Risk Factors for Acute Postoperative Sort Throat (POST) After Supraglottic Device Use

Background. To identify risk factors for acute postoperative sort throat (POST) after general anesthesia using a supraglottic airway device (SAD) in adults undergoing elective tympanoplasty. Methods. The medical records of 1424 adults who underwent elective tympanoplasty under general anesthesia using an SAD were reviewed retrospectively. Patients received one of four SADs at the discretion of the anesthesiologists: exible reinforced LMA (FLMA), Ambu AuraFlex FLMA, or two local brand devices (Tuoren FLMA or Tuoren Esophageal Drainage LMA). POST upon discharge from the postanesthesia care unit (PACU) was measured using a visual analog sale (VAS) and categorized as no pain, mild pain, and moderate to severe pain. Data regarding potential risk factors for POST were collected from the medical records. Results. The mean patient age was 43.24 years; 622 patients were male and 802 were female. The overall incidence of POST during the PACU stay was 38.1%. Female sex, certain types of SAD, higher intracuff pressure, and longer duration of surgery were independent risk factors for POST on multivariate logistic regression analysis. Compared to FLMA, both the Tuoren FLMA and the Tuoren Esophageal Drainage LMA were risk factors for POST. The Tuoren Esophageal Drainage LMA was the strongest independent risk factor for moderate to severe POST. elective tympanoplasty anesthesia. The Tuoren Esophageal Drainage LMA was the strongest predictor of moderate to severe POST.


Data collection
This study was a retrospective chart review analysis of adults, aged 18 to 70 years, who were received general anesthesia with an SAD from July 2018 to December 2019 at the otologic center of our institution. The study included 1424 patients with an American Society of Anesthesiologists physical status of I-II who underwent elective tympanoplasty under general anesthesia with an SAD during the study period. Patients with preoperative sore throat or those whose had a tracheal tube were excluded from the study.

Anesthesia
All patients received sevo urane-based or propofol-based general anesthesia with an SAD. Inhaled nitrous oxide was not used. The type of anesthesia was based on the anesthesiologists' preference.
Mechanical ventilation was used throughout the procedure.
No premedication was administered before anesthesia induction. Routine monitoring, including noninvasive blood pressure, electrocardiography (lead II), pulse oximetry, and end-tidal carbon dioxide, was used for each patient. All patients underwent preoxygenation, followed by intravenous induction with remifentanil 0.5 µg kg − 1 and propofol 1.5-2 mg kg − 1 . Rocuronium 0.6 mg kg − 1 or cisatracurium 0.1 mg kg − 1 was administered at induction to facilitate SAD insertion. Following induction, a SAD was placed when the consultant determined that an adequate depth of anesthesia (e.g., no motor response to jaw thrust) had been reached. All SADs were inserted by an anesthesiologist with more than 1-year experience using SADs. Lidocaine gel was applied to the posterior surface of the SAD as a lubricant prior to insertion. The technique for SAD insertion was guided by both the manufacturers' recommendations and the anesthesia providers' routine practice. Following insertion, all SADs were in ated with a 20-mL B. Braun™ syringe (B. Braun Medical Inc., Bethlehem, PA, USA) to achieve an effective airway seal, as determined by listening for an audible leak during manual ventilation up to 20-cm H 2 O airway pressure. After a seal was achieved, the anesthesia providers reattached the B. Braun™ syringe to the SAD cuff and allowed the plunger to equilibrate with the SAD cuff pressure. The resulting cuff pressure was measured by manometry (VBM, Medizintechnik, Suiz, Germany) with the patients' head in the supine, neutral position (cephalic neutral position). The cuff pressure was then measured after positioning the patients' head for surgery, with the head turned to the side opposite the operative ear (cephalic lateral position).
Adequacy of ventilation was assessed by observing thoracoabdominal movement during manual ventilation and by visualizing the capnography end-tidal carbon dioxide tracing. If the SAD was not seated properly or an adequate seal was not achieved, the device was repositioned.

Outcomes
Relevant information was retrieved from the medical records. Data were collected regarding factors that may contribute to POST: age, sex, height, weight, type of SAD, size of SAD, use of an oral airway device at any time, number of SAD insertion attempts, SAD intracuff pressure, oropharyngeal leak pressure, volume of gas injected into the SAD cuff, anesthesia regimen (inhalation or intravenous), and duration of surgery. We also collected information about pharyngolaryngeal complications. This information was obtained by nurses in the postanesthesia care unit (PACU), who conducted interviews in all patients at the time of discharge from the PACU.
The primary outcome was the incidence of POST at discharge from the PACU. The severity of POST was measured with a visual analog sale (VAS) ranging from 0 to 10. We de ned a VAS of 0 as no pain, 1 to 3 as mild pain, and 4 to 10 as moderate to severe pain. The secondary outcomes were SAD intracuff pressures (after allowing passive recoil of the syringe attached to the SAD cuff) with the head in both the cephalic neutral and cephalic lateral positions and other pharyngolaryngeal complications, such as dysphagia or dysphonia. Dysphagia was de ned as di culty in, or pain provoked by, swallowing (yes/no). Dysphonia was de ned as di culty speaking or pain while speaking (yes/no).

Statistical analysis
Continuous data were described as mean ± standard deviation, and categorical data were described as number (percentage). Age, weight, height, body mass index (BMI), and volume of gas injected into the SAD cuff were compared between POST groups using analysis of variance. Cuff pressure in the cephalic neutral position, cuff pressure in the cephalic lateral position, duration of surgery, airway seal pressure in the cephalic neutral position, and airway seal pressure in the cephalic lateral position were compared using the Wilcoxon rank-sum test. Cochran-Mantel-Haensel test was used to explore associations between POST and sex, SAD type, SAD size, number of SAD insertion attempts, and anesthesia regimen.
For multivariate analysis, a partial proportional odds model based on logistic regression analysis was used because SAD type was considered to not meet the assumption of proportional odds. All variables with a P value < 0.05 in univariate analysis were included in the multiple logistic regression analysis. In the multivariate regression model, the dependent variable was POST, divided into three groups: no pain, mild pain, and moderate to severe pain. Age, sex, BMI, height, SAD type, SAD size, duration of surgery, cuff pressure in the cephalic lateral position, and cuff pressure in the cephalic neutral position were the independent variables. Regression coe cients and odds ratios (ORs) with 95% con dence intervals (95% CIs) were determined. All statistical analyses were performed using SAS 9.4. (SAS Institute, Inc., Cary, NC, USA).

Results
Of the 1424 adults included in the study, 622 were male and 802 were female. The mean age was 43.2 ± 14.4years. Most patients (99.1%) received a size 3 or 4 SAD, with 13 (0.9%) receiving a size 2.5 or 5 SAD.
The mean BMI was 18.6 kg/m 2 for the 3 patients receiving a size 2.5 SAD and 24.6 kg/m 2 for the 10 patients receiving a size 5 SAD.
The incidence of POST at discharge from the PACU was 38.1%. A total of 877 patients (61.6%) had no sore throat, 474 patients (33.3%) had a mild sore throat, and 73 patients (5.1%) had a moderate to severe sore throat. No dysphagia or dysphonia was recorded.
In univariate analyses, eight POST risk factors differed signi cantly between the three groups: age, sex, height, SAD type, SAD size, cuff pressure in the cephalic lateral position, cuff pressure in the cephalic neutral position, and duration of surgery (Table 1).

Discussion
The overall incidence of POST upon discharge from the PACU was 38.4% after the use of one of four types of SADs in patients undergoing elective tympanoplasty. When classi ed by severity, the incidence of mild POST was 33.3% and the incidence of moderate to severe POST was 5.1%. No patient developed dysphonia or dysphagia. Female sex, certain types of SAD, higher intracuff pressure in the cephalic neutral position, and longer duration of surgery were independent risk factors for developing POST.
Sore throat is a common complaint after surgery. It affects patient satisfaction and can in uence activity after discharge. SADs offers an alternative to traditional tracheal intubation, with the potential bene t of reducing the risk of sore throat [5][6][7]. However, the overall incidence of POST after using an SAD can be as high as 42% [8]. Our 38.4% incidence of POST of any severity at discharge from the PACU is consistent with this previously reported percentage.
Various pre-and intraoperative factors in uence the incidence of postoperative pharyngolaryngeal complications. Tracheal intubation, female sex, younger age, pre-existing lung disease, prolonged duration of anesthesia, and presence of a blood-stained tracheal tube on extubation are associated with the greatest risk in adults. Insertion technique, SAD choice, and intracuff pressure may contribute to the incidence of sore throat after anesthesia with an SAD [9]. Careful LMA insertion techniques are of paramount importance for preventing airway trauma and POST [6]. In our study, all SADs were inserted by experienced anesthesiologists using standard insertion techniques, and we found that risk factors for POST in the early postoperative period (at PACU discharge) were female sex, higher intracuff pressure in the cephalic neutral position, longer duration of surgery, and certain types of SAD.
Previous studies have shown that women have a higher risk of POST than men after use of a tracheal tube [10][11][12]. Likewise, Jaensson et al. reported a higher incidence of POST after LMA use in women than in men (26% vs. 6%) [13]. Our study con rms these ndings. By contrast, Grady et al. [14] reported similar incidences of POST after LMA use between men and women. The lower incidence of POST in men may be explained by differences in pharyngeal anatomy between sexes [15]. The design of currently used SADs may be better suited to the anatomical features of the male pharynx rather than the female pharynx.
Overall, the incidence of POST differs minimally between rst-and second-generation SADs, with the exception of a lower incidence with the i-gel, which may be attributed to the absence of an in atable cuff [9,16,17]. In the current study, we found that both local brand SADs, the rst-generation Tuoren FLMA and the second-generation Tuoren Esophageal Drainage LMA, contributed to POST. Compared to FLMA, both Tuoren devices were signi cant independent risk factors for POST. The Tuoren Esophageal Drainage LMA was also the strongest independent risk factor for moderate to severe POST, with an OR of 10.75 (95% CI, 4.82-23.99). These results may be at least partially explained by the composition of the SADs.
The FLMA and Ambu AuraFlex LMA are both made of polyvinyl chloride (PVC), while the Tuoren SADs are composed of medical-grade silicone. Our ndings suggest that silicone may cause more mucosal irritation than PVC. The Tuoren Esophageal Drainage LMA also has a relatively xed and curved structure with an in atable airway cuff, and as a second-generation SAD, it has two drain channels that emerge proximally as separate ports and enter a chamber beside the cuff bowl. It has a more strongly tapered leading tip than the other SADs used in our study, which may have caused more mucosal injury.
An increasing body of evidence has con rmed that high SAD intracuff pressures contribute signi cantly to laryngopharyngeal complications, including POST [8,9,16,18], and adjustment of intracuff pressure reduces the incidence of moderate POST [8,19,20]. As SAD intracuff pressure increases, perfusion of the airway mucosa progressively decreases, resulting in postoperative pharyngolaryngeal complications [21]. Use of manometry can reduce pharyngolaryngeal complications by 70%, compared with routine care without manometry [4]. In our study, the mean cuff pressures in the cephalic neutral position were 27.23 cm H 2 O, 28.75 cm H 2 O, and 30.33 cm H 2 O in the no pain, mild sore throat, and moderate to severe sore throat groups, respectively. The differences between these pressures were statistically signi cant, suggesting that as intracuff pressure increases, the incidence of POST also increases. Multivariate analysis also identi ed higher intracuff pressure in the cephalic neutral position as a signi cant risk factor for POST, although it appeared to be a weak predictor (OR, 1.02; 95% CI, 1.00-1.04).
Grady et al. [14] reported that use of a large SAD was associated with a higher incidence of POST in both sexes. In their study, patients were randomized to receive either a large SAD (size 5 in males and size 4 in females) or a small SAD (size 4 in males and size 3 in females). Use of a large SAD was associated with a four-fold increased risk of developing sore throat on the rst postoperative day, despite the measured SAD intracuff air pressure being higher in patients with a small LMA. In our study, we found no signi cant association between POST and SAD size, which is likely because all patients were managed with a small SAD (almost exclusively size 4 in males and size 3 in females). Furthermore, intracuff pressures were relatively low (< 35 cm H 2 O) and mechanical ventilation was used in our study, in contrast to the higher intracuff pressures (> 60 cm H 2 O) and use of spontaneous ventilation in the Grady et al study. Grady et al. [14] also reported that males with POST underwent longer procedures than men who did not develop POST. When considering all patients (males and females), the risk of developing POST increased by 33% (OR, 1.33; 95% CI, 1.03-1.71) for each 10-min increase in surgery duration. In the current study, we likewise identi ed longer surgery duration as a risk factor for POST.
Although pressure neuropraxia from the SAD cuff can result in dysphonia, SAD-related dysphonia is rare [22][23][24]. Seet  This study has some weaknesses. First, since data were obtained from a single center, in which local brand (Tuoren) devices were used, the results may not be applicable to other regions. Second, the incidence of POST was assessed only at discharge from the PACU. POST may persist for 2 to 3 days and require additional attention from healthcare providers [26]. However, Seet et al. found that the incidence of POST was low (3.1%) in their pressure-limiting group at 24 hours postoperatively [4]. Third, we did not monitor intracuff pressures intraoperatively. Burgard et al. [15] found that LMA cuff pressures increased signi cantly during the rst 60 minutes of anesthesia because of nitrous oxide diffusion into the cuff. In our study, we avoided nitrous oxide and used only an air/oxygen mixture. Vasanth Karthik et al. observed no change in intracuff pressures 1 hour after SAD insertion when using an air/oxygen mixture [27]. Furthermore, it has been suggested that use of intermittent positive pressure ventilation rather than intracuff pressure may be responsible for postoperative pharyngolaryngeal complications [28]. Further study regarding the association between different ventilation modes and POST is warranted.

Conclusion
In summary, this study provides additional information to assist practitioners, especially in China, in choosing the most appropriate SAD. The ideal SAD would be easy to use and provide a high airway leak pressure and cause no POST. If selection criteria are based on POST, our results suggest that the Tuoren Esophageal Drainage LMA is the least preferable among the four SADs evaluated.

Declarations
Acknowledgements. The authors express special gratitude to the physicians and nurses of the departments that cooperated with this study.
Availability of data and materials. The datasets used /or analyzed during the current study available from the corresponding author on reasonable request.
Author Contributions. YZH and KZC wrote the manuscript and were involved in data gathering, analysis, and interpretation. They made equal contribution. SX conceived of the study, and participated in its design and coordination and the drafting of manuscript. All authors read and approved the nal manuscript.