ORIGINAL ARTICLE
Accuracy of Advanced Endoscopy and Fecal Calprotectin for Prediction of Relapse in Ulcerative Colitis: A Prospective Study Aranzazu Jauregui-Amezaga, MD,* María López-Cerón, MD, PhD,* Montserrat Aceituno, MD,* Mireya Jimeno, MD,† Cristina Rodríguez de Miguel, RN,* Susana Pinó-Donnay, RN,* Michel Zabalza, MD,* Miquel Sans, MD, PhD,* Elena Ricart, MD, PhD,* Ingrid Ordás, MD,* Begoña González-Suárez, MD, PhD,* Miriam Cuatrecasas, MD, PhD,† Josep Llach, MD, PhD,* Julian Panés, MD, PhD,* and María Pellise, MD, PhD*
Background: The aim of this study was to determine the accuracy of advanced endoscopy for prediction of relapse in ulcerative colitis, in comparison with serum and fecal biomarkers.
Methods: Patients with ulcerative colitis with sustained clinical remission defined as absence of blood in stool for a minimum of 3 months and Mayo endoscopic subscore of 0 were included. High-resolution rectosigmoidoscopy was performed at baseline and at the end of study (week 52 or relapse), assessing mucosal pit pattern by chromoendoscopy and narrow band imaging as well as vascular pattern by narrow band imaging. Histology was evaluated at baseline and at the end of the study. Follow-up for 1 year or until relapse with clinical evaluations and serum and fecal biomarkers every 3 months was established. Relapse was defined as presence of blood in stool and a Mayo endoscopic subscore $1 with histologic confirmation.
Results: Seventeen out of 64 patients (27%) relapsed during the follow-up period. Baseline clinical characteristics in patients who relapsed and those who did not were similar. Neither pit or vascular pattern nor histology was significantly different between relapsers and nonrelapsers. Among serum biomarkers, high platelet count was significantly associated with higher relapse rates. Fecal calprotectin was predictor of relapse within 3- and 12-month period with high specificity but low sensitivity. Conclusions: Advanced endoscopy and histology do not predict relapse over 1-year period in patients with ulcerative colitis. Fecal calprotectin can predict relapse in 3- and 12-month period with low accuracy. (Inflamm Bowel Dis 2014;20:1187–1193) Key Words: ulcerative colitis, chromoendoscopy, narrow band imaging, fecal calprotectin, predictor, relapse
A
growing body of evidence shows that in patients with ulcerative colitis (UC), achieving clinical remission and mucosal
Received for publication February 25, 2014; Accepted April 3, 2014. From the *Department of Gastroenterology, Institut de Malalties Digestives i Metabòliques, Hospital Clínic, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain; and † Department of Pathology, Hospital Clínic de Barcelona, Barcelona, Spain. J. Panés and M. Pellise shared senior authorship. Supported in part by grants from Fondo de Investigaciones Sanitarias (FIS Proyectos de Evaluación de Tecnologías Sanitarias PI07/90174). Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD) is funded by Instituto de Salud Carlos III. Prototype endoscopic equipment was supplied by Olympus Medical Systems, Europe. Fecal calprotectin analysis was supplied by Cerba Internacional, Spain. Guarantor of the article: M. Pellise, MD, PhD. M. López-Cerón was a research fellow from the FIS (Rio Hortega contract). C. Rodríguez de Miguel is a research nurse supported by Olympus Medical Systems, Europe. The remaining authors have no conflicts of interest to disclose. Reprints: María Pellise, MD, PhD, Department of Gastroenterology, Institut de Malalties Digestives i Metabòliques, CIBERehd, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain (e-mail: [email protected]). Copyright © 2014 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1097/MIB.0000000000000069 Published online 28 May 2014.
Inflamm Bowel Dis Volume 20, Number 7, July 2014
healing are associated with improved quality of life and lower relapse rates, hospitalizations, and surgery requirements.1–5 However, subclinical mucosal inflammation is also considered a risk factor for relapse and for the development of tissue damage, dysplasia, and cancer.6–9 Accurate prediction of relapse may be helpful to personalize therapy in patients with UC. New biomarkers have risen as an alternative to endoscopy, and among those, fecal calprotectin (FC) has been studied most widely and appears as one of the most promising.10,11 Assessment of FC levels has been proposed as a noninvasive test for evaluation of intestinal inflammation in patients with inflammatory bowel disease and has also been studied to identify those in remission who are at risk of early relapse, although inter- and intra-assay variability, and lack of established cutoff levels are the major limitations of this biomarker.12–17 Modern endoscopic techniques of “virtual histology” are very useful for detection of subtle mucosal changes not apparent with conventional endoscopy. Indeed, advanced techniques such as magnifying endoscopy and chromoendoscopy (CE), which were initially used for the diagnosis of neoplasia, have also demonstrated to be useful for the assessment of disease activity in patients with UC.18–22 Nevertheless, the www.ibdjournal.org |
1187
Jauregui-Amezaga et al
findings of advanced endoscopic techniques that are associated with active disease or that predict relapse have not been yet clearly defined. Nishio et al22 demonstrated that rectal pit pattern assessed by high-resolution magnifying methylene blue CE in patients with quiescent UC was a predictor of relapse. However, this result remains to be confirmed and compared with the predictor value of biomarkers. Whereas CE applies staining to assess pit pattern characteristics, narrow band imaging (NBI) is a novel technology that enables observing the pit pattern and intramucosal vascular network without the use of dye. The aim of this study was to determine the accuracy of high-resolution CE (HRE-CE) and NBI (HRENBI) for the prediction of relapse in quiescent UC and compare it with serum and fecal biomarkers.
MATERIALS AND METHODS Inclusion and Exclusion Criteria Clinically inactive patients with UC assessed by a partial Mayo score #1 defined as normal stool frequency or 1 to 2 more stools than normal, no rectal bleeding, and normal physician global assessment during at least 3 consecutive months were prospectively recruited from the outpatient clinic of the Gastroenterology Department at the Hospital Clinic of Barcelona (Barcelona, Spain).23 Patients who received induction treatment or dose adjustment including oral or topical steroids (any dose), oral mesalamine at doses of 3 g/day or higher, rectal 5-aminosalicylic acid at doses higher than 3 g/week, or anti-tumor necrosis factor within the previous 3 months were excluded, as well as patients with severe concomitant diseases, pregnant women, or those who denied participating in the study. Subsequently, endoscopic activity was assessed performing a baseline rectosigmoidoscopy, and only patients with a baseline Mayo endoscopic subscore of 0, with normal vascular pattern, no erythema, friability, erosions, or ulcers were included. Mucosal scaring or presence of pseudopolyps was not exclusion criteria. All patients gave their written informed consent to participate in the study after being approved by ethics committee of the hospital. The study was performed according to the
Inflamm Bowel Dis Volume 20, Number 7, July 2014
good clinical practice guidelines of the European Medicines Agency (CPMP/ICH/135/95, July 2002).
High-resolution Rectosigmoidoscopy and Histology High-resolution rectosigmoidoscopy (Olympus H180; Olympus Europe, Hamburg, Germany) was performed at baseline and at the end of follow-up. Bowel preparation was done with 2 rectal sodium phosphate dibasic and sodium phosphate monobasic enemas (Enema Casen 250 mL; Casen Fleet Laboratory, Madrid, Spain). Mucosa was assessed in the rectosigmoid segments in all cases. Colonic mucosa was stained with 0.1% methylene blue after removing the mucus with water and Nacetylcysteine. Mucosal pit pattern was assessed using HRE-CE and HRE-NBI according to Nishio et al22 classification into 4 grades on the basis of size, shape, and arrangement of pits. Because the presence of Nishio grade 4 was considered exclusion criteria, patients were classified into 3 grades in our series as shown in Figure 1. Mucosal vascular pattern intensity was evaluated with HRE-NBI and also classified into 3 grades of intensity (Fig. 1). When characteristics of more than 1 grade were present in the same patient, the highest grade was recorded. Endoscopic characteristics were assessed in situ by 1 experienced endoscopist and post hoc by 2 different endoscopists using video recordings and still pictures. Final pit pattern classification was based on the agreement of at least 2 of the examiners. A consensus grading was achieved in case of total disagreement between the 3 endoscopists. Biopsy samples were obtained from rectal mucosa for histologic analysis. For histologic assessment, severity of inflammatory cell infiltrate and other microscopic changes in the colonic mucosa were assessed according to the classification by Matts24 and Riley et al.6 Matts classification divides mucosal inflammation into 5 grades, according to cellular infiltration, presence of crypt abscesses, erosions or ulcerations. In the classification by Riley et al, 6 histologic features are assessed, 3 of them corresponding to acute inflammation markers (acute inflammatory cell infiltrate, crypt abscesses, and mucin depletion) and other 3 corresponding to chronic inflammation (breached surface epithelium, chronic inflammatory cell infiltrate, and crypt architectural irregularities). Each of them is divided
FIGURE 1. Mucosal pit pattern was assessed by HRE-CE and HRE-NBI according to Nishio and divided in 3 grades. Mucosal vascular pattern was evaluated by HRE-NBI and classified into 3 grades of intensity.
1188
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Advanced Endoscopy and FC
in 4 grades of severity (none, mild, moderate, or severe). Basal plasmacytosis, defined as a dense infiltration of plasma cells in the lower third of the mucosa, and eosinophilic infiltrate in the lamina propria were evaluated, assessing the intensity of the infiltrate in 4 grades (none, mild, moderate, or severe).24,25
Follow-up Patients were followed up by a dedicated nurse blinded to endoscopy findings and biomarker tests every 3 months until completing a 12-month period or until relapse. Clinical activity was assessed using the Mayo score.23 Relapse was defined as presence of macroscopic blood in stool confirmed by a Mayo endoscopic subscore $1 and histologic corroboration. After the standard clinical practice recommendations, stool cultures with Clostridium difficile toxin assay were performed to exclude enteric infection. Classical serum biomarkers (reactive C protein, erythrocyte sedimentation rate, leukocyte counts, platelet counts, albumin, and hemoglobin) and FC were determined at inclusion and every 3 months until the end of the study. FC was measured on frozen fecal sample by quantitative enzyme-linked immunosorbent assay using a commercial kit (Cerba Internacional, Sabadell, Spain).
Statistics This was a pilot study and a formal sample size calculation was not performed. Statistical analysis was performed using the statistical package SPSS 18. Numerical data with normal distribution were analyzed by Student’s t test (or analysis of variance for more than 2 groups), and results are presented as mean and SD. Nonparametric tests (Mann–Whitney or Kruskal–Wallis) were applied in variables with no normal distribution, and results were obtained as median and interquartile range. Correlation between 2 quantitative variables was assessed using Spearman correlation coefficient. Categorical data were compared using chi-square test. Linear regression was used to analyze relations between variables. Interobserver agreement was tested using Kappa coefficient. Test characteristics are given as sensitivity, specificity, positive and negative predictive value, and overall accuracy. The level of significance was set at 0.05 (two-tailed).
RESULTS Patients Eighty-two patients with UC in clinical remission were screened. Twelve were excluded due to endoscopic activity (Mayo endoscopic subscore $ 1), and 70 were included in the study. Of these, 6 patients were lost to follow-up. From the remaining 64 patients who completed the study, 17 (27%) relapsed within the 12-month follow-up period (see flow diagram, Fig. 2). Globally, mean time to relapse was 186 days and the number of patients who relapsed in each trimester was 3, 5, 6, and 3 respectively. Patients’ baseline characteristics including gender, age, disease duration, extent of disease according to Montreal classification,26 treatment at inclusion, and smoking habit were similar in the group of patients who relapsed and those who did not relapse (Table 1).
FIGURE 2. Enrollment and outcomes of patients. 82 UC patients in clinical remission were screened, 12 were excluded due to endoscopic activity and 6 dropped off during the follow up. Only patients with complete follow-up (n=64) were analyzed. 17/64 patients (27%) relapsed within the follow-up period.
None of the patients was receiving topical treatment, 34 (49%) were under stable doses of oral mesalamine (,3 g/d), and 19 (27%) were under immunosuppressant or anti-tumor necrosis factor maintenance treatment. Although it was not a prospectively
TABLE 1. Patients’ Clinical Baseline Characteristics Patient Characteristics (n ¼ 64)
Relapse (n ¼ 17)
No Relapse (n ¼ 47)
Male, no. (%) Age, yra Disease duration, yra Extension, no. (%) Proctitis Left-sided colitis Extensive colitis Smoker Treatment, no. (%) Mesalazine AZA Anti-TNF
8 (47%) 46 6 13.6 9 6 4.3
22 (47%) 46 6 15.9 9 6 7.3
0 (0%) 10 (59%) 7 (41%) 3 (18%)
6 (13%) 20 (42%) 21 (45%) 6 (13%)
0.62
7 (41%) 6 (35%) 0 (0%)
27 (57%) 12 (26%) 1 (2%)
0.25 0.44 0.54
P 0.99 0.9 0.82 0.61
Patients’ clinical baseline characteristics including gender, age, disease duration, extent of disease according to Montreal classification, smoking habit, and treatment at inclusion are collected in this table. Patients’ baseline characteristics were similar in relapsers and nonrelapsers groups. a Expressed as mean 6 SD. Anti-TNF, anti-tumor necrosis factor; AZA, azathioprine.
www.ibdjournal.org |
1189
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Jauregui-Amezaga et al
TABLE 2. Endoscopic Findings Using High-resolution Rectosigmoidoscopy Endoscopic Findings CE pit pattern, n (%) Small and round pits Large and somewhat irregular pits Very irregular pits NBI pit pattern, n (%) Small and round pits Large and somewhat irregular pits Very irregular pits NBI vascular intensity, n (%) Weak Normal Strong
Relapse (n ¼ 17)
No Relapse (n ¼ 47)
7 (41) 9 (53) 1 (6)
24 (51) 19 (40) 4 (9)
12 (71) 3 (17) 2 (12)
33 (70) 13 (28) 1 (2)
2 (12) 12 (70) 3 (18)
2 (4) 38 (81) 7 (15)
P 0.69
0.57
0.72
Rectal pit pattern was assessed by CE and NBI, and rectal vascular pattern was evaluated by NBI. No statistically significant differences were found between relapsers and nonrelapsers.
established selection criteria, for the purpose of interpreting the results of the study, it should be noted that all included patients had previous evidence of endoscopic involvement of the rectal area.
High-resolution Rectosigmoidoscopy and Histology As shown in Table 2, a minority of patients displayed irregular pit pattern with HRE-CE and HRE-NBI. The proportion of patients with large and somewhat irregular pits (grade 2) was higher with HRE-CE than with HRE-NBI. Regarding vascular
pattern, most patients showed a normal vascular pattern with HRE-NBI. Globally, there were no differences in the Nishio classification determined by HRE-CE or HRE-NBI between relapsers and nonrelapsers, and vascular pattern assessed by HRE-NBI was not significantly associated with relapse. Interobserver agreement for the endoscopic features was mild (Kappa ¼ 0.3). Regarding histology, 57 of 64 patients (82% of relapsers and 92% of nonrelapsers) showed Matts grade 3 cellular infiltration in basal rectal biopsies. Results for the Riley classification are shown in Table 3. Overall, baseline histology features were not different between relapsers and nonrelapsers. Basal plasmacytosis was found in 38 (59%) patients but only 7 of these (18%) relapsed, whereas 10 of the 24 (38%) patients without basal plasmacytosis relapsed, so basal plasmacytosis did not appear as a predictor of relapse (P ¼ 0.07). Finally, presence of eosinophils in rectal mucosa was observed in 32 (50%) patients but only 9 (28%) of these relapsed, similarly to the group without eosinophils where 8 of the 32 (25%) relapsed (P ¼ 0.78). Presence of an eosinophil infiltrate was significantly associated with the presence of chronic inflammatory activity (P ¼ 0.03).
Biomarkers Baseline serum and fecal biomarkers in relapsers and nonrelapsers over the 1 year follow-up are shown in Table 4. Among serum biomarkers, only platelet count was significantly associated with relapse at 1 year (P ¼ 0.04) with low specificity (0.7) and sensitivity (0.65). Baseline FC levels were numerically higher in relapsers than in nonrelapsers, although differences were not significant (200 [20–329] mg/g versus 75 [53–191] mg/g, respectively; P ¼ 0.75). However, the proportion of patients who relapsed was significantly higher in the group of patients with FC $250 mg/g (7/14 [50%]) compared with patients with FC ,250 mg/g (10/49 [20%]), odds ratio ¼ 3.9 (95% confidence interval [CI]: 1.1–13.7). The overall accuracy of this cutoff was
TABLE 3. Histologic Assessment According to Riley Classification Acute Inflammatory Activity Acute Inflammatory Cell Infiltrate Patients, % Grades 0 1 2 3
23.4 R 13 2 2 0 P ¼ 0.82
NR 36 11 0 0
Crypt Abscesses 4.7 R 15 1 1 0 P ¼ 0.11
NR 46 1 0 0
Chronic Inflammatory Activity
Mucin Depletion 59.4 R 7 7 2 1 P ¼ 0.72
NR 19 24 4 0
Breached Surface Epithelium 98.4 R 0 14 2 1 P ¼ 0.76
NR 1 38 8 0
Chronic Inflammatory Cell Infiltrate 6.2 R 14 2 1 0 P ¼ 0.024
NR 46 1 0 0
Crypt Architectural Irregularities 75 R NR 4 12 10 31 2 4 1 0 P ¼ 0.51
Riley et al6 evidenced in a previous study a high prevalence of histologic abnormalities in quiescent patients with UC, but association between inflammatory infiltrate severity and frequency of UC relapse was only observed in acute inflammatory infiltrates and not in chronic ones. In our study, neither acute nor chronic inflammatory activity is related to UC relapse. NR, nonrelapsers; R, relapsers.
1190
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Advanced Endoscopy and FC
TABLE 4. Baseline Biomarkers Basal Activity Biomarkers; Normal Ranges CRP (mg/dL); ,1 ESR (mm/hr) Leucocytes (U/L); 4–11 · 109 Platelets (U/L); 130–400 · 109 Albumin (g/L); 30–48 Hemoglobin (g/L); 120–170 FC (mg/g)
Relapse (N ¼ 17); Median (P25–P75) 0.2 8 7000 292,000 48 143 200
(0.1–0.4) (4–10) (5780–7980) (248,000–327,000) (45–50) (136–146) (20–329)
No Relapse (N ¼ 47); Median (P25–75) 0.2 9 6700 251,000 45 139 75
(0.1–0.4) (5–20) (5700–8150) (212,000–292,000) (45–48) (131–153) (53–191)
P 0.81 0.40 0.88 0.04 0.14 0.46 0.13
Among the baseline serum biomarkers analyzed (CRP, ESR, leukocytes, platelets, albumin, hemoglobin) and fecal biomarkers (FC), only platelet count reached statistical significance for prediction of relapse in 1-yr period. CRP, C-reactive protein; ESR, erythrocyte sedimentation rate.
0.73, with a sensitivity, specificity, and negative predictive value of 0.41, 0.85, and 0.80, respectively. When a cutoff level of 100 mg/g was chosen, sensitivity was 0.5 and specificity was 0.7. Using a cutoff level of 150 mg/g, similar results were obtained. When occurrence of relapse over the 3 months after each calprotectin measurement was assessed, FC levels were significantly higher in relapsers than in nonrelapsers (204 mg/g [44–486] versus 65 mg/g [40–123]; P ¼ 0.02). A cutoff level of 100 mg/g (odds ratio ¼ 2.2 [95% CI: 1.1–4.5]) or 250 mg/g (odds ratio ¼ 4.5 [95% CI: 2.1–9.5]) significantly predicted relapse. Global accuracy, sensitivity, specificity, and negative predictive values were 0.64, 0.53, 0.67, and 0.88 respectively, for a cutoff 100 mg/g and 0.78, 0.45, 0.85, and 0.88 for a cutoff 250 mg/g. FC levels were not correlated with endoscopic lesion severity (P ¼ 0.88 for pit pattern by CE, P ¼ 0.32 for pit pattern by NBI, and P ¼ 0.06 for vascular pattern by NBI) or histologic activity assessed according to Riley or Matts scores (P ¼ 0.4 and P ¼ 0.09, respectively).
DISCUSSION The results of this prospective cohort study do not confirm the usefulness of advanced endoscopy or histology for prediction of relapse in patients with UC in remission, by contrast thrombocytosis, and increased FC were significantly associated with higher relapse rates, although accuracy of these predictors is low and the usefulness of the biomarkers in clinical practice has limitations. The capability of advanced endoscopy for the identification of histologic persistence of inflammation and prediction of relapse could be a valuable, simple, and practical tool that could potentially improve management of patients with UC, and a landmark for mucosal healing assessment in the context of clinical trials. Nishio et al22 evaluated pit pattern in rectal mucosa of 113 clinically quiescent patients with UC, using highresolution video-magnifying methylene blue CE for prediction of relapse. A positive correlation was found between magnifying CE grades and histologic inflammation, and the endoscopic
classification was a predictor of relapse. Other studies had demonstrated that CE with and without magnification was more accurate than white light conventional endoscopy for the assessment of extent and degree of mucosal alterations in UC.27 Our prospective study in a homogeneous population of patients with UC in clinical and endoscopic remission, defined as a Mayo endoscopy subscore of 0, shows that HRE-CE findings are not associated with the risk of relapse. Also, HRE-NBI could not detect differences in pit pattern or in vascular pattern intensity between relapsers and nonrelapsers. The discrepancy with the results of Nishio et al22 can be explained by several factors. Probably the most important factor is that the study by Nishio et al includes not only patients in endoscopic remission (Mayo endoscopy subscore of 0) but also those with granularity or erythema at endoscopic examination (Mayo endoscopic subscore of 1). By contrast, very strict inclusion criteria were followed in our study, accepting exclusively patients with Mayo endoscopic subscore of 0, who are those really in endoscopic remission. Indeed, the objective of this study was to determine if mucosal alterations only detectable by advanced endoscopic techniques such as CE or NBI had a predictive value. Actually, previous evidence has demonstrated that minimal endoscopic activity detected by conventional endoscopy, such as a Mayo endoscopic subscore of 1, is associated with higher risk of relapse in comparison with a score of 0.8 Second, patients under immunosuppressant therapy were not included in the study by Nishio et al; these patients may have a lower risk of relapse, despite the presence of endoscopic activity. Finally, Japanese authors used a magnification endoscope that could be technically superior to the one we used, and this could represent a potential drawback to our study. However, magnification endoscopes are expensive and not widely available in western countries, whereas the newly commercialized high-resolution endoscopes represent a reasonable alternative. Additionally, they are supplemented by electronic chromoscopy technology such as NBI that provides the capacity of enhancing mucosal superficial vascular network and as a consequence could be very useful to detect subtle inflammatory changes. www.ibdjournal.org |
1191
Jauregui-Amezaga et al
Regarding the pathology, results from our study showed that, despite complete absence of endoscopic activity, most of patients had persistent histologic inflammatory lesions (89% according to Matts classification). In this study, histologic changes did not appear as a predictor of relapse. In 1991, Riley et al6 evidenced a high prevalence of histologic abnormalities in quiescent patients with UC, but association between inflammatory infiltrate severity and frequency of UC relapse was only observed for acute inflammatory infiltrates and not for chronic changes. More recently, Bitton et al25 reported that basal plasmacytosis was a predictive of relapse in patients with UC in clinical remission. However, in our series, only 18% of patients with basal plasmacytosis relapsed, in contrast with the 38% of patients without basal plasmacytosis. Once again, we consider that the strict selection criteria of our study could be determinant to explain this disagreement. In fact, the above-mentioned studies included patients with mild endoscopic activity, which could account for their positive results. In addition, we only included patients in stable remission and required at least a 3-month period after induction treatment or dose adjustment, whereas in previous studies, patients with shorter periods of remission were included.25 It is remarkable that, among serum biomarkers, high platelet count was a significant predictor of relapse in 1-year period. This is a finding that needs to be confirmed because previous studies did not show an ability of serum biomarkers to predict relapse in UC.28–30 With regard to fecal biomarkers, several studies have shown that FC is a valid biomarker for differentiating active and inactive UC and its levels correlate with endoscopic disease severity.29,31 FC levels have also been studied as predictors of relapse: Costa et al12 found that FC was a strong predictor of clinical relapse in UC and a cutoff level of 150 mg/g had the highest sensitivity and specificity. Nevertheless, this study has some limitations that prevent its results to be reproduced in our setting. First, they included both patients with Crohn’s disease and UC. Second, relapse was diagnosed based only on clinical criteria; presence of lesions was not assessed at inclusion and relapse was not corroborated by endoscopic examination. Our results showed that FC was able to identify patients who would relapse in the next 3 or 12 months with high specificity and negative predictive value. Therefore, FC can be used in clinical practice to select patients who need colonoscopy to confirm the relapse, which would limit the number of invasive procedures and would also result in savings on health resources. Given the high specificity of high FC levels for prediction of relapse, it remains to be seen if a management strategy based solely on FC measures, without endoscopic confirmation of activity, can improve disease course in the long-term and is cost-effective. Although previous studies suggest that FC is correlated with intestinal inflammation and presence of leucocytes in epithelium, we did not find a relation between histologic findings and FC levels.10 The strengths of this study lie on its prospective design, the power, the strict inclusion criteria, and an exhaustive follow-up. Moreover, it is the first study that compares 2 promising and novel strategies such as advanced endoscopy and fecal biomarkers for prediction of relapse in UC.
1192
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Assessment of endoscopic lesions in this study was limited to the rectosigmoid area. Rectal sparing occurs in some patients with UC, particularly those with associated primary sclerosing cholangitis.32 In our cohort, all included patients had evidence of involvement of the rectum in previous endoscopic examinations, including 2 patients having concomitant primary sclerosing cholangitis. Patients receiving treatments for UC could also have rectal sparing. However, this condition appears to be rare since in an ongoing study to assess correspondence between endoscopic activity in the rectosigmoid and more proximal colonic segments, more severe lesions in proximal segments were found in only 2.5% of cases, so its impact on the overall results should be minimal (Julian Panés, MD, PhD, personal communication, March 2014). Another limitation of this study could be the use of a highresolution endoscope without magnification. However, the evaluation of pit pattern and vascular intensity with these high-resolution endoscopes has demonstrated to be highly accurate for the diagnosis of malignancy.33 Moreover, our results suggest that these technical aspects are not a limitation because there were no microscopic architectural changes predictive of relapse. In fact, taking into account the results of this study, we hypothesize that more subtle changes in the mucosa should be studied as potential predictors of long-term outcome. Actually, preliminary results have shown that confocal endomicroscopy could be a useful tool for prediction of relapse in UC.34 In conclusion, high-resolution endoscopy with CE and NBI and histology were not useful for prediction of 1-year relapse in patients with UC in clinical and endoscopic remission. By contrast, FC could predict relapse in 3- and 12-month period with high specificity but low sensitivity. Future studies evaluating biomarkers and new endoscopic technology should be accomplished to progress in the development of a useful and simple tool for the assessment of mucosal changes in patients with UC that may guide management decisions.
ACKNOWLEDGMENTS Author contributions: M. Pellise designed the study; M. Aceituno, J. Panés, M. Sans, E. Ricart, and I. Ordás conducted patient recruitment; M. Lopez-Cerón, M. Aceituno, B. González, J. Llach, C. Rodriguez de Miguel, S. Pino, M. Zabalza, A. Jauregui-Amezaga, and M. Pellise worked on the acquisition of data, endoscopic examinations, and follow-up; M. Jimeno and M. Cuatrecasas conducted the pathological examinations; and A. Jauregui-Amezaga, J. Panés, and M. Pellise were in charge of the analysis and interpretation of data. All the authors helped on the critical revision of the article.
REFERENCES 1. Meucci G, Fasoli R, Saibeni S, et al. Prognostic significance of endoscopic remission in patients with active ulcerative colitis treated with oral and topical mesalazine: a prospective, multicenter study. Inflamm Bowel Dis. 2012;18:1006–1010.
Inflamm Bowel Dis Volume 20, Number 7, July 2014
2. Ardizzone S, Cassinotti A, Duca P, et al. Mucosal healing predicts late outcomes after the first course of corticosteroids for newly diagnosed ulcerative colitis. Clin Gastroenterol Hepatol. 2011;9:483–489.e483. 3. Froslie KF, Jahnsen J, Moum BA, et al. Mucosal healing in inflammatory bowel disease: results from a Norwegian population-based cohort. Gastroenterology. 2007;133:412–422. 4. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology. 2009;137:1250–1260. quiz 1520. 5. Solberg IC, Lygren I, Jahnsen J, et al. Clinical course during the first 10 years of ulcerative colitis: results from a population-based inception cohort (IBSEN Study). Scand J Gastroenterol. 2009;44:431–440. 6. Riley SA, Mani V, Goodman MJ, et al. Microscopic activity in ulcerative colitis: what does it mean? Gut. 1991;32:174–178. 7. Rutter M, Saunders B, Wilkinson K, et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology. 2004;126:451–459. 8. Colombel JF, Rutgeerts P, Reinisch W, et al. Early mucosal healing with infliximab is associated with improved long-term clinical outcomes in ulcerative colitis. Gastroenterology. 2011;141:1194–1201. 9. Gupta RB, Harpaz N, Itzkowitz S, et al. Histologic inflammation is a risk factor for progression to colorectal neoplasia in ulcerative colitis: a cohort study. Gastroenterology. 2007;133:1099–1105. quiz 1340–1091. 10. Roseth AG, Schmidt PN, Fagerhol MK. Correlation between faecal excretion of indium-111-labelled granulocytes and calprotectin, a granulocyte marker protein, in patients with inflammatory bowel disease. Scand J Gastroenterol. 1999;34:50–54. 11. Tibble J, Teahon K, Thjodleifsson B, et al. A simple method for assessing intestinal inflammation in Crohn’s disease. Gut. 2000;47:506–513. 12. Costa F, Mumolo MG, Ceccarelli L, et al. Calprotectin is a stronger predictive marker of relapse in ulcerative colitis than in Crohn’s disease. Gut. 2005;54:364–368. 13. Tibble JA, Sigthorsson G, Bridger S, et al. Surrogate markers of intestinal inflammation are predictive of relapse in patients with inflammatory bowel disease. Gastroenterology. 2000;119:15–22. 14. D’Inca R, Dal Pont E, Di Leo V, et al. Can calprotectin predict relapse risk in inflammatory bowel disease? Am J Gastroenterol. 2008;103:2007–2014. 15. Gisbert JP, Bermejo F, Perez-Calle JL, et al. Fecal calprotectin and lactoferrin for the prediction of inflammatory bowel disease relapse. Inflamm Bowel Dis. 2009;15:1190–1198. 16. Garcia-Sanchez V, Iglesias-Flores E, Gonzalez R, et al. Does fecal calprotectin predict relapse in patients with Crohn’s disease and ulcerative colitis? J Crohns Colitis. 2010;4:144–152. 17. D’Haens G, Ferrante M, Vermeire S, et al. Fecal calprotectin is a surrogate marker for endoscopic lesions in inflammatory bowel disease. Inflamm Bowel Dis. 2012;18:2218–2224. 18. Kudo T, Matsumoto T, Esaki M, et al. Mucosal vascular pattern in ulcerative colitis: observations using narrow band imaging colonoscopy with special reference to histologic inflammation. Int J Colorectal Dis. 2009; 24:495–501.
Advanced Endoscopy and FC
19. Matsumoto T, Kuroki F, Mizuno M, et al. Application of magnifying chromoscopy for the assessment of severity in patients with mild to moderate ulcerative colitis. Gastrointest Endosc. 1997;46:400– 405. 20. Fujiya M, Saitoh Y, Nomura M, et al. Minute findings by magnifying colonoscopy are useful for the evaluation of ulcerative colitis. Gastrointest Endosc. 2002;56:535–542. 21. Esaki M, Kubokura N, Kudo T, et al. Endoscopic findings under narrow band imaging colonoscopy in ulcerative colitis. Dig Endosc. 2011;23 (suppl 1):140–142. 22. Nishio Y, Ando T, Maeda O, et al. Pit patterns in rectal mucosa assessed by magnifying colonoscope are predictive of relapse in patients with quiescent ulcerative colitis. Gut. 2006;55:1768–1773. 23. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med. 1987;317:1625–1629. 24. Matts SG. The value of rectal biopsy in the diagnosis of ulcerative colitis. Q J Med. 1961;30:393–407. 25. Bitton A, Peppercorn MA, Antonioli DA, et al. Clinical, biological, and histologic parameters as predictors of relapse in ulcerative colitis. Gastroenterology. 2001;120:13–20. 26. Satsangi J, Silverberg MS, Vermeire S, et al. The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut. 2006;55:749–753. 27. Kiesslich R, Fritsch J, Holtmann M, et al. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology. 2003;124:880– 888. 28. Lewis JD. The utility of biomarkers in the diagnosis and therapy of inflammatory bowel disease. Gastroenterology. 2011;140:1817–1826. e1812. 29. Schoepfer AM, Beglinger C, Straumann A, et al. Ulcerative colitis: correlation of the Rachmilewitz endoscopic activity index with fecal calprotectin, clinical activity, C-reactive protein, and blood leukocytes. Inflamm Bowel Dis. 2009;15:1851–1858. 30. Osada T, Ohkusa T, Okayasu I, et al. Correlations among total colonoscopic findings, clinical symptoms, and laboratory markers in ulcerative colitis. J Gastroenterol Hepatol. 2008;23(suppl 2):S262–S267. 31. Xiang JY, Ouyang Q, Li GD, et al. Clinical value of fecal calprotectin in determining disease activity of ulcerative colitis. World J Gastroenterol. 2008;14:53–57. 32. Loftus EV Jr, Harewood GC, Loftus CG, et al. PSC-IBD: a unique form of inflammatory bowel disease associated with primary sclerosing cholangitis. Gut. 2005;54:91–96. 33. Rex DK. Narrow-band imaging without optical magnification for histologic analysis of colorectal polyps. Gastroenterology. 2009;136: 1174–1181. 34. Kiesslich R, Duckworth CA, Moussata D, et al. Local barrier dysfunction identified by confocal laser endomicroscopy predicts relapse in inflammatory bowel disease. Gut. 2012;61:1146–1153.
www.ibdjournal.org |
1193
Accuracy of Advanced Endoscopy and Fecal Calprotectin for Prediction of Relapse in Ulcerative Colitis: A Prospective Study Aranzazu Jauregui-Amezaga, MD,* María López-Cerón, MD, PhD,* Montserrat Aceituno, MD,* Mireya Jimeno, MD,† Cristina Rodríguez de Miguel, RN,* Susana Pinó-Donnay, RN,* Michel Zabalza, MD,* Miquel Sans, MD, PhD,* Elena Ricart, MD, PhD,* Ingrid Ordás, MD,* Begoña González-Suárez, MD, PhD,* Miriam Cuatrecasas, MD, PhD,† Josep Llach, MD, PhD,* Julian Panés, MD, PhD,* and María Pellise, MD, PhD*
Background: The aim of this study was to determine the accuracy of advanced endoscopy for prediction of relapse in ulcerative colitis, in comparison with serum and fecal biomarkers.
Methods: Patients with ulcerative colitis with sustained clinical remission defined as absence of blood in stool for a minimum of 3 months and Mayo endoscopic subscore of 0 were included. High-resolution rectosigmoidoscopy was performed at baseline and at the end of study (week 52 or relapse), assessing mucosal pit pattern by chromoendoscopy and narrow band imaging as well as vascular pattern by narrow band imaging. Histology was evaluated at baseline and at the end of the study. Follow-up for 1 year or until relapse with clinical evaluations and serum and fecal biomarkers every 3 months was established. Relapse was defined as presence of blood in stool and a Mayo endoscopic subscore $1 with histologic confirmation.
Results: Seventeen out of 64 patients (27%) relapsed during the follow-up period. Baseline clinical characteristics in patients who relapsed and those who did not were similar. Neither pit or vascular pattern nor histology was significantly different between relapsers and nonrelapsers. Among serum biomarkers, high platelet count was significantly associated with higher relapse rates. Fecal calprotectin was predictor of relapse within 3- and 12-month period with high specificity but low sensitivity. Conclusions: Advanced endoscopy and histology do not predict relapse over 1-year period in patients with ulcerative colitis. Fecal calprotectin can predict relapse in 3- and 12-month period with low accuracy. (Inflamm Bowel Dis 2014;20:1187–1193) Key Words: ulcerative colitis, chromoendoscopy, narrow band imaging, fecal calprotectin, predictor, relapse
A
growing body of evidence shows that in patients with ulcerative colitis (UC), achieving clinical remission and mucosal
Received for publication February 25, 2014; Accepted April 3, 2014. From the *Department of Gastroenterology, Institut de Malalties Digestives i Metabòliques, Hospital Clínic, Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Institut d’Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), University of Barcelona, Barcelona, Spain; and † Department of Pathology, Hospital Clínic de Barcelona, Barcelona, Spain. J. Panés and M. Pellise shared senior authorship. Supported in part by grants from Fondo de Investigaciones Sanitarias (FIS Proyectos de Evaluación de Tecnologías Sanitarias PI07/90174). Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBEREHD) is funded by Instituto de Salud Carlos III. Prototype endoscopic equipment was supplied by Olympus Medical Systems, Europe. Fecal calprotectin analysis was supplied by Cerba Internacional, Spain. Guarantor of the article: M. Pellise, MD, PhD. M. López-Cerón was a research fellow from the FIS (Rio Hortega contract). C. Rodríguez de Miguel is a research nurse supported by Olympus Medical Systems, Europe. The remaining authors have no conflicts of interest to disclose. Reprints: María Pellise, MD, PhD, Department of Gastroenterology, Institut de Malalties Digestives i Metabòliques, CIBERehd, Hospital Clinic, Villarroel 170, Barcelona 08036, Spain (e-mail: [email protected]). Copyright © 2014 Crohn’s & Colitis Foundation of America, Inc. DOI 10.1097/MIB.0000000000000069 Published online 28 May 2014.
Inflamm Bowel Dis Volume 20, Number 7, July 2014
healing are associated with improved quality of life and lower relapse rates, hospitalizations, and surgery requirements.1–5 However, subclinical mucosal inflammation is also considered a risk factor for relapse and for the development of tissue damage, dysplasia, and cancer.6–9 Accurate prediction of relapse may be helpful to personalize therapy in patients with UC. New biomarkers have risen as an alternative to endoscopy, and among those, fecal calprotectin (FC) has been studied most widely and appears as one of the most promising.10,11 Assessment of FC levels has been proposed as a noninvasive test for evaluation of intestinal inflammation in patients with inflammatory bowel disease and has also been studied to identify those in remission who are at risk of early relapse, although inter- and intra-assay variability, and lack of established cutoff levels are the major limitations of this biomarker.12–17 Modern endoscopic techniques of “virtual histology” are very useful for detection of subtle mucosal changes not apparent with conventional endoscopy. Indeed, advanced techniques such as magnifying endoscopy and chromoendoscopy (CE), which were initially used for the diagnosis of neoplasia, have also demonstrated to be useful for the assessment of disease activity in patients with UC.18–22 Nevertheless, the www.ibdjournal.org |
1187
Jauregui-Amezaga et al
findings of advanced endoscopic techniques that are associated with active disease or that predict relapse have not been yet clearly defined. Nishio et al22 demonstrated that rectal pit pattern assessed by high-resolution magnifying methylene blue CE in patients with quiescent UC was a predictor of relapse. However, this result remains to be confirmed and compared with the predictor value of biomarkers. Whereas CE applies staining to assess pit pattern characteristics, narrow band imaging (NBI) is a novel technology that enables observing the pit pattern and intramucosal vascular network without the use of dye. The aim of this study was to determine the accuracy of high-resolution CE (HRE-CE) and NBI (HRENBI) for the prediction of relapse in quiescent UC and compare it with serum and fecal biomarkers.
MATERIALS AND METHODS Inclusion and Exclusion Criteria Clinically inactive patients with UC assessed by a partial Mayo score #1 defined as normal stool frequency or 1 to 2 more stools than normal, no rectal bleeding, and normal physician global assessment during at least 3 consecutive months were prospectively recruited from the outpatient clinic of the Gastroenterology Department at the Hospital Clinic of Barcelona (Barcelona, Spain).23 Patients who received induction treatment or dose adjustment including oral or topical steroids (any dose), oral mesalamine at doses of 3 g/day or higher, rectal 5-aminosalicylic acid at doses higher than 3 g/week, or anti-tumor necrosis factor within the previous 3 months were excluded, as well as patients with severe concomitant diseases, pregnant women, or those who denied participating in the study. Subsequently, endoscopic activity was assessed performing a baseline rectosigmoidoscopy, and only patients with a baseline Mayo endoscopic subscore of 0, with normal vascular pattern, no erythema, friability, erosions, or ulcers were included. Mucosal scaring or presence of pseudopolyps was not exclusion criteria. All patients gave their written informed consent to participate in the study after being approved by ethics committee of the hospital. The study was performed according to the
Inflamm Bowel Dis Volume 20, Number 7, July 2014
good clinical practice guidelines of the European Medicines Agency (CPMP/ICH/135/95, July 2002).
High-resolution Rectosigmoidoscopy and Histology High-resolution rectosigmoidoscopy (Olympus H180; Olympus Europe, Hamburg, Germany) was performed at baseline and at the end of follow-up. Bowel preparation was done with 2 rectal sodium phosphate dibasic and sodium phosphate monobasic enemas (Enema Casen 250 mL; Casen Fleet Laboratory, Madrid, Spain). Mucosa was assessed in the rectosigmoid segments in all cases. Colonic mucosa was stained with 0.1% methylene blue after removing the mucus with water and Nacetylcysteine. Mucosal pit pattern was assessed using HRE-CE and HRE-NBI according to Nishio et al22 classification into 4 grades on the basis of size, shape, and arrangement of pits. Because the presence of Nishio grade 4 was considered exclusion criteria, patients were classified into 3 grades in our series as shown in Figure 1. Mucosal vascular pattern intensity was evaluated with HRE-NBI and also classified into 3 grades of intensity (Fig. 1). When characteristics of more than 1 grade were present in the same patient, the highest grade was recorded. Endoscopic characteristics were assessed in situ by 1 experienced endoscopist and post hoc by 2 different endoscopists using video recordings and still pictures. Final pit pattern classification was based on the agreement of at least 2 of the examiners. A consensus grading was achieved in case of total disagreement between the 3 endoscopists. Biopsy samples were obtained from rectal mucosa for histologic analysis. For histologic assessment, severity of inflammatory cell infiltrate and other microscopic changes in the colonic mucosa were assessed according to the classification by Matts24 and Riley et al.6 Matts classification divides mucosal inflammation into 5 grades, according to cellular infiltration, presence of crypt abscesses, erosions or ulcerations. In the classification by Riley et al, 6 histologic features are assessed, 3 of them corresponding to acute inflammation markers (acute inflammatory cell infiltrate, crypt abscesses, and mucin depletion) and other 3 corresponding to chronic inflammation (breached surface epithelium, chronic inflammatory cell infiltrate, and crypt architectural irregularities). Each of them is divided
FIGURE 1. Mucosal pit pattern was assessed by HRE-CE and HRE-NBI according to Nishio and divided in 3 grades. Mucosal vascular pattern was evaluated by HRE-NBI and classified into 3 grades of intensity.
1188
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Advanced Endoscopy and FC
in 4 grades of severity (none, mild, moderate, or severe). Basal plasmacytosis, defined as a dense infiltration of plasma cells in the lower third of the mucosa, and eosinophilic infiltrate in the lamina propria were evaluated, assessing the intensity of the infiltrate in 4 grades (none, mild, moderate, or severe).24,25
Follow-up Patients were followed up by a dedicated nurse blinded to endoscopy findings and biomarker tests every 3 months until completing a 12-month period or until relapse. Clinical activity was assessed using the Mayo score.23 Relapse was defined as presence of macroscopic blood in stool confirmed by a Mayo endoscopic subscore $1 and histologic corroboration. After the standard clinical practice recommendations, stool cultures with Clostridium difficile toxin assay were performed to exclude enteric infection. Classical serum biomarkers (reactive C protein, erythrocyte sedimentation rate, leukocyte counts, platelet counts, albumin, and hemoglobin) and FC were determined at inclusion and every 3 months until the end of the study. FC was measured on frozen fecal sample by quantitative enzyme-linked immunosorbent assay using a commercial kit (Cerba Internacional, Sabadell, Spain).
Statistics This was a pilot study and a formal sample size calculation was not performed. Statistical analysis was performed using the statistical package SPSS 18. Numerical data with normal distribution were analyzed by Student’s t test (or analysis of variance for more than 2 groups), and results are presented as mean and SD. Nonparametric tests (Mann–Whitney or Kruskal–Wallis) were applied in variables with no normal distribution, and results were obtained as median and interquartile range. Correlation between 2 quantitative variables was assessed using Spearman correlation coefficient. Categorical data were compared using chi-square test. Linear regression was used to analyze relations between variables. Interobserver agreement was tested using Kappa coefficient. Test characteristics are given as sensitivity, specificity, positive and negative predictive value, and overall accuracy. The level of significance was set at 0.05 (two-tailed).
RESULTS Patients Eighty-two patients with UC in clinical remission were screened. Twelve were excluded due to endoscopic activity (Mayo endoscopic subscore $ 1), and 70 were included in the study. Of these, 6 patients were lost to follow-up. From the remaining 64 patients who completed the study, 17 (27%) relapsed within the 12-month follow-up period (see flow diagram, Fig. 2). Globally, mean time to relapse was 186 days and the number of patients who relapsed in each trimester was 3, 5, 6, and 3 respectively. Patients’ baseline characteristics including gender, age, disease duration, extent of disease according to Montreal classification,26 treatment at inclusion, and smoking habit were similar in the group of patients who relapsed and those who did not relapse (Table 1).
FIGURE 2. Enrollment and outcomes of patients. 82 UC patients in clinical remission were screened, 12 were excluded due to endoscopic activity and 6 dropped off during the follow up. Only patients with complete follow-up (n=64) were analyzed. 17/64 patients (27%) relapsed within the follow-up period.
None of the patients was receiving topical treatment, 34 (49%) were under stable doses of oral mesalamine (,3 g/d), and 19 (27%) were under immunosuppressant or anti-tumor necrosis factor maintenance treatment. Although it was not a prospectively
TABLE 1. Patients’ Clinical Baseline Characteristics Patient Characteristics (n ¼ 64)
Relapse (n ¼ 17)
No Relapse (n ¼ 47)
Male, no. (%) Age, yra Disease duration, yra Extension, no. (%) Proctitis Left-sided colitis Extensive colitis Smoker Treatment, no. (%) Mesalazine AZA Anti-TNF
8 (47%) 46 6 13.6 9 6 4.3
22 (47%) 46 6 15.9 9 6 7.3
0 (0%) 10 (59%) 7 (41%) 3 (18%)
6 (13%) 20 (42%) 21 (45%) 6 (13%)
0.62
7 (41%) 6 (35%) 0 (0%)
27 (57%) 12 (26%) 1 (2%)
0.25 0.44 0.54
P 0.99 0.9 0.82 0.61
Patients’ clinical baseline characteristics including gender, age, disease duration, extent of disease according to Montreal classification, smoking habit, and treatment at inclusion are collected in this table. Patients’ baseline characteristics were similar in relapsers and nonrelapsers groups. a Expressed as mean 6 SD. Anti-TNF, anti-tumor necrosis factor; AZA, azathioprine.
www.ibdjournal.org |
1189
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Jauregui-Amezaga et al
TABLE 2. Endoscopic Findings Using High-resolution Rectosigmoidoscopy Endoscopic Findings CE pit pattern, n (%) Small and round pits Large and somewhat irregular pits Very irregular pits NBI pit pattern, n (%) Small and round pits Large and somewhat irregular pits Very irregular pits NBI vascular intensity, n (%) Weak Normal Strong
Relapse (n ¼ 17)
No Relapse (n ¼ 47)
7 (41) 9 (53) 1 (6)
24 (51) 19 (40) 4 (9)
12 (71) 3 (17) 2 (12)
33 (70) 13 (28) 1 (2)
2 (12) 12 (70) 3 (18)
2 (4) 38 (81) 7 (15)
P 0.69
0.57
0.72
Rectal pit pattern was assessed by CE and NBI, and rectal vascular pattern was evaluated by NBI. No statistically significant differences were found between relapsers and nonrelapsers.
established selection criteria, for the purpose of interpreting the results of the study, it should be noted that all included patients had previous evidence of endoscopic involvement of the rectal area.
High-resolution Rectosigmoidoscopy and Histology As shown in Table 2, a minority of patients displayed irregular pit pattern with HRE-CE and HRE-NBI. The proportion of patients with large and somewhat irregular pits (grade 2) was higher with HRE-CE than with HRE-NBI. Regarding vascular
pattern, most patients showed a normal vascular pattern with HRE-NBI. Globally, there were no differences in the Nishio classification determined by HRE-CE or HRE-NBI between relapsers and nonrelapsers, and vascular pattern assessed by HRE-NBI was not significantly associated with relapse. Interobserver agreement for the endoscopic features was mild (Kappa ¼ 0.3). Regarding histology, 57 of 64 patients (82% of relapsers and 92% of nonrelapsers) showed Matts grade 3 cellular infiltration in basal rectal biopsies. Results for the Riley classification are shown in Table 3. Overall, baseline histology features were not different between relapsers and nonrelapsers. Basal plasmacytosis was found in 38 (59%) patients but only 7 of these (18%) relapsed, whereas 10 of the 24 (38%) patients without basal plasmacytosis relapsed, so basal plasmacytosis did not appear as a predictor of relapse (P ¼ 0.07). Finally, presence of eosinophils in rectal mucosa was observed in 32 (50%) patients but only 9 (28%) of these relapsed, similarly to the group without eosinophils where 8 of the 32 (25%) relapsed (P ¼ 0.78). Presence of an eosinophil infiltrate was significantly associated with the presence of chronic inflammatory activity (P ¼ 0.03).
Biomarkers Baseline serum and fecal biomarkers in relapsers and nonrelapsers over the 1 year follow-up are shown in Table 4. Among serum biomarkers, only platelet count was significantly associated with relapse at 1 year (P ¼ 0.04) with low specificity (0.7) and sensitivity (0.65). Baseline FC levels were numerically higher in relapsers than in nonrelapsers, although differences were not significant (200 [20–329] mg/g versus 75 [53–191] mg/g, respectively; P ¼ 0.75). However, the proportion of patients who relapsed was significantly higher in the group of patients with FC $250 mg/g (7/14 [50%]) compared with patients with FC ,250 mg/g (10/49 [20%]), odds ratio ¼ 3.9 (95% confidence interval [CI]: 1.1–13.7). The overall accuracy of this cutoff was
TABLE 3. Histologic Assessment According to Riley Classification Acute Inflammatory Activity Acute Inflammatory Cell Infiltrate Patients, % Grades 0 1 2 3
23.4 R 13 2 2 0 P ¼ 0.82
NR 36 11 0 0
Crypt Abscesses 4.7 R 15 1 1 0 P ¼ 0.11
NR 46 1 0 0
Chronic Inflammatory Activity
Mucin Depletion 59.4 R 7 7 2 1 P ¼ 0.72
NR 19 24 4 0
Breached Surface Epithelium 98.4 R 0 14 2 1 P ¼ 0.76
NR 1 38 8 0
Chronic Inflammatory Cell Infiltrate 6.2 R 14 2 1 0 P ¼ 0.024
NR 46 1 0 0
Crypt Architectural Irregularities 75 R NR 4 12 10 31 2 4 1 0 P ¼ 0.51
Riley et al6 evidenced in a previous study a high prevalence of histologic abnormalities in quiescent patients with UC, but association between inflammatory infiltrate severity and frequency of UC relapse was only observed in acute inflammatory infiltrates and not in chronic ones. In our study, neither acute nor chronic inflammatory activity is related to UC relapse. NR, nonrelapsers; R, relapsers.
1190
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Advanced Endoscopy and FC
TABLE 4. Baseline Biomarkers Basal Activity Biomarkers; Normal Ranges CRP (mg/dL); ,1 ESR (mm/hr) Leucocytes (U/L); 4–11 · 109 Platelets (U/L); 130–400 · 109 Albumin (g/L); 30–48 Hemoglobin (g/L); 120–170 FC (mg/g)
Relapse (N ¼ 17); Median (P25–P75) 0.2 8 7000 292,000 48 143 200
(0.1–0.4) (4–10) (5780–7980) (248,000–327,000) (45–50) (136–146) (20–329)
No Relapse (N ¼ 47); Median (P25–75) 0.2 9 6700 251,000 45 139 75
(0.1–0.4) (5–20) (5700–8150) (212,000–292,000) (45–48) (131–153) (53–191)
P 0.81 0.40 0.88 0.04 0.14 0.46 0.13
Among the baseline serum biomarkers analyzed (CRP, ESR, leukocytes, platelets, albumin, hemoglobin) and fecal biomarkers (FC), only platelet count reached statistical significance for prediction of relapse in 1-yr period. CRP, C-reactive protein; ESR, erythrocyte sedimentation rate.
0.73, with a sensitivity, specificity, and negative predictive value of 0.41, 0.85, and 0.80, respectively. When a cutoff level of 100 mg/g was chosen, sensitivity was 0.5 and specificity was 0.7. Using a cutoff level of 150 mg/g, similar results were obtained. When occurrence of relapse over the 3 months after each calprotectin measurement was assessed, FC levels were significantly higher in relapsers than in nonrelapsers (204 mg/g [44–486] versus 65 mg/g [40–123]; P ¼ 0.02). A cutoff level of 100 mg/g (odds ratio ¼ 2.2 [95% CI: 1.1–4.5]) or 250 mg/g (odds ratio ¼ 4.5 [95% CI: 2.1–9.5]) significantly predicted relapse. Global accuracy, sensitivity, specificity, and negative predictive values were 0.64, 0.53, 0.67, and 0.88 respectively, for a cutoff 100 mg/g and 0.78, 0.45, 0.85, and 0.88 for a cutoff 250 mg/g. FC levels were not correlated with endoscopic lesion severity (P ¼ 0.88 for pit pattern by CE, P ¼ 0.32 for pit pattern by NBI, and P ¼ 0.06 for vascular pattern by NBI) or histologic activity assessed according to Riley or Matts scores (P ¼ 0.4 and P ¼ 0.09, respectively).
DISCUSSION The results of this prospective cohort study do not confirm the usefulness of advanced endoscopy or histology for prediction of relapse in patients with UC in remission, by contrast thrombocytosis, and increased FC were significantly associated with higher relapse rates, although accuracy of these predictors is low and the usefulness of the biomarkers in clinical practice has limitations. The capability of advanced endoscopy for the identification of histologic persistence of inflammation and prediction of relapse could be a valuable, simple, and practical tool that could potentially improve management of patients with UC, and a landmark for mucosal healing assessment in the context of clinical trials. Nishio et al22 evaluated pit pattern in rectal mucosa of 113 clinically quiescent patients with UC, using highresolution video-magnifying methylene blue CE for prediction of relapse. A positive correlation was found between magnifying CE grades and histologic inflammation, and the endoscopic
classification was a predictor of relapse. Other studies had demonstrated that CE with and without magnification was more accurate than white light conventional endoscopy for the assessment of extent and degree of mucosal alterations in UC.27 Our prospective study in a homogeneous population of patients with UC in clinical and endoscopic remission, defined as a Mayo endoscopy subscore of 0, shows that HRE-CE findings are not associated with the risk of relapse. Also, HRE-NBI could not detect differences in pit pattern or in vascular pattern intensity between relapsers and nonrelapsers. The discrepancy with the results of Nishio et al22 can be explained by several factors. Probably the most important factor is that the study by Nishio et al includes not only patients in endoscopic remission (Mayo endoscopy subscore of 0) but also those with granularity or erythema at endoscopic examination (Mayo endoscopic subscore of 1). By contrast, very strict inclusion criteria were followed in our study, accepting exclusively patients with Mayo endoscopic subscore of 0, who are those really in endoscopic remission. Indeed, the objective of this study was to determine if mucosal alterations only detectable by advanced endoscopic techniques such as CE or NBI had a predictive value. Actually, previous evidence has demonstrated that minimal endoscopic activity detected by conventional endoscopy, such as a Mayo endoscopic subscore of 1, is associated with higher risk of relapse in comparison with a score of 0.8 Second, patients under immunosuppressant therapy were not included in the study by Nishio et al; these patients may have a lower risk of relapse, despite the presence of endoscopic activity. Finally, Japanese authors used a magnification endoscope that could be technically superior to the one we used, and this could represent a potential drawback to our study. However, magnification endoscopes are expensive and not widely available in western countries, whereas the newly commercialized high-resolution endoscopes represent a reasonable alternative. Additionally, they are supplemented by electronic chromoscopy technology such as NBI that provides the capacity of enhancing mucosal superficial vascular network and as a consequence could be very useful to detect subtle inflammatory changes. www.ibdjournal.org |
1191
Jauregui-Amezaga et al
Regarding the pathology, results from our study showed that, despite complete absence of endoscopic activity, most of patients had persistent histologic inflammatory lesions (89% according to Matts classification). In this study, histologic changes did not appear as a predictor of relapse. In 1991, Riley et al6 evidenced a high prevalence of histologic abnormalities in quiescent patients with UC, but association between inflammatory infiltrate severity and frequency of UC relapse was only observed for acute inflammatory infiltrates and not for chronic changes. More recently, Bitton et al25 reported that basal plasmacytosis was a predictive of relapse in patients with UC in clinical remission. However, in our series, only 18% of patients with basal plasmacytosis relapsed, in contrast with the 38% of patients without basal plasmacytosis. Once again, we consider that the strict selection criteria of our study could be determinant to explain this disagreement. In fact, the above-mentioned studies included patients with mild endoscopic activity, which could account for their positive results. In addition, we only included patients in stable remission and required at least a 3-month period after induction treatment or dose adjustment, whereas in previous studies, patients with shorter periods of remission were included.25 It is remarkable that, among serum biomarkers, high platelet count was a significant predictor of relapse in 1-year period. This is a finding that needs to be confirmed because previous studies did not show an ability of serum biomarkers to predict relapse in UC.28–30 With regard to fecal biomarkers, several studies have shown that FC is a valid biomarker for differentiating active and inactive UC and its levels correlate with endoscopic disease severity.29,31 FC levels have also been studied as predictors of relapse: Costa et al12 found that FC was a strong predictor of clinical relapse in UC and a cutoff level of 150 mg/g had the highest sensitivity and specificity. Nevertheless, this study has some limitations that prevent its results to be reproduced in our setting. First, they included both patients with Crohn’s disease and UC. Second, relapse was diagnosed based only on clinical criteria; presence of lesions was not assessed at inclusion and relapse was not corroborated by endoscopic examination. Our results showed that FC was able to identify patients who would relapse in the next 3 or 12 months with high specificity and negative predictive value. Therefore, FC can be used in clinical practice to select patients who need colonoscopy to confirm the relapse, which would limit the number of invasive procedures and would also result in savings on health resources. Given the high specificity of high FC levels for prediction of relapse, it remains to be seen if a management strategy based solely on FC measures, without endoscopic confirmation of activity, can improve disease course in the long-term and is cost-effective. Although previous studies suggest that FC is correlated with intestinal inflammation and presence of leucocytes in epithelium, we did not find a relation between histologic findings and FC levels.10 The strengths of this study lie on its prospective design, the power, the strict inclusion criteria, and an exhaustive follow-up. Moreover, it is the first study that compares 2 promising and novel strategies such as advanced endoscopy and fecal biomarkers for prediction of relapse in UC.
1192
| www.ibdjournal.org
Inflamm Bowel Dis Volume 20, Number 7, July 2014
Assessment of endoscopic lesions in this study was limited to the rectosigmoid area. Rectal sparing occurs in some patients with UC, particularly those with associated primary sclerosing cholangitis.32 In our cohort, all included patients had evidence of involvement of the rectum in previous endoscopic examinations, including 2 patients having concomitant primary sclerosing cholangitis. Patients receiving treatments for UC could also have rectal sparing. However, this condition appears to be rare since in an ongoing study to assess correspondence between endoscopic activity in the rectosigmoid and more proximal colonic segments, more severe lesions in proximal segments were found in only 2.5% of cases, so its impact on the overall results should be minimal (Julian Panés, MD, PhD, personal communication, March 2014). Another limitation of this study could be the use of a highresolution endoscope without magnification. However, the evaluation of pit pattern and vascular intensity with these high-resolution endoscopes has demonstrated to be highly accurate for the diagnosis of malignancy.33 Moreover, our results suggest that these technical aspects are not a limitation because there were no microscopic architectural changes predictive of relapse. In fact, taking into account the results of this study, we hypothesize that more subtle changes in the mucosa should be studied as potential predictors of long-term outcome. Actually, preliminary results have shown that confocal endomicroscopy could be a useful tool for prediction of relapse in UC.34 In conclusion, high-resolution endoscopy with CE and NBI and histology were not useful for prediction of 1-year relapse in patients with UC in clinical and endoscopic remission. By contrast, FC could predict relapse in 3- and 12-month period with high specificity but low sensitivity. Future studies evaluating biomarkers and new endoscopic technology should be accomplished to progress in the development of a useful and simple tool for the assessment of mucosal changes in patients with UC that may guide management decisions.
ACKNOWLEDGMENTS Author contributions: M. Pellise designed the study; M. Aceituno, J. Panés, M. Sans, E. Ricart, and I. Ordás conducted patient recruitment; M. Lopez-Cerón, M. Aceituno, B. González, J. Llach, C. Rodriguez de Miguel, S. Pino, M. Zabalza, A. Jauregui-Amezaga, and M. Pellise worked on the acquisition of data, endoscopic examinations, and follow-up; M. Jimeno and M. Cuatrecasas conducted the pathological examinations; and A. Jauregui-Amezaga, J. Panés, and M. Pellise were in charge of the analysis and interpretation of data. All the authors helped on the critical revision of the article.
REFERENCES 1. Meucci G, Fasoli R, Saibeni S, et al. Prognostic significance of endoscopic remission in patients with active ulcerative colitis treated with oral and topical mesalazine: a prospective, multicenter study. Inflamm Bowel Dis. 2012;18:1006–1010.
Inflamm Bowel Dis Volume 20, Number 7, July 2014
2. Ardizzone S, Cassinotti A, Duca P, et al. Mucosal healing predicts late outcomes after the first course of corticosteroids for newly diagnosed ulcerative colitis. Clin Gastroenterol Hepatol. 2011;9:483–489.e483. 3. Froslie KF, Jahnsen J, Moum BA, et al. Mucosal healing in inflammatory bowel disease: results from a Norwegian population-based cohort. Gastroenterology. 2007;133:412–422. 4. Sandborn WJ, Rutgeerts P, Feagan BG, et al. Colectomy rate comparison after treatment of ulcerative colitis with placebo or infliximab. Gastroenterology. 2009;137:1250–1260. quiz 1520. 5. Solberg IC, Lygren I, Jahnsen J, et al. Clinical course during the first 10 years of ulcerative colitis: results from a population-based inception cohort (IBSEN Study). Scand J Gastroenterol. 2009;44:431–440. 6. Riley SA, Mani V, Goodman MJ, et al. Microscopic activity in ulcerative colitis: what does it mean? Gut. 1991;32:174–178. 7. Rutter M, Saunders B, Wilkinson K, et al. Severity of inflammation is a risk factor for colorectal neoplasia in ulcerative colitis. Gastroenterology. 2004;126:451–459. 8. Colombel JF, Rutgeerts P, Reinisch W, et al. Early mucosal healing with infliximab is associated with improved long-term clinical outcomes in ulcerative colitis. Gastroenterology. 2011;141:1194–1201. 9. Gupta RB, Harpaz N, Itzkowitz S, et al. Histologic inflammation is a risk factor for progression to colorectal neoplasia in ulcerative colitis: a cohort study. Gastroenterology. 2007;133:1099–1105. quiz 1340–1091. 10. Roseth AG, Schmidt PN, Fagerhol MK. Correlation between faecal excretion of indium-111-labelled granulocytes and calprotectin, a granulocyte marker protein, in patients with inflammatory bowel disease. Scand J Gastroenterol. 1999;34:50–54. 11. Tibble J, Teahon K, Thjodleifsson B, et al. A simple method for assessing intestinal inflammation in Crohn’s disease. Gut. 2000;47:506–513. 12. Costa F, Mumolo MG, Ceccarelli L, et al. Calprotectin is a stronger predictive marker of relapse in ulcerative colitis than in Crohn’s disease. Gut. 2005;54:364–368. 13. Tibble JA, Sigthorsson G, Bridger S, et al. Surrogate markers of intestinal inflammation are predictive of relapse in patients with inflammatory bowel disease. Gastroenterology. 2000;119:15–22. 14. D’Inca R, Dal Pont E, Di Leo V, et al. Can calprotectin predict relapse risk in inflammatory bowel disease? Am J Gastroenterol. 2008;103:2007–2014. 15. Gisbert JP, Bermejo F, Perez-Calle JL, et al. Fecal calprotectin and lactoferrin for the prediction of inflammatory bowel disease relapse. Inflamm Bowel Dis. 2009;15:1190–1198. 16. Garcia-Sanchez V, Iglesias-Flores E, Gonzalez R, et al. Does fecal calprotectin predict relapse in patients with Crohn’s disease and ulcerative colitis? J Crohns Colitis. 2010;4:144–152. 17. D’Haens G, Ferrante M, Vermeire S, et al. Fecal calprotectin is a surrogate marker for endoscopic lesions in inflammatory bowel disease. Inflamm Bowel Dis. 2012;18:2218–2224. 18. Kudo T, Matsumoto T, Esaki M, et al. Mucosal vascular pattern in ulcerative colitis: observations using narrow band imaging colonoscopy with special reference to histologic inflammation. Int J Colorectal Dis. 2009; 24:495–501.
Advanced Endoscopy and FC
19. Matsumoto T, Kuroki F, Mizuno M, et al. Application of magnifying chromoscopy for the assessment of severity in patients with mild to moderate ulcerative colitis. Gastrointest Endosc. 1997;46:400– 405. 20. Fujiya M, Saitoh Y, Nomura M, et al. Minute findings by magnifying colonoscopy are useful for the evaluation of ulcerative colitis. Gastrointest Endosc. 2002;56:535–542. 21. Esaki M, Kubokura N, Kudo T, et al. Endoscopic findings under narrow band imaging colonoscopy in ulcerative colitis. Dig Endosc. 2011;23 (suppl 1):140–142. 22. Nishio Y, Ando T, Maeda O, et al. Pit patterns in rectal mucosa assessed by magnifying colonoscope are predictive of relapse in patients with quiescent ulcerative colitis. Gut. 2006;55:1768–1773. 23. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. A randomized study. N Engl J Med. 1987;317:1625–1629. 24. Matts SG. The value of rectal biopsy in the diagnosis of ulcerative colitis. Q J Med. 1961;30:393–407. 25. Bitton A, Peppercorn MA, Antonioli DA, et al. Clinical, biological, and histologic parameters as predictors of relapse in ulcerative colitis. Gastroenterology. 2001;120:13–20. 26. Satsangi J, Silverberg MS, Vermeire S, et al. The Montreal classification of inflammatory bowel disease: controversies, consensus, and implications. Gut. 2006;55:749–753. 27. Kiesslich R, Fritsch J, Holtmann M, et al. Methylene blue-aided chromoendoscopy for the detection of intraepithelial neoplasia and colon cancer in ulcerative colitis. Gastroenterology. 2003;124:880– 888. 28. Lewis JD. The utility of biomarkers in the diagnosis and therapy of inflammatory bowel disease. Gastroenterology. 2011;140:1817–1826. e1812. 29. Schoepfer AM, Beglinger C, Straumann A, et al. Ulcerative colitis: correlation of the Rachmilewitz endoscopic activity index with fecal calprotectin, clinical activity, C-reactive protein, and blood leukocytes. Inflamm Bowel Dis. 2009;15:1851–1858. 30. Osada T, Ohkusa T, Okayasu I, et al. Correlations among total colonoscopic findings, clinical symptoms, and laboratory markers in ulcerative colitis. J Gastroenterol Hepatol. 2008;23(suppl 2):S262–S267. 31. Xiang JY, Ouyang Q, Li GD, et al. Clinical value of fecal calprotectin in determining disease activity of ulcerative colitis. World J Gastroenterol. 2008;14:53–57. 32. Loftus EV Jr, Harewood GC, Loftus CG, et al. PSC-IBD: a unique form of inflammatory bowel disease associated with primary sclerosing cholangitis. Gut. 2005;54:91–96. 33. Rex DK. Narrow-band imaging without optical magnification for histologic analysis of colorectal polyps. Gastroenterology. 2009;136: 1174–1181. 34. Kiesslich R, Duckworth CA, Moussata D, et al. Local barrier dysfunction identified by confocal laser endomicroscopy predicts relapse in inflammatory bowel disease. Gut. 2012;61:1146–1153.
www.ibdjournal.org |
1193
