Comparison of the LMA-Classic ™ with the New Disposable Soft Seal Laryngeal Mask in Spontaneously Breathing Adult Patients

After institutional review board approval (Catharina Hospital Eindhoven) and informed patient consent, we prospectively studied 200 unselected, consecutive, supine anesthetized patients who were scheduled to undergo elective surgery with the use of a laryngeal mask. Patients aged 18–80 yr were included in the study. Patients were excluded if they had an American Society of Anesthesiologists physical status class of III to V; required surgery in a nonsupine position or had to undergo oral or nasal surgery; had a preoperative sore throat; had a known or predicted difficult airway, chronic obstructive pulmonary disease, or respiratory tract pathology; or were considered unsuitable for surgery with the use of a laryngeal mask.

Patients were allocated randomly, using computer-generated tables of random numbers, to a size 4 laryngeal mask and received either the silicone rubber reusable LMA-Classic ™ or the disposable Portex Soft Seal LM. All laryngeal masks were inserted for a wide range of routine general surgery, orthopedic, urologic, gynecologic, and plastic surgery. Patients were unaware of the airway device used. The LMA-Classic ™ masks used were normal clinical-use products in our hospital. The Soft Seal LMs were all new, as they were used only once. At our institution, the LMA-Classic ™ is the standard supraglottic airway device. All laryngeal masks in both groups were inserted by the same anaesthesiologist (A. V. Z.; personal experience: more than 2,500 LMA-Classic ™ masks and 80 Soft Seal LMs inserted before the study).

Routine preinsertion tests of the cuff for leaks, as recommended by the manufacturers, were done immediately before use. The tip of the laryngeal mask as well as the posterior aspect of the mask was coated with a water-soluble lubricant with 2% lidocaine (Instillagel; Farco-Pharma, Cologne, Germany), which is the normal practice in our institution. The cuff was inflated before insertion with ambient room air at 10 mmHg (15 cm H²O) above atmospheric pressure, which is the standard practice in our institution and advocated by many authors. 3,11–18 

No premedication was used, and all patients were preoxygenated for 3 min before intravenous induction. Intravenous induction of general anesthesia consisted of 1 μg/kg fentanyl and 3 mg/kg propofol. The lungs were ventilated manually with sevoflurane (end-tidal 2%-3%) added to a mixture of 66% N²O in oxygen administered via  a facemask without the use of an oral airway. The laryngeal mask was inserted 1 min after completion of induction and after loss of lash reflex and the relaxation of the jaw. 19An insertion attempt was defined as placement of the laryngeal mask in the mouth and withdrawal of the device from the mouth. Successful placement of an effective airway and adequate ventilation was confirmed by the presence of resistance to further downward movement, observing chest wall movement with manual ventilation, listening to the escape of gas from the mouth, detection of a square wave trace capnograph during manually assisted ventilation, and observation of movement of the reservoir bag on spontaneous ventilation. The ease of insertion of the laryngeal mask was graded as (1) very easy insertion at the first attempt with no resistance; (2) easy insertion at the first attempt with little resistance; (3) some difficulties, but successful at the second attempt; or (4) not successful.

When the black line (LMA-Classic ™) or blue line (Soft Seal LM) was centrally positioned, the laryngeal mask pilot balloon cuff pressure was adjusted to 45 mmHg (60 cm H²O) as recommended, 20–24using a hand-held pressure gauge to monitor the initial cuff pressure in cm H²O (Endotest; Rüsch, Kernen, Germany) and a pressure transducer (AS/3; Datex-Ohmeda, Helsinki, Finland) to measure the cuff pressure continuously in mmHg. The laryngeal mask was securely fixed using tape with a bite block. Subsequently, the seal around the larynx by the laryngeal mask was checked for any air leak by ventilating the patient against an outlet valve pressure of 15 mmHg (20 cm H²O). 25For that reason, we closed the expiratory valve of the circle system at a fixed gas flow of 3 l/min and noted the airway pressure at which equilibrium was reached. Any air entering the stomach was noted when measuring oropharyngeal leak pressure by listening over the epigastrium with a stethoscope. When the laryngeal mask was considered to function adequately, allowing the patient to breathe spontaneously (sevoflurane–nitrous oxide–oxygen) with end-tidal carbon dioxide levels between 4.0 and 6.0, a fiberscope (StyletScope; Nihon Kohden, Tokyo, Japan) was inserted with its tip located at the inner aperture of the laryngeal mask. The position of the epiglottis was assessed using a standardized 25–26four-point scale (1–4) as follows: grade 4, only vocal cords seen; grade 3, vocal cords and posterior epiglottis seen; grade 2, vocal cords and anterior epiglottis seen; grade 1, vocal cords not seen. Grade 3 and 4 positions were considered correct, whereas grades 1 and 2 were considered suboptimal positions. 25All fiberoptic scoring was made within approximately 2 min of the laryngeal mask insertion. The presence of any aspirated or regurgitated material in the hypopharynx and the incidence of laryngeal spasm or other adverse events (coughing, retching, breath holding) were noted.

Anesthesia was maintained with a fresh gas flow of 3 l/min using 66% N²O in oxygen and sevoflurane (end-tidal sevoflurane levels between 1.5% and 2%). Fentanyl, 0.05 mg, was added intravenously as required. When surgically indicated, but well after the insertion of the laryngeal mask airway and fiberoptic score, 0.5 mg/kg rocuronium was used as the neuromuscular blocking drug. Standard patient monitoring consisted of inspiratory and expiratory concentrations of nitrous oxide, oxygen, and sevoflurane; electrocardiogram; heart rate; capnography; noninvasive arterial blood pressure; peripheral oxygen saturation measured by pulse oximetry; respiratory rate; and tidal volume.

At the end of surgery, nitrous oxide and sevoflurane were discontinued to allow the patient to resume full recovery. When the protective reflexes had returned to normal, the laryngeal mask was removed synchronously with deflation of the cuff to 45 mmHg (60 cm H²O) to avoid secretions entering the larynx or provoking laryngeal spasms. 21On removal of the laryngeal mask, the presence or absence of blood was noted. Any blood on the mask or in the secretions on the mask was considered positive. Patients were requested to grade any sore throat at 2 and 24 h after surgery. Postoperative evaluations, including the evaluation of blood on the laryngeal mask, were performed by another anesthesiologist.

The following parameters of each patient were collected: sex, age, weight, height, type of surgery, time to insert the laryngeal mask, number of insertion attempts to achieve a satisfactory airway, and duration of laryngeal mask in situ . Laryngeal mask cuff pressures were monitored continuously, while intracuff pressure limitation was not practiced, with time 0 the moment the pilot cuff was inflated to a pressure of 45 mmHg until the moment nitrous oxide was switched off.

We estimated the final cuff pressure difference between the LMA-Classic ™ and the Soft Seal LM to be 35% lower in the latter. Sample size calculation was performed using Sampsize version 2.0 (Sample Size tables for clinical studies; Blackwell Science Ltd., Oxford, United Kingdom). Ho:μ1 =μ2 (no difference between population means, two-sided hypothesis). The study had a power of 90% (β= 10%; α= 0.05), and the standardized difference was 0.35 (SD, 0.86). The estimated sample size was 88 patients per group. Two groups of 100 patients were chosen. Statistical analysis was performed using SPSS version 9.0 for Windows (SPSS Inc., Chicago, IL). Mann–Whitney U and Kruskal-Wallis nonparametric data analyses were used for analysis of variables accordingly in addition to the chi-square test. All data are presented as mean ± SD. Statistical significance was considered at P < 0.05.

The two studied groups were similar in age, sex, weight, and height (table 2). Facemask ventilation before the laryngeal mask insertion in all patients was easy, and no patient needed an oral airway to maintain the airway. In two patients (one in each group), a continuous leak of the airway existed because a good seal could not be obtained, and intubation was chosen. Data from these patients have been excluded, and the cases were repeated. In all other patients, adequate insertion of the laryngeal mask was obtained. The overall laryngeal mask insertion time was within 20 s and successful on the first insertion attempt in 97% of the LMA-Classic ™ group and 95% of the Soft Seal LM group (table 2). In 3% of the LMA-Classic ™ group and 5% of the Soft Seal LM group, laryngeal mask insertion was only successful at the second attempt. The insertion of the Soft Seal LM became easier with time and experience. The laryngeal mask could be passed in the hypopharynx in all patients. In no case in either group did insertion of the laryngeal mask provoke laryngeal spasm or retching. No significant differences were seen in the hemodynamic and respiratory data in patients of both groups, and there were no events of desaturation during the trial. None of the patients needed a muscle relaxant at the time of insertion of the laryngeal mask. However, in 8% of the patients in the LMA-Classic ™ group and 9% of the patients in the Soft Seal LM group, muscle relaxants were used because of surgical indications, but only well after the insertion and the clinical and fiberoptic evaluation of the laryngeal mask. Ventilation was well tolerated, and its use did not show any adverse effect, nor did it affect the results.

The mean laryngeal mask cuff pressure increased from 45 mmHg to 100.3 ± 23.6 mmHg in the LMA-Classic ™ group and from 45 mmHg to 46.8 ± 3.6 mmHg in the Soft Seal LM group (P < 0.001) at the end of surgery (table 2). The presence of macroscopic blood on removal of the mask (4%) was only seen in the LMA-Classic ™ group (table 2).

Sore throat at 2 h in the postoperative period was seen more frequently (P < 0.05) in the LMA-Classic ™ group (20.5%vs.  10.2%;table 3). Mild sore throat, including dryness of the mouth, not requiring any treatment, was mentioned in 11 patients in the LMA-Classic ™ group versus  10 in the Soft Seal group. In the LMA-Classic ™ group, six patients had moderate pain, and three patients had severe pain of the throat. No statistically significant difference between the two groups at 24 h was observed (table 3).

At the time we checked the fiberoptic anatomic position of the laryngeal mask, we did not see any suspected or proved episodes of aspiration, regurgitation, or vomiting, although frequently, material was emerging from the trachea lumen into the hypopharynx, and saliva could be seen collecting in the hypopharynx. Although in 6% of the LMA-Classic ™ and 4% of the Soft Seal LM cases the vocal cords could not be made visible by fiberoptic examination, satisfactory anesthesia was never a problem. The outcome of the fiberoptic examination did not reach statistical difference between the two groups (table 4).

This is the first report on the use of a new disposable laryngeal mask, the Soft Seal LM, in which we conducted a clinical comparison of the reusable LMA-Classic ™ and the new disposable Soft Seal LM size 4 and found that both laryngeal masks provided an adequate airway and similar clinical performance in spontaneously breathing patients. The Soft Seal LM was less traumatic, causing less sore throat in the early postoperative period.

It is a common practice to inflate the laryngeal mask cuff to the maximal recommended volume, 3while intracuff pressures are rarely measured and maintained in clinical practice. This often results in high intracuff pressures, up to 250 mmHg, 17possibly leading to increased mucosal pressure, 2,27pharyngolaryngeal trauma, 13–14,27and a suboptimal seal. 3In the past, we used this high recommended cuff volume practice until we found out that the majority of our patients had cuff pressures greater than 120 cm H²O (90 mmHg) immediately after insertion of the laryngeal mask. Many authors now recommend inserting the laryngeal mask with the cuff partially inflated. 3,11–18 

During anesthesia, a significant increase in cuff pressure is seen because of the presence of nitrous oxide, carbon dioxide, or other gases and because of the warming up of these gases. Avoiding high intracuff pressures and volumes during nitrous oxide anesthesia results in a better outcome for the patient (more optimal seal and less pharyngolaryngeal trauma). 3,14–17Some investigators show that high laryngeal mask pressures do not increase pharyngeal mucosal injury in dogs, 28while others suggest reducing the cuff volume to a “just seal” situation. 29 

We, too, believe that insertion of the laryngeal mask airway with the cuff partially inflated is equally successful to insert the laryngeal mask and results in a lower incidence of sore throat and pharyngeal mucosal bleeding. This might be the result of the presentation of a softer leading edge to the posterior pharyngeal wall.

The incidence of sore throat using the LMA-Classic ™ in our study was 20%, which is identical to the findings of Brimacombe et al.  15when the cuff is partially inflated, compared to 42% if the cuff was fully inflated after insertion. Our incidence of sore throat with the Soft Seal LM is even lower (i.e. , 10%), which is similar to the incidence of sore throat after the use of a facemask (i.e. , 8%) in the studies of Brimacombe et al.  15and Dingley et al.  30 

Using the standard approach technique, with the insertion of a noninflated LMA-Classic ™, the reported incidence of sore throats ranges between 21.4% and 30% according to Wakeling et al. , 1328.5% in the study of Dingley et al. , 30and up to 42% as measured by Brimacombe et al.  15The sore throat incidence is less when the cuff is partially inflated compared to noninflated cuffs, i.e. , 4.1%versus  21.4% as demonstrated by Wakeling et al.  13 

The Soft Seal LM is translucent, allowing easier visualization of any secretions inside it. On the disposable laryngeal mask, the size of the product and maximum cuff inflation volume are printed on the pilot balloon, allowing the clinician to recognize its size once in situ  (fig. 1). Although the IDs of the LMA-Classic ™ and the Soft Seal LM are identical (11.0 mm), in vitro  a 7.0-mm cuffed endotracheal tube can be inserted into the disposable laryngeal mask, while only a 6.0-mm endotracheal tube can be inserted into the shaft of the LMA-Classic ™.

Patients with a “high” or “anterior” larynx did not always show a complete view of the larynx inlet by fiberoptic examination through the laryngeal mask in our study. However, the number of patients whose vocal cords were not visible was not significantly different between the two groups, confirming that insertion of both laryngeal masks results at least in the same position. All laryngeal masks used in this study resulted in adequate ventilation. This confirms that the absence of epiglottic bars have no deleterious effects on the performance of the laryngeal mask. 31Blockage of the airway tube by the epiglottis may be avoided by the deeper bowl, as is the case with the disposable laryngeal mask.

We only compared size 4 laryngeal masks because this was the only size of disposable laryngeal mask available at the time of our study. For obvious financial reasons, we compared new disposable laryngeal masks with reusable LMA-Classic ™. The LMA-Classic ™ may become more permeable to nitrous oxide the more it is used. The anatomic position of the LMA by fiberoptic evaluation was restricted to the view obtained from the tube of the laryngeal mask, limiting the results. We did not evaluate whether the use of a different lubricant on the LMA before insertion resulted in more or less pharyngolaryngeal trauma.

In conclusion, disposable laryngeal masks are an acceptable device to replace the reusable LMA-Classic ™, resulting in a good laryngeal seal and similar clinical performance. The absence of epiglottic bars seems not to have deleterious effects on the performance of the laryngeal mask. Our study also shows that measuring cuff pressures of laryngeal masks may be advisable when reusable LMA-Classic ™ masks are used, but this is unnecessary with the disposable Soft Seal LM.

The authors thank Marlies Schijven, M.D., Ph.D., Hsc. (Statistician, Department of Surgery, Catharina Hospital, Eindhoven, The Netherlands), for statistical analysis; Berry Bijsterveld and Jos Verbeek (Technicians, Catharina Hospital) for technical assistance; Guy Van Dael and Willem Binnendijk (Registered Medical Photographers, Audiovisual Department, Catharina Hospital) for audiovisual assistance; and the operating certified registered nurse anesthetists (Catharina Hospital) for their general assistance in this study.