]. According to these criteria, smallbowel CD should be suspected in patients with at least one gastrointestinal symptom (diarrhea, abdominal pain, weight loss) + one extra intestinal sign/symptom and/or positive biochemical inflammatory marker and/or positive abdominal imaging findings. Clinical features are presented in Table I. Previous work-up contained at a minimum gastroscopy with duodenal biopsies and recent colonoscopy with biopsies that
1085
ruled out ongoing colonic inflammation. Bacterial fecal tests were negative. Small-bowel inflammation was graded by two of the authors using the Lewis score (LS), which has been developed [14] and used [15] for grading inflammatory mucosal lesions detected by CE. The LS gives points for mucosal edema, ulcers, and strictures. Only the most affected tertile of the small bowel is counted. Ulcers are defined as mucosal breaks with white or yellow bases surrounded by red or pink collars. A score of 790 represents moderate to severe inflammation. The inclusion criteria were inflammatory lesions with an LS ‡135 diagnosed by CE, known or suspected CD, and recent colonoscopy that ruled out active inflammatory bowel disease in the colon. The exclusion criteria were use of non-steroidal antiinflammatory agents, active inflammation in another part of the gastrointestinal tract, and known or suspected small-bowel stricture. Patients’ symptoms were graded using the Harvey– Bradshaw index (HBI), which is a validated symptombased scale for grading CD activity [16]. C-reactive protein (CRP) is known to correlate well with inflammatory activity in the bowel in CD [17] and was measured as a blood marker of inflammation. Fecal calprotectin (FC) is a protein released from the granulocytes in the bowel mucosa and the levels correlate with inflammatory activity in CD [18,19]. FC was used as a fecal marker of inflammation and measured using an enzyme-linked immunosorbent assay test (Bühlmann Laboratories, Switzerland). The patients were examined and asked to fill in a questionnaire about their symptoms and medication use. The questions were based on the HBI. Patients were also requested to provide a blood sample for CRP analysis and a stool test for FC. Patients referred for CE by a gastroenterologist were followed up for treatment and/or further investigations as required. Patients referred from primary care were given an appointment with a gastroenterologist at the clinic. Decisions about treatment and/or further investigations, usually DAE or a second ileocolonoscopy for new biopsies, were based on the clinical picture and were not standardized for this study. Treatment included standard drugs used for CD in clinical practice: corticosteroids, 5-aminosalicylic acid, thiopurines, and tumor necrosis factor-a blockers. Watchful waiting, rather than active treatment, was chosen in patients with mild disease activity and uncertain diagnosis. Data about treatment are shown in Table I. All patients, irrespective of their chosen therapy, were followed up after approximately 9 months by
1086
C. M. Höög et al.
Table I. Clinical data, results of CEs, medication during the study period, and final diagnosis.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
Before inclusion
At inclusion
Study period
Follow-up
Symptom at capsule endoscopy referral
LS 0
CRP 0
FC 0
PAD
Treatment for CD
LS 1
CRP 1
FC 1
Final diagnosis
D+P+E D+P+E D+P+E B+P D+P+E D+P+E P+E D+P+E D+P D+P+E D+P P+E D+P D+E D D+P+E P+E D+P+E D+P+E D+P D+P D+P D+P D+E D+P D+P+E D+P B+E D+P B+P
412 3120 3593 143 1968 233 515 675 4128 225 1690 450 303 675 393 458 233 768 225 562 412 790 1690 143 1068 361 450 1350 1690 2764
1 7 13 1 4 1 53 13 15 6 15 1 3 4 1 4 1 39 1 12 3 3 2 6 12 13 1 21 8 3
121 520 524 144 900 117 17 231 1508 115 725 98 50 151 68 67 61 214 1324 124 1200 930 130 423 1145 90 67 689 1598
CD CD CD CD CD CD CD CD CD CD CD CD CD -
Budesonide Budesonide Thiopurin 5-ASA Anti-TNF Budesonide 5-ASA Thiopurin Anti-TNF Thiopurin Steroids Anti-TNF Thiopurin 5-ASA Anti-TNF Steroids 5-ASA Steroids Budesonide Steroids Budesonide Anti-TNF Thiopurin Budesonide Budesonide
337 3664 233 233 0 0 0 900 768 0 225 135 0 225 135 0 1368 225 562 143 562 1217 225 450 112 143 135 1630 225
2 2 9 2 1 1 5 1 14 4 4 1 1 4 1 6 2 147 1 4 4 2 7 6 11 7 1 3 6 2
159 848 480 156 327 15 15 15 160 488 15 40 15 72 15 20 15 271 89 153 15 1073 460 396 444 42 40 191
CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD CD
Abbreviations: 0 = at inclusion; 1 = at follow-up; D = diarrhea; B = gastrointestinal bleeding; P = abdominal pain; E = extraintestinal symptoms; LS = Lewis score; CRP = C-reactive protein; FC = fecal calprotectin; CD = Crohns disease; PAD = pathologic anatomic diagnosis; 5-ASA = 5-aminosalicylic acid; Anti-TNF = tumor necrosis factor blockers.
repeat CE, and assessments of HBI, CRP, and FC. Patients diagnosed with a narrow stricture at the first CE underwent MRI. Figure 1 shows a flowchart of the study. The sample size of 30 was based on the assumption of a correlation >0.49 with a power of 80%. Spearman’s rank sum test was used to calculate the correlations between LS, CRP, HBI, and FC. This test was chosen because HBI is based partly on nominal data and because of the presence of skewed data with outliers. Means were compared between patients with and without biopsy-verified diagnoses using Mann–Whitney U-tests. The t-test was not applied because of skewed data. The two-tailed significance level was set at 0.05 for all analyses. All calculations were made using SPSS Statistics 20 software (IBM, Somers, NY, USA). The study was approved by the Swedish Committee of Ethics: Etikprövningsnämnden Stockholm (no. 2011-446-31/4).
Results A total of 30 patients were included in this study. Their median age was 37 years (range 19–77) and 12 patients were female. The distribution of smallbowel lesion was extensive in 13 (43%), upper tertile in 1 (3%) and lower tertile in 16 (53%) of the patients. All patients completed the whole study except for one woman who did not receive a second CE because of several narrow strictures detected at the first CE. MRI was performed instead in this patient and showed normal findings despite the known strictures. Medians, means, and ranges for LS, HBI, and CRP and FC levels at inclusion (0) and at follow-up (1) are shown in Table II. LS was correlated with FC both at inclusion (p = 0.003, r = 0.54) and follow-up (p < 0.001, r = 0.83) (Figures 2 and 3). LS was correlated with CRP at inclusion (p = 0.006, r = 0.49) (Figure 4), but not at
CE findings correlate to FC and CRP
follow-up (p = 0.11, r = 0.30). There was no correlation between LS and HBI either at inclusion (p = 0.211, r = 0.24) or follow-up (p = 0.98, r = 0.04). CD was confirmed in 13 out of 30 patients by histopathological diagnosis (Table I). The lesions in the other 17 patients were not possible to reach, not remaining at the time of endoscopy, or biopsy resulted in an inconclusive histological diagnosis. LS and FC levels were significantly higher among patients with a histological diagnosis of CD (p = 0.026 and p = 0.012, respectively) compared with patients without a conclusive histopathological diagnosis. Final diagnosis of small-bowel Crohns disease had been set in 21 of the 30 study patients (including those 4 patients with an earlier diagnosis of CD) at a follow-up 1 year after the closure of the study (Table I). Diagnosis was based on a combination of clinical, endoscopical, histopathological, and biochemical results.
Clinical suspicion of CD in the small bowel
Referral for CE Lewis score > 135 Inclusion in study
Decision of further investigations and/or treatment is made by the patients clinician No treatmentwhatchful waiting
Follow up: CE, CRP, FC, HBI
Discussion Figure 1. Flowchart of the study.
This study demonstrated a correlation between inflammation detected by CE and inflammatory parameters in the blood and feces in patients with Table II. Medians, means, and ranges for Lewis score, Harvey–Bradshaw index, and C-reactive protein and fecal calprotectin at inclusion (0) and at follow-up (1). LS 0 N Mean Median Range
30 1050 538 135–3985
HBI 0
CRP 0
30 5 4 0–12
30 9 4 1–52
FC 0
LS 1
29 461 151 15–1581
29 478 225 0–3664
HBI 1
CRP 1
FC 1
30 4 4 0-13
30 9 4 1–146
28 215 121 15–1058
Abbreviations: LS = Lewis score; HBI = Harvey–Bradshaw index; CRP = C-reactive protein; FC = fecal calprotectin.
R2 Linear = 0.348
2000
1500 Calprotectin
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CRP, FC, HBI
Treatment
1087
1000
500
0 0
1000
2000 3000 Lewis score
4000
5000
Figure 2. Correlation between Lewis score and calprotectin at inclusion.
1088
C. M. Höög et al. R2 Linear = 0.413
1200
Calprotectin
1000 800 600 400
0 0
1000
2000 Lewis score
3000
4000
Figure 3. Correlation between Lewis score and calprotectin at follow-up.
R2 Linear = 0.011
60 50
40 CRP
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
200
30
20 10 0 0
1000
2000 3000 Lewis score
4000
5000
Figure 4. Correlation between Lewis score and C-reactive protein at inclusion.
suspected CD, but no correlation with gastrointestinal symptoms. The study patients had been admitted for CE because of gastrointestinal symptom. Previous examinations, including colonoscopy, had been negative and use of non-steroidal anti-inflammatory drugs had been excluded, thus raising the suspicion of smallbowel CD. CE showed inflammatory lesions in the small bowel, which in combination with clinical symptoms were the inclusion criteria. Biochemical parameters were tested and the patients were followed up with new examinations after 9 months. A histological diagnosis of CD was achieved in approximately half of the patients.
A lack of positive biopsy findings in the small bowel is not uncommon in clinical practice, due to several reasons (mentioned in the Results section), and the diagnosis must therefore be based on other parameters or remain unclear. Also, inflammatory lesions are known to be present in up to 13% of healthy individuals [20,21]. It is thus vital to learn more of the clinical implications of inflammatory CE findings. In this study, patients with biopsy-confirmed CD had higher LS and FC levels, possibly reflecting the fact that more lesions indicated a higher possibility of true CD. These findings also suggest that the combination of a high LS and high FC value implicates CD and strengthens the indication for moving further in the
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
CE findings correlate to FC and CRP diagnostics, for example, by performing DAE with biopsies for histology, or moving on directly to medical treatment. HBI was not significantly correlated with endoscopic inflammation. This may have been because HBI did not fully reflect the symptoms of strict small-bowel CD, or because some of the patients may have suffered from irritable bowel syndrome, which also commonly coexists with CD [22]. The results are in line with a recent study where no correlation between the endoscopic and clinical picture was found [23]. CD usually follows a relapsing-remitting course, and the endoscopic picture may therefore vary over time to reflect this, or in line with the effects of medication. To confirm the reliability of the data, repeat CE, HBI, CRP, and FC measurements were performed after 9 months follow-up. Correlations between LS and FC but not to CRP remained at follow-up, indicating an association between CE inflammation and FC persisting over time and disease activity. This result supports that in the treatment of patients with small-bowel CD, tests with FC alone will facilitate the evaluation of disease activity. At 1 year follow-up after the closure of the study, 21 of 30 patients had been diagnosed with smallbowel CD. CD is a clinical diagnosis that should integrate all findings [4] and it is our opinion that in clinical practice it is sometimes necessary to follow the course of the disease and the responses to medical treatment before a conclusive diagnosis are made. The rigorous patient follow-up and low drop-out rate represented strengths of this study. The fact that not all patients were biopsied for histopathological diagnosis at the affected part of the small bowel was a limitation. The study do not give an answer to how to interpret all inflammatory lesions shown by CE but may contribute to knowledge of the relations between the shown small-bowel lesions and biochemical parameters in patients with symptoms of an inflammatory bowel disease. In conclusion this study demonstrated a correlation between the severity of inflammation in the small bowel detected by CE and biochemical inflammatory parameters (FC and CRP) in patients with a clinical picture suggestive of small-bowel CD. Correlation to FC remained at follow-up. No correlation between CE findings and symptoms was found any time. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
1089
References [1] Iddan G, Meron G, Glukhovsky A, Swain P. Wireless capsule endoscopy. Nature 2000;405:417. [2] Fisher L, Lee Krinsky M, Anderson MA, Appalaneni V, Banerjee S, Ben-Menachem T, et al. The role of endoscopy in the management of obscure GI bleeding. Gastrointest Endosc 2010;72:471–9. [3] Bourreille A, Ignjatovic A, Aabakken L, Loftus EV, Eliakim R, Pennazio M, et al. Role of small-bowel endoscopy in the management of patients with inflammatory bowel disease: an international OMED-ECCO consensus. Endoscopy 2009;41:618–37. [4] Baumgart DC, Sandborn WJ. Crohn’s disease. Lancet 2012; 380:1590–605. [5] Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, et al. Toward an integrated clinical, molecular and serological classification of inflammatory bowel disease: Report of a Working Party of the 2005 Montreal World Congress of Gastroenterology. Can J Gastroenterol 2005;19:5–36. [6] Jensen MD, Nathan T, Rafaelsen SR, Kjeldsen J. Diagnostic accuracy of capsule endoscopy for small bowel Crohn’s disease is superior to that of MR enterography or CT enterography. Clin Gastroenterol Hepatol 2011;9:124–9. [7] Höög CM, Bark L, Arkani J, Gorsetman J, Broström O, Sjöqvist U. Capsule retentions and incomplete capsule endoscopy examinations: an analysis of 2300 examinations. Gastroenterol Res Pract 2012;2012:518718. [8] Rondonotti E, Marmo R, Petracchini M, de Franchis R, Pennazio M. The American Society for Gastrointestinal Endoscopy (ASGE) diagnostic algorithm for obscure gastrointestinal bleeding: Eight burning questions from everyday clinical practice. Dig Liver Dis 2013;45:179–85. [9] Yamamoto H, Sugano K. A new method of enteroscopy– the double-balloon method. Can J Gastroenterol 2003;17: 4.273–4. [10] Morgan D, Upchurch B, Draganov P, Binmoeller KF, Haluszka O, Jonnalagadda S, et al. Spiral enteroscopy: prospective U.S. multicenter study in patients with small-bowel disorders. Gastrointest Endosc 2010;72:992–8. [11] May A, Färber M, Aschmoneit I, Pohl J, Manner H, Lotterer E, et al. Prospective multicenter trial comparing push-and-pull enteroscopy with the single- and doubleballoon techniques in patients with small-bowel disorders. Am J Gastroenterol 2010;105:575–81. [12] Girelli CM, Porta P, Malacrida V, Barzaghi F, Rocca F. Clinical outcome of patients examined by capsule endoscopy for suspected small bowel Crohn’s disease. Dig Liver Dis 2007;39:148–54. [13] Mergener K, Ponchon T, Gralnek I, Pennazio M, Gay G, Selby W, et al. Literature review and recommendations for clinical application of small-bowel capsule endoscopy, based on a panel discussion by international experts. Consensus statements for small-bowel capsule endoscopy, 2006/2007. Endoscopy 2007;39:895–909. [14] Gralnek IM, Defranchis R, Seidman E, Leighton JA, Legnani P, Lewis BS. Development of a capsule endoscopy scoring index for small bowel mucosal inflammatory change. Aliment Pharmacol Ther 2008;27:146–54. [15] Rosa B, Moreira MJ, Rebelo A, Cotter J. Lewis Score: a useful clinical tool for patients with suspected Crohn’s Disease submitted to capsule endoscopy. J Crohns Colitis 2012;6:692–7.
1090
C. M. Höög et al.
Scand J Gastroenterol Downloaded from informahealthcare.com by University of Utah on 10/15/14 For personal use only.
[16] Best WR. Predicting the Crohn’s disease activity index from the Harvey-Bradshaw Index. Inflamm Bowel Dis 2006;12: 304–10. [17] Vermeire S, Van Assche G, Rutgeerts P. C-reactive protein as a marker for inflammatory bowel disease. Inflamm Bowel Dis 2004;10:661–5. [18] Gaya DR, Lyon TD, Duncan A, Neilly JB, Han S, Howell J, et al. Faecal calprotectin in the assessment of Crohn’s disease activity. QJM 2005;98:435–41. [19] Jensen MD, Kjeldsen J, Nathan T. Fecal calprotectin is equally sensitive in Crohn’s disease affecting the small bowel and colon. Scand J Gastroenterol 2011;46:694–700. [20] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Zlotnick S, Fort JG, et al. Video capsule endoscopy to prospectively assess small bowel injury with celecoxib, naproxen plus
omeprazole, and placebo. Clin Gastroenterol Hepatol 2005;3:133–41. [21] Goldstein JL, Eisen GM, Lewis B, Gralnek IM, Aisenberg J, Bhadra P, et al. Small bowel mucosal injury is reduced in healthy subjects treated with celecoxib compared with ibuprofen plus omeprazole, as assessed by video capsule endoscopy. Aliment Pharmacol Ther 2007;25:1211–22. [22] Halpin SJ, Ford AC. Prevalence of symptoms meeting criteria for irritable bowel syndrome in inflammatory bowel disease: systematic review and meta-analysis. Am J Gastroenterol 2012;107:1474–82. [23] Yang L, Ge ZZ, Gao YJ, Li XB, Dai J, Zhang Y, et al. Assessment of capsule endoscopy scoring index, clinical disease activity, and C-reactive protein in small bowel Crohn’s disease. J Gastroenterol Hepatol 2013;28:829–33.
