Efficacy of Proton Pump Inhibitor Therapy for Eosinophilic Oesophagitis in 630 Patients

Results From the EoE Connect Registry

Emilio J. Laserna-Mendieta; Sergio Casabona; Danila Guagnozzi; Edoardo Savarino; Antonia Perelló; Antonio Guardiola-Arévalo; Jesús Barrio; Isabel Pérez-Martínez; Anne Lund Krarup; Javier Alcedo; Susana de la Riva; Esther Rey-Iborra; Cecilio Santander; Ángel Arias; Alfredo J. Lucendo


Aliment Pharmacol Ther. 2020;52(5):798-807. 

In This Article


Study Population

On the search date, January 30, 2020, 842 patients were registered on EoE CONNECT as having demographical data completed. Of those, 630 had PPI as an induction treatment, either as first-line therapy (n = 600) or as second-line therapy after failure of other treatments (n = 30). Among these 630 patients, PPI therapy was subsequently used to maintain remission in 172 of them either by modifying the dose or changing the PPI drug. Maintenance therapy with a reduced dose of PPI was prescribed in 138 of the 172 patients.

Patients were recruited at 13 hospitals in Spain, Italy and Denmark. Table 1 summarises the main demographic and clinical characteristics of the EoE patients included in this study.

PPIs as Initial Therapy to Induce Remission in EoE

First-line treatment including PPIs alone or in combination represented 83.6% of all initial therapies for patients with EoE registered on EoE CONNECT. Six-hundred patients (94.3%) received PPI therapy alone and 25 (3.9%) used it in association with swallowed topic corticosteroids. PPIs were rarely associated with diets or dilation as the initial option to treat EoE (Table S1). Additionally, PPI monotherapy was prescribed as second-line therapy in 30 patients who failed dietary treatments (n = 15), swallowed topic corticosteroids (n = 14) and systemic steroids (n = 1). Subsequent descriptive data and analyses for effectiveness were related exclusively to the 630 patients receiving PPIs alone to induce remission of EoE.

Omeprazole was the most commonly prescribed first-line PPI in patients with EoE (48.4%), while rabeprazole was the least prescribed option (4.0%) (Table S2). Double doses (ie omeprazole 40 mg daily and equivalent) or higher were used in the vast majority of cases (87.1%), either split as a twice daily dose (97.4%) or as a once daily intake (2.6%). While the daily dosages for the most prescribed drugs of omeprazole (40 mg), pantoprazole (80 mg) and lansoprazole (60 mg) were double, esomeprazole was most frequently prescribed as a quadruple dosage (80 mg; in 74.0% of EoE patients receiving esomeprazole).

The median and IQR duration of PPI therapy up to evaluation was 72 (62–98) days. As shown in Table 2, most patients (37.8%) were treated between 56 and 70 days (ie 8 to 10 weeks), in agreement with the recommendation of clinical practice guidelines of a minimum of 8 weeks PPI treatment before endoscopic evaluation.[1,35] A minority of patients (10.5%) were evaluated before the 8th week or beyond the 6th month (5.0%). The remaining patients had PPI effectiveness evaluated between 71 and 180 days (10–26 weeks), including 19.9% for > 10 to 12 weeks and 26.8% within 3–6 months.

Effectiveness of PPI Therapy to Induce Remission in EoE

Overall, PPI therapy reduced eosinophil density below the diagnostic threshold of 15 eos/hpf in 48.8% of patients, with 37.9% of patients achieving deep histological remission. A further decrease in eosinophil count from baseline was documented in 4.1% of patients despite no histological remission. Regarding clinical response, PPI therapy induced symptomatic improvement in 71.0% of patients (Figure 1 and Table S3). When both responses were considered together, clinico-histological remission was achieved in 48.9% of the 569 patients who were fully evaluated.

Figure 1.

Bar chart for histological (A) and symptomatic (B) responses for proton pump inhibitor (PPI) mono-therapy to induce and maintain remission in patients with eosinophilic oesophagitis. For induction of remission, patients were classified according to the PPI dosage prescribed: high dose was double dosage or higher, and low dose was standard dosage or lower. For maintenance therapy, only patients with dosage reduction from that used for induction were included. eos/hpf: eosinophils per high power field

Effectiveness rates of PPI induction therapy depended on the PPI dosages prescribed. Histological remission rate was higher for patients treated with high compared to standard or low doses (50.7% vs 36.7%; P = 0.038; OR = 1.77), and the same happened for symptomatic improvement (73.9% vs 54.6%; P < 0.001; OR = 2.36). Overall, the likelihood of achieving clinico-histological remission was greater for high compared to standard or low PPI doses (50.8% vs 35.8%, respectively; P = 0.027; OR = 1.85).

Among patients with high PPI doses, a greater proportion of them responded to quadruple (58.1%; n = 105) than to double doses (49.2%; n = 390), although this difference did not reach statistical significance (P = 0.124).

No significant differences were found among the different PPI drugs in achieving clinico-histological remission when used at high doses (P = 0.091). However, esomeprazole and omeprazole tended to provide higher effectiveness (55.8% and 54.5%, respectively) compared to pantoprazole, rabeprazole and lansoprazole (46.3%, 38.9% and 37.0%, respectively) (Table S4). This trend to a higher effectiveness for esomeprazole and omeprazole in clinico-histological response was not due to use of quadruple doses among these patients, since patients with double doses displayed similar response ratios (57.1% and 53.5%, respectively).

PPI therapy was significantly less effective in inducing clinico-histological remission when it was prescribed after failure of a first-line therapy with diet or swallowed topic corticosteroids, compared with being used as the initial therapy for EoE (27.6% vs 50.0%, respectively; P = 0.022; OR = 2.6). The reduced number of patients evaluated in this sub-group (n = 29) means this difference should be interpreted cautiously, as it could also be identifying a subgroup of patients with low adherence to any therapy.

Finally, we assessed whether PPI treatment length influenced the effectiveness of achieving clinico-histological remission of EoE. An 8 to 10-week (56-0 days) PPI treatment length provided 50.4% remission rate, which increased to 65.2% when treatment was prolonged to between 71 and 90 days (>10 to 12 weeks). However, treating patients with PPI beyond the 3rd month (>90 days) decreased effectiveness to 44.1%, possibly because longer treatments might reduce adherence (Table 2). When effectiveness was compared among these three groups of patients, statistical significance was detected (P = 0.022).

PPI Therapy to Maintain EoE in Remission

In total, 172 patients who were amongst those who had received PPIs as induction therapy were also treated with PPIs to maintain remission of their disease (Table S5). Maintenance treatment strategies consisted of reducing the PPI dose in 138 patients, increasing the PPI dose in 20 patients, and switching to equivalent dosage of a different PPI drug in 14 patients. The remaining patients, mostly those not responding to PPIs, were treated with dietary interventions (149 patients), STC (92 patients) and a combination of therapies involving PPIs (31 patients). Subsequent descriptive data and analyses for effectiveness refer exclusively to the 138 patients who received reduced PPI doses as single therapy to maintain EoE in remission.

Among PPI-responsive EoE patients, the most and the least prescribed drugs were omeprazole (58.7%) and rabeprazole (4.3%), respectively, which was also the case in induction of remission therapy (Table S2). Standard PPI doses were preferred to maintain EoE in remission (110 patients, 79.7%), while high PPI doses were still used in 21 patients (15.2%). The remaining 7 patients (5.1%) used half the standard doses.

Accurate data on effectiveness and evaluation date of PPI therapy to maintain EoE in remission was available for 90 patients (Table 2), assessed after a median of 117 days (IQR: 90–207) at the treatment institution.

Effectiveness of PPI Therapy to Maintain EoE in Remission

A reduced dose of PPI from that used for induction was effective in maintaining remission of EoE in 72 patients (69.2%), with 62 of them (59.6%) being in deep histological remission. As for symptoms, 98 patients (84.5%) reported any clinical improvement from baseline, with 84 of these patients (72.4%) being in clinical remission (Figure 1 and Table S3). Taken together, PPI therapy was effective in maintaining EoE in clinico-histological remission in 72 of the 103 (69.9%) patients who had both responses fully assessed.

As with induction therapy, PPI dose, drug type and length of treatment were analysed to identify potential differences in effectiveness, with no significant associations being found. The limited number of patients within subgroups however could have prevented identifying differences. In this sense, the clinico-histological maintenance of remission rates among patients who used quadruple doses of induction and switched to double doses for maintenance (n = 15) was 80%, while it was 68.2% among patients who received standard doses or lower for maintenance of remission (n = 88).

No PPI drug was found to be significantly superior to any other in maintaining EoE in remission when used at standard doses or lower. However, lansoprazole tended to be the least effective drug in terms of clinico-histological remission (57.1%) (Table S4). As for treatment length, effectiveness was higher when assessed after 2–3 months (75%), than when it was done between 3 and 6 months (55.2%) or beyond the 6th month (69.0%) (Table 2).

Finally, the effectiveness of PPIs to maintain EoE in clinico-histological remission tended to be related to their effectiveness to induce disease remission, which was greater when deep remission (<5 eosinophils per hpf) was reached (n = 83) than when eosinophil count reduction was between 6 and 15 eosinophils per hpf (n = 14) after induction treatment (73.5% vs 50.0%; P = 0.112).

Determinants for PPI Therapy Effectiveness in Inducing EoE Remission

In order to identify demographic and clinical variables associated with the effectiveness of PPI therapy to induce and maintain clinico-histological remission of EoE, uni- and multivariate analyses were performed. The variables and the categories compared are described in Table S6.

Apart from PPI dose and treatment length, EoE phenotype and presence of fibrotic changes at baseline endoscopy (rings and/or strictures) were identified as being significantly associated with effectiveness of PPI to induce clinico-histological remission of EoE. A multivariate analysis was performed with data from the 257 patients who had been assessed for all the four variables. EoE phenotype and treatment length remained statistically significant, providing evidence that patients with an inflammatory instead of stricturing phenotype had higher chances of accomplishing clinico-histological remission of EoE after PPI therapy (OR 3.7; 95% CI, 1.4–9.5); and a length of PPI treatment between 71 and 90 days provided significantly higher remission rates than that which lasted between 56 and 70 days (OR 2.7; 95% CI, 1.3–5.3) (Table 3).

Regarding PPI therapy in maintaining EoE remission, no variable was found to be significant for clinico-histological effectiveness in univariate analysis, although phenotype was close to statistical significance, with PPI therapy showing more than twice the effectiveness in inflammatory compared to stricturing phenotype (Table S7).

In addition, although no statistical differences in effectiveness were detected between children and adults, an increased clinico-histological response rate was observed for adults for both induction and maintenance therapies with PPIs (Table S7).