ACCURACY OF THE EOSINOPHILIC ESOPHAGITIS ENDOSCOPIC REFERENCE SCORE IN CHILDREN

To assess the efficacy of applying the endoscopic reference score for EoE (EREFS) in children with symptoms of esophageal dysfunction naïve to proton pump inhibitor (PPI) therapy.

Methods:

An observational cross-sectional study was conducted by reviewing reports and photographs of upper gastrointestinal endoscopies (UGE) and esophageal biopsies of patients with symptoms of esophageal dysfunction. Patients who were treated with PPI or had other conditions that may cause esophageal eosinophilia were excluded.

Results:

Of the 2,036 patients evaluated, endoscopic findings of EoE were identified in 248 (12.2%) and more than one abnormality was observed in 167 (8.2%). Among all patients, 154 (7.6%) presented esophageal eosinophilia (≥15 eosinophils per high power field) (P<0.01). In this group, 30 patients (19.5%) had normal endoscopy. In patients with EoE, edema (74% vs 6.5%, P<0.01) and furrows (66.2% vs 2.4%, P<0.01) were more prevalent than in the control group. Association of edema and furrows was more frequent in patients with EoE than in the control group (29.2% vs 1.6%, P<0.01, OR=24.7, CI=15.0-40.5). The presence of more than one endoscopic finding had sensitivity of 80.5%, specificity of 93.4%, positive predictive value (PPV) of 50%, negative predictive value (NPV) of 98.3%, and accuracy of 92.4%.

Conclusion:

In conclusion, this study showed that endoscopic features suggestive of EoE had high specificity and NPV for diagnosing EoE in children naïve to PPI therapy. These findings highlight the importance of the EREFS in contributing to early identification of inflammatory and fibrostenosing characteristics of EoE, making it possible to identify and to avoid progression of the disease.

Keywords:
Eosinophils; severity of illness index; child; endoscopy gastrointestinal; eosinophilic esophagitis

HIGHLIGHTS

• The EoE endoscopic reference score (EREFS) was developed and validated in adults and has been demonstrated to be an adequate tool for diagnosing and assessing treatment response in children.

• The presence of more than one endoscopic finding stronglysuggests EoE.

• The EoE endoscopic reference score presents high specificity and negative predictive value for diagnosing EoE in children naïve to proton pump inhibitor (PPI) therapy.

• Endoscopic findings suggestive of EoE in patients naïve to treatment may be useful to characterize disease phenotype and individualize treatment according to the initial clinical presentation.

Excluded procedures (n=924)
Endoscopy for control of treatment - EoE or GERD	730
PPI use up to 12 months before endoscopy	34
Esophageal surgery	60
Esophageal atresia (n=57) / Esophageal atresia with gastric transposition (n=2) / Esophageal duplication (n=1)
Complications related to GERD	8
Barrett’s esophagus (n=3) / peptic stricture (n= 5)
Anti-reflux surgery	38
Nissen fundoplication (n=36) / Bianchi surgery (n=2)
Caustic stricture	7
Other causes of esophageal eosinophilia	26
Achalasia (n=3) / Connective tissue disorders (n=2) / Crohn’s disease and unclassified colitis (n=18) / Celiac disease (n=3)
Infectious esophagitis	2
Cytomegalovirus (n=1) / candidiasis (n=1)
Incomplete medical records (endoscopy or histological report)	19

	Eosinophilia
	Absent (n=1,882)	Present (n=154)	P-value*
Female (n=1,010)	961 (51.0%)	49 (31.8%)	<0.01
Male (n=1,026)	921 (49.0%)	105 (68.2%)
Age at diagnosis (in months) **	97.4 (2.3-223.6)	98.8 (4.3-229)	0.72
Endoscopic findings of EoE (n=248)	124 (6.6%)	124 (80.5%)	<0.01
No endoscopic findings of EoE (n=1,788)	1,758 (93.4%)	30 (19.5%)
EREFS ***	0.10±0.42	1.98±1.33	<0.01
Eosinophil count/HPF ***	0.46±1.88	32.63±14.29	<0.01

	Eosinophilia
	Absent (n=1.882)	Present (n=154)	OR	CI95%	P-value
Edema (n=236)	122 (6.5%)	114 (74.0%)	41.1	27.4-61.6	<0.01
Concentric rings (n=8)	0 (0.0%)	8 (5.2%)	90.2	11.0-738.4	<0.01
Whitish exudates <10% (n=63)	20 (1.1%)	43 (27.9%)	36.1	20.5-63.4	<0.01
Whitish exudates >10% (n=19)	2 (0.1%)	17 (11.0%)	116.6	26.7-510.0	<0.01
Vertical lines (n=147)	45 (2.4%)	102 (66.2%)	80.1	51.2-125.1	<0.01
Stricture (n=8)	4 (0.2%)	4 (2.6%)	12.5	3.1-50.5	0.02
Mucosal fragility (n=13)	10 (0.5%)	3 (1.9%)	3.7	1.0-13.6	0.07

	Eosinophilia
	Absent (n=1,882)	Present (n=154)	OR	CI95%	P -value
ED + VL + WE <10% (n=39)	9 (0.5%)	30 (19.5%)	50.3	23.4-108.4	<0.01
ED + VL + WE >10% (n=14)	2 (0.1%)	12 (7.8%)	79.4	17.6-358.4	<0.01
ED + VL (n=76)	31 (1.6%)	45 (29.2%)	24.6	15.0-40.5	<0.01
ED + WE (n=13)	9 (0.5%)	4 (2.6%)	5.5	1.7-18.2	<0.01
Concentric rings + others (n=8)	0 (0.0%)	8 (5.2%)	90.2	11.0-738.4	<0.01
ED + VL + Stricture (n=7)	4 (0.2%)	3 (1.9%)	9.3	2.1-42.0	<0.01
No endoscopic findings of EoE (n=1,788)	1,758 (93.4%)	30 (19.5%)	58.6	37.8-90.8	<0.01

	Sensitivity	Specificity	PPV	NPV	Accuracy
Edema	74.0%	93.5%	48.3%	97.8%	92.0%
Concentric rings	5.2%	100.0%	100.0%	92.8%	92.8%
Whitish exudates	38.9%	98.8%	73.2%	95.2%	94.3%
Vertical lines	66.2%	97.6%	69.4%	97.2%	95.2%
Stricture	2.6%	99.8%	50.0%	92.6%	92.4%
Mucosal fragility	1.9%	99.5%	23.1%	92.5%	92.1%

Instituto Brasileiro de Estudos e Pesquisas de Gastroenterologia e Outras Especialidades - IBEPEGE. Rua Dr. Seng, 320, 01331-020 São Paulo - SP Brasil, Tel./Fax: +55 11 3147-6227 - São Paulo - SP - Brazil
E-mail: secretariaarqgastr@hospitaligesp.com.br

Acompanhe os números deste periódico no seu leitor de RSS

[1] Corresponding author: Luciana Ribeiro. E-mail: gastroped@hpp.org.br