Introduction

Total joint arthroplasty (TJA) is the most effective treatment for treating end-stage joint diseases, which can relieve pain, restore joint function, and improve quality of life1,2. The demand for TJA is rapidly rising, with projected increases of 71% and 85% for primary knee and hip arthroplasties, respectively, by 20303. The elderly population constitutes the majority of TJA patients, presenting with complex comorbidities such as hypertension and impaired lung function4,5. These factors contribute to substantial perioperative management challenges, particularly in the context of anesthesia and airway management.

Endotracheal intubation (EI) is the standard method for securing ventilation during general anesthesia and is generally effective in ensuring intraoperative safety6,7. However, EI is associated with pressure-induced stimulation that can trigger increases in catecholamine levels, leading to tachycardia and hypertension8,9,10. Excessive hemodynamic fluctuations during anesthesia can result in postoperative impairment of other organ functions, including acute kidney injury and myocardial injury, especially in elderly patients11,12,13. Additionally, EI and mechanical ventilation during anesthesia will cause postoperative discomfort, including extreme thirst, pharyngeal pain, and swallowing difficulties14,15. These adverse effects can negatively impact recovery and contribute to patient dissatisfaction after surgery16.

Laryngeal mask airway (LMA) has emerged as a viable alternative for routine and difficult airway management, which can alleviate the hemodynamic response during insertion and removal and reduce the incidence of postoperative odynophagia17,18. However, air leakage and displacement of the LMA during surgery are still concerns for anesthesiologists, particularly in the lateral position and extensive surgical maneuvers of TJA19. LMA has been increasingly assessed as an alternative to endotracheal intubation (EI) across surgical specialties. A 2024 RCT by Liu B et al. (n = 180) using lung ultrasound found that LMA significantly reduced early post-operative atelectasis compared with EI during lower-limb orthopaedic procedures and other surgeries, although haemodynamic data were not reported20. Hartmann B et al. 2004 further showed that LMA shortened anaesthesia-controlled time in outpatient orthopaedic cases, but pharyngeal morbidity was not assessed21. Consequently, whether the benefits of LMA on airway-related morbidity and hemodynamic stability persist in the specific context of total joint arthroplasty remains uncertain.

To address these concerns, we conducted this randomized controlled trial to compare the effects of LMA and EI on hemodynamic stability, postoperative pharyngeal discomfort, and other complications in patients undergoing TJA. By evaluating these outcomes, this study aims to provide valuable insights into optimizing airway management for improving perioperative care and patient satisfaction in TJA. By integrating the Glasgow Voice Disorder Assessment Scale (GRBAS) voice-quality scale with continuous hemodynamic monitoring, our study is the first to evaluate airway strategies within the specific context of TJA. Exclusion of patients with anticipated difficult airways enhances the external validity of our findings and facilitates translation into routine clinical practice.

Materials and methods

Patients

This single-center, randomized controlled trial was approved by the Ethics Review Board of our institution (19/07/2023/, No.2023–1090) and registered on the Chinese Clinical Trial Registry (14/08/2023, ChiCTR2300074701). A consecutive series of patients undergoing scheduled total hip or knee arthroplasty at our institution between August 2023 and August 2024 for the treatment of end-stage joint disease were enrolled. Eligible participants were aged between 18 and 80 years and provided written informed consent.

Patients were excluded if they met any of the following criteria: (1) preoperative C-reactive protein levels > 10 mg/L or interleukin-6 levels > 15 pg/mL; (2) preoperative hemoglobin < 110 g/L or abnormal coagulation; (3) preoperative American Society of Anesthesiologists classification > III; (4) preoperative level of transaminase or creatinine more than three times the upper limit of normal; (5) cardio- or cerebrovascular incidents during the previous six months; (6) difficult airway identified during preoperative assessment based on Mallampati grade, modified Cormack-Lehane scale, and Intubation Difficulty Scale score ≥ 5 points (All evaluations were performed by the same anesthesiologist with more than 10 years of experience); (7) history of pharyngeal or laryngeal surgery; (8) pharyngeal or laryngeal diseases; or (9) systemic diseases requiring treatment prior to TJA.

Patients were withdrawn from the study after enrollment if they or their legal representatives withdrew consent, or if a serious adverse event occurred.

Randomization, blinding and interventions

After enrollment, each patient will draw an opaque envelope containing a computer-generated random number. The envelope is given to a researcher, who records the patient’s grouping but does not disclose it to the patient. On the day of surgery, the researcher will inform the anesthesiologist of the patient’s group assignment and establish the corresponding artificial airway. The EI group received endotracheal intubation for anesthesia, while the LMA group used a laryngeal mask airway to maintain intraoperative ventilation during anesthesia. Data collection will be performed by other researchers who are not involved in the randomization process. In this way, the data collectors and analysts, as well as the patients themselves, are unaware of the group assignments. The attending anesthesiologist could not be blinded to airway allocation; this represents a potential performance bias, although outcome assessors and the statistician were unaware of group assignment.

Anesthesia and surgery

All patients in our study received general anesthesia. Induction was performed with intravenous bolus injections of propofol (1.5–2.5 mg/kg), sulfentanyl (0.25–0.5 μg/kg), and atracurium (0.15–0.2 mg/kg). Anesthesia was maintained with continuous intravenous infusion of remifentanil (0.05–0.12 μg/kg·min) and continuous inhalation of sevoflurane. Sulfentanyl (5ug) was administered every hour during the procedure. Preoxygenation and denitrification were achieved by administering 100% oxygen for 3–5 min after induction, maintaining expired oxygen concentration ≥ 90% using a tight-fitting face mask. In the EI group, a video laryngoscope was used for visual endotracheal intubation (Sheridan, Teleflex Medical Co., Ltd., USA), while the LMA group used a LMA (SLIPA-2G Laryngeal Mask Airway, Hangzhou Fushan Medical Appliances Co.,LTD., China) made of latex-free thermoplastic elastomer with an integrated bite block and gastric-drainage portal to establish the artificial airway. For the EI group, a size ID 7.0–8.0 endotracheal tube was typically used for women (inserted to a depth of approximately 21 cm), while a size ID 7.5–8.5 tube was used for men (inserted to a depth of approximately 22 cm). The choice of LMA size was based on body weight: size 3.0 for patients weighing 30–50 kg, size 4.0 for those weighing 50–70 kg, size 5.0 for those weighing 70–100 kg, and size 6.0 for those weighing > 100 kg. LMA insertion was performed using the standard index-finger insertion technique by the same attending anesthesiologist with over 100 LMA insertions. Correct placement was confirmed by bilateral chest rise, end-tidal carbon dioxide waveform, and peak airway pressure < 20 cmH2O. The LMA was then secured to the patient’s cheek using a 3 M elastic tape crisscross technique. All artificial airways were established by the same experienced anesthesiologist. None of the patients received surface anesthesia prior to airway establishment. Both groups received the same ventilation mode. Blood pressure was continuously monitored during induction and recorded every 5 min throughout the procedure using an electrocardiogram and an upper arm sphygmomanometer. Fingertip oxygen saturation was also recorded continuously. Target blood pressure was maintained using intravenous hydroxylamine, ephedrine, and nicardipine.

All patients were administered intravenous tranexamic acid (2 g) 15–30 min prior to surgery, followed by a second dose of intravenous tranexamic acid (1 g) 8 h postoperatively. In addition, all patients received intravenous cefuroxime (1500 mg) 30 min before surgery for infection prophylaxis. Patients allergic to cefuroxime were administered 600 mg of clindamycin instead. All TJAs were performed by the same surgeon and his team, using a posterolateral approach with an uncemented acetabular component and a biologic femoral stem for hip arthroplasty, and a medial parapatellar approach with a cemented prosthesis for knee arthroplasty. No patient received a urinary catheter during the procedure or a drainage tube postoperatively. Postoperative analgesic strategies were consistent for all patients.

Outcomes and follow-up

The primary outcome was the overall incidence of sore throat at 6 h after surgery assessed by the 4-point visual rating scale (VRS) (score ≥ 1). All other dimensions (GRBAS sub-scores, pharyngeal irritation, 24 h and 48 h data) were secondary/descriptive endpoints and were not included in the sample-size calculation. Hoarseness was assessed using the Grbas scale, pharyngeal pain was evaluated using a 4-point VRS, and pharyngeal irritation was also evaluated using a 4-point scale. The GRBAS scale is composed of the following components22: Grade (G) reflects the severity of hoarseness; Roughness (R) describes a rasping or rattling voice; Breathiness (B) refers to a whispery voice; Asthenia (A) denotes a weak voice; and Strain (S) indicates an effortful or constricted voice. Each parameter in the GRBAS scale is rated on a 4-point scale: 0 (normal), 1 (slight), 2 (moderate), or 3 (severe). The 4-point scale was defined as follows: 0, no pain; 1, mild pain; 2, moderate pain; and 3, severe pain.

Secondary outcomes included hemodynamics parameters, the incidence of airway complications, and other postoperative discomfort including nausea, vomiting, dyspnea, and dysphagia. Since there is no standardized definition for hemodynamic instability, we defined it as either a minimum mean arterial pressure less than 55 mmHg, a blood pressure changes of more than 30% from baseline for 5 min, or a blood oxygen saturation less than 92%23. Data on outcomes were collected before and multiple times after surgery, up to 14 days after discharge.

Statistical analysis

Data were analyzed using SPSS version 27.0 (IBM, Armonk, NY, USA) using P < 0.05 as a definition of statistical significance. Continuous variables were presented as mean ± standard deviation while categorical variables were presented as n (%). Pairwise comparisons of continuous variables were performed using the Mann–Whitney U test if the data were skewed or exhibited unequal variances, and the independent-samples t-test was used otherwise. Pairwise comparisons of categorical variables were performed using the chi-squared test or Fisher’s exact test.

The minimal sample size was estimated based on a previous report, which indicated that the incidence of postoperative sore throat 2 h after surgery was 48%24. Sample size was calculated solely for the 6-h sore-throat incidence, assuming a 48% event rate in the EI group and a clinically relevant absolute reduction of 24% (to 24%) in the LMA group. Sample size estimation using PASS 2021 software revealed that at least 42 patients were required to detect this minimal difference with a test power of (1 − β) = 0.8 and a significance level of α = 0.05. To compensate for 10% dropouts, 50 patients were enrolled per group.

Results

Among the 103 patients initially screened for enrollment, 3 were excluded due to meeting the exclusion criteria, resulting in a final cohort of 100 patients, with 50 patients in the EI group and 50 patients in the LMA group (Fig. 1). Baseline characteristics were comparable between the two groups (Table 1).

Fig. 1
figure 1

Flowchart of patient enrollment and inclusion in analysis. EI, Endotracheal intubation; LMA, Laryngeal mask airway.

Full size image
Table 1 Demographic, surgical, and anesthetic characteristics of the patients.
Full size table

The incidence of hemodynamic instability was significantly higher in the EI Group compared to the LMA group at the time of artificial airway establishment (40% to 4%, P < 0.001), at the start of surgery (62% to 14%, P < 0.001), during the post-extubation stage (78% to 24%, P < 0.001) (Table 2). Hypertension was the most common type of hemodynamic instability in both groups. No significant differences were observed in hypotension, hypoxemia, arrhythmias, or heart rate fluctuations between the two groups.

Table 2 Hemodynamic parameters and types of hemodynamic instability of two groups during anesthesia induction, surgery and recovery stages.
Full size table

The EI Group showed a significantly higher mean heart rate during anesthesia induction (P < 0.001) (Table 2). During surgery, the LMA group had significantly higher minimum intraoperative diastolic blood pressure (P = 0.029). No other significant differences in hemodynamic parameters were observed between the two groups.

The EI Group exhibited significantly higher incidences of hoarseness (P = 0.003), roughness (P < 0.001), asthenia (P = 0.021), strain (P = 0.022), and pharyngeal irritation (P = 0.033) at 6 h postoperatively compared to the LMA group (Table 3). At 24 h postoperatively, the EI group also showed significantly higher rates of hoarseness (P = 0.021), roughness (P = 0.014), and breathiness (P < 0.001). No significant differences were observed in other parameters.

Table 3 Incidence and severity of post-operative pharyngeal discomfort and GRBAS voice-quality scores at 6 h, 24 h and 48 h after total joint arthroplasty.
Full size table

The EI Group had a significantly higher incidence of dysphagia (P = 0.001) at 6 h postoperatively compared to the LMA group (Table 4), while no significant differences were found in other adverse events.

Table 4 Secondary adverse events within the first 48 h after total joint arthroplasty: nausea, vomiting, dysphagia, dyspnoea and airway complications.
Full size table

Discussion

This randomized controlled trial suggests that the use of LMA in TJA can provide more stable hemodynamics during anesthesia, as well as reduce postoperative pharyngeal discomfort. Our results also indicate that the safety of LMA use in TJA is comparable to that of EI. These findings may provide an improved ventilation strategy for airway management in orthopedic anesthesia.

Postoperative sore throat is a common complaint, affecting 10–70% of intubated surgical patients25,26,27. It is primarily caused by intubation-induced trauma to the laryngeal and airway mucosa, as well as stretching of neck muscles and ligaments28. In our study, the incidence of sore throat after EI was as high as 46%, consistent with previous research. Postoperative sore throat can negatively impact patient satisfaction and leave lasting discomfort even after discharge29,30. In the context of enhanced recovery after surgery, postoperative pharyngeal discomfort can delay oral intake, prolong hospital stay, and hinder the patient’s return to normal life31,32,33. While interventions such as topical steroids or lidocaine have been explored to alleviate postoperative pharyngeal discomfort, their effectiveness remains controversial34,35,36. Murugaiyan et al. reported that anesthetics cream reduced the incidence of sore throat, but Tanaka et al. conducted a meta-analysis and found that the effect of lidocaine is affected by multiple factors such as concentration and usage37,38. Topical glucocorticoids also have unclear dose–effect relationships and potential risks of hyperglycemia39. Therefore, this study did not use these interventions to eliminate confounding factors. Furthermore, the hemodynamic changes induced by tracheal intubation pose greater risks, including myocardial and acute kidney injury, and may even increase postoperative mortality40,41. Given that most TJA patients are elderly with multiple comorbidities, they are particularly susceptible to organ damage caused by hemodynamic instability.

LMA, as a supraglottic airway device, offers greater comfort for patients. In our study, the incidence of postoperative sore throat with LMA was 44% in patients who used the LMA, and the incidence of sore throat was reduced to 12% 6 h after TJA, which dropped to 12% at 6 h postoperatively, indicating that LMA causes less pharyngeal irritation. LMAs are less invasive, easier to insert, and associated with reduced airway trauma, making them well-suited for enhanced recovery pathways42. Hemodynamic stability during LMA insertion and removal is particularly advantageous for elderly patients, especially those with cardiovascular and cerebrovascular conditions43,44. Our study corroborates these benefits, demonstrating that LMA has a lesser impact on hemodynamics compared to EI. In addition, LMA use has been shown to reduce the risk of postoperative atelectasis, further supporting its utility in elderly patients20,45.

Despite these advantages, LMA leakage remains a concern. The leakage rate for LMAs is approximately 1.1%, rising to 5% in obese patients46,47. Over 60% of patients experiencing LMA leakage develop significant hypoxia, hypercapnia, or airway obstruction, with 42% suffering from leak-related hypoventilation46. Postural changes during surgery also increase the risk of LMA failure46. In TJA, concerns arise from patient positioning and surgical maneuvers, such as lateral positioning during total hip arthroplasty and the impact of percussive force when stabilizing prostheses. These factors can potentially cause LMA displacement or leakage. However, our study found that TJA does not significantly increase the risk of LMA leakage or dislocation. Contrary to these concerns, our data indicate that TJA per se does not increase the incidence of LMA leakage or dislodgement. The absence of device displacement or clinically relevant air leak in the present cohort is most likely attributable to four peri-operative measures: (1) rigorous selection of the SLIPA-2G size based on body weight; (2) secure crisscross tape fixation; (3) maintenance of adequate anaesthetic depth (bispectral index 40–60) to minimize laryngeal tone; and (4) deliberate avoidance of excessive patient movement by the surgical team. It is plausible that omission of similar precautions in earlier investigations accounts for their higher rates of LMA-related morbidity.

Our findings should be interpreted with caution in light of several limitations. One is that our sample was relatively small, which may limit the generalizability of our results. We also did not explore the effects of different modes of ventilation in TJA. Finally, this study did not include patients with difficult airway, thus the conclusions are not applicable to this population. A further limitation is that the attending anesthesiologist was necessarily aware of airway allocation, which may introduce performance bias; however, outcome data were collected by independent, blinded assessors. These shortcomings should be addressed in future work, preferably with larger samples from multiple centers.

Conclusions

The use of the LMA in TJA significantly reduces postoperative pharyngeal discomfort and provides better hemodynamic stability compared to the use of the EI. Additionally, LMA ventilation does not increase the incidence of airway complications during TJA. These findings offer valuable clinical insights that could enhance post-operative recovery quality and improve patient satisfaction, highlighting LMA as a viable and effective alternative to EI for airway management in orthopedic surgeries.