A Composite Score of Physiomechanical Esophageal Function Using Functional Lumen Imaging Probe Panometry in Eosinophilic Esophagitis

Abstract

Background and Aims:

The evaluation provided by functional lumen imaging probe (FLIP) Panometry includes esophageal distensibility/compliance (mechanics) of the esophageal body and esophagogastric junction (EGJ) and esophageal motility (secondary peristalsis). We aimed to develop a composite score using these parameters to characterize physiomechanical function in patients with eosinophilic esophagitis (EoE).

Methods:

215 adult patients with EoE that completed FLIP Panometry during sedated endoscopy with esophageal biopsies were included. FLIP metrics of esophageal body Compliance, Contractile response, Distensibility plateau, and maximum EGJ Diameter (C2D2) were scored as 0 for normal, vs 1 or 2 for increasing degree of abnormality. Scores were summed to calculate the composite “C2D2” score.

Results:

The C2D2 score had a significant positive correlation with mucosal eosinophil count (rho=0.241) and total endoscopic (EREFS) score (rho=0.467). Among patients off treatment at baseline evaluation (n=46), future proton pump inhibitor (PPI)-responders (i.e. achieved mucosal eosinophil count <15 per hpf after PPI treatment) had lower C2D2 scores than PPI-non-responders (median (IQR) 2 (1–3) vs 4 (2–6); P=0.003). A regression model (that controlled for age, sex, and baseline eosinophil count) showed a C2D2 score ≤3 had an odds ratio (95% confidence interval) of 14.5 (2.6–85) to predict future PPI-response. However, total EREFS scores (P=0.142) and baseline eosinophil count (P=0.480) did not differ between PPI-responders and PPI-non-responders.

Conclusions:

This composite score of FLIP Panometry metrics, the C2D2 score, may facilitate characterizing physiomechanical function in EoE and serve as an objective outcome measure.

Introduction

Eosinophilic esophagitis (EoE) is a clinicopathologic disorder characterized by esophageal symptoms, often dysphagia, and mucosal eosinophilia. The chronic inflammation of EoE is thought to result in fibrostenotic remodeling of the esophagus causing characteristic rings and strictures.^1–5 These remodeling changes can also be objectively assessed using functional lumen imaging probe (FLIP) to quantify esophageal distensibility and compliance of the esophageal wall.^{6, 7} Reduced esophgeal distensibility on FLIP was associated with risk for food impaction and/or requirement for therapeutic dilation in patients with EoE.^{6, 8} Additionally, FLIP appears to be a promising tool to monitor treatment effects and progression of remodeling over time in EoE.^{9, 10}

We recently described a model that incorporated the various physiomechanical patterns from FLIP Panometry to classify esophageal function in EoE.¹¹ While this classification scheme related to clinical features of EoE disease severity and also predicted histologic response to proton pump inhibitor (PPI), we recognized that strict classification labels of some borderline cases could be a limiting factor. Hence, given the multiple objective measures of esophageal function provided by the FLIP Panometry approach, we sought to develop a composite score that could utilize key FLIP Panometry outputs to facilitate defining esophageal function in EoE on a continuous scale, and thus could potentially complement the classification scheme.

Among key FLIP Panometry metrics, the distensibility plateau (DP), which represents the narrowest region within the esophageal body that ceases to distend with stepwise filling of the FLIP bag and increased intrabag pressure, was among the initial FLIP metrics described for use in EoE.⁶ While the clinical relevance of this metric was demonstrated with its association with risk for food impaction, it can be driven by a focal stricture and thus may not reflect global changes of esophageal remodeling in EoE.⁸ To address this potential limitation of DP, we recently described a FLIP metric of esophageal body compliance to reflect the relationship of esophagela body volume with distensive pressure, and thus provide more of a global assessment of esophageal body remodling than the DP.⁷ EGJ distensibility and maximum EGJ diameter are also evaluated during FLIP Panometry also further compliment the evaluation of esophageal function in EoE.¹² Additionally, we demonstrated that dysmotility in EoE could be assessed as the esophageal contractile response to distension (secondary peristalsis) using FLIP, which has facilitated evaluating the relationship of esophageal remodeling (i.e. reduced esophageal distensibility) with esophageal dysmotility.¹³

Ultimately, we hypothesized that a composite score that incorporated a graded scale from each of these key FLIP Panometry measures of esophageal function could be developed to characterizing the severity of physiomechanical dysfunction using FLIP Panometry. By parameterizing physiomechanical function on a continuous scale, this score could become an independent measure of esophageal function for EoE and/or be utilized to complete the recently described PhysioMechancial classification scheme.¹¹ Herein, we conceptualized a score incorporating measures of esophageal body Compliance, Contractile response, Distensibility plateau, and maximum EGJ Diameter, the C2D2 score. This study aimed to test the relationship of this FLIP Panometry composite score against characteristics of EoE disease severity, as well as test prediction of histologic treatment response.

Methods

Subjects

This patient cohort has been previously described.¹¹ Consecutive adult (age ≥18 years) patients with a diagnosis of EoE that completed FLIP during sedated endoscopy between January 2015 and December 2021 were included. Patients were prospectively evaluated per a standard of care approach and data was maintained in an observational data registry. This study then utilized retrospective analysis of this data for an observational, cross-sectional study focused on the time of the endoscopy with FLIP. If a patient with EoE had completed more than one FLIP, the initial FLIP encounter was utilized for analysis. Patients with technically limited FLIP, with previous foregut surgery, or without EoE were excluded. FLIP was typically performed for patients completing endoscopy with known EoE or for a primary complaint of dysphagia and EoE was suspected because of endoscopic features of EoE (i.e. longitudinal furrows, rings, or strictures).

Patients were diagnosed with EoE per consensus guidelines meaning ≥15 eosinophils/hpf on esophageal biopsies (during previous endoscopy, or endoscopy with FLIP included in this study).¹⁴ Additional clinical features were obtained from patients’ electronic medical records including time of onset of esophageal dysphagia or chest pain, time of EoE diagnosis, EoE-related treatment at the time of endoscopy with FLIP (PPI and/or topical steroid or elimination diet) and history of previous endoscopic dilation. When reported, the onset of symptoms was applied as the onset of disease (EoE) while diagnostic delay was calculated as the difference between onset of symptoms and EoE diagnosis.

Additionally, subsets of patients had histologic PPI or topical corticosteroid (TCS) response assessed; Figure 1. PPI-response was assessed in patients that completed a baseline endoscopy with FLIP off of EoE treatment (including PPI) that then completed a follow-up endoscopy after 8 weeks of treatment with PPI. TCS-response was assessed in patients with active EoE ((i.e. ≥15 eos/hpf while on PPI) when patients completed a follow-up endoscopy after 8 weeks of treatment with either swallowed fluticasone or budesonide. PPI-response or TCS-response was defined by achieving histologic remission (<15 eos/hpf) on the respective treatment.

Figure 1. Patient flow.

The C2D2 score was defined on FLIP Panometry performed at the baseline endoscopy with FLIP. Subgroups of patients had treatment response to proton pump inhibitor (PPI) and/or topical corticosteroid (TCS) defined on a future endoscopy after treatment. Response to treatment was defined as achieving an esophageal peak mucosal eosinophil density of <15 eos/hpf after treatment.

“*” = with or without proton pump inhibitor (PPI).

The study protocol was approved by the Northwestern University Institutional Review Board as minimal risk with a waiver of informed consent for analysis of deidentified, coded patient data.

Endoscopic, histologic, and symptom assessment

Subjects underwent upper endoscopy in the left lateral decubitus position. During endoscopy (same encounter as FLIP), four esophageal mucosal biopsies were obtained at approximately 5-cm and 15-cm proximal to squamocolumnar junction. Histologic evaluation of biopsy specimens was performed by gastrointestinal pathologists. The peak number of eosinophils per hpf (0.196 mm²) was recorded for each patient. Histologic remission was defined by an eosinophil count <15/hpf, as opposed to active eosinophilia when eosinophil count was ≥15/hpf.

Endoscopic features of EoE (Edema (scored 0–1), Rings (scored 0–3), Exudate (scored 0–2), Furrows (scored 0–1), and Stricture (scored 0–1)) were graded during the upper endoscopy to calculate a validated endoscopic assessment instrument, the Endoscopic EoE Reference Score (EREFS).⁵

Some patients completed validated dysphagia symptom severity scores, the Brief Esophageal Dysphagia Questionnaire (BEDQ) and Visual Dysphagia Question (VDQ) portion of the EoE Symptom Activity Index, on the day of the FLIP test.^{15, 16} For BEDQ (range 0–40) and VDQ (range 0–24), greater scores indicate greater symptom severity.

FLIP Study Protocol

The FLIP study using 16-cm FLIP (EndoFLIP^® EF-322N; Medtronic, Inc, Shoreview, MN) was performed during sedated endoscopy as previously described.¹⁷ With the endoscope withdrawn and after calibration to atmospheric pressure, the FLIP was placed transorally and positioned within the esophagus with 1–3 impedance sensors beyond the EGJ, maintaining this positioning throughout the FLIP study. Stepwise 10-ml FLIP distensions beginning with 30–40 ml and increasing to a target volume of 70 ml were performed; each stepwise volume was maintained for 30–60 seconds. Also in some studies, the FLIP was partially emptied and repositioned to the proximal esophagus, where stepwise FLIP filling was then again performed for distensibility measurement of the proximal esophagus. In some studies, excessive pressures (typically >60 mmHg, though no formal threshold was applied) limited the maximum FLIP distension to <70ml.

FLIP Panometry Analysis

FLIP data were exported to a customized program to generate FLIP Panometry plots for analysis (program available for free, open-source download at http://www.wklytics.com/nmgi). FLIP Panometry analysis was performed blinded to clinical characteristics as previously described.¹¹ Briefly specific features and patterns of esophageal contractility were identified applied to assign a contractile response pattern (Table S1). Analysis of esophageal distensibility included measurement of the DP and compliance of the esophageal body and the maximum EGJ diameter. Areas of the FLIP Panometry output that were affected by dry catheter artifact (i.e. artifact that impacts diameter measurement when occlusion of the FLIP balloon disrupts impedance planimetry circuit) were omitted from analysis. When both the distal and proximal esophageal body were evaluated (n=105; 51% of the cohort), the lower of the two DPs was applied for analysis. The DP was measured as the narrowest (after excluding esophageal contractions), fixed diameter that was observed in response to increasing FLIP volume and pressure.^{11, 13}

An estimate of esophageal body compliance was measured from the 8-cm length segment of the distal esophagus (3–11cm proximal to the EGJ) using an additional program that utilized a customizable ‘smart box tool’ during the 60ml FLIP fill volume (or maximum FLIP fill volume if FLIP filling protocol was to <60ml).¹⁸ The tool measured esophageal body volume (V2), which was normalized to 8-cm of axial length, and FLIP pressure during the measurement window (P2). Esophageal body compliance was then calculated as (V2-V1)/(P2-P1), V1 and P1 set to 0ml and 0mmHg, respectively.

The maximum EGJ diameter was measured at the greatest diameter achieved across the EGJ during the 60ml or 70ml FLIP fill volume (or maximum FLIP fill volume if <60ml).

The C2D2 score

To compute the C2D2 score, each of four FLIP Panometry features (compliance; contractile response; DP; maximum EGJ diameter) were scored from 0–2 with a score of 0 representing ‘normal’ function and increasing scores (1 or 2) indicative of more severe abnormality as defined by thresholds from healthy controls or clinically relevant outcomes among patients (Table 1).^{8, 12, 19, 20} The C2D2 score was the sum of four features, thus ranging from 0 to 8, with greater scores indicative of greater degrees of dysfunction.

Table 1.

The C2D2 score for FLIP Panometry assessment of physiomechanical esophageal function in EoE.

Feature	Score: Definition	n (%) of cohort
Compliance	0: >450 mm3/mmHg	138 (64)
	1: 300–450 mm3/mmHg	37 (17)
	2: <300 mm3/mmHg	40 (19)
Contractile response (CR)	0: Normal or borderline CR	138 (64)
	1: Impaired-disordered or spastic-reactive CR	58 (27)
	2: Absent CR	19 (9)
Distensibility plateau	0: >17 mm	113 (53)
	1: 14–17 mm	63 (29)
	2: <14 mm	39 (18)
Diameter of EGJ (maximum EGJ diameter)	0: >16 mm	87 (41)
	1: 12–16 mm	100 (47)
	2: <12 mm	28 (13)

A DP threshold of >17mm was applied based on its previously demonstrated association with lowered risk for food impaction (and is also the lower threshold of healthy controls); the 14mm was selected to reflect severe abnormality.^{8, 19} A compliance threshold of 450 mm³/mmHg was observed as the lower threshold from healthy controls.^{7, 18} A maximum EGJ diameter >16mm was determined as the relevant threshold to exclude relevant EGJ obstruction and <12mm was previously described as an additional critically relevant threshold for EGJ obstruction.¹² Finally, normal or borderline contractile response are observed among healthy controls and the majority of patients with normal primary peristalsis, whereas an absent contractile response was associated with the greatest frequency of failed primary peristalsis.²⁰

Statistical Analysis

Summary statistics were reported as mean (standard deviation; SD), or median (interquartile range; IQR) depending on data distribution. Correlation of continuous variables was assessed using Spearman’s rho. Groups were compared using Chi-square test for categorical variables and ANOVA/t-tests or Kruskal-Wallis/Mann-Whitney U for continuous variables, depending on data distribution. Intra-subject comparisons were performed using paired t-tests or Wilcoxon signed-rank tests, depending on data distribution. Additionally, logistic regression models with PPI-response as the dependent variable were assessed. These models included fixed effects for age, sex, and mucosal eosinophil density, as well as the C2D2 score or the total EREFS score. Statistical tests were two-tailed and significance set at 5%. Post-hoc comparison testing, as appropriate, was completed using a Bonferroni correction.

Results

Patient Characteristics

215 patients with EoE (mean age 38 (SD 12) years; 31% female) were included (Table 2). At the time of endoscopy with FLIP, 162 (75%) patients were on treatment, which included only PPI in 110 patients (51% of cohort), topical steroid in 24 patients (11%; 12 patients also on PPI), and elimination diet in 28 patients (13%; 13 also on PPI); Figure 1. Among patients on treatment, 61 (38%) were in histologic remission (eosinophil density <15 eos/hpf) whereas the remaining patients had ≥15 eos/hpf. There were 74 patients (66% of cohort) that had had a previous dilation; previous dilation occurred at a median (IQR) 29 (8–54) months prior to endoscopy with FLIP.

Table 2.

Cohort characteristics.

		Total cohort
N		215
Age, years, mean (SD)		38 (12)
Female, n (%)		66 (31)
Age of symptom onset, years, mean (SD)a		28 (14)
Symptom duration, years, median (IQR)a		7 (3–15)
Diagnostic delay, years, median (IQR)a		4 (1–12)
Off treatment, n (%)		53 (25)
On proton pump inhibitor, n (%)		135 (63)
On topical steroid, n (%)		24 (11)
On elimination diet, n (%)		28 (13)
On >1 treatment, n (%)		25 (12)
Previous dilation, n (%)		73 (36)
Visual Dysphagia Question score		5 (1–9)
[n completed VDQ]		[84]
BEDQ score		4 (1–8)
BEDQ≥1 FI, n(%)		49 (50)
BEDQ>1 ER visit for FI, n(%)		13 (13)
[n completed BEDQ]		[99]
Eosinophil count (eos/hpf), mean (SD)b		37 (33)
Eosinophil count <15/hpf, n/n on treatment (%)		61/162 (38)
Eosinophil count ≤6/hpf, n/n on treatment (%)		50/162 (31)
Endoscopic score (EREFS)
Edema score, n (%)	0	63 (29)
	1	152 (71)
Ring score, n (%)	0	28 (13)
	1	115 (54)
	2	63 (29)
	3	9 (4)
Exudate score, n (%)	0	115 (54)
	1	76 (35)
	2	24 (11)
Furrow score, n (%)	0	50 (23)
	1	165 (77)
Stricture, n (%)	0	43 (20)
	1	172 (80)
Total EREFS score, median (IQR)		4 (3–5)
EREFS fibrotic score, median (IQR)		2 (1–3)
EREFS inflammatory score, median (IQR)		2 (1–3)

Data reflects cross-sectional assessment at time of baseline endoscopy with FLIP.

^aAvailable in 171 patients.

^bIncluded all patients (on and off treatment).

All 53 patients off treatment had >15 eos/hpf. Among these patients, subsequent PPI-response was defined on follow-up endoscopy after ≥8 weeks of PPI in 46 patients: 24 (52%) patients were PPI-responders and 22 (48%) patients were PPI-non-responders. Among the entire cohort, there were 104 PPI-non-responders, 46 of whom completed a follow-up endoscopy on TCS: 27 (59%) were TCS-responders and 19 (41%) were TCS-non-responders.

The C2D2 score and association with EoE disease features

Among the entire cohort, the median (IQR; 5–95^th percentile) C2D2 score was 2 (1–3; 0–7). The frequency of scores for each item among the cohort are listed in Table 1.

The C2D2 score was significantly positively correlated with symptom duration, diagnostic delay, mucosal eosinophil density, and EREFS scores (total, fibrotic, and inflammatory), as were each of the individual items (with exception of contractile response with diagnostic delay, eosinophil count, or EREFS inflammatory score); Table 3; Figure S1. Hence, greater C2D2 scores were observed with greater duration of symptom duration (P<0.001) and diagnostic delay (P<0.001); Figure 2.

Table 3.

Correlation of C2D2 scores with clinical characteristics of EoE.

	C2D2 (total)	Compliance score	Contractile response score	Distensibility plateau score	Diameter at EGJ score
Age	0.117 s1	−0.038	0.219*	−0.038	0.204*
Symptom duration	0.396**s1	0.306**	0.238**	0.366**	0.302**
Diagnostic delay	0.336**	0.283**	0.122	0.340**	0.263**
BEDQ score [n=95 completed BEDQ]	0.010	0.013	−0.116	0.093	0.033
VDQ score [n=82 completed VDQ]	0.102	0.178	0.042	0.198	0.003
Eosinophil count (eos/hpf)	0.241** s1	0.252**	0.075	0.202**	0.179**
Total EREFS score	0.467** s1	0.423**	0.239**	0.474**	0.284**
Total EREFS-Fibrotic score	0.530**	0.453**	0.315**	0.558**	0.303**
Total EREFS-Inflammatory score	0.264**	0.259**	0.099	0.264**	0.159*

Values reflect Spearman’s rho.

^**P<0.01

^*P<0.05.

^s1Scatter plots displayed in Figure S1.

BEDQ – brief esophageal dysphagia questionnaire; VDQ - Visual Dysphagia Question

Figure 2. C2D2 score by EoE disease duration.

* Indicates significant difference on pairwise comparison (after Bonferroni correction) between subgroups of<5 years (A) or <1 year (B) with each longer duration.

The C2D2 score did not differ between patients on treatment (median (IQR) C2D2 score 2 (1–3)) or off treatment (C2D2 score 2 (1–4), P=0.185. However, the C2D2 score was lower in patients on treatment in histologic remission (median (IQR) 1 (0–2.5) than in patients with active eosinophilia whether they were on treatment (median (IQR) 2 (1–4)) or off treatment (2 (1–4)); P<0.010. The C2D2 score was also greater in patients with previous dilation (median (IQR) 2 (1–5)) than in those without (2 (0–3)); P=0.014.

The C2D2 scores differed across EREFS-item scores of edema (P=0.003); rings (P<0.001), exudate (P=0.002), furrows (P=0.037), and stricture (P<0.001); Figure 3. Among patients with stricture, the C2D2 score also differed relative to endoscopic estimates of stricture size with greater C2D2 scores in patients with greater severity of luminal stenosis (P<0.001); Figure 3.

Figure 3. C2D2 score by endoscopic features defined using the EREFS score.

There were differences in C2D2 score among each of the EREFS items: A) edema, B) rings, C) exudates, D) furrows, and stricture (E, F). “*” reflects a significant difference on pairwise testing (after Bonferroni correction when applicable).

The C2D2 scores (total and individual item scores), however, were not significantly correlated with symptom scores (BEDQ or VDQ). The eosinophil count (BEDQ rho = 0.123; P=0.227; VDQ 0.140 (P=0.206) and EREFs scores (total EREFS:BEDQ rho=0.034; P=0.735; VDQ rho=0.177; P=0.107) were also not significantly correlated with symptom scores.

Predictors of PPI or TCS response

Baseline factors for future PPI or TCS response were assessed among patients that complete the baseline endoscopy with FLIP with active EoE and then had PPI and/or TCS response defined on a future endoscopy; Figure 1. Among the PPI-responders (n=24), baseline C2D2 score was lower than in PPI-non-responders (n=22); P=0.003; Figure 4. However, total EREFS score (P=0.142) and baseline peak eosinophil count (P=0.480) did not differ between PPI responders and PPI non-responders (Figure 4). Binary regression analysis was performed using baseline features to assess predictor for PPI-response (Table 4). When controlling for age, sex, and baseline mucosal eosinophil density, the C2D2 score remained a significant predictor for future PPI response (odds ratio (95^th confidence interval) 1.6 (1.1–2.3); P=0.008. Further, when applying a C2D2 score threshold, a C2D2 score of ≤3 carried an odds ratio of 14.5 (95^th CI 2.6–85) to predict a PPI-response; Table 4.

Figure 4. Comparison of baseline features relative to future PPI (A-C) or topical corticosteroid (TCS; D-F) response.

C2D2 score (A, D), EREFS score (B, E), and peak mucosal eosinophil density (C, F) were assessed at the baseline endoscopy with FLIP in these subgroups of patients with active EoE at baseline assessment. PPI-response or TCS-response was defined at a separate endoscopy by achieving mucosal eosinophil density <15 eos/hpf after ≥8 weeks of treatment. “*” reflects a significant difference (P<0.05).

Table 4.

Binary logistic regression analysis for predictors of proton-pump inhibitor (PPI) response.

	Univariable			Multivariable model			Multivariable model - 2
	B (SE)	OR (95% CI)	P	B (SE)	OR (95% CI)	P	B (SE)	OR (95% CI)	P
Age, years, mean (SD)	−0.04 (0.03)	0.97 (0.9–1.0)	0.233	−0.03 (0.03)	0.97 (0.9–1.0)	0.353	−0.03 (0.04)	0.97 (0.9–1.0)	0.398
Sex (male)	0.75 (0.8)	2.1 (0.5–9.8)	0.338	0.3 (0.9)	1.4 (0.2–7.8)	0.733	0.3 (0.9)	1.3 (0.2–7.7)	0.773
Eosinophil count, continuous	0.01 (0.01)	1.0 (1.0–1.0)	0.502	0.004 (0.01)	1.0 (1.0–1.0)	0.748	0.01 (0.01)	1.0 (1.0–1.0)	0.531
Total EREFS score, continuous	0.35 (0.26)	1.4 (0.9–2.3)	0.166
C2D2 score, continuous	0.45 (0.2)	1.6 (1.2–2.3)	0.004	0.47 (0.2)	1.6 (1.1–2.3)	0.008
C2D2 score ≤3	2.8 (0.9)	15.9 (3.0–85)	0.010				2.7 (0.9)	14.5 (2.6–82)	0.002

PPI-response, defined as achieving mucosal eosinophil density <15 eos/hpf at a follow-up endoscopy after ≥8 weeks of PPI treatment, was utilized as the dependent outcome. Model inputs were obtained at the baseline evaluation (which was performed off of treatment in the subgroup of patients included in the regression analysis; n=46).

For prediction of TCS-response, there was not a difference in baseline C2D2 score (P=0.409) or total EREFS score (P=0.356); Figure 4. Baseline eosinophil count was greater in TCS-non-responders than in was in TCS responders (P=0.003).

Discussion

The key findings of this study was that a composite score of esophageal function based on FLIP Panometry parameters of esophageal wall mechanics and motility, the C2D2 score (Compliance; Contractile response; Distensibility plateau; Diameter of EGJ), was correlated with features of EoE disease severity. Additionally, the C2D2 score was also predictive of histologic response to PPI therapy, whereas the EREFS score and baseline mucosal eosinophil density were not. Thus, the C2D2 score could be useful to quantify EoE disease severity on a continuous scale and may also have prognostic implications regarding response to PPI treatment.

While mucosal eosinophilic inflammation is a key defining feature and primary therapeutic target of EoE, there is significant heterogeneity in the consequences of this chronic inflammation in EoE, potentially involving focal stricture, isolated to the EGJ or esophageal body and or diffuse, narrow caliber esophagus.²¹ This may relate to patchy histologic involvement of EoE, overlap with GERD, or disease duration (i.e. progressive, fibrostenotic remodeling). FLIP Panometry offers the unique opportunity to quantify these various perturbations associated with EoE. The DP measures the area of most severe stenosis in the esophageal body with implications for risk of food impaction, but DP does not differentiate between focal or global stenosis. The FLIP compliance measure compliments the DP by providing a more global measure of esophageal body remodeling. Additionally, stenosis can be isolated to the EGJ or be involved in concert with stenosis in the esophageal body. Further, esophageal dysmotility can be observed in EoE, and appears to be associated with remodeling (more than eosinophilia), and represents an additional component of functional impairment in EoE that is uniquely evaluated with FLIP Panometry.¹³ Overall, the present study aimed to incorporate the various components of the FLIP Panometry evaluation into a scoring scheme that could be utilized to quantify esophageal function in EoE. In so doing, this study demonstrated that the C2D2 score was associated with various features of EoE disease severity and could offer a tool to predict future response to PPI treatment.

We recently described applying these FLIP Panometry metrics to classify physiomechanical function among this cohort of EoE patients.¹¹ This model, which was based on the typical patterns of FLIP Panometry observed among patients with EoE, utilized the FLIP metrics to classify EoE patients into phenotypes of physiomechanical function and similar to the C2D2 score (as expected given the overlap), carried associations related to other measures of EoE disease severity and prediction of PPI-response. While the two approaches have shared features, the composite C2D2 score parameterized physiomechanical function onto a continuous scale and we anticipate that this will we provide a useful complement to the PhysioMechanical classification scheme. The C2D2 score may be of particular use in ‘borderline’ cases in which applying strict classification criteria may carry limitations. The respective roles for the composite score and classification scheme will likely be clarified and refined with future study.

Other methods have been described to grade disease severity in EoE.^{5, 22, 23} Among these, the EREFS score, was developed to facilitate standardized reporting of endoscopic findings of EoE and has been incorporated as an outcome measure in clinical treatment trials.^{5, 24} There is correlation between FLIP measures (including the C2D2 score) with the EREFS score, but FLIP may enhance the assessment by more precisely quantifying the degree of remodeling, particularly when incorporating the array of measures reflected in the C2D2 score. FLIP also assesses the extent of remodeling, whereas the EREFS score may not reflect global vs isolated involvement. Further, unlike other EoE disease severity scoring tools, FLIP also assesses dysmotility. Ultimately, FLIP and the C2D2 score demonstrated their promise by being predictive of future PPI response, whereas the EREFS score or degree of mucosal eosinophil density did not. However, eosinophil density differed between TCS responders and non-responders, whereas the C2D2 score and EREFS scores did not. As a severe narrow caliber esophagus was previously described as a risk factor for lack of TCS response it is unclear why neither the C2D2 score nor EREFS score would predict TCS response in this cohort, though it is possible that the available sample size for this subgroup analysis was susceptible to a potential type II error.³ Potential incorporation of this approach into future clinical trials will be useful to further test its ability to predict treatment response. However, the C2D2 score could still potentially help improve the efficiency of EoE management by helping to avoid treatment trials with a low likelihood for success (e.g. forego PPI trials in patients with high C2D2 scores).

Symptoms scores are also used to assess EoE disease severity and although important with regard to patient quality of life, there is a disconnect between symptom severity and objective measures of EoE disease activity (e.g. eosinophilia; endoscopic findings) observed both in the present and prior studies.^{25, 26} A recent scheme proposed a scoring scheme (the EoE Severity Index) that incorporated patient reported symptoms (including food impactions) and other aspects of EoE, including endoscopic and histologic features to reflect EoE disease severity.²³ Hence, a future direction may be to include incorporating patient reported outcomes with FLIP measures and endoscopic/histologic features to further facilitate characterize and monitor disease activity in EoE.

While this study carries strengths related to its rigorous evaluation of a sizable cohort of EoE patients, there are limitations as well. The study cohort was heterogeneous in regard to treatment status, thus somewhat limiting evaluation of the impact of treatment effects on the C2D2 score. While the heterogeneity of this EoE cohort may be viewed as a limitation, this also provided a broad range of EoE disease severity which facilitated analysis of associations of the C2D2 score against EoE disease severity. This also suggests that this FLIP Panometry approach can be utilized across the disease and treatment course of EoE. While clinical treatment outcomes were assessed for predictive analysis, the observational and primary cross-sectional design carries inherent limitations; additional study of the longitudinal effects of treatment on FLIP parameters and the C2D2 score is in progress. Additionally, the peak mucosal eosinophil count was the only histologic measure utilized in this study. Histologic scoring that incorporates fibrosis, such as the EoE Histology Scoring System, may enhance correlation with the biomechanical assessment of FLIP.²² Additionally, we recognize the FLIP may not be available at all centers. However, this study supports the increased use of FLIP to assess esophageal diseases and describes an approach that may facilitate its broader use.

In conclusion, this study described the C2D2 score that incorporates multiple measures of esophageal biomechanics and motility to facilitate the application of FLIP Panometry to assess EoE disease severity. The C2D2 score was associated with various indicators of EoE disease severity, but also demonstrated a potential advantage over commonly used ones (eosinophil count and EREFS score) to predict histologic PPI response. While future prospective treatment trials are needed to determine how changes in C2D2 score track with other clinical outcomes, this study provides further support that FLIP Panometry is a useful tool to help assess EoE disease severity and direct clinical management.

Supplementary Material

Figure S1

Figure S1. Correlation of C2D2 score with clinical characteristics of EoE. Darker marked indicate overlapping data points.

Abbreviations:

BEDQ

Brief Esophageal Dysphagia Questionnaire

CR

contractile response

DP

distensibility plateau

EREFS

Endoscopic EoE Reference Score

EGJ

esophagogastric junction

EoE

eosinophilic esophagitis

eos/hpf

Eosinophils per high powered field

FLIP

functional lumen imaging probe

IDCR

impaired/disordered contractile response

IQR

interquartile range

LES

lower esophageal sphincter

PPI

proton pump inhibitor

RACs

repetitive antegrade contractions

SD

standard deviation

SRCR

spastic-reactive contractile response

VDQ

Visual Dysphagia Question

Data sharing statement:

Data including the patient functional lumen imaging probe output for each patient and interpretations are included as an Appendix. Additional data that support the findings of this study are available from the corresponding author upon reasonable request and completion of necessary privacy and ethical approvals.