Distension-Contraction Profile of Peristalsis in Patients with Nutcracker Esophagus

Abstract

Introduction:

High amplitude peristaltic esophageal contractions, i.e., nutcracker esophagus was originally described in association with “angina-like pain” of esophageal origin. However, significant number of nutcracker patients also suffer from dysphagia. High resolution esophageal manometry (HRM) assesses only the contraction phase of peristalsis. The degree of esophageal distension during peristalsis is a surrogate of relaxation, and can be measured from the intraluminal esophageal impedance measurements.

Aims:

Determine the amplitude of distension and temporal relationship between distension and contraction during swallow-induced peristalsis in nutcracker patients.

Methods:

HRM impedance (HRMZ) studies were performed and analyzed in 24 Nutcracker, and 30 normal subjects in the Trendelenburg position. A custom-built software calculated the numerical data of the amplitudes of distension and contraction, the area under the curve (AUC) of distension and contraction, and the temporal relationship between distension and contraction.

Results:

In normal subjects, the distension peaks, similar to contraction traverse sequentially the esophagus. The amplitude of contraction is greater in the nutcracker esophagus but the amplitude of distension and area under the curve of distension are smaller in patients compared to controls. Distension peaks are aligned closely with contraction in normal subjects, but in patients the bolus travels faster to the distal esophagus, resulting in greater time interval between the distension peak and contraction peak. Receiver operative characteristics (ROC) curve reveal high sensitivity and specificity of the above parameters in patients.

Conclusion:

Abnormalities in the distension phase of peristalsis is a possible mechanism of dysphagia in patients with nutcracker esophagus.

INTRODUCTION

High amplitude peristaltic esophageal contractions or nutcracker esophagus was originally described in association with “angina-like” chest pain of esophageal origin¹. Significant number of patients with nutcracker esophagus, in addition to chest pain, also suffer from dysphagia^{2, 3}. By definition, the esophageal contractions in nutcracker esophagus are peristaltic with a normal latency of distal esophageal contraction. It is not clear why these patients should have dysphagia symptom. In general, the dysphagia symptom is associated with incomplete esophageal emptying or poor bolus clearance. However, studies show that the bolus clears normally with swallows in patients with nutcracker esophagus⁴. Studies by Omari and colleague indicate that abnormal bolus flow pattern during peristalsis may be the cause of dysphagia in non-obstructive esophageal dysphagia^{5, 6}. Whether patients with nutcracker esophagus have alteration in the bolus flow patterns is not known.

Each swallow induces a wave of inhibition followed by contraction, both of which travel from the top to the bottom of the esophagus in a sequential or peristaltic fashion. During inhibition phase, there is relaxation of the esophageal musculature and opening of the esophageal lumen that allows bolus to travel the esophagus with minimal resistance. Esophageal inhibition is followed by contraction, which propels the bolus towards the stomach. The duration of inhibition increases from cranial to caudal direction in the esophagus resulting in delay of contraction in the distal esophagus (latency period⁷). initial inhibition followed by contraction is the basis of peristalsis or the “law of intestine”, throughout the gastrointestinal tract, as described by Bayless and Starling, 120 years ago^{8, 9}. Whether patients with nutcracker esophagus have abnormalities of the relaxation phase of peristalsis or esophageal distension is not known.

High resolution esophageal manometry (HRM) assesses only the contraction phase of peristalsis. The degree of esophageal distension ahead of contraction is a surrogate of relaxation and can be measured from the intraluminal esophageal impedance measurements of the HRMZ recordings^10–12. We recently described the characteristics of swallow-induced distension-contraction profiles in normal healthy subjects; the amplitude and duration of distension increases from proximal to distal direction in the esophagus. Furthermore, there is a unique temporal relationship between distension and contraction, i.e., former travels in close relation with the later, more so in the Trendelenburg position than in the supine position¹². We have developed a computer software program that can generate distension-contraction profiles of the esophagus during peristalsis, quantify the amplitude of distension, and the temporal relationship between distension-contraction waveforms, from the HRMZ studies ¹³.

The goal of our study was to compare the distension-contraction profile of patients with nutcracker esophagus and normal asymptomatic healthy subjects.

Methods

Study Population:

We studied 24 patients with nutcracker esophagus (18 males), mean age 36 years (range 21-74) and 30 healthy subjects (12 male, mean age, 33 year (range 24-62). The data from some of the normal subjects were published in an earlier report, defining the characteristics of distension-contraction profile¹⁴. The nutcracker patients were referred to the esophageal motility laboratory of UCSD for assessment of chest pain and/or dysphagia symptoms. Nutcracker esophagus was defined when the mean contraction amplitude of 10 swallows on HRM, at any location in the distal esophagus, was greater than 200mmHg. The human investigation committee of the University of California, San Diego approved the study protocol (IRB # 182156).

HRMZ Recordings

All subjects were studied using a catheter assembly that consisted of an HRMZ catheter (4.2mm diameter; Medtronic Inc., MN), equipped with 36 pressure transducers (1 cm apart), and 18 impedance electrodes (2 cm apart). Viscous lidocaine (2% lidocaine hydrochloride topical solution, USP) was administered orally and nasally for local anesthesia, followed by placement of the HRMZ catheter assembly through the nose. Eight to 10 swallows using 10 ml of 0.5N saline were performed in each subjects in the Trendelenburg position. The saline used for swallows was warmed to 37 degree Celsius (body temperature) because temperature has significant effect on the conductivity of saline. The latter was checked prior to its use. The stretcher was tilted to minus −15° degree (head-end of the subject lower than feet, i.e., Trendelenburg position). The rationale for using Trendelenburg position in these studies is described in our earlier publications¹¹, but briefly it separates swallowed air from the saline bolus, which is important in determination of the accurate luminal cross sectional area from the impedance measurements¹¹.

Data Analysis

Whole HRMZ recordings were exported as text files and imported into a MATLAB based program (Dplots, Motilityviz, La Jolla, CA) for further analysis. The esophageal luminal cross sectional area (CSA) was calculated using principles described in the earlier study¹². The graphical user interface allowed the user to import studies, and select individual swallows to carry out the analysis. The region of interest (ROIs) for each swallow was extracted from the HRMZ data by using an interactively placed rectangle superimposed on the HRM topograph by the user, starting from the onset of the upper esophageal sphincter relaxation to 2 seconds after the return of LES pressure back to baseline at the end of peristaltic contraction. The selected ROI was located between the lower edge of UES and the upper edge of LES. The selected ROI was divided into 4 equal segments of the esophagus and following parameters were extracted for each segment, (1) peak amplitude and area under the curve (AUC) of distension, (2) peak amplitude and AUC of contraction, (3) time period between the onset of UES relaxation and peak distension (same as nadir impedance) (T1), and (4) time between T1 and peak contraction (T2) at each channel location in the esophagus.

Statistical Analysis

Quantitative data are reported as mean ± SD. The normality of the distributions was checked by the Shapiro-Wilk test. For, across the subjects comparisons we used a non-parametric Wilcoxon signed rank test for all comparisons, and Wilcoxon rank sum test in the case of non-paired data. P < 0.05 was deemed significant. Receiver operator characteristics (ROC) curves were generated to determine sensitivity and specificity of the parameters in distinguishing normal subjects from patients with nutcracker esophagus.

Results

Patient demographics, manometry findings and dysphagia score are shown in Table 1. Figure 1 shows examples of distension-contraction profile in a normal subject during 2 separate swallows, and figure 2 in a patient with the nutcracker esophagus (in order to show the intra-swallow variation within each group). In normal subjects, the distension and contraction wave travel sequentially (in a peristaltic manner) along the length of the esophagus. The distension amplitude and duration are greater in the distal as compared to the proximal esophagus. The distension waveform in normal subjects is a bell shaped curve, with varying height and widths at different locations in the esophagus. On the other hand, in nutcracker patients the distension amplitude is lower, and more irregularly shaped as compared to controls. In patients, the distension occurs in a fragmented manner with more than one peak at a given location. Furthermore, there is a difference in the temporal relationship between amplitude of distension and contraction, with T1 being shorter in the nutcracker patients. Figure 3 shows the distension-contraction pattern at different time points during a single swallow in a normal subject (from Figure 1A), and a nutcracker patient (from Figure 2A). Note, close temporal correlation between distension and contraction in normal subject. On the other hand, in the nutcracker patient, bolus is fragmented along the length of the esophagus, and peak distension occurs early resulting in alteration in the temporal correlation between distension and contraction.

Table 1.

Nutcracker Patient Demographic

Patient #	Gender	Age	Distal Latency	IRP	DCI	Dysphagia Score
1	M	65	7.1	5	4068	1
2	F	61	6.9	16	5625	13
3	F	60	8.6	12	5971	2
4	F	63	6	16	6212	6
5	M	53	6.1	3	2861	4
6	F	65	8.1	2	4514	4
7	F	71	7.4	3	7436	6
8	M	58	5.4	26	4133	9
9	F	55	5.5	14	9263	11
10	F	70	9.4	23	10503	3
11	M	78	7.1	28	6902	5
12	M	70	12	4	8031	14
13	F	51	7.1	4	14053	7
14	M	63	4.6	14	6214	0
15	F	52	7.6	0	6820	10
16	F	48	4.7	12	3796	8
17	M	66	6.4	5	4769	0
18	F	66	5.5	14	3343	1
19	F	51	6.4	3	2986	15
20	M	63	6	26	5479	2
21	F	55	3.9	7	5810	23
22	F	76	4.7	20	6155	19
23	M	66	8.8	3	4959	19
24	F	48	7.4	6	4092	21

Figure 1:

Sample distension-contraction plots of 2 swallows in a normal subject with 10ml, 0.5N saline, in Trendelenburg position. Panels A and C showing distension as a waveform superimposed on the pressure topograph, and panels B and D, show an alternate display, distension and contraction both as waveforms.

Figure 2:

Distension-contraction plots of 2 swallows in a nutcracker patient with 10ml swallow of 0.5N saline in the Trendelenburg position. Panels A and C showing distension as a waveform superimposed on the pressure topograph and panels B and D, show an alternate display, i.e., distension and contraction both as waveforms. Note, differences in the distension waveform in nutcracker patient as compared to control show in Figure 1. Amplitude of distension is smaller, distension peak earlier and distension is fragmented in the patient.

Figure 3:

Schematic of distension contraction along the length of the esophagus in normal subject (swallow from Fig 1 A) and a nutcracker patient ( swallow from Fig 2 A). Note, the lower amplitude of distension, earlier of arrival of bolus in the distal esophagus and fragmented distension in the esophagus in the nutcracker patient.

Quantitative differences between the 2 groups are summarize in Figure 4. The mean data show that the peak pressures are higher in the nutcracker group in segments 2, 3 and 4 as compared to normal. The mean contractions were 107 ± 42 vs 176 ± 54 mmHg (p< 0.01) for segment 2, 144 ± 60 mmHg vs 247 ± 50 mmHg (p< 0.01) for segment 3, and 152 ± 62 mmHg vs 274 ± 59 mmHg (p< 0.01) for segment 4, in normal subjects and nutcracker patients, respectively. In contrast to the contraction amplitude, peak distension amplitude were lower in the nutcracker patients, in esophageal segments 3 and 4 (distal segments). In segment 3, the mean distension values were 152 ± 29 mm² for normal, and 125 ± 36 mm² for nutcracker patients, (p< 0.01). In segment 4, peak distension values were 195 ± 23 mm² and 165 ± 37 mm² , (p< 0.01), for normal subjects and nutcracker patients, respectively.

Figure 4:

Mean data extracted from the distension contraction plots in controls and nutcracker patients, (A) peak contraction amplitude, (B) peak distension amplitude, (C) T1 = time difference between the onset of swallow and distension amplitude, (D) T2 = time between the peak distension and peak contraction amplitude, (E) Area under the curve (AUC) of distension and (F) AUC of pressure. Note, pressures are higher in the nutcracker patients but the distension is smaller compared to controls.

The AUC for pressure and distention was different between the two groups in segments 3 and 4 (distal part of the esophagus) of the esophagus. Contraction values were higher and distension values lower in patients as compared to normal counterparts. Contraction values 6697± 4612 vs 11423 ± 3824, (P<0.01) for segments 3, and 7685 ± 4655 vs 16793 ± 7146 for segment 4 in normal and patient groups, respectively. The AUC of distension for segment 3 and 4 were higher in the normal group, being 11439 ± 2751 vs 8647 ± 2695, and 17286 ± 5266 vs 10485 ± 29167 for the normal and nutcracker respectively.

The T1 (time between the onset of swallow and peak distension) was significantly different between two groups in segments 3 and 4. For segment 3, the mean values were 1.28 ± 0.24 vs 0.99 ± 0.24 secs, p< 0.01, for normal and patients, respectively. Similar values for segment 4 were 1.94 ± 0.38 vs 1.5 ±0.22, p< 0.01 for normal and patient, respectively. The T2 time, which is the time interval between peak distension and peak contraction was not different in any of the segments between the two groups.

Figure 5 shows the ROC curves for 6 variables tested to determine their sensitivity and specificity to differentiate between nutcracker patients and normal subjects. The area under the curve of ROC for, peak pressure, peak distension and AUC of distension were perfect, i.e., 1 suggesting high sensitivity and specificity. The area under the curve of ROC for the T1 was also high 0.826 and 0.882 for esophageal segments 3 and 4 respectively.

Figure 5:

Receiver operator characteristics curves for the six parameters measured in the study.

Figure 6 shows the relationship between patient dysphagia symptom scores and some of the measured parameters. There was no statistically significant correlation between the dysphagia score and; 1) AUC of distension, 2) AUC contraction, 3) amplitude of distension and 4) T1. The same was also true (no linear correlation) between dysphagia score with contraction amplitude and DCI (data not shown). However, there were two patients in our group with zero dysphagia score. Interestingly, their distension profiles had normal appearance, (Figure 7).

Figure 6:

Scatter plot of dysphagia score vs other measured parameters, (A) area under the curve (AUC) pressure, (B) AUC distension, (C) Peak amplitude of distension and (D) T1 in the nutcracker group (missing dysphagia scores for 3 patients). Also shown are the Spearman correlation coefficient (r), and the coefficient of determination (R²⁾.

Figure 7:

Distension contraction profile in 2 patients with nutcracker esophagus with zero dysphagia score. The distension profile in both of these patients resemble those of normal asymptomatic subject.

DISCUSSION

Our earlier studies determined the methodology to measure luminal CSA from the intraluminal impedance recordings of the esophagus and validated with intraluminal ultrasound images as the ground truth^10–12. The US imaging allows measurements at one^{15, 16} or two locations¹⁷ in the esophagus. However, using impedance measurement one can detect luminal CSA at every two centimeters along the whole length of the esophagus. The latter has allowed us to define important characteristics of the distension waveforms along the length of the esophagus during swallow-induced peristalsis, similar to what is measured with the closely spaced pressure transducer in the HRM. The protocol for measuring luminal CSA from impedance recordings is slightly different from the one used in clinical practice for the routine HRM. Swallowed bolus is the same, i.e., 0.5 N saline but it is warmed to body temperature because latter has significant influence on the electrical conductivity of the swallowed saline. One needs to know the precise conductivity of swallowed saline for the accurate CSA calculation by the impedance technique. Furthermore, studies were performed in the Trendelenburg position, because it allows separation of swallowed air from liquid bolus, as the two travel along the length of the esophagus during peristalsis¹¹. Air can have major effects on the conductivity of swallowed saline. Therefore, separation of air from the liquid is an essential step for the accurate CSA measurement from the impedance recordings. In our earlier studies we determined the effect of Trendelenburg position on the distension and contraction waveforms during peristalsis¹². We used 10ml bolus swallows, instead of 5ml bolus used in clinical studies because esophageal distension is bigger with 10ml, and more likely to detect distension abnormalities between patients and normal.

Important characteristics of distension waveform along the length of the esophagus are very similar to the contraction waveform in normal subjects ¹⁴, i.e., 1) distension peak travels sequentially along the length of the esophagus, 2) amplitude of distension is greater in the distal as compared to the proximal esophagus, 3) distension waveforms generally have a single prominent peak and, 4) at a given location in the esophagus, there is a close temporal correlation between the distension and contraction waveforms, i.e., end of distension wave occurs with the onset of contraction wave. In the current study, we found several differences in the distension profile during peristalsis in patients with nutcracker esophagus as compared to controls. These differences are; 1) the distension amplitude and AUC of distension waveform are smaller in patients than normal subjects and these difference are present in the distal esophagus, 2) bolus travels and arrives in the distal esophagus of patients faster compared to normal subjects, 3) unlike normal subjects, the esophagus distends in a fragmented manner during peristalsis in the nutcracker esophagus patients. In our earlier studies, we described bolus flow patterns in 3 achalasia subtypes¹⁸. In patients with achalasia esophagus type 2, swallow-induced esophageal pressurization related to longitudinal muscle contraction of the esophagus causes propulsion of bolus into the stomach against a closed LES with only partial emptying of the esophagus. In majority of patients with achalasia esophagus type 3, esophageal emptying is complete^{19, 20}. However, bolus arrives in the distal esophagus much later after the swallow onset, as compared to normal subjects, because of the luminal closure, rather than luminal opening of the distal esophagus²⁰. Similar findings, i.e., delayed arrival of bolus in the distal esophagus has been described in patients with non-obstructive dysphagia by Nugyen et al⁵.

There is an inverse relationship between luminal CSA and impedance values during bolus transport in the esophagus. Lower nadir impedance values imply higher luminal CSA, and vice versa. Nadir impedance value has been used in the calculation of dysphagia index in patients with non-obstructive dysphagia who have higher nadir values than normal subjects⁵. Our finding of low amplitude of distension in patients with nutcracker esophagus is in accordance with the earlier studies by Omari and colleagues. A lower AUC of distension in patients is reflective of lower amplitude and lower duration of distension in nutcracker patients. Bolus fragmentation during peristalsis is also an important feature in patients with esophageal motor disorders, and has been well described on barium swallow studies in association with diffuse esophageal spasm and achalasia type 3 esophagus. We observed bolus fragmentation in patients with achalasia type 3 esophagus in our earlier study²⁰.

The nature of defect that leads to the bolus flow abnormality in nutcracker esophagus can’t be determined from our study. However, the defect is clearly in the initial or the inhibition/distension phase of peristalsis. The excitatory phase of peristalsis, i.e., sequential nature of contraction along the esophagus is present, and the amplitude of contraction is actually greater in the nutcracker patients than normal. A low amplitude of distension is an indicator of one or more of the following possibilities, 1) impaired inhibitory innervation of the esophagus which is also a hallmark of achalasia esophagus, diffuse esophageal spasm^{21, 22} and recently described in patients with the Jackhammer/nutcracker esophagus²³, 2) thicker muscularis propria seen in patients with nutcracker esophagus²⁴ that is likely to render the esophagus less compliant, and 3) excessive stretch of the esophageal wall in the distended segment of esophagus related to excessive longitudinal muscle contraction in the contracted segment (above the distended segment). All of the above factors will result in a esophagus that is narrower than normal during bolus flow. The pharyngeal pump is generally responsible for propelling bolus to the mid and distal esophagus very quickly after the onset of swallow^{25, 26}. Trendelenburg position delays the pharyngeal pump related bolus propulsion to the distal esophagus in normal subjects because bolus has to move against gravity¹². Based on the principles of fluid flow, as defined by “Poiseuille law of physics”, one would expect a greater velocity of bolus flow and faster arrival of bolus in the distal esophagus in a narrow tube, as compared to a wider tube, which is exactly what we found in the patients with nutcracker esophagus, i.e., a shorter T1. A higher velocity of bolus flow along the length of esophagus may explain the apparent paradox of lower amplitude of distension in patients, even though volume entering and leaving the esophagus is same (10ml) in patients and normal subjects.

One may argue that some of the differences observed between normal subjects and patients are relatively small and can they explain the cause of dysphagia in patients? Brief esophageal dysphagia questionnaire used in our study is a validated tool to assess the severity of dysphagia²⁷. Standard HRMZ recordings utilize liquid bolus to study esophageal peristalsis, which may not be a significant challenge to test the distensibility of esophagus. Most of the patients with dysphagia have difficulty swallowing solid food. Along those line, Fox et all have championed the use of solid food challenge during the HRMZ recording. They found that many more patients have abnormal HRMZ recordings during solid food than the liquid swallows²⁸. Lack of correlation between subjective (dysphagia score) and objective (distension parameters) findings during HRMZ recordings remains a challenge²⁹ because patients may have different sensitivity. Patients may modify their diet and report lower dysphagia scores on the questionnaire. A recent study found no differences in the dysphagia score between patients with achalasia esophagus, distal esophageal spasm, nutcracker (jackhammer) esophagus and esophago-gastric junction outflow obstruction (EGJOO)³⁰, which is surprising because one would expect higher dysphagia score in achalasia esophagus given greater objective esophageal motor abnormalities (impaired LES relaxation and loss of peristalsis) than in patients with isolated esophagogastric junction outflow obstruction (impaired LES relaxation and normal peristalsis).

We used an older definition of the nutcracker esophagus, i.e., contraction amplitude of > 200mmHg in the distal esophagus. The current definition, based on the expert consensus in the field, is distal contractile integral of > 5000 mmHg/cm/sec³¹. We reanalyzed our data based on the DCI of > 5000 (n=14) and less than 5000 (n=10) in our patient population, and found no difference in the dysphagia scores as well as any of the measured parameters. In summary, poor clearance of bolus is generally associated with esophageal dysphagia. We propose that a strong contraction of the esophageal musculature, pushing bolus through a narrower than normal tube is the cause of dysphagia sensation in patients with the nutcracker esophagus.