Compartmentalized pressurization is a novel prognostic factor for hypercontractile esophagus

Abstract

Background

Hypercontractile esophagus (HE) is a disorder of increased esophageal body contractile strength on high-resolution esophageal manometry (HREM). Compartmentalized pressurization (CP) is a pattern with an isobaric contour of >30 mmHg extending from the contractile front to the lower esophageal sphincter on HREM. The relevance of CP to HE has yet to be explored.

Methods

A retrospective review was performed on 830 HREM studies of patients to identify HE. HE patients’ CP status and symptoms by Eckardt score (ES) were reviewed. Diagnoses were made using Chicago Classification (CC) v4.0.

Results

47 patients (5.6%) were identified as having HE by CCv3, 30 (3.6%) of which had HE by CCv4. 11/30 HE patients had CP, and 19/30 did not. CP was associated with chronic opioid use (36.4% vs 5.3% p=0.047). Presenting ES was greater for HE patients with CP (7 vs 4). 7 HE patients with CP and 11 without CP were managed medically. ES after medical therapy was higher in HE patients with CP compared to those without CP (9 vs 0). No HE patients with CP responded to medical therapy. Kaplan-Meier analysis demonstrated significance of this association over time. 83% of all HE patients had all cause symptom remission.

Conclusions

HE patients with CP are associated with a higher presenting ES. HE patients with CP do not respond to medical therapy, while HE patients without CP frequently do respond. CP in HE may have prognostic value in determination of treatment strategy for patients with HE.

Graphical Abstract

Compartmentalized pressurization is a pattern on high resolution esophageal manometry that signifies improper pressure coordination below the peristaltic wave and above the esophagogastric junction, similar to intrabolus pressure. Found abundantly in patients with hypercontractile esophagus, it may be associated with low distensibility index on endoluminal functional lumen imaging probe. The presence of compartmentalized pressurization in hypercontractile esophagus patients appears to predict medical treatment failure, thereby identifying candidates for endoscopic therapy.

Introduction

High-resolution esophageal manometry (HREM) is the gold standard for diagnosing esophageal motility disorders.¹ Chicago Classification version 4.0 (CCv4) redefined the disorder of abnormal elevation in peristaltic vigor, and is termed hypercontractile esophagus (HE). New criteria in CCv4 for HE include symptoms of dysphagia or chest pain, and absence of achalasia and mechanical obstruction.² By CCv3 criteria, increased peristaltic vigor was detected on 1.4–3.9% of all HREM studies.^3,4

There is a heterogeneity of patterns of HE on HREM, which may be different clinical entities, but there are no current guidelines for how this should affect patient management. The multipeaked pattern (MPP) is described as repetitive, prolonged contractions of the esophageal body.^2,5,6 Other patterns described include a classic, single-peaked swallow, and the hypertensive lower esophageal sphincter (LES) pattern.^2,7,8 MPP is associated with the presence of dysphagia compared to those with a classic single peaked swallow.⁶ However, the hypertensive LES pattern did not show any difference in presentation or response to treatment.⁸

Compartmentalized pressurization (CP) signifies improper pressure coordination below the peristaltic wave, which may impede proper bolus transit and cause dysphagia.^9,10 CP is a qualitative finding of vertical pressure greater than 30 mmHg extending from the contractile front to the LES.¹⁰ CP is highly predictive of symptoms of dysphagia in patients with esophagogastric junction outflow obstruction (EGJOO).⁹ A cohort of patients with EGJOO with 96% endoscopic treatment success rate were found to have CP 100% of the time, indicating a high positive predictive value for detecting clinically relevant EGJOO.¹¹ The presence and relevance of CP to HE has yet to be studied.

CP and intrabolus pressure (IBP) signify an increased pressure above the EGJ during deglutition. IBP is a quantitative measure of pressure 1 centimeter above the proximal border of the EGJ during the time of esophageal transit to the LES. It is the maximal average pressure over 3 nonconsecutive seconds.^12–14 However, it is not available on some manometric software.¹⁰ CP is a qualitative finding that is representative of increased pressure in this same region during deglutition. It is sensitive, but not specific for IBP >30 mmHg. CP’s benefit is that it can be measured on all HREM software using the isobaric contour tool.

Endoluminal functional lumen imaging probe (FLIP) has been used to study HE. Currently, low distensibility index (DI) on FLIP helps diagnose EGJOO.^15–17 One study showed that 8/15 patients with HE on HREM also had spastic-reactive pattern of the body on FLIP. This pattern was found ubiquitously across other CCv4 diagnoses.¹⁸ Though it is not used to diagnose HE, there is potential for supportive diagnostic information based on secondary peristaltic patterns on FLIP.

Opioid induced esophageal disorder can cause HE, and if suspected, chronic opioid cessation is recommended.¹⁹ Gastroesophageal reflux disease (GERD) is present in 43% of HE patients by CCv3.²⁰ Dysphagia improvement occurred independent of presence of GERD and proton pump inhibitor (PPI) treatment.²⁰

Medical treatment strategies for HE include calcium channel blockers (CCB), nitrates, peppermint oil, selective serotonin reuptake inhibitors (SSRIs), and other neuromodulators. CCBs, nitrates, and peppermint have been demonstrated to lower esophageal body pressure, with CCBs inducing symptom resolution in patients in a randomized control trial.^21–23 SSRIs have reduced dysphagia and chest pain in patients with esophageal body contraction abnormalities, but without affecting the contractile vigor.^24,25 Mean follow up time of 2.8 years in patients with increased esophageal body pressure show that 73% of patients have symptom improvement after 3 years.²⁶ HE patients may have spontaneous resolution of their symptoms without treatment.^3,26

Endoscopic management has demonstrated efficacy through Botulinum toxin (Botox) injections of the LES and peroral endoscopic myotomy (POEM) with a pooled clinical success rate of 79% on meta-analysis, and 82% on POEM alone, with clinical superiority to medical management.⁴ Botox injections demonstrated efficacy in retrospective studies in management of HE, but a randomized control trial with a double-blind sham control of Botox in HE demonstrated that although 23 patients had significantly improved symptoms, this was independent of Botox treatment.^27,28 POEM has demonstrated efficacy in treating HE without randomization.^29–31 Reported clinical success rates are as high as 92% at 1 year and 82% at 4 years.^29,31

Conservative management for HE with a step-wise approach is recommended.² HE can be managed medically or procedurally, with 74% experiencing symptom improvement.⁴ Medical therapy has a pooled success rate of 63%, although some patients treated would not meet current CCv4 criteria for diagnosis of HE.^2,4

Currently, there is no ability to predict which HE patients will go into remission with medication or spontaneously, and which will require procedural intervention for timely symptom resolution. The aim of our study is to characterize patients with HE by contractile patterns including the presence or absence of CP, evaluate the association of contractile patterns to clinical presentation, analyze treatment response to medical therapy by contractile pattern, and evaluate outcomes in patients with HE with consideration to the effect of time.

Methods

Ethical Considerations

This study was approved by the Institutional Research Board of both the University of Maryland Medical Center (UMMC) and the Baltimore Veterans Affairs Medical Center (BVAMC). As a retrospective study, this was considered exempt for the need for informed consent from participating patients.

Patient Cohort

This historical cohort included all patients undergoing HREM from 9/1/2017 to 12/31/22 at both UMMC and BVAMC. All HREM studies were reviewed to ensure patients met CC criteria for HE, for demographics, and for medical history including endoscopy reports. Patient symptoms were evaluated using the Eckardt Score (ES).³² Date of referral for symptoms, HREM, medical and procedural intervention, and follow up were recorded. Symptoms by ES were recorded at each clinical visit, as well as type of intervention.

Medical treatment strategies were made by the patient’s gastroenterologist and the patient through shared decision making. A conservative approach to management of HE was taken. Medical therapy was encouraged. Patients with more severe symptoms (ES greater than or equal to 4) were offered consultation for POEM or surgical myotomy. All POEMs were performed by a single, experienced, advanced endoscopist.²⁹ One patient had a surgical myotomy performed by an experienced surgeon.

Exclusion criteria are age under 18, presence of achalasia or EGJOO, and esophageal structural abnormalities such as an active stricture, mass, or diverticulum found on endoscopic assessment.

HREM procedures

All HREM was performed with Diversatek Ultima or Insight g3 catheters (Milwaukee, Wisconsin, USA) and were interpreted with ZVU v2.3 software. Patients gargled 10 mL of 2% lidocaine solution for 10 seconds. 4% viscous lidocaine solution was administered trans nasally via a syringe. After catheter insertion, a baseline recording was taken. 10 5 mL saline swallows were then recorded. Double swallows and swallows with retching were excluded from measurement. Distal contractile integral (DCI), integrated relaxation pressure (IRP), and distal latency (DL) were recorded and interpreted according to the existing CC at the time of the study.^2,33 Patients with elevated IRP and at least one normal esophageal body swallow underwent five upright swallows. IRP of 20 mmHg supine and 15 mmHg upright were cutoffs for elevation for detection of LES relaxation impairment.^9,10 Diagnosis of HE was according to CC, requiring at least 20% of swallows to have DCI greater than 8000 mmHg*s*cm, with normal IRP and DL.^2,10,34 In addition, Diagnosis of HE in this cohort required: 1) a supine median IRP less than or equal to 20, 2) if supine median IRP was greater than 20, an upright median IRP less than or equal to 15. Therefore, none of the patients in this study fits the CCv4 criteria for EGJOO.

All HREM was reinterpreted for this study. CP and MPP were evaluated. Neither were commented on during initial HREM interpretation. CP was determined using an isobaric contour of 30 mmHg. Extension of vertical pressure from the contractile front to the LES without breaks on a single hypercontractile swallow was required to establish CP (Figure 1). This follows previously established protocols.^9,11 MPP was defined by two contraction peaks, with a pressure trough that is greater than zero, separation of pressure peaks by at least 1 second, and pressure peaking at more than 10 mmHg. MPP was evaluated using conventional manometry.^5,6,35

Figure 1.

Demonstration of compartmentalized pressurization (CP; black arrow) in two different patients with HE. An isobaric contour of 30 mmHg was applied to both panels. The left panel demonstrates the absence of CP. The right panel demonstrates the presence of CP between the two black arrows.

Protocol for multiple rapid swallows was followed as previously described in literature.^36,37 All patients with hypercontractile esophagus diagnosed in 2021 or later were evaluated with multiple rapid swallows, consistent with a change in HREM protocol in CCv4.

Esophageal pH monitoring

Ambulatory catheter (24 hours) and wireless Bravo (48–96 hour) pH monitoring findings were recorded when available. Esophageal acid exposure was considered abnormal when pH was less than 4 for greater than 6% of the time. Acid hypersensitivity was determined when the symptomatic index is over 50% and the symptom association probability was greater than 95%. All tests were performed off PPI for at least 7 days.

FLIP procedures

FLIP (Minneapolis, Minnesota, USA) was performed by four experienced gastroenterologists during esophagogastroduodenoscopy using an EF-322N balloon. Indications for FLIP included pre-POEM assessment, and evaluation and exclusion of EGJOO. All FLIPs were secondary to HREMs. Stepwise inflation with measurement of pressure and DI (mm² mmHg⁻¹) occurred from 30 mL to 70 mL after observing for 60 seconds at each level. Median EGJ DI was determined at the maximum diameter of the LES. Point-of-care FLIP interpretation was made consistent with methods described in literature.³⁸ Diameter was considered at point-of-care to rule out achalasia. Maximum diameter was not always recorded in the medical chart. Diameter is therefore not reported in this study. All DIs were reported at 60 mL.

Definitions of covariates

Diagnosis of GERD was made by either reflux esophagitis on esophageal biopsy or based on results of the pH monitoring study. Potential for PPI responsiveness included diagnosis of GERD and patients with reflux hypersensitivity that received PPIs. Chronic opioid use is defined as use of an opioid medication at least daily for at least 3 months. Medical therapy included use of CCB, nitrates, SSRIs, or peppermint.

Change in ES was calculated by subtracting the presenting ES from the ES after therapy. Symptom improvement was defined as ES reduction to less than or equal to 3 as previously defined.³⁹ For patients presenting with an ES of 3 or fewer, reduction of score by greater than half was required to establish symptom improvement.⁹ This cut point was established to effectively perform a Kaplan-Meier analysis.

Statistical analysis

Quantitative variables are displayed as means and standard deviations. Qualitative variables are displayed as percentages. Ordinal variables, the Eckardt scores, are displayed as medians with an interquartile range (IQR). Quantitative, qualitative, and ordinal variables are evaluated with students t-tests, Fisher’s exact tests, and Kruskal-Wallis tests respectively. Multivariable analysis was conducted using multivariable linear regression.

A sensitivity analysis was conducted to select variables chosen for the final model. Effect modification was evaluated by looking for a statistically significant p-value of an interaction term, with a threshold set at less than 0.05. Confounding was evaluated for by adding potential confounders to the multivariable linear regression model and was determined by a threshold percent change in the beta coefficient of greater than 10%.

Survival analysis was conducted using Kaplan-Meier analysis. Statistical significance is set at p-value less than 0.05 using the log-rank test. All tests were conducted using SAS studio (SAS, Cary, NC, USA).

Results

Cohort characteristics

830 patients underwent HREM. Average age of patients were 56 ± 14 years and 52% female. 47 patients (5.6%) met CCv3 criteria for HE. Eight of these patients did not have symptoms, three had esophageal strictures, and six patients did not have any symptoms recorded in their medical chart (Figure 2).

Figure 2.

Flow diagram of patient selection.

HE on presentation

30 patients met CCv4 criteria for HE and had clinical evaluation, a 3.6% prevalence. These patients were on average 57 years old and 40% were female. 27% had the potential for acid suppressant medication responsiveness, four of which had findings on both pH testing and positive biopsy for reflux associated changes of the esophagus, and four of which only had positive biopsies for reflux associated changes in the esophagus. 12 patients had pH testing, three with reflux hypersensitivity, and one with acid reflux. 16.7% of patients were chronic users of opioid medication. 50% met criteria for the MPP. 47 ± 23% of all swallows were hypercontractile (Table 1).

Table 1.

characteristics of all patients with HE and patients with HE by CP presence. Data is presented by time period: on presentation, after medical therapy, and after procedural therapy.

Time period		All HE	CP	No CP	p-value
Presentation	N	30	11 (37%)	19 (63%)	-
	Age	57.0±11.0	54.6±10.4	58.4±5.4	0.38
	Gender (% female)	12 (40%)	5 (36%)	7 (45%)	0.71
	Chronic opioid use	5 (16.7%)	4 (36.4%)	1 (5.3%)	0.047
	Potential for acid suppression responsiveness	8 (26.7%)	2 (18.2%)	6 (31.6%)	0.67
	Multipeaked	15 (50%)	7 (63.4%)	8 (42.1%)	0.45
	Mean DCI	10213±11223	14051±17863	7991±3383	0.29
	Median IRP	17.4±5.4	18.9±6.6	16.5±4.5	0.24
	% hypercontractile swallows	47.4±22.5	48.0±20.0	47.0±24.4	0.91
	Multiple Rapid Swallow IRP	13.4±4.3	19	12.7 ± 4.0	-
	Multiple Rapid Swallow DCI	9486 ± 8383	23534	7730±6972	-
	Peristaltic reserve	33% (3/9)	100% (1/1)	25% (2/8)	-
	Total time observed (days)	543±443	828±567	245±379	<0.01
	Total presenting ES	4.5 (4–7)	7 (6–9)	4 (3–5)	<0.01
	Presenting dysphagia score	2 (1–3)	2 (2–3)	2 (1–2)	0.12
	Presenting chest pain score	1 (1–2)	2 (0–3)	1 (1–2)	0.48
	Presenting regurgitation score	1 (0–1)	1 (0–2)	1 (0–1)	0.07
	Presenting weight loss score	0 (0–3)	3 (0–3)	0 (0–0)	<0.01
After medical therapy	N	18	7†	11‡
	Age	57.2±11.9	52.9±8.9	59.9±13.1	0.23
	Gender (% female)	9 (50%)	4 (57.1)	5 (45.6%)	1.0
	Chronic opioid use	4 (22.2%)	3 (16.7%)	1 (9.1%)	0.25
	Potential for acid suppression responsiveness	4 (22.2%)	0 (0%)	4 (36.4%)	0.12
	Multipeaked	8 (44.4%)	4 (57.1%)	4 (36.4%)	0.26
	DCI	10977±14271	15813±22427	7900±4202	0.39
	IRP	16.8±6.0	18.3±6.9	15.9±5.6	0.43
	% hypercontractile swallows	50.0±24.9	52.5±22.5	48.5±27.3	0.75
	Total time observed (days)	551±429	797±528	393±274	0.048
	Total time on medical therapy	178±181	189±157	171±200	0.84
	Total post-medication ES	1 (0–9)	9 (7–10)	0 (0–1)	<0.01
	Post-medication dysphagia score	1 (0–2)	3 (2–3)	0 (0–1)	<0.01
	Post-medication chest pain score	0 (0–2)	2 (2–3)	0 (0–0)	<0.01
	Post-medication regurgitation score	0 (0–1)	2 (1–3)	0 (0–0)	<0.01
	Post-medication weight loss score	0 (0–3)	3 (3–3)	0 (0–0)	<0.01
	Total post-medication changes in ES	2 (0–3)	0 (−2–0)	3 (2–4)	<0.01
	Post-medication changes in dysphagia score	0 (0–1)	0 (0–0)	0 (0–2)	<0.01
	Post-medication changes in chest pain score	1 (0–2)	0 (0–0)	2 (1–2)	<0.01
	Post-medication changes in regurgitation score	0 (0–0)	0 (0–0)	0 (0–0)	0.16
	Post-medication changes in weight loss score	0 (0–0)	0 (0–0)	0 (0–0)	0.16
Procedural therapy	N	10	8§	2¶	-
	Age	53.6±8.9	52.9±9.3	56.5±9.2	-
	Gender (% female)	4 (40%)	4 (50%)	0 (0%)	-
	Chronic opioid use	3 (30%)	3 (37.5%)	0 (0%)	-
	Potential for acid suppression responsiveness	1 (10%)	1 (12.5%)	0 (0%)	-
	Multipeaked	5 (50%)	4 (50%)	1 (50%)	-
	Mean DCI	14338±18510	15440±20774	9932±3760	-
	Median IRP	19.3±6.78	19.4±7.7	19±1.4	-
	% hypercontractile swallows	49.7%±21.8%	49.7%±22.3%	50%±28.3%	-
	Total time observed (days)	739±474	812±495	448±310	-
	Number (%) tried on medical therapy	6 (60%)	6 (75%)	0 (0%)	-
	Total time on medical therapy	216±153	216±153	0	-
	Total post-procedure ES	1 (0–3)	1 (0–3)	1.5 (0–3)	-
	Post-procedure dysphagia score	0.5 (0–1)	0.5 (0–1)	0 (0–2)	-
	Post-procedure chest pain score	0.5 (0–1)	0.5 (0–1)	0.5 (0–1)	-
	Post-procedure regurgitation score	0 (0–0)	0 (0–1)	0 (0–0)	-
	Post-procedure weight loss score	0 (0–0)	0 (0–0)	0 (0–0)	-
	Total post-procedure changes in ES	7 (4–7)	7 (4–7)	4.5 (2–5)	-

^†:6 treated with CCBs, 1 treated with nitrates

^‡:7 treated with CCBs, 1 treated with CCBs and SSRIs, 1 treated with SSRIs, 1 treated with nitrates, 1 treated with peppermint

^§:7 treated with POEM, 1 treated with surgical myotomy

^¶:1 treated with POEM, 1 treated with Botox

11 (37%) patients had CP and 19 (63%) did not. There were no significant differences in age, gender, potential for acid suppressant medication responsiveness, presence of MPP, or HREM measurements. Patients with CP were more likely to have chronic opioid medication usage compared to those without CP (36.4% vs 5.3%, p = 0.047). Patients with CP had a higher median ES compared to those without CP (7 vs 4, p < 0.01) (Table 1).

There was no association between total presenting ES with the MPP, the potential for response to acid suppressant medication and gender. Chronic opioid use predicts a greater presenting ES (7 vs 4, p = 0.045) (Table 2). There was no correlation between age, IRP, DCI, and percent of hypercontractile swallows with presenting ES (Table 3).

Table 2.

Associations of covariates with the Eckardt score on presentation, after medical therapy, and the change in Eckardt score after therapy.

ES	Covariate	Subcategory	Eckardt Score	p-value
Presenting ES	Gender	Male	5 (3–7)	1.0
		Female	4 (4–7)
	Multipeaked	Presence	6 (4–7)	0.12
		Absence	4 (3–5)
	Potential for improvement with acid suppression medication	Presence	4.5 (3–7)	0.72
		Absence	4.5 (4–5)
	Chronic opioid use	Presence	7 (6–9)	0.045
		Absence	4 (3–6)
Post medication ES	Gender	Male	1 (0–9)	0.86
		Female	2 (0–7)
	Multipeaked	Presence	4 (0–9)	0.93
		Absence	1 (0–7)
	Potential for improvement with acid suppression medication	Presence	0.5 (0–1.5)	0.21
		Absence	4 (0–9)
	Chronic opioid medication	Presence	9.5 (5–10)	0.049
		Absence	1 (0–7)
Post medication ES change	Gender	Male	2 (0–3)	0.82
		Female	2 (0–3)
	Multipeaked	Presence	1 (0–3)	0.62
		Absence	2 (0–3)
	Potential for improvement with acid suppression medication	Presence	3 (2.5–3.5)	0.09
		Absence	1 (0–3)
	Chronic opioid medication	Presence	0 (−1.5–3)	0.14
		Absence	2 (0–3)
Post procedure ES	Gender	Male	1.5 (0–3)	0.72
		Female	1 (0–2.5)
	Multipeaked	Presence	3 (2–3)	0.11
		Absence	0 (0–0)
	Potential for improvement with acid suppression medication	Presence	3 (3–3)	0.25
		Absence	0 (0–3)
	Chronic opioid medication	Presence	3 (0–3)	0.38
		Absence	0 (0–3)
Post procedure ES change	Gender	Male	6 (4–7)	0.30
		Female	7 (5.5–8)
	Multipeaked	Presence	7 (4–7)	1.0
		Absence	7 (5–7)
	Potential for improvement with acid suppression medication	Presence	4 (4–4)	0.26
		Absence	7 (5–7)
	Chronic opioid use	Presence	4 (4–9)	0.71
		Absence	7 (5–7)

Table 3.

Table demonstrating correlations covariates with the total presenting Eckardt score using Spearman’s correlation.

	Spearman coefficient	p-value
Total presenting ES vs age	−0.044	0.82
Total presenting ES vs IRP	−0.01	0.94
Total presenting ES vs DCI	0.19	0.32
Total presenting ES vs % hypercontractile swallow	0.13	0.50
Total post medication ES vs age	−0.24	0.34
Total post medication ES vs IRP	0.02	0.93
Total post medication ES vs DCI	0.04	0.88
Total post medication ES vs % hypercontractile swallow	−0.06	0.82
Total post medication changes in ES vs age	0.41	0.09
Total post medication changes in ES vs IRP	−0.04	0.85
Total post medication changes in ES vs DCI	0.04	0.88
Total post medication ES vs % hypercontractile swallow	0.10	0.68
Total post procedure ES vs age	0.30	0.40
Total post procedure ES vs IRP	−0.11	0.75
Total post procedure ES vs DCI	0.30	0.40
Total post procedure ES vs % hypercontractile swallow	0.34	0.34
Total post procedure changes in ES vs age	0.31	0.38
Total post procedure changes in ES vs IRP	−0.42	0.22
Total post procedure changes in ES vs DCI	−0.27	0.45
Total post procedure ES vs % hypercontractile swallow	0.16	0.66

A sensitivity analysis was performed to evaluate for effect modification and confounding using multiple linear regression. There was no effect modification, nor confounding by age, gender, chronic opioid use, MPP, or potential for acid suppression therapy response. Therefore, the final model used did not include these variables. The multiple linear regression model showed that the presence of CP increased the predicted ES of a patient with HE by average of 3.3. The intercept for this model was 3.95, indicating that all patients with HE had a predicted ES of 3.95 if they do not have CP.

FLIP

Nine patients underwent FLIP prior to intervention, six with CP and three without CP. Eight of these patients (six with CP and two without CP) received FLIP prior to POEM as part of the standard protocol for POEM at our institution. Patients with CP had a median DI on presentation of 1.34 mm² mmHg⁻¹. Two patients had repetitive antegrade contractions (RACs), one had a spastic-reactive pattern (sustained LES contraction), one had absent contractility, and one had a borderline contractile response. Patients without CP had a median DI at presentation of 1.93 mm² mmHg⁻¹. All three of these patients had RACs (table 4). Examples of FLIPs are displayed in Figure 3.

Table 4.

FLIP on presentation prior to any procedural intervention by average ± standard deviation

		All HE	CP	No CP	p-value
N		9	6	3	-
Average median DI (mm2 mmHg−1)		1.54 ± 0.72	1.34 ± 0.75	1.93 ± 0.44	0.26
Maximum pressure		58.9 ± 3.33	70.9 ±14.0	72.5 ± 34.4	0.17
Contractile patterns	Repetitive antegrade contractions	5	2	3	-
	Repetitive retrograde contractions	1	1	0
	Absent	1	1	0
	Borderline contractile response	1	1	0
	Spastic-reactive contractile response	1	1	0

Figure 3.

FLIP of a HE patient without CP (left) and a HE patient with CP (right) at 60 mL of insufflation. The left patient with HE and without CP on HREM demonstrates repetitive antegrade contractions, and a normal DI. The right patient had both HE and CP on HREM and demonstrates a spastic reactive contractile pattern in the form of sustained LES contraction and a low DI.

HE after medical therapy

18 patients met CCv4 criteria for HE and received subsequent medical management. These patients were average age 57 years old and 50% female. 22% have the potential for acid suppressant medication responsiveness. 22% chronically use opioid medication. 44% meet criteria for the MPP. 50±25% of all swallows were hypercontractile. Patient follow-up was 551 ± 429 days, and patients spent 178 ± 181 days on medical therapy (table 1).

Seven HE patients with CP and 11 HE patients without CP underwent medical treatment. Of the patients with CP, six were treated with CCBs and one was treated with nitrates. Of the patients without CP, seven were treated with CCBs, one with both CCB and SSRI, one with SSRI, one with nitrates, and one with peppermint. There were no significant differences in age, chronic opioid use, gender, potential for acid suppressant medication responsiveness, presence of MPP, or HREM measurements. Patients with CP had a higher median ES compared to those without CP (9 vs 0, p<0.01) after medical therapy, and patients with CP had less reduction in ES after medical therapy compared to those without CP (0 vs 3, p<0.01). Patients with CP were observed on average for a greater duration of time (797 ± 528 days vs 393 ± 274 days, p<0.048), but on average spent similar amounts of time on medical therapy for HE (189 ± 157 days vs 171 ± 200 days, p=0.84) (Table 1).

An association of total post medication ES with the MPP, the potential for response to acid suppressant medication, and gender was not detected. Chronic opioid use also predicted a greater post medication ES (9.5 vs 1, p = 0.049) (Table 2). Change in ES was not predicted by gender, MPP, potential response to acid suppressant medication, or chronic opioid use. A correlation of age, IRP, DCI, and percent of hypercontractile swallows with ES after medication or change in ES after medication were not detected (Table 3).

A sensitivity analysis was used to evaluate for effect modification and confounding using multiple linear regression. There was no effect modification nor confounding by age, gender, chronic opioid use, MPP, or potential for acid suppression therapy response. Therefore, the final model used did not include these variables. The final model used showed that the presence of CP increased the predicted ES of a patient with HE after medical therapy by 8.6. The intercept for this model is 0.5, indicating that all patients without HE had a predicted ES of 0.5 after medical therapy.

HE after procedural intervention

10 patients proceeded to procedural intervention. Eight of these patients had CP. Of these eight patients, six failed medical therapy. Seven of these patients had POEM and one had a surgical myotomy. Two patients without CP had procedural intervention. None of these two patients were willing to try medical therapy. One patient had a POEM and one had a Botox injection to the LES. All patients had symptom improvement. Median change in ES for all patients was 7 (IQR 4–7). No statistical testing was conducted due to the small sample size of this group.

Analysis for symptom remission

None of the seven patients with CP achieved symptom remission with medication alone. 10 of 11 patients without CP achieved symptom remission with medication alone and were more likely to achieve symptom remission (p < 0.01). Total time observed from diagnosis to completion of medical therapy was 551 ± 479 days for all HE patients. Patients with CP when compared to patients without CP were afforded more time (797 ± 528 vs 393 ± 274 days, p = 0.048) from diagnosis to completion of therapy. They overall spent similar amounts of time on medication (189 ± 157 vs 171 ± 200 days, p=0.84). Kaplan-Meier analysis was conducted to evaluate the effect of time on symptom remission. The likelihood of symptom remission was significantly greater for patients without CP when evaluated in days from the time of HE diagnosis (p < 0.01) (Figure 4).

Figure 4.

Kaplan-Meier analysis for probability of symptom remission. The left panel (p<0.01) is the percent of HE patients that are symptomatic by time since diagnosis in days for those that had medical treatment only. The right panel (p<0.01) is the percent of HE patients that are symptomatic by time since diagnosis in days for all therapies, including medical treatment and procedural intervention. In both figures, the blue line represents HE patients without CP and the red line represents HE patients with CP.

All 30 patients were also reviewed for their likelihood to achieve symptom remission with respect to time. 83% of patients achieved symptom remission from any cause. There was no detected difference in overall likelihood in achieving symptom remission, with 8/11 with CP achieving symptom remission, and 17/19 without CP achieving symptom remission (p = 0.33). All five patients not achieving symptom remission were never deemed refractory to treatment but were lost to follow-up. Average time to symptom remission for all patients was 543 ± 443 days. Patients with CP took longer on average to undergo symptom remission compared to those without CP (827 ± 567 vs 379 ± 245 days, p = 0.03). Kaplan-Meier analysis was conducted to evaluate the effect of time on symptom remission. The likelihood of symptom remission over time was significantly greater for patients without CP (p < 0.01) when using the initial time point of HE diagnosis (Figure 4).

Discussion

In this historical cohort, we describe manometric patterns of MPP, single contractile peaks, and a pattern yet to be described in HE, CP, by the current CCv4 criteria. This study demonstrated that CP, a novel pattern described in HE, presented with greater symptoms by ES (7 vs 4), independent of age, gender, presence of GERD, the MPP, or chronic opioid use. This difference diverges further after attempting medical therapy, independent of age, gender, GERD presence, the MPP, and chronic opioid use. No patients in this study with CP responded to medical therapy, while 91% of patients without CP responded to treatment with medical therapy. Patients with CP had a median ES of 9 after medical therapy, compared to a median ES of 0 in those without CP. This effect was also considered with respect to time, which fortified this association. All patients in this cohort who had procedural intervention achieved symptom remission. A subset of the patients underwent FLIP as part of their diagnostic work up. Patients with CP appear to have DI and FLIP patterns typically associated with achalasia, suggesting that the CP pattern on HREM may be associated with LES dysfunction, lending more credence to FLIP’s utility in patients with HE.

The pattern of CP was previously identified as having high prevalence in patients with clinically relevant EGJOO.⁹ In a cohort of patients with EGJOO by CCv4, 100% of patients with EGJOO had CP.¹¹ As such, it appeared to be a marker for obstruction of the EGJ. HE is a disorder which is conventionally understood and described to have normal behavior of the EGJ. However, 37% of patients presented with CP in this cohort. We theorize that the CP pattern may signify EGJ obstruction, even when LES measurements on HREM are normal.

In essence, patients in this cohort with CP behave more like patients with abnormalities of the EGJ rather than the esophageal body, where the treatment strategy is with myotomy rather than medication.² CP requires an isobaric contour of 30 mmHg reaching from the contractile front to the LES.¹⁰ By definition, it requires contraction of the LES to reach an isobaric contour of 30 mmHg. IRP measures the lowest pressure at the LES for 4 contiguous or non-contiguous seconds during the swallow. Here, we hypothesize that in the case of patients with HE, IRP, which is measuring the period of relaxation during the phase of deglutitive inhibition, may not have the proper sensitivity to detect improper contraction of the LES. This improper contraction could disrupt bolus transit, cause symptoms, yet not be detected by the IRP measurement, since this pressure increase could not be part of the lowest pressure over 4 non-contiguous or continuous seconds or have a large effect on the median IRP. We hypothesize that CP may be a marker that can be used to increase sensitivity for detecting EGJ dysfunction in the setting of HE. This theory is consistent with our findings that HE patients with CP were refractory to medical therapy, yet responsive to procedures targeting LES.

Pertinently, no patients were classified as EGJOO. There was no statistically significant difference in IRP by CP. Six patients had elevated supine IRP, but all upright IRPs were normal for these patients. All patients were therefore classified as HE by CCv4. This is a crucial distinction, as in this context, CP appeared to detect EGJ dysfunction that current CCv4 would not fully elucidate.

CP is likely highly correlated to IBP. Its benefit is that it can be measured on all HREM software using an isobaric contour tool. Although IBP is available on the ManoView software (ManoScan System, Medtronic), it is not obtainable on all software, including on software used in this study.¹⁰ In this study, CP presence always indicated an IBP greater than 30 mmHg. CP is sensitive for an IBP greater than 30 mmHg, but IBP greater than 30 mmHg may not be specific for CP. CP extends from the contractile front to the LES, while IBP is measured 1 cm above the EGJ. Patients with elevated IBP may not always have extension of pressure from the contractile front to the EGJ. Direct comparison of the two measurements is a warranted area for future study.

With heightened criteria to correlate HREM with symptoms, there would be an anticipated decrease in prevalence of HE. HE prevalence in a prior meta-analysis ranged from 0.7–3.9% by CCv3.^3,4 In this cohort, the CCv3 prevalence of HE was 5.6%, and 3.6% according to CCv4. The reported CCv3 prevalence would be among the highest reported in literature.⁴

In this study, the MPP was not demonstrated to present with greater symptoms than patients with a single contraction peak, and it also was not associated with CP. The association, or lack of association, of the MPP and symptoms may be dependent on catheter type. The prevalence of the MPP in this study was 50%; this is lower than the reported rate of 81% using the ManoView software.⁵ It is congruent with other cohorts described on Sandhill software, which described a 43% prevalence.⁴⁰ This may be due to the display of the conventional manometry plots using the Diversatek’s ZVU software. The ManoView software depicts 12 conventional lines for interpretation of the esophageal body.^5,6 The ZVU software in this study has seven conventional plot lines available for analysis, four of which study the esophageal body. This may decrease the sensitivity for detecting the MPP pattern, as more subtle repetitive contractions may be missed with greater spacing between conventional plot lines.¹⁰ Under detection of MPP could result in a misclassification bias. Our study raises the importance of validating HREM patterns on different catheters and software. Further exploration of the MPP pattern is a warranted direction for future study. In this study, there was no association of the MPP with symptoms, nor with CP.

GERD is a comorbidity with HE as defined by CCv3 43% of the time, but improvement of symptoms in these patients occurs independent of GERD treatment.²⁰ Nonetheless, GERD was evaluated as a potential confounder in this study due to the potential for overlap in symptoms of these patients as measured by ES. In this study, only 7% of patients by CCv4 and 11% of patients by CCv3 met criteria for GERD. The same definition was used in this study as the one used by Mallet et al.²⁰ Nonetheless, this study has a lower incidence of comorbidity of HE and GERD, which did not confound these results.

Chronic opioid use appeared to be associated with CP, as well as more symptoms by ES on presentation, and symptoms after medical therapy. After multiple linear regression modeling, it did not appear that chronic opioid use predicted either presenting ES or ES after medical therapy among patients with HE. Nonetheless, chronic opioid use is associated with the presence of the CP pattern. In this study, chronic opioid use was not part of the exclusion criteria for HE. All patients that were chronic opioid users were recommended cessation of their medication as first line therapy, but they were unable to do so. Three achieved symptom remission with POEM, one with medication, and another was lost to follow up prior to attempting therapy. POEM in patients with chronic opioid use has been reported as successful, previously, though our approach is still cautious.⁴¹ The effect of chronic opioid use is accounted for statistically and did not bias results of this study. However, patients with chronic opioid use still warrant careful consideration for management.

FLIP was measured on a subset of patients with HE. Indications for FLIP in this study were for pre-POEM measurements in eight patients and for ruling out EGJOO in the other. As such, these patients were pre-selected as having the most severe and/or refractory symptoms, or highest IRP. Therefore, this data should be interpreted cautiously, as it is not representative of all patients with HE. However, the CP group frequently had FLIPs (8/11). These patients trended toward lower DI, and their FLIPs did not rule out achalasia. A limitation in this study is that maximum diameter is not reported. The diameter was considered when making point-of-care assessments in patient diagnosis for ruling out achalasia but was not available for chart review for this study. This is due to difficulty in measuring the maximum diameter when there are RACs, as the location of the LES is not easily discerned in these instances, due to appropriate LES relaxation; in these patients, it was interpreted clinically as a larger diameter, while the DI was still reported. FLIP data in HE patients with CP profiled more similarly toward values consistent with those found in achalasia and EGJOO, which supports the theory that CP is detecting EGJ dysfunction.^17,38 The role of FLIP in HE warrants future study, with particular attention to comprehensive reporting of FLIP data, including the maximum diameter.

Symptoms are often used to guide therapy in HE. In this cohort, the absence of dysphagia was 100% specific for medication induced symptom remission, and 36% sensitive. All patients without dysphagia had symptom remission with mediation, but many patients with dysphagia had symptom remission with medication. Absence of weight loss was 91% sensitive and 83% specific for predicting medication induced symptom remission. CP was the greatest predictor of medication induced symptom remission, as the absence of CP was both 100% sensitive and specific.

Strengths of this study include evaluation of patients with HE according to CCv4 with respect to time at multiple phases of care. Only one prior study evaluated HE with respect to follow-up time.²⁶ HE duration provides valuable prognostic information regarding management of patients with HE. Another strength of this study is the CP pattern itself. While only described on Diversatek catheters on ZVU software, it is a pattern that should be able to be detected on all systems, as only an isobaric contour tool is needed for detection on an HREM system. This contrasts with the MPP, which may have varying sensitivity for detection based on the catheters and software used to perform HREM.

Weaknesses in this study include small sample size. However, of the 11 cohorts considered in the most recent meta-analysis of HE by CCv3, this would be the third largest.⁴ Another limitation is the grouping of multiple medications together as medical therapy. This was necessary to preserve group size for comparison. Procedural intervention similarly was grouped together, and these patients were not compared statistically due to small sample size.

In conclusion, CP is a novel pattern described in the setting of HE. Patients with CP presented with a greater symptom score by ES and were less likely to respond to medical therapy. DI on FLIP may reinforce the CP pattern on HREM, and there may be an increasing role for its utility in HE. The finding of CP on HREM holds prognostic value, as it appeared to predict which patients will have symptom remission in 6 months on average, and which patients would be refractory to conservative medical therapy.

Abbreviations:

HREM

high-resolution esophageal manometry

CCv4

Chicago Classification version 4

HE

hypercontractile esophagus

MPP

multipeaked pattern

LES

lower esophageal sphincter

CP

compartmentalized pressurization

IBP

intrabolus pressure

EGJOO

esophagogastric junction outflow obstruction

FLIP

functional lumen imaging probe

DI

distensibility index

GERD

gastroesophageal reflux disease

PPI

proton pump inhibitor

CCB

calcium channel blocker

SSRI

selective serotonin reuptake inhibitor

POEM

peroral endoscopic myotomy

UMMC

University of Maryland Medical Center

BVAMC

Baltimore Veterans Affairs Medical Center

ES

Eckardt score

WL

weight loss

DCI

distal contractile integral

IRP

integrated relaxation pressure

DL

distal latency

IQR

interquartile range

RACs

repetitive antegrade contractions

DATA AVAILABILITY STATEMENT:

“The data that support the findings of this study are available from the corresponding author upon reasonable request.”