Laryngeal mask airway vs endotracheal tube to facilitate bedside percutaneous tracheostomy in critically ill patients: a prospective comparative study.SP Ambesh, PK Sinha, M Tripathi, P Matreja
Department of Anaesthesiology and Critical Care Medicine, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow - 226 014, India. , India
Correspondence Address: Source of Support: None, Conflict of Interest: None PMID: 12082319
Source of Support: None, Conflict of Interest: None
BACKGROUND: Inadvertent puncture of endotracheal tube (ETT) cuff and accidental tracheal extubation are the potential life threatening complications during percutaneous dilatational tracheostomy (PDT). As an alternative to ETT, the laryngeal mask airway (LMA) has been used sporadically but no large study is available on its use. AIMS: To study the use of LMA during PDT on controlled ventilation in critically sick patients and compare its advantages and disadvantages in relation to ETT. SETTINGS AND DESIGN: This prospective, randomised, comparative study was conducted in intensive care unit of a super-speciality, tertiary care hospital. METHODS AND MATERIAL: The bedside PDT was performed in 60 critically ill patients using Ciaglia's kit. The patients received controlled ventilation of lungs either through ETT or LMA. The feasibility of the procedure and complications were compared in two groups. RESULTS: In LMA group, 33% of patients suffered with potentially catastrophic complications, e.g., loss of airway, inadequate ventilation of lungs leading to significant hypoxia, gastric distension and regurgitation. In ETT group there were 6.6% incidence of ETT impalement, 6.6% cuff puncture and 3.3% accidental tracheal extubation. Despite the technical difficulties (cuff puncture, etc.) the ETT technique was basically safe, whereas despite of the theoretical advantages the LMA technique rendered some patients at considerable risks of hypoxia and gastric regurgitation-aspiration. CONCLUSION: The LMA does not provide safe patent airway to facilitate bedside PDT in critically sick population on controlled ventilation. The ETT is safer for controlled ventilation and should be continued to secure the airway for this purpose until a better alternative is available.
Keywords: Adolescent, Adult, Aged, Comparative Study, Critical Illness, Female, Human, Intensive Care Units, Intubation, Intratracheal, instrumentation,methods,Laryngeal Masks, Male, Middle Age, Prospective Studies, Trachea, injuries,Tracheostomy, adverse effects,instrumentation,
Percutaneous dilational tracheostomy (PDT) has gained wide acceptance as a bedside procedure in the management of critically ill patients in intensive care unit (ICU). The procedure is based on the Seldinger Technique that involves entering tracheal lumen with a needle and cannula, passing a guidewire through the cannula and creating a tracheal stoma using either the Portex guidewire dilating forceps (the Griggs method) or a series of graded tapered dilators (the Ciaglia Method). Various studies,,,,,, have demonstrated the superiority of the guidewire technique over conventional surgical tracheostomy. Inadvertent puncture of the endotracheal tube (ETT) cuff or impalement of ETT with the tracheal puncture needle is perhaps the greatest concern during the formation of PDT. The standard recommendation before formation of PDT is that the ETT be withdrawn until the cuff lies immediately below the vocal cords. In spite of this common practice, endotracheal cuff punctures and impalement of ETT have been reported.,,,,, The use of laryngeal mask airway (LMA) has been suggested as an alternative to ETT and has been used sporadically,, but no large study is available to compare its efficacy in providing safe and patent airway during the PDT.
In a prospective and randomised study we have used the LMA to study whether laryngeal mask techniques would
indeed facilitate PDT in critically ill patients and what the respective advantages and disadvantages of the compared technique are? The experiences of PDT under LMA in relation to ETT are presented.
The institute’s ethics committee for human studies approved the study. Following written informed consent from the next of kin, sixty consecutive patients of both genders, aged 18 to 75 years receiving ventilatory support for a variable period were randomly allocated into two groups of 30 each. Randomisation of patients was done using computer generated table of random numbers. The patients with unstable cervical spine, gross thyroid swelling, severe coagulopathies (prothrombin time >20 seconds) and positive end expiratory pressure (PEEP) support of more than 5 cm H2O were excluded from the study. Before conducting this study the authors had experience of successful placement of LMA in over one thousand patients in anaesthetic practice and more than two hundred cases of successful bedside PDTs in intensive care unit of this institute.
A night before the PDT nasogastric (NG) tube feed was stopped. At the start of procedure the gastric contents were aspirated and the NG tube was removed. The monitoring included were continuous electrocardiogram lead II, invasive/ non-invasive arterial blood pressure, airway pressure, pulse oximetry, end-tidal carbon-di-oxide (EtcO2) and respiratory plethysmography. Midazolam (2-5 mg), sodium thiopental (200-300mg) and vecuronium (4-6 mg) were administered intravenously to all of our patients. During the procedure all patients breathed 100% oxygen on pressure-controlled ventilation (Servo Ventilator 300; Siemens Elema, Sweden) of lungs. Patients were positioned supine with a sandbag under the shoulders to make the neck moderately extended and relevant landmarks easily identifiable.
In group I (the ETT group) the ETT was readjusted under direct laryngoscopy to lay the endotracheal cuff immediately below the vocal cords and one of the anaesthetist colleagues held the ETT in-situ. In-group II (the LMA group) patients, the ETT was replaced with the LMA (size 3 in females and size 4 in males) under cricoid pressure. The LMA cuff was inflated with approximately 20 to 30 ml of air and adequacy of seal between cuff and pharyngeal wall was judged by the auscultation of anterior neck and ballottement of pilot balloon. The air entry into the lungs was checked with respiratory plethysmography and chest auscultation on intermittent positive pressure ventilation.
The PDT was performed using Ciaglia’s technique3 with commercially available Ciaglia percutaneous tracheostomy introducer set (William Cook Europe A/S, Denmark). Following placement of tracheal cannula the ETT was moved about 1 cm up and down. If the movement of ETT was associated with the movement of tracheal cannula it was thought that the impalement of endotracheal tube has occurred. The cannula was withdrawn and reinserted after adjustment of the ETT. In the event of ETT cuff puncture the pressure control was increased to adjust the loss of tidal volume. During the procedure, observations were made for complications like puncture of ETT cuff, impalement of ETT, accidental extubation of trachea or displacement of LMA, gastric insufflation of air and regurgitation of gastric contents. The ETT cuff puncture was detected on appearance of bubbling sounds from the oral cavity and loss of tidal volume. Sudden decrease in airway pressure, loss of tidal volume and fall in Etco2 were the signs of loss of airway (e.g. ETT or LMA dislodgement). Gastric air insufflation of stomach was diagnosed on percussion and auscultation of epigastric region on control ventilation. Regurgitation was diagnosed by presence of gastric contents in oral cavity. Pulse oximetry was used to detect arterial oxygen desaturation (SpO2 <95%). Following extubation, the ETT was examined for puncture marks. After successful placement of the tracheostomy tube the controlled ventilation of lungs was resumed through tracheostomy tube and all patients were subjected for bedside chest X-ray to look for surgical emphysema or pneumothorax.
The results were analysed using ANOVA, student’s ‘t’ test and ‘Z’ test. Statistical significance was accepted at the 95% confidence level (p<0.05).
Age, gender, body mass indexes (BMI), nature of illness and pattern of ventilatory support were comparable between the two groups [Table - 1]. In ETT group, two (6.6 %) patients had perforation of the cuff during the initial tracheal puncture; however, the PDT was successfully performed after increasing the pressure control. In two (6.6 %) patients, the ETT was impaled with the tracheal puncture needle and required withdrawal of the needle and readjustment of the ETT for fresh tracheal puncture. Following initial tracheal puncture, one (3.3%) patient had intratracheal bleeding that was detected with the sudden spurt of blood in the ETT. The cuff was immediately deflated and the ETT was readjusted in such a way so that the endotracheal cuff lay at the level of tracheal puncture. The cuff was then inflated to produce the tamponade effect locally. The bleeding stopped within few minutes and PDT was performed in the lower space uneventfully. In one (3.3%) patient the ETT had slipped out of larynx during middle of the procedure and required reintubation.
In LMA group three (9.9%) patients had displacement of LMA during the procedure that were detected immediately and the LMA were repositioned back. In five (16.6%) patients, satisfactory lung ventilation was not possible through the LMA as most of the tidal volume leaked out from its periphery and SpO2 decreased to less than 90%. In these patients, reintubation was done with cuffed ETT and the PDTs were performed uneventfully. In two (6.6%) patients, gastric distension due to air escape led to regurgitation of gastric contents. Fortunately, none of these patients had pulmonary aspiration. However, no difficulty was encountered in dilating the tracheal stoma in any of the patients.
While performing PDT, the tracheal cuff puncture might cause loss of significant amount of tidal volume on controlled ventilation, which can be dangerous in patients having low arterial oxygen tension (paO2<60 mm. Hg). The withdrawal of the endotracheal cuff just below the vocal cords does not guarantee the prevention of tracheal cuff puncture. Even with such a practice cuff perforation and lacerations have been reported.,,,, This is because the anatomic length of the adult human larynx varies from approximately 3.4-4.4 cm. Since, the average length of the ETT (8.0-9.0 mm) cuffs for adults is about 3 cm, and length of the tube beyond cuff also measures about 3 cm, it is possible that tracheal needle puncture can damage the endotracheal cuff or tube. Schwann has found useful to deflate the ETT cuff and reposition the tube so that the inflated cuff sits just below the level of the vocal cords. The cuff is re-deflated for the needle puncture and maintained deflated until the needle is withdrawn from the trachea, leaving only the J-wire in-situ. However, again this technique does not eliminate cuff damage by the subsequent insertion of dilators.
Various authors,, have performed bronchoscopic guided percutaneous tracheostomy. Using direct bronchoscope or bronchoscopic video imaging, the tip of the ETT is retracted to the level of the cricoid cartilage well above the proposed tracheostomy site. The advantage is that both the ends of endotracheal tube as well as the needle puncture site in the midline of trachea can be visualised and PDT under direct vision is performed. But again the maintenance of adequate minute ventilation to the patient is challenging due to leaks around the bronchoscope and endotracheal tube adapter resulting into waste of ventilation and hypoxia. Apart from this, there is potential risk of damage to the bronchoscope optical fibres with the tracheal puncture needle or the dilators, which may not make the procedure cost effective.
In anaesthetic practice, the LMA has been used successfully for spontaneous as well as controlled ventilation of lungs for various surgical procedures and resection of high tracheal stenosis. Tarpy and Lynch have successfully used the LMA in two patients during the PDT procedure. In our experience, the LMA has failed to provide clear airway in significant number of patients (p<0.05) on controlled ventilation. Though, none of our patients had aspiration but there was potential danger of aspiration of gastric contents as two (6.6%) of our patients had regurgitation in tracheostomy position. Brimacombe et al have successfully performed endoscopic percutaneous tracheostomy using size 3 LMA in a sedated and paralysed adult female patient on controlled ventilation but they did not rule out the potential danger of regurgitation-aspiration, loss of airway (dislodgement of LMA) and inadequate ventilation. It is well known that the LMA does not provide airtight seal around the glottis during controlled ventilation of lungs. Some of the air is likely to escape leading to waste of ventilation and subsequent hypoxia in susceptible patients. Further, during the PDT the position of the patient is such where LMA displacement is quite likely. In spite of gastric tube aspiration the stomach of critically ill patients cannot be guaranteed empty and therefore these patients carry high risk for regurgitation and pulmonary aspiration. On the other hand, head and neck extension with cricoid pressure may make insertion of LMA more difficult.
In our study we used LMA size-3 in females and size-4 in males. The selection of correct size LMA can be difficult since the relationship between gender, weight, height and upper airway geometry appears inconsistent. So far no method has been identified that give an easy measurement of potential pharyngeal volume directly correlating with appropriate size of LMA. The instruction manual of LMA recommends that the size-4 should be used in adults of average or large size, and size-3 in children or small adults of more than 25 Kg in weight. Later, following introduction of LMA size-5, the inventor recommended using as large a size of mask as possible. Despite these instructions, size-3 mask is commonly used in females and the size-4 in males., We have used 20-30 ml of air for optimal cuff inflation for a given patient. The amount of air injected was not decided on the basis of gender of the patient or the size of LMA used, rather adequacy of seal achieved around the mask was the sole criterion. Adequacy of seal was decided on the basis of auscultation of anterior neck and ballottement of pilot balloon.
Recently, Asai et al have used larger size of masks (size-4 in females and size-5 in males) in order to avoid air leak from the gap between the mask and pharynx. Though, with this practice, the incidence of air leak was decreased significantly; still 17% of females and 13% of males had the air leak. Voyagis et al used LMA size-4 in females and size-5 in males of European population to achieve better laryngeal seal and recommended gender related formula for proper selection of the size of LMA. The use of LMA size-4 in females and size-5 in males, may decrease the incidence of air leak and complications like pharyngeal wall ischemia and lingual nerve injury due to overinflation of smaller size LMA cuff. But even with such a practice the potential risks of airway loss, gastric distension-regurgitation and hypoxia due to inadequate ventilation in low compliant lungs of this critically sick population cannot be ruled out.
In this study, none of our patients of ETT group had gastric distension or regurgitation due to escape of air into the stomach. The ETT provided safe, patent airway in all patients except one without producing hypoxia. Another important advantage of ETT that we experienced in one of our case is that in the event of inadvertent puncture of a blood vessel, the subsequent intratracheal bleeding can be controlled by repositioning of the endotracheal cuff just at the site of bleeder. The inflated cuff provides tamponade effect locally and prevents further bleeding and thereby safeguards against soiling of lungs with the blood and subsequent hypoxia.
In conclusion, the use of LMA during PDT in critically sick patients renders them susceptible to more potentially
catastrophic complications e.g., hypoxia, airway loss or regurgitation-aspiration. The ETT is superior to LMA for this procedure. The use of LMA to facilitate bedside PDT in critically sick patients on controlled ventilation is not safe and cannot be recommended.
[Table - 1], [Table - 2]