INTRODUCTION
Lung transplantation (LTx) is a viable option for most patients with end-stage lung diseases 1. Transplant failure is frequently caused by the development of bronchiolitis obliterans syndrome (BOS), a devastating complication that accounts for 30% of all deaths within a three-year period post transplantation 1. Risk factors associated with the development of BOS include cytomegalovirus infection and positive antibodies to class I HLA. Gastroesophageal reflux disease (GERD) is associated with other pulmonary diseases such as asthma, cystic and pulmonary fibrosis and obstructive sleep apnea syndrome 2. Therefore, pH, bile acid and trypsin levels in GERD have been studied as biomarkers for chronic allograft dysfunction 3. GERD is also associated with esophageal motor disorders (EMD), although it is not clear if GERD is the cause or the consequence of the esophageal motility impairment 4. Although GERD has been extensively studied as a factor associated with LTx rejection, there are some gaps with regard to the potential contribution to the abnormal motility of the esophagus. Esophageal dysmotility can impair esophageal clearance and could favor retrograde movement of the esophageal content to the pharyngeal area. Therefore, the refluxate material is not necessarily the acid that comes from the stomach. In this respect, experimental studies performed in rat lung transplant models (WKY-to-F344) showed that both acid (pH 2.5) and neutralized acid pH (pH 7.4) caused lesions consistent with obliterative bronchiolitis 5. Findings show that impaired esophageal peristalsis and hypotensive lower esophageal sphincter (LES) are more prevalent in LTx candidates with GERD 6 and hypercontractile motor disorders have been reported post-LTx 7. Patients who have undergone a bilateral lung transplantation or a re-transplantation have a more pronounced delay in gastric emptying, impaired esophageal motility and delayed esophageal acid clearance as compared to patients who had undergone a unilateral lung transplantation 8.
Esophageal high-resolution manometry (HRM) represents an unquestionable advance in esophageal manometry and allows a better characterization of EMD and esophagogastric junction (EGJ) morphology 9. The Chicago classifications include the new data that HRM provides for the study of EMD 10.
The aim of the study was to assess esophageal motor disorders by high resolution manometry (HRM) both pre-LTx and six months post-LTx, in patients with and without organ rejection.
MATERIAL AND METHODS
Patients
This was a single-center retrospective analysis of a prospectively-collected database of 57 consecutive patients over 18 years of age referred to our esophageal motility laboratory both pre-LTx and six months post-LTx, between October 2009 and November 2016. Inclusion criteria were: a) signed informed consent; 2) a minimum age of 18 years; and 3) indication for LTx. The exclusion criteria was: a) patients who refused to sign the informed consent.
Gastrointestinal (GI) symptoms were evaluated both pre and six months post-LTx using the modified DeMeester Questionnaire that evaluated heartburn, regurgitation and dysphagia. These were scored between 0-3 (0: none; 1: mild; 2: moderate; 3: very severe) 11.
Acute rejection was defined as a clinical, radiological or functional episode that was compatible with a histological demonstration according to the international classification of the Lung Rejection Study Group, or without histological demonstration but with a response to the specific treatment for rejection 12. Chronic rejection was defined according to the international guidelines. This was defined as a drop in forced expiratory volume in one second (FEV1) < 80% of the maximum volume achieved after transplant. This was on the condition that other etiologies such as infections, airway stenosis or native lung complications in single-lung transplant recipients were excluded 12. Chronic rejection can present at any time after the transplant.
The study was performed following the protocol proposed by the Lung Transplant Program approved by the Hospital Universitario 12 de Octubre.
HRM protocol
HRM was performed both pre-LTx and six months post-LTx in all patients according to the protocol established at our hospital. HRM was performed using a solid-state manometric assembly with 36 circumferential sensors with an outer diameter of 4.2 mm spaced at 1-cm intervals (Given Scientific Instruments Inc., Los Angeles, CA, USA). The transducers were calibrated at 0 and 300 mmHg by applying an external pressure prior to recording. The previously described standard protocol was used 13. All subjects were studied following an overnight fast and after discontinuing medication within 48 hours (except for immunosuppressant therapy post-LTx) that could affect the examination results. However, all patients received a double dose of proton pump inhibitors (PPI) post-LTx as per protocol and when the HRM was performed six months post-LTx. After calibrating the equipment, the recording probe was introduced nasally and the procedure was performed with the patient in the supine position 14. Subsequently, the recording was started with a 30 second basal period (landmark) with no deglutitions in order to obtain the sphincter pressures, followed by ten deglutitions of 5 ml every 20 seconds. If the first landmark were not satisfactory, a second landmark was performed at the end of the study. Furthermore, if the patient swallowed twice or the interval between swallows was under 20 seconds, these deglutitions were not considered as valid for the analysis. The final analysis required ten valid deglutitions 15.
24h pH-monitoring protocol
Twenty-four hours ambulatory pH-monitoring without PPI was carried out pre-LTx. The sensor was placed 5 cm above the upper limit of the lower esophageal sphincter (LES) determined by HRM. The study catheter was attached to an ambulatory recorder (Matla Systems Inc, Madrid, Spain). Post-LTx, pH-monitoring was performed with a double dose of PPI as per protocol (Lung Transplantation Unit). The proximal sensor was placed 5 cm above the upper limit of the LES and the distal sensor on the stomach. Patients were encouraged to carry out their normal diet and daily activities.
Data analysis
Manometric data were analyzed using the Mano ViewTM analysis software version 2.1 (Given Scientific Instruments Inc. Los Angeles, CA, USA) and reanalyzed with the new version 3.3. Two experienced researchers performed HRM and reviewed manually all the deglutitions of each patient. Finally, a manometric diagnosis was reached using the analysis algorithm recommended by the Chicago Classification. Esophageal motor function was evaluated according to the Chicago Classification v. 3.0 10. Subsequently, the EMD were classified into the following subtypes: normal, achalasia, hypercontractile esophagus, distal esophageal spasm, absent and ineffective peristalsis 16.
The following manometric measurements were analyzed: basal pressure of the LES (normal values: 10-35 mmHg; minimum pressure obtained with the sleeve device) 17, integrated relaxation pressure (IRP) (four seconds IRP normal values [IRP4s]: < 15 mmHg) 10, pressurization presence or absence 9, distal contractile integral (DCI) (normal values: 450-8,000 mmHgm.s.cm) 10, and distal latency (DL) (normal values: > 4.5 s) 10.
The normal values (thresholds) applied were based on the Chicago Classification, except for the cut-offs for LES resting pressure, as this was not considered in this classification. LES was considered as normal (basal pressure was between 10 and 35 mmHg and IRP-4s < 15 mmHg), hypotensive (basal LES pressure was < 10 mmHg and IRP-4s < 15 mmHg) and hypertensive (basal pressure ≥ 35 and IRP-4s < 15 mmHg).
HRM diagnosis was divided into the following groups: normal, ineffective peristalsis (≥ 50% weak and failed contractions; DCI < 450 mmHg.cm.s and DCI < 100 mmHg.cm.s respectively), absent peristalsis (100% failed contraction and IRP-4s < 15 mmHg), EGJ outflow obstruction (EGJOO) (IRP-4s ≥ 15 mmHg), achalasia (100% of failed contractions and IRP ≥ 15 mmHg), distal esophageal spasm (≥ 20% of premature contractions; distal latency < 4.5 s) and hypercontractile esophagus (≥ 20% of hypercontractile contractions; DCI ≥ 8000 mmHg.cm.s) 10.
EGJ morphology was classified depending on the distance between the LES and crural diaphragm (CD) as follows: type I (normal), for distances < 1 cm; type II, for distances between 1-2 cm; and type III (hiatal hernia), for distances over 2 cm 10,18.
Patients were classified as having GERD when the total time at pH < 4 was over 4.5% according to the DeMeester reference values when the procedure was performed without PPI 19 and 1.6% when performed with PPI 20. The association with reflux symptoms was not used in the final diagnosis for GERD as most patients were asymptomatic. pH data were not compared pre-LTx and post-LTx.
Statistical analysis
HRM quantitative data were described using the mean and confidence intervals (CI) for parametric variables and the median and interquartile range (IQR) for non-parametric variables. Qualitative data were expressed as absolute frequencies. Acute and chronic rejection patients were grouped for analysis purposes. Data were stratified for patients as rejection and non-rejection and the distribution was compared with the Mann-Whitney U test (non-normal distribution) or Chi-squared test (categorical variables) or Fisher's exact test, as appropriate. The statistical significance of paired differences between pre-LTx and post-LTx HRM data was calculated using the Wilcoxon signed rank test (non-normal distribution) or McNemar's test (categorical variables). The data analysis was generated using the IBM SPSS statistics 22 software.
RESULTS
Fifty-seven patients were included and analyzed both pre-LTx and post-LTx. Forty-four (77.2%) did not experience a rejection, nine (15.8%) had acute rejection and four (7%), chronic rejection. Most patients had obstructive type end-stage lung disease both in the non-rejection and rejection groups (53.2% and 46.2%). GI symptoms were very infrequent and the symptom score was similar in both the pre-LTx and post-LTx groups. Overall, 17.5% and 15.8% of patients had symptoms pre-LTx and post-LTx, respectively; heartburn and regurgitation were the most frequent. None of the patients had dysphagia either pre-LTx or six months post-LTx. However, two patients presented with dysphagia during follow up after the initial six month period. Patient characteristics are shown in Table 1.
BMI: body mass index; LTx: lung transplantation; GI: Gastrointestinal; Nd: no data (patients did not report symptoms). Statistical method: Mann-Whitney U, Chi-squared test. *Double channel esophageal pH monitoring pre-LTx was performed without PPI and double channel esophago-gastric pH monitoring post-LTx, with PPI as per protocol. pH data pre-LTx and post-LTx were not compared.
Analysis pre-LTx vs. post-LTx
HRM parameters and diagnosis are shown in Table 2. EMD were found in 33.3% and 49.1% of patients pre-LTx and post-LTx respectively (p = 0.018); hypercontractile esophagus was the most relevant EMD post-LTx (p = 0.018). Accordingly, DCI was significantly higher post-LTx (p < 0.001) (Fig. 1). Of the eleven patients with hypercontractile esophagus post-LTx, three (27.3%) had abnormal pH monitoring, two (18.2%) had EGJOO and one (9%) had abnormal pH monitoring with EGJOO. There was no evidence of other possible causes in the remaining five (45.5%) cases.
HRM: high-resolution manometry; LES: lower esophageal sphincter; IRP: integral relaxation pressure; EGJ: esophagogastric junction; DCI: distal contractile integral; EGJOO: esophagogastric junction outflow obstruction; IQR: interquartile range. Statistical methods: Wilcoxon signed rank test and McNemar's test. *Two patients with hypercontractile esophagus also had EGJOO post-LTx.
An EGJ morphology change was observed pre-LTx and post-LTx; type I (normal) was more predominant post-LTx (63.2% and 82.5% respectively, p = 0.007). Accordingly, EGJ length was shorter post-LTx due to a decrease in the intrathoracic length. Hypotensive LES was more frequent pre-LTx than post-LTx (33.3% and 17.5% respectively, p = 0.017) (Table 2). There were no differences in the frequency of EMD according to the type of LTx (bilateral or unilateral); 19 (51.4%) patients had bilateral LTx and ten (50%) patients with unilateral had normal peristalsis (NS).
Analysis rejection vs. non-rejection group
EMD were more frequent post-LTx in both the non-rejection and rejection groups, but particularly in the rejection group (43.2% and 69.2%, respectively; p = 0.09). Hypercontractile esophagus was observed in 23% of the cases post-LTx in the rejection group (Table 3). Accordingly, DCI tended to be higher in the rejection group (Fig. 2). EGJ was more frequently classed as type I (normal) in the non-rejection compared to the rejection group (86.4% vs. 69.2% respectively, NS) (Table 3).
HRM: high-resolution manometry; EMD: esophageal motor disorder; EGJOO: esophagogastric junction outflow obstruction. *Comparison of post-LTx between the non-rejection and rejection group (normal vs. EMD, p = 0.09). Statistical methods: Chi-squared test or Fisher's exact test, as appropriate.
The HRM results were as follows in the nine patients with acute rejection, three cases had no abnormalities, three had hypercontractile esophagus (one of them had also EGJOO), one had EGJOO and two, distal spasm. The three patients with hypercontractile esophagus still had the same disorder one year post-LTx. Two of these patients presented with dysphagia during follow up after six months and were treated with endoscopic botulinum toxin injection (Fig. 3). In addition, four patients had chronic rejection, two had weak peristalsis and one had EGJOO.
DISCUSSION
Patients with end stage lung disease pre-LTx have a high prevalence of EMD and hypotensive LES with a reported frequency of around 76.6% using HRM (1). We found a similar frequency in our series of 66.6% (including EMD and hypotensive LES). According to the Chicago Classification 3.0, the frequency of EMD pre-LTx in our series was 33.3%, and these alterations increased post-LTx up to 49.1%. The most frequent disorders found were ineffective peristalsis, hypercontractile esophagus and EGJOO. Previous studies performed with water perfused manometry have reported dysmotility post-LTx in up to 44% of the patients. Although the results are not strictly comparable between conventional and HRM, the authors found similar findings, including ineffective motility and nutcracker esophagus 21. In addition, we found significant changes in EGJ morphology pre-LTx and post-LTx; more type I (normal) EGJ was observed post-LTx, particularly in the non-rejection group. These changes could be related to the surgical changes and the improvement of pulmonary function. However, these findings were not present in the rejection group (Table 3).
There is some evidence that demonstrates a link between GERD and LTx rejection 8,22,23 that can occur early post-LTx 21, but the potential implication of esophageal dysmotility in LTx rejection has not yet been clarified. The importance of EGJOO for chronic rejection has been recently reported 24. In our series, EGJOO was observed pre-LTx in five patients, two related with abnormal acid reflux. The disorder persisted post-LTx in the same five patients and only one had abnormal pH-monitoring. When the data was analyzed considering the rejection vs. non-rejection group, EGJOO was more frequent in the rejection group. On the other hand, hypercontractile esophagus was diagnosed more frequently than expected in the post-LTx group, which was 19.3% (18.2% in the non-rejection and 23% in the rejection group).
The physiopathology of hypercontractile esophagus remains unclear and different hypotheses have been suggested, including mechanical EGJOO, occurrence secondary to reflux disease, primary esophageal muscle hypercontractility 25 and eosinophilic infiltration into the muscularis propria 26. In this regard, EMD was observed in one patient four weeks post-LTx, who nevertheless recovered without any specific treatment besides PPI 7.
In our series, all the patients received double doses of PPI as per protocol post-LTx and HRM was performed after six months. Therefore, the disorder was not corrected with PPI use. Furthermore, the small group of patients with complications post-LTx and EMD were followed up over six months and the disorder persisted in all cases. Therefore, we do not consider this as a transient problem. It was not possible to clarify the etiology of the disorder in 45.5% of the patients. In this respect, and according to previous reports 25, the disorder could have been initiated by neuromuscular stress induced by surgical trauma and perhaps, complicated by partial vagotomy during LTx leading to esophageal dysmotility. Some medications such as azithromycin can influence esophageal motility 27, but none of the patients were under the effects of this medication during the procedure as this treatment was stopped at least 48 hours previously.
The mechanism could be related to abnormal acid reflux or EGJOO in the remaining 54.5% of patients. An abnormal pH monitoring result was found in 27.3% of cases. EGJOO was a possible mechanism in 18.2% of cases with hypercontractile esophagus, and abnormal GERD was also found in one of these cases. Therefore, some of the complications post-LTx could be related to GERD and others could be a consequence of retained material in the esophagus due to an impaired esophageal clearance. A very recent finding supports our theory, namely, that poor esophageal clearance is due to an abnormal post-reflux swallow-induced peristaltic wave index associated with obstructive chronic lung allograft dysfunction 24. PPI treatment is necessary in these types of patients due to the high prevalence of GERD, although it would not be useful to prevent the aspiration of esophageal content. Even though antireflux surgery has been demonstrated to be safe post-LTx and leads to a stabilization of lung function in patients with reflux, there is no definitive data supporting the hypothesis that it improves the survival rate 28. On the other hand, antireflux surgery has not prevented acute rejection episodes 28. Although some other factors can influence rejection, it is important to appreciate that esophageal dysmotility which can impair bolus transit, may also contribute to LTx failure.
The main limitation of our study is the small number of patients in the rejection group, although we still observed significant changes pre-LTx and post-LTx and between the rejection vs. non-rejection group. Another limitation is the fact that HRiM and 24h pH/impedance was not performed in all of the patients.
In conclusion, significant changes in esophageal motility can be observed pre-LTx and particularly post-LTx; hypercontractile esophagus is a frequent EMD. EMD were more frequent in the rejection group compared to the non-rejection group. Esophageal motility alterations leading to impaired esophageal clearance should be considered as another factor that might contribute to LTx failure.