Although the risk of death and morbidity due to the pulmonary aspiration of gastric contents may have decreased by 50% in the past two decades [1-3], this remains a feared complication during any anesthesia, especially in patients undergoing emergency procedures. The two most common methods of securing an unsusceptible airway during induction of anesthesia are still rapid sequence induction and the Sellick maneuver . However, these prophylactic procedures also entail certain risks and dangers such as cardiovascular instability from the sudden administration of high doses of anesthestics and/or succinylcholine, esophageal rupture, as well as failure of tracheal intubation, which can result in hypoxemia and subsequent aspiration [2,5,6].
Gastroesophageal regurgitation occurs in anesthetized patients due to an increase of intragastric pressure and/or relaxation of the lower esophageal sphincter (cardia sphincter). Therefore, we recently developed a nasogastric tube to prevent the reflux of gastric contents by blocking the cardia with a balloon. The purpose of the present animal investigations was to test the ability of the new tube to prevent experimental gastroesophageal regurgitation. In addition, the function of the nasogastric balloon catheter was tested on awake, healthy volunteers under various conditions to provoke vomiting. Further tests were performed on patients with an increased risk of aspiration to determine whether the conventional induction of anesthesia with ventilation via a mask is possible with the tube in place.
The disposable and traction-elastic nasogastric tube (Aspisafe Registered Trademark, Braun, Melsungen, Germany) counteracts gastric reflux by means of a gastric balloon which is inflated after the transnasal insertion of the nasogastric tube into the stomach and the aspiration of the gastric contents and is then placed under tension at the cardia in such a way that the reflux of residual gastric contents into the esophagus is prevented Figure 1. The position of the nasogastric tube is fixed using a foam-cushioned nose stopper which can be displaced and fixed along the aspiration tube, so that the tension adjusted tube ensures a close fit of the balloon at the cardia. The existent tension of the aspiration tube or the pressure of the balloon on the cardia is controlled by a pressure-monitoring instrument, such as a conventional cuff pressure measuring device Figure 1 or a small disposable manometer. The appropriate tensile stress of the nasogastric tube (200 g) and the ellipsoid form and size of the gastric balloon were tested in preliminary investigations on anatomic models and in the context of autopsies.
The experimental animal investigations were conducted with animal ethics committee approval. A total of 20 pigs weighing 23 +/- 1 kg served as experimental animals. The anesthesia was induced in the animals by administration of metomidate (Hypnodil Registered Trademark, Janssen, Neuss, Germany, 15 mg/kg body weight intraperitoneally) and azaperone (Stressnil Registered Trademark, Bayer, Leverkusen, Germany, 30-50 mg intraperitoneally). After intubation, the intragastric tube was inserted transorally into the stomach. Then the gastric balloon was inflated in 12 animals, and the balloon was apposed onto the cardia under the proper tension. The pressure in the gastric balloon was measured with a cuff pressure device. The position of the probe was fixed by pushing the nasal stopper forward against two intraorally fixed towel clamps which simulated the nasal orifice. In eight additional pigs (control group), the cardia was not blocked. After the positioning of the nasogastric tube, a microtip catheter was pushed forward into the stomach via the main lumen of the nasogastric tube to measure the intragastric pressure (IGP). A second microtip catheter was inserted into the region of the lower esophageal sphincter under endoscopic control to measure the intraesophageal pressure (LEP). To detect a gastroesophageal reflux, fluids which served as gastric contents and which had been stained with methylene blue dye were instilled into the stomach via the nasogastric tube and an endoscope was placed into the upper part of the terminal esophagus to allow continuous imaging of the region by means of a video camera and monitor. Afterward, four different provocation maneuvers were performed in the following sequence: 1) gastric fluid filling with 0.5 L of a 0.9% sodium chloride solution stained with methylene blue dye (M1); 2) 45 degrees head-down positioning (M2); 3) external gastric compression (M3); 4) drug-induced vomiting (M4) depending on the effect obtained with ipecacuana (20 mL via nasogastric tube), apomorphine, and/or carbachol. The animals, which initially breathed spontaneously, were then mechanically ventilated, and anesthesia was continued with fentanyl, metomidate (0.3-0.5 mg/kg via perfusor), and nitrous oxide (60%-70%). Laparatomy was performed in the animals and the pylorus was ligated with suture material. The laparotomy wound was then closed again. After this preparation, two further provocation maneuvers were performed: 1) gastric fluid filling with 2 L of a 0.9% NaCl stained with methylene blue dye (M5); 2) external gastric compression (M6).
With approval of the local ethics commission, 26 healthy, adult, and awake test subjects with maintained protective reflexes (14 women, 12 men; aged 21-37 yr) participated in the study. All test subjects received written (detailed information sheet) and oral information in advance concerning the scope and purpose of the investigation and had given their written consent to participate in the investigations. After topical anesthesia of the nasopharyngeal space with lidocaine gel, the nasogastric balloon tube was inserted in the conventional manner via the nose into the stomach of the test subjects who had previously consumed at least 1 L of mineral water. A microtip catheter was present in the lumen of the tube with a manometer at the gastric end. After having placed the nasogastric tube into the stomach, the balloon was inflated and apposed to the cardia under tension as previously described. The balloon pressure was monitored with a conventional cuff pressure measurement device. To test the functional capability of the nasogastric tube, each test subject self-induced vomiting several times by digital stimulation of the posterior wall of the pharynx while lying down on the side and in the sitting position. The intragastric pressure was displayed on a monitor by means of the microtip catheter in the stomach. At the end of the experiment, the gastric balloon was deflated and the provocation of vomiting was repeated in the same way. In 6 of the 26 test subjects, the provocation of vomiting was performed using five different tube tensions ranging from 0-700 g. The tensile stress of the suction tube between nose and cardia was measured by a tension measurement probe placed around the tube between nose stopper and nose. In these subjects the balloon pressure was measured using conventional Statham transducers. The analog data were recorded simultaneously and stored on magnetic tape.
With approval of the local ethics commission, 42 patients (aged 19-89 yr) have been investigated. These patients all had an increased risk of aspiration and were scheduled for various general and emergency surgery procedures with endotracheal intubation. After thorough information and written consent, the nasogastric tube was positioned in the usual manner and an attempt was then made to aspirate any gastric contents via the tube prior to inflating the balloon and blocking the cardia in the conventional way. After the patients breathed oxygen for a few minutes, anesthesia was initiated with etomidate (0.2 mg/kg intravenously [I.V.]), fentanyl (2 micro gram/kg I.V.), and vecuronium (0.1 mg/kg I.V.). After cessation of spontaneous breathing, the patients were manually ventilated with 100% oxygen via a transparent anesthetic mask until an adequate relaxation had occurred (>2 min). Care was taken to keep the inspiratory ventilation pressure under 20 cm H2 O. The patients were then endotracheally intubated without any haste, the endotracheal cuff was inflated, and afterward the tension on the nasogastric tube was released and the gastric balloon was deflated. The nasogastric tube was left in place during the operation. In the termination phase, the cardia was blocked once again with the technique described and this protective mechanism maintained until tracheal extubation was completed.
The values presented are means +/- SEM. The data were analyzed with repeated-measurement analysis of variance, followed by paired and unpaired Student's t-test with Bonferroni corrections. A P value of <0.05 was considered statistically significant.
Under the various provocation maneuvers, reflux of gastric contents or blue staining of the lower esophagus could not be observed in any of the animals investigated (n = 12) with cardia blockade. Prior to the provocation measures (control), the LEP and IGP did not differ significantly (Table 1, Group A). After filling the stomach with 0.5 L of fluid (M1), the IGP increased significantly. A reversal of the pressure gradient (pressure difference significant) resulted from the filling of the stomach, since the LEP increased simultaneously by an average of only 2 mm Hg (not significant). The subsequent head-down positioning of the animals (M2) brought about a further increase of IGP with an increased pressure difference (on average 5 mm Hg) between IGP and esophageal pressure. The IGP increased to an average of 84 mm Hg due to external compression of the stomach (M3), whereas the LEP increased only to an average of 48 mm Hg. IGP and LEP still differed significantly during this provocation maneuver. The pressure conditions were similar under provoked vomiting (M4), although a reflux of gastric contents into the esophagus was not observed even under these conditions. After ligature of the pylorus and filling of the stomach with 2 L of fluid (M5), the IGP exceeded the LEP significantly. The additional compression of the stomach under these conditions (M6) brought about a marked increase in IGP, whereas the LEP increased significantly less. In the group without cardia blockade (n = 8), LEP and IGP reacted similarly (Table 1; Group B). However, in the head-down position, the LEP and IGP did not differ significantly. Thirty-seven of 48 provocation maneuvers performed with an unblocked cardia led to gastroesophageal reflux.
In 20 of the 26 test subjects investigated, the vomiting caused a significant increase in intragastric pressure from 10 +/- 3 to 59 +/- 18 (28-107) mm Hg. A reflux of gastric contents was not observed in any of the test subjects. After deflation of the balloon, a reflux of gastric contents could be induced using the same provocation maneuvers in 18 of the 20 test subjects. The intragastric pressure increased significantly from 10 +/- 3 to 51 +/- 11 (24-93) mm Hg. In the six test subjects equipped with a tension monitoring device to measure the tensile stress directly, the condition of the nasogastric balloon under tension at the pressure level of 50 cm H2 O yielded a tensile stress of 209 +/- 24 g. The balloon and intragastric pressures measured simultaneously under five different tension levels are summarized in Table 2. A reflux of gastric contents was not observed in any of the test subjects at a tension level >or=to 150 g. At a tensile tension of > 300 g, all test subjects had a sensation of discomfort.
In the 42 patients investigated with a balloon nasogastric tube, ventilation via a mask upon induction of anesthesia could be performed without any difficulties. During the ventilation via a mask, the inspiratory pressure could be maintained < 20 cm H2 O in all patients. The time between induction of anesthesia and endotracheal intubation was 171 +/- 9 s. There were no indications for regurgitation or aspiration during the induction and termination phase in any patients. Complications associated with insertion, residence, or withdrawal of the tube were not observed in the patients investigated.
The present investigation on anesthetized pigs has shown that the nasogastric balloon tube prevents gastroesophageal reflux of stomach contents under provocation of vomiting and regurgitation. In comparison, the control study on animals with an unblocked cardia showed that a reflux of stomach contents into the esophagus could be induced by approximately 75% of the provocation maneuvers. Predominantly significant pressure differences in terms of a gastroesophageal pressure gradient, one of the prerequisites for gastroesophageal reflux to occur, were encountered in both groups under the different and aggressive provocation maneuvers. Although only an IGP in excess of 20 mm Hg is regarded as critical for triggering reflux , regurgitation can also occur within this limit, as documented by the findings of the control group. The high overall reflux incidence of the control group may be partly explained by the resident nasogastric tube (with deflated balloon), which probably served as a track for the regurgitation of stomach contents.
The increases in IGP occurring under the various provocation maneuvers were associated with slight increases of pressure in the region of the lower esophageal sphincter. These pressure increases probably occur secondarily to the increases of intragastric and intraabdominal pressure, and are thought to be based on a physiologic protective mechanism [8,9]. A reactive increase in pressure in the region of the lower esophageal sphincter has also been measured in response to an increase of IGP due to muscle fasciculations after the administration of succinylcholine [10,11]. Without provocation maneuvers, a significant difference between IGP and LEP was not detected in the two groups, even though the average pressure was higher in the region of the lower esophageal sphincter. Similar findings have been recorded by Smith et al. , who observed a decrease of both the LEP as well as the IGP under anesthesia. Although there was a greater decrease in the LEP, it still remained slightly above the IGP. In the awake and healthy patient, the lower esophageal sphincter is the major barrier to gastroesophageal reflux. Here, the resting pressure is approximately 20 mm Hg higher than in the gastric fundus .
Since the results of animal experiments can be extrapolated to humans only with reservation, the functional capability of the device was tested under provocation of vomiting on awake and healthy volunteers. The experiments show that the insertion of the nasogastric balloon tube in awake subjects prevents a reflux of stomach contents under the provocation of vomiting. The IGP observed under the provocation of vomiting corresponded both to those observed in animal experiments and the values described in the literature . Our early clinical experiences indicate that ventilation via a mask with the nasogastric tube is practical, since it provides the prerequisites for unhurried induction by avoiding interference with the circulation and hypoxia. In ventilation via a mask, care must be taken so that the inspiratory ventilation pressure does not exceed 20 cm H2 O. However, this recommendation is based only on theoretical considerations according to which higher ventilation pressures counteract the sealing pressure of the balloon on the cardia by inadvertent esophageal ventilation, and may limit the function of the balloon.
We obviously do not know whether any of the patients we tested would have vomited, regurgitated, or aspirated during routine rapid sequence or awake intubation anesthetic sequencing. A power analysis using the recent literature figures for the incidence of aspiration [2,3,12] shows that we cannot conclude, or even hint or imply from the small number of our patients, that the tube reliably prevents either vomiting, regurgitation, or aspiration of gastric contents during induction or any time during anesthesia. Our first clinical results showing no aspirations are completely consistent (P > 0.9) with the small incidence of clinically important aspirations among patients with a raised risk of aspiration varying from 11.2/10,000 to 22.8/10,000 cases [2,3,12]. In order to show a reduction of the aspiration incidence up to zero with the new device, subsequent studies with a minimum of 5000 patients are required. Unfortunately, the recent literature does not specify the incidence of intraoperative vomiting or regurgitation without aspiration. This is probably higher since reflux does not always result in aspiration.
The likelihood of inducing vomiting during insertion of the device might be the same as that of conventional nasogastric tubes, because the cuff consists of a soft and pliable material which fits closely around the catheter shaft during the insertion of the device. The risk of aspiration during insertion is given only when the protective reflexes of the upper respiratory tract are impaired or abolished. The nasogastric tube should not be used in patients with sliding hiatal hernias or stomach tumors. All complications which may also occur by the insertion of a conventional stomach tube, particularly mucosal injuries and/or bleeding of the nose, the pharynx, the esophagus, and the stomach, are also to be expected by the insertion of the nasogastric tube presented here. As is also shown by the present investigation, vomiting is accompanied by brief high IGP, which does not result in rupture of the stomach. Even in the absence of pressure compensation, and under very high IGP in excess of 100 mm Hg, ruptures did not occur as shown by the animal experiments.
Mucosal damage which may theoretically result from the appositional contact pressure of the balloon on the cardia and may lead to a subsequently reduced perfusion of the mucosa in the region of the cardia is improbable, since pressure is exerted on the mucosa via the balloon only for a short time. A too-intense pressure of the balloon on the gastric mucosa, or an excessive tension on the cardia, is prevented by the external pressure monitoring device. The manometer may also aid in detecting an improper positioning in the esophagus, balloon leakages, or other factors which may result in inadequate appositional contact pressure of the balloon on the cardia. The tensile elasticity of the nasogastric tube compensates for any displacements of position owing to the flexibility and mobility of the parts of the body situated between the nasal stopper and the balloon.
To summarize, the present experimental findings in animals and test subjects document the function of the new nasogastric balloon tube and its ability to prevent gastroesophageal reflux by occluding the cardia with a balloon under controlled tension. Initial clinical experiences show that ventilation via a mask in combination with this tube enables the induction of anesthesia with little discomfort and stress for the patient. However, the first clinical studies could not demonstrate a decrease in the incidence of aspiration of gastric contents due to the small number of patients investigated. Further extensive clinical investigations are required to show whether the new nasogastric tube can indeed improve the outcome of patients in danger of aspiration by preventing negative outcomes related to vomiting, regurgitation and aspiration. The tube might then allow a routine low-risk induction of anesthesia and endotracheal intubation in patients with an increased risk of aspiration of gastric contents and may also be used during the termination of the anesthesia.
The technical assistance of Andrea Oldag and Doris Drose is gratefully acknowledged. The author is also grateful to Volker Klute of the B. Braun Co., Melsungen, for his valuable technical support during the investigation of the test subjects.