A asthma is now well recognized [l], although the

Transcription

1 Simultaneous Tracheal and Esophageal ph Monitoring: Investigating Reflux-Associated Asthma Raymund J. Donnelly, FRCSE, Richard G. Berrisford, FRCS, Catherine I. A. Jack, MRCP, Jane A. Tran, BTech, and Chistopher C. Evans, FRCP The Cardiothoracic Centre, Liverpool, United Kingdom Aspiration of gastric acid into the trachea may cause asthma in some patients who have gastroesophageal reflux. Antireflux surgery has been advocated for such patients, but lack of an objective test for acid aspiration makes patient selection difficult. We report a new technique for demonstrating acid aspiration, simultaneous tracheal and esophageal ph monitoring. Tracheal ph was measured with a 1.0-mm ph electrode introduced through the cricothyroid membrane under bronchoscopic vision. A standard esophageal ph electrode was placed in the usual position. Tracheal and esophageal ph were monitored over a 24-hour period. Peak expiratory flow rate was measured hourly while the patient was awake. We present data obtained in 3 patients with severe asthma and symptomatic gastroesophageal reflux. All 3 patients demonstrated a decrease in tracheal ph to less than 5.5, coinciding with a decrease in esophageal ph to less than 4.0. The test was repeated after antireflux operation and showed that significant decreases in esophageal ph no longer lowered tracheal ph. Asthmatic symptoms were improved, and medication was reduced in 2 of the 3 patients. (Ann Thorac Surg 1993;56: ) n association between gastroesophageal reflux and A asthma is now well recognized [l], although the pathophysiology underlying the association has been difficult to investigate. Two theories are proposed: aspiration of refluxed gastric contents and neurogenic reflex bronchoconstriction stimulated by esophageal exposure to acid. Evidence for and against these theories has been reviewed recently [2]. Several studies have shown that surgical treatment of gastroesophageal reflux improves asthma in some patients [3], but there has been no objective test to evaluate pretreatment and posttreatment aspiration. Radionuclide studies have not been satisfactory to date [4]. We describe a new technique for demonstrating acid aspiration, simultaneous 24-hour esophageal and tracheal ph monitoring. We present our findings in 3 patients with severe asthma and gastroesophageal reflux treated surgically. Material and Methods The procedure had Ethical Committee approval, and all patients gave informed consent. Patients We have studied 3 patients, all complaining of severe nocturnal asthma and symptomatic gastroesophageal reflux. Presented at the Twenty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Antonio, TX, Jan 25-27, Address reprint requests to MI Donnelly, The Cardiothoradc Centre, Thomas Drive, Liverpool, United Kingdom L14 3PE. Simultaneous Esophageal and Tracheal ph Monitoring Administration of all H,-blocking medication was stopped for 48 hours before ph testing. Fiberoptic esophagoscopy was performed under general anesthesia and endotracheal intubation. A standard 2.1-mm adult esophageal ph probe (Model ; Synectics, Dallas, TX) was introduced into the esophagus through the nasopharynx to a point 5 cm above the high-pressure zone previously identified on manometry. To achieve this, it was sometimes necessaryto remove the endotracheal tube for a short time. Ventilation was continued through a rigid bronchoscope introduced to a point just below the vocal cords. Under sterile conditions, a 14F cannula (Wallace, Colchester, Essex, UK) was introduced into the trachea through the cricothyroid membrane. It was important to introduce the cannula under bronchoscopic vision to avoid damaging the posterior tracheal wall. A specially made 1.0-mm ph electrode (Synectics) was passed into the trachea through the cannula and guided, under bronchoscopic vision, to a point 2 cm above the canna. The cannula was then removed, and the ph electrode was taped to the patient's neck. We have recently found that the standard 1.5-mm pediatric ph electrode (Model 0012; Synectics) introduced through a 13F cannula is more sturdy and equally satisfactory. Placement of the tracheal ph probe is also possible using local anesthesia and flexible bronchoscopy. The esophageal and tracheal ph probes and a sternal reference electrode were calibrated with buffer solutions (ph 1.0 and ph 7.0) and connected to a Digitrapper MkIII by The Society of Thoracic Surgeons /93/$6.00

2 1030 DONNELLY ET AL Ann Thorac Surg Table 1. Symptoms and Medication for Asthma and Gastroesophageal Repup Asthma Patient Symptoms Asthma Medication Reflux Symptoms Antireflux Medication Patient 1 42-y-old woman Asthmatic since age 11 years Recurrent pneumonia Skin testing positive Patient 2 35-y-old woman Asthmatic since age 6 months Skin testing negative Patient 3 41-y-old man Asthmatic since age 2 years Skin testing negative Sleep regularly interrupted No nocturnal symptoms Sleep regularly interrupted Morning hoarseness No nocturnal symptoms Sleep regularly interrupted Morning hoarseness Unchanged a Postoperative symptoms and medications are shown in bold type. Salbutamol inhalation every Beclomethasone inhalation every Intermittent prednisolone Salbutamol inhalation every Salbutamol inhalation every Beclomethasone inhalation every Prednisolone 60 mg daily Aminophylline 225 mg every 8 hours Salbutamol inhalation every Prednisolone 5 mg every Salmeterol inhalation every Beclomethasone inhalation every Ipratropium nebulizer every 4 hours Terbutaline subcutaneously 10 mg/24 h Domiciliary oxygen Unchanged Severe Unable to lie flat None Moderate Worse at night Resistant to medical treatment None Severe Worse at night Refractory to medical treatment Moderate improvement Ranitidine 300 mg every Asilone Nil Omeprazole 20 mg daily Nil Omeprazole 20 mg daily Ranitidine 150 mg every 8 hours Ranitidine 150 mg every 8 hours recorder using a three-probe adapter (Model ). Twenty-four-hour ph data from the two recording channels were downloaded and analyzed using the Synectics Liberty System and EsopHogram software. Asthma Assessment All 3 asthmatic patients had moderate-to-severe symptoms that were partially or fully reversible with bronchodilator therapy. All had decreases in peak expiratory flow rate of more than 20%. Particular note was made of nocturnal symptoms, medication, history of allergy, and results of skin testing. Symptoms of coughing and wheezing were noted, and hourly peak expiratory flow rate recordings were made during 24-hour tracheal and esophageal ph monitoring. Gastroesophageal Reflux Assessment Symptoms of heartburn, acid regurgitation, and dysphagia were noted as well as antireflux medication. In addition to 24-hour ph recording, esophageal manometry was performed using a standard water-perfused catheter and an Arndorfer infusion pump. All patients underwent esophagoscopy and biopsy. Grading of esophagitis was based on the Savary-Miller and DeMeester classifications 151. Antireflux Operation Belsey Mark IV antireflux procedures were performed on the 3 asthmatic patients in whom tracheal aspiration was occurring at the same time as gastroesophageal reflux. Manometry, esophagoscopy, and 24-hour tracheal and esophageal ph monitoring were repeated 6 to 12 weeks postoperatively. Results Technique of Simultaneous Tracheal and Esophageal ph Monitoring We found no difficulties in placing the tracheal electrode through the cricothyroid membrane. The first electrodes we used (1 mm) were not sturdy and only stood up to two or three uses. We have subsequently used the standard 1.5-mm pediatric esophageal electrode in the trachea. All patients tolerated the procedure, finding that an initial foreign body sensation wore off quickly. Lateral radiographs showed that none of the tracheal electrodes were coughed up. As the system uses a third electrode as a reference point, faults with this could be misinterpreted as simultaneous decrease in tracheal and esophageal ph. However, such interference was usually obvious, and the time period it covered was blanked with the esophogram software. Glass electrodes using their own internal standard are as yet far too large (5 mm) for use in the trachea. Clinical Results and Diagnostic Studies Clinical data are summarized in Table 1, and results from diagnostic studies are listed in Table 2. Pathologic gastro-

3 Ann Thorac Surg DONNELLY ET AL 1031 Table 2. Results From Diaanostic Studies ~~ Patient Endoscopy 24-Hour Esophageal ph Simultaneous Tracheal and Esophageal ph 1 Grade 2 esophagitis Barrett s esophagus No hiatal hernia No esophagitis 2 Grade 1 esophagitis No hiatal hernia No esophagitis Unable to lie flat off treatment DeMeester score on treatment 36.4 DeMeester score 11.0 DeMeester score 53.8 DeMeester score 11.1 Simultaneous drop in tracheal and esophageal ph below 5.5 at 2 to 4 AM (Fig 1A). During this time the patient woke wheezing and peak expiratory flow rate decreased by more than 20% No decrease in tracheal ph to less than 5.5 Tracheal ph 6.3 (mean), 6.5 (median)(fig 1B) Tracheal ph < 5.5 on two occasions, both coinciding with an esophageal ph < 4.0 (Fig 2 4 No corresponding decrease in peak expiratory flow rate Tracheal ph did not decrease below 6.0 No wheezing. Peak expiratory flow rate did not decrease >20% (Fig 2B) 3 Grade 2 esophagitis DeMeester score 9.1 Tracheal ph decreased below ph 5.5 during a dip Small sliding hiatal hernia (Manometry showed short low- in esophageal ph to below 3.5. This No esophagitis pressure LES) corresponded to a decrease in peak expiratory DeMeester score not available as flow rate 220% (Fig 3A) upright component of trace Tracheal ph remained >6.0 (Fig 38) failed to record a Results of investigations performed postoperatively are shown in bold. esophageal reflux is signified by a DeMeester score greater than [6]. Segments from simultaneous tracheal and esophageal ph recordings are shown in Figures 1 to 3. The position of the patient during episodes of gastroesophageal reflux is not the same in Figures 1A and 2A (preoperative) and Figures 1B and 2B (postoperative). It would appear that the results are biased; this is not so because tracheal ph did not decrease to less than 5.5 in any patient postoperatively whether upright or supine. Comment Several authors have shown that antireflux operations can improve respiratory symptoms and reduce medication in asthmatics with gastroesophageal reflux [ However, about 80% of asthmatic patients have significant gastroesophageal reflux on 24-hour ph monitoring [18], and about 40% have esophagitis on endoscopy [19]. How can we select those patients whose respiratory symptoms may be relieved by an antireflux operation, particularly if their reflux symptoms alone do not warrant operation? Significant criteria in the patient s history include nocturnal asthma, nocturnal coughing and wheezing fits, morning hoarseness, recognition of simultaneous reflux and asthmatic symptoms, onset of reflux symptoms preceding asthmatic symptoms, lack of an allergic component to their asthma, and a history of recurrent pneumonia [1, 10, 11, 13, 16, Historical features may be the best predictor of outcome after operation in these patients PI. The Bernstein acid infusion test may provoke asthmatic symptoms if bronchospasm occurs as a reflex to esophageal exposure to acid [22]. However, variables that reduce bronchial reactivity, such as daytime hours, may obscure a positive result [23]. Acid perfusion of the distal esophagus is a weak stimulus for an acute asthmatic attack, although it may heighten a histamine challenge test [24, 251. Scintigraphy has been used to demonstrate aspiration of gastric acid, but results obtained with this technique have been inconsistent [4], and the test does not establish the relation between aspiration and reflux episodes. Twenty-four-hour esophageal ph monitoring should be done to document significant abnormal reflux. Wheezing or coughing coinciding with a reflux episode is particularly significant. Monitoring ph in both distal and proximal esophagus has been done, but results have been conflicting. Ramenofsky and Leape [26] found that children with pneumonia who required antireflux operation had twice as many episodes of upper esophageal reflux as those with pneumonia who were successfully treated medically. They used a historical control group. More recently, Gustafsson and co-workers [27] could demonstrate very few episodes of high esophageal reflux in children with asthma and concluded that acid aspiration was not an important trigger for asthma. Double-probe esophageal ph monitoring has not been a particularly helpful investigation [20] and is not currently listed as an investigation of choice [22]. By combining 24-hour tracheal ph monitoring with simultaneous esophageal ph monitoring, we can establish whether or not ph changes in the two viscera are associated. If microaspiration does cause bronchospasm in these patients, this technique would seem to be suitable for establishing which patients are microaspirating because of gastroesophageal reflux. Acidification of the trachea in cats has been shown to provoke bronchospasm [28]. Argument as to the exact

4 DONNELLY ET AL Ann Thorac Surg trigger for bronchospasm in these patients, whether exposure of the esophagus to acid or reagins, or microaspiration, is fruitless as probably all three are implicated [2]. However, the results presented here do suggest that microaspiration of acid is important. We stress that this is a pilot study. We cannot draw any conclusions with this small number of patients. Patients 1 and 2 certainly benefitted from antireflux operation, and the findings on simultaneous tracheal and esophageal ph monitoring were a good predictor of outcome. Respiratory symptoms in patient 3 were not improved. He was believed to have sufficient reflux symptoms and manometric abnormality to warrant operation but did not have 8... r r...!... ;... A ~ P Pa -P 0 q- D... B Fig 2. (A) Segment from the preoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 2. The patient did not complain of wheezing during this recording, although tracheal ph decreased parallel to esophageal ph. The patient was supine. (B) Segment from the postoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 2. The decrease in esophageal ph represents a normal episode of reflux (DeMeester score was 11.1). Tracheal ph did not decrease to less than 5.5 at any point during the 24-hour recording; neither did the patient complain of wheezing. The patient was upright and postprandial. B Fig 1. (A) Segment from the preoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 1. The patient was awake and wheezing during this time and required an inhaled bronchodilator. The patient was supine. (B) Segment from the postoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 1. This shows a normal episode of gastroesophageal reflux; the postoperative DeMeester score was 11.O. Tracheal ph did not decrease to less than 5.5 at any point during the 24-hour recording; neither did the patient complain of wheezing. The patient was upright. significant gastroesophageal reflux on 24-hour ph monitoring. In addition, his chronic asthma was very severe, and his bronchial reactivity may have been blunted. Further work is needed to evaluate this technique more fully. It would be useful to use tracheal ph monitoring alongside other techniques such as proximal esophageal ph monitoring and scintigraphic scanning of the lungs to evaluate these alternative techniques further. We intend to screen consecutive patients presenting with asthma to establish the prevalence of tracheal microaspiration and then randomize such patients to medical antireflux treatment, operation, or no antireflux treatment. With greater patient numbers it should be possible

5 Ann Thorac Surg DONNELLY ET AL 1033 A B 0' t I / 0' Fig 3. (A) Segment from the preoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 3. During this period severe wheezing developed such that peak expirato y flow rate was unrecordable, and the patient required nebulized bronchodilators. The patient was upright. (B) Segment from the postoperative tracheal ph (broken line) and esophageal ph (solid line) recording from patient 3. Tracheal ph did not decrease to less than 5.5 at any point during the 24-hour recording, although there were dips in esophageal ph to less than 4.0. However, the patient still complained of wheezing and required a bronchodilator inhaler during the night. to establish a normal range for tracheal ph as well as the incidence of tracheal ph dips that are not associated with gastroesophageal reflux. This work was funded by the British Lung Foundation. References 1. Pope CE 11. Respiratory complications of gastro-oesophageal reflux. Scand J Gastroenterol 1989;24: Mansfield LE. Interactions, associations, and relationships between the lungs and the esophagus. Clin Rev Allergy 1990;8: Sontag SJ. Gut feelings about asthma. The burp and the wheeze. Chest 1991;99: Stein MR. A practical approach to gastroesophageal reflux and asthma. J Asthma 1989;26: Little AG, DeMeester TR, Kirchner PT, OSullivan GC, Skinner DB. Pathogenesis of esophagitis in patients with gastroesophageal reflux. Surgery 1980;88: DeMeester TR. Prolonged oesophageal ph monitoring. In: Read NW, ed. Gastrointestinal motility: which test? Wrightson Biomedical, 1989: Overholt RH, Voorhees RJ. Esophageal reflux as a trigger in asthma. Dis Chest 1966;49:464&. 8. Urschel HC, Paulson DL. Gastroesophageal reflux and hiatal hernia. J Thorac Cardiovasc Surg 1967;53: Lomasney TL. Hiatus hernia and the respiratory tract. Ann Thorac Surg 1977;24:44& Pellegrini CA, DeMeester TR, Johnson LF, Skinner DB. Gastroesophageal reflux and pulmonary aspiration: incidence, functional abnormality, and results of surgical therapy. Surgery 1979;86: Perrin-Fayolle M, Be1 A, Braillon G, et al. Asthme et reflux gastro-oesophagien. Poumon Coeur 1980;36: Johnson DG, Syme WC, Matlak ME, Black RE, Herbst SJ. Gastro-oesophageal reflux and respiratory disease: the place of the surgeon. Aust N Z J Surg 1984; Sontag S, OConnell S, Greenlee H, et al. Is gastroesophageal reflux a factor in some asthmatics? Am J Gastroenterol 1987;82: Tardif C, Nouvet G, Denis P, Tombelaine R, Pasquis P. Surgical treatment of gastroesophageal reflux in ten patients with severe asthma. Respiration 1989;56: Ribet M, Pruvot FR, Mensier E, Ghoch K, Rousseau B. Gastro-oesophageal reflux and respiratory disorders treated by Hill's procedure. Eur J Cardiothorac Surg 1989;3: Perrin-Fayolle M, Gormand F, Braillon G, et al. Long-term results of surgical treatment for gastroesophageal reflux in asthmatic patients. Chest 1989;96: Larrain A, Carrasco E, Galleguillos F, Sepulvida R, Pope CE 11. Medical and surgical treatment of nonallergic asthma associated with gastroesophageal reflux. Chest 1991;6: 133CL Sontag S, OConnell S, Khandelwal S, et al. Most asthmatics have gastroesophageal reflux with or without bronchodilator therapy. Gastroenterology 1990;99: Sontag SJ, Schnell TG, Miller TQ, et al. Prevalence of esophagitis in asthmatics. Gut 1992;33: Castell DO. Asthma and gastroesophageal reflux. Chest 1989;96: Perrin-Fayolle M. Gastroesophageal reflux and chronic respiratory disease in adults. Clin Rev Allergy 1990;8: Mansfield LE. Gastroesophageal reflux and respiratory disorders. Compr Ther 1992;18:& Johnson LF, Rajagopal KR. Does intraesophageal acid trigger bronchial asthma? No, but maybe yes. Chest 1989;96: Tan WC, Martin RJ, Pandey R, Ballard RD. Effects of spontaneous and simulated gastroesophageal reflux on sleeping asthmatics. Am Rev Respir Dis 1990;141: Ekstrom T, Tibbling L. Esophageal acid perfusion, airway function, and symptoms in asthmatic patients with marked bronchial hyperreactivity. Chest 1989;96: Ramenofsky ML, Leape LL. Continuous upper esophageal ph monitoring in infants and children with gastroesophageal reflux, pneumonia and apneic spells. J Pediatr Surg 1981;16: Gustafsson PM, Kjellman N-IM, Tibbling L. Bronchial asthma and acid reflux into the distal and proximal oesophagus. Arch Dis Child 1990;65: Tuckman DN, Boyle IT, Pack AI, et al. Comparison of airway responses following tracheal or esophageal acidification in the cat. Gastroenterology 1984;

6 1034 DONNELLY ET AL Ann Thorac Surg DISCUSSION DR TOM R. DeMEESTER (Los Angeles, CA): I congratulate Dr Donnelly and colleagues on an excellent study, nicely presented, on the very important topic of diagnosing chronic aspiration from occult gastroesophageal reflux. To keep us current in this field, it is important that these presentations are made in this forum. I have some questions. First, I am pleased to see that you have control data regarding normal subjects and would like you to expand on that a little. Obtaining these control studies is hard. I am sure that getting individuals to volunteer to have their tracheal ph monitored is difficult, but very critical to your conclusions. How confident are you of your control data and this consistent tracheal ph? Is alkaline reflux that occurs after surgical procedures on the stomach or the esophagus also a problem, and is your technique applicable to alkaline reflux as well? Give us a little more insight as to how well the patients tolerate the probe. DR BERRISFORD: Thank you very much for your comments, Dr DeMeester. Regarding the controls, this was a study done by the chest physicians, not by us. In fact, I think most of these patients had asthma, but they did not have severe reflux symptoms. They were not entirely normal patients, and I agree it is going to be very difficult to find a normal group who will subject themselves to tracheal ph monitoring. As regards the alkaline reflux, what we would like to do now is to run as large a population as we can through this test, and as you yourself did, stratify the tracheal ph and find out what really is abnormal and what really is normal. We know that about 80% of asthmatic patients have significantly abnormal gastroesophageal reflux and about 40% have esophagitis, but we do not know how many asymptomatic refluxers would be improved by antireflux operation. We need to do this study next, and as you say, controls will be very important. The procedure is tolerated very well. The patients do get an initial foreign tody sensation and some coughing after they have woken up, but within an hour these patients are very comfortable and none of them have complained of serious problems afterward. DR THOMAS R. J. TODD (Ottawa, Canada): I also enjoyed your report and congratulate you on exploring alternative ways of evaluating these people to determine if indeed they are aspirating. I have been interested in that for a long time, particularly in patients receiving assisted ventilation in the intensive care unit, but there are a couple of things that bother me that I think you and your colleagues are in a unique position to examine. First, if we look at the patients you showed us, it appears as if they are still refluxing after operation-perhaps not for the same prolonged period, and maybe that is the determinant of aspiration, but it makes you begin to wonder if indeed patients with asthma plus gastroesophageal reflux do not have a glottic dysfunction that you have not actually measured. I think you have a unique opportunity to examine that. The other point is that it would be interesting to know what is the time course of decreases in ph in the trachea after you do the procedure. We know that general anesthesia does lead to glottic dysfunction that can last for several hours, some people think up to 18 to 24 hours. Finally, we know that glottic dysfunction may be exacerbated by the presence of even a small catheter through the cricothyroid membrane. Before doing your studies, have you examined by use of thin barium swallows whether these patients have a glottic dysfunction induced by your catheter? DR BERRISFORD: To answer the first question, the patients have normal episodes of gastroesophageal reflux; that is, the total time the esophageal ph is less than 4 is not abnormal, and they did not have abnormal DeMeester scores postoperatively. So they are still refluxing but not pathologically. Regarding the second question, I have heard this problem of glottic dysfunction mentioned before. I do not know that barium studies would help a lot, because microaspiration has not been shown previously on radiographic studies. The slide I put up of 10 control patients did show an initial lower ph, but that was because these ph probes were put in under local anesthesia. We have not found any such changes at the beginning of the recording if the probes were inserted under general anesthesia, despite an endotracheal tube being passed initially. To answer the question whether or not the ph probe causes glottic dysfunction, we really need to have another leg to the control studies to find out whether or not this is the case.