Postoperative pulmonary complications (PPCs) have been inconsistently defined in clinical trials, but they usually include pneumonia, atelectasis, respiratory failure, and reintubation [1]. The incidence of PPCs reportedly varies from less than 2% to 70%, depending on the type of surgery, surgical population, and definition of PPCs [2–6]. PPCs are the most important and independent determinants of 30-day mortality, and nearly 25% of the mdeaths occurring within 1 week after surgery are related to PPCs [3,7]. Additionally, because PPCs cause an increase in morbidity, medical costs, and duration of hospitalization, the identification of its risk factors and application of strategies to reduce PPCs are important clinically.

Many physicians struggle to reduce the risk of increased morbidity and mortality in patients undergoing surgery. Most factors that contribute to the occurrence of PPCs are related to surgery or the patients [8]. Moreover, as risk factors for PPCs, the strength of the evidence for anesthesia-related factors is weaker than that for surgery- or patient-related factors [3,8]. Nevertheless, strong evidence has shown that residual neuromuscular block (NMB) is associated with an increased risk of PPCs [1]. Therefore, it is important to understand the factors related to the field of anesthesia and make efforts to prevent them.

Among the anesthesia-related factors, the use of NMB agents (NMBAs) is known to be associated with increased PPCs as well as residual NMB, since the first report of its contribution to postoperative mortality in 1954. However, the relationship between PPCs and the type of reversal agents for NMB, such as conventional anticholinesterases and the relatively new sugammadex, remains debatable. This review discusses the important research findings on PPCs in the field of neuromuscular research.

PPCs are not a new concept. They have existed since long and remain a subject of interest for all anesthesiologists, physicians, and surgeons involved in perioperative medicine [9–11]. Since there is no fixed definition for PPCs, various definitions have been used over time. The commonly accepted definitions of PPCs include the European Perioperative Clinical Outcome (EPCO) definitions published in 2015 and a new definition published in 2018 [12,13]. In the EPCO definitions, PPCs comprise respiratory infections, respiratory failure, pleural effusion, atelectasis, pneumothorax, and aspiration pneumonitis [12]. On the other hand, the new definition of PPCs announced in 2018 comprises atelectasis detected on computed tomography or chest radiography, pneumonia diagnosed according to the US Centers for Disease Control criteria, acute respiratory distress syndrome according to the Berlin consensus definition, and pulmonary aspiration with no clinical history or radiological evidence [13].

The causes of PPCs are multifactorial, and the related risk factors can be classified as preoperative or intraoperative. Preoperative factors include advanced age, American Society of Anesthesiologists physical status (ASA PS) ≥ II, frailty, chronic obstructive pulmonary disease, recent upper respiratory tract infection, obstructive sleep apnea (OSA), serum albumin level < 30 g/L, alcohol use, delirium, and abnormal chest radiography findings [1,8]. Intraoperative factors include surgical factors such as surgical site (e.g., abdominal, thoracic, neurosurgery, head and neck, or vascular), duration of surgery (> 2 h), and emergency surgery- or anesthesia-related factors such as general anesthesia or regional anesthesia, use of NMBAs, neostigmine administration, residual NMB, sugammadex with supraglottic airway, and no use of peripheral nerve stimulator [1,8,14].

Several risk score models have been proposed to predict PPCs, but there is no “one-size-fits-all” model for the risk stratification for PPCs, and most of them have certain limitations [7,14]. To date, the Assess Respiratory Risk in Surgical Patients in Catalonia (ARISCAT) risk score is the only predictive model that has shown sufficient predictive power in external validation [14]. Notably, only few models among the independent variables of several predictive models for PPCs, including the ARISCAT model, include anesthesia-related factors such as the use and reversal of NMBAs and anesthesia techniques (Table 1) [15–21]. This is probably because surgical factors, such as surgical site, and emergency surgery- and patient-related factors, such as age and underlying diseases, have a greater effect on PPCs than do the anesthetic factors [8,22]. Nevertheless, since the anesthesia-related factors, such as residual NMB, are modifiable and preventable unlike other factors such as age, surgical site, emergency surgery conditions, and surgical duration, understanding the effects of NMB and its reversal on PPCs is clinically relevant.

Residual NMB, often defined by a train-of-four (TOF) ratio < 0.9, is one of the well-known anesthesia-related risk factors for PPCs. It is almost clear that residual NMB is associated with postoperative upper airway muscle dysfunction [23–25]. The main mechanisms of PPCs induced by residual NMB due to the NMBAs used during general anesthesia are respiratory muscle dysfunction, impairment of the hypoxic ventilatory response, and inability to protect the airway during swallowing. These pathological mechanisms may lead to adverse respiratory events such as atelectasis, hypoxia, aspiration, pneumonia, and reintubation [24].

Two representative strategies that anesthesiologists implement to avoid this potentially dangerous residual NMB include quantitative neuromuscular monitoring and appropriate reversal of the NMB. Neuromuscular monitoring in the perioperative period is essential to determine the degree of recovery from NMB through quantitative evaluation and to confirm the presence of residual NMB. A minimal level of NMB (TOF ratio, 0.7–0.9) cannot be detected without quantitative monitoring of the TOF [24,26]. To exclude clinically significant residual NMB, the TOF ratio must exceed 0.9, when recorded with mechanomyography or electromyography, and exceed 1.0 when using acceleromyography (AMG) [27].

Additionally, appropriate NMB reversal requires the administration of a titrated dose of the reversal agent according to the level of the block under quantitative neuromuscular monitoring [28]. Reversal agents for NMB are divided into the following two major categories: anticholinesterases (e.g., neostigmine and pyridostigmine) and sugammadex. Anticholinesterases bind to acetylcholinesterase, which can prolong the effects of acetylcholine and competitively antagonize NMBAs. However, it has a ceiling effect, with no effect above a certain dose, and NMB over a deep block cannot be reversed with anticholinesterases. In previous studies, proper administration of neostigmine under neuromuscular monitoring reduced the PPCs associated with the use of NMBAs; however, there were reports of inappropriate reversal using neostigmine, such as neostigmine administration not guided by the TOF ratio or administered dose > 60 µg/kg [29,30]. Thus, it should be used only when the twitch height is ≥ 20% of the control or when the minimum TOF count is two [31]. In contrast, sugammadex is a relatively new reversal agent approved in 2008 in Europe, 2012 in Korea, and 2015 in the United States, which can reverse the NMB at any stage by encapsulating only the aminosteroidal NMBAs [32]. Various studies have shown that sugammadex is associated with a lower incidence of residual NMB than that of neostigmine [33–35]. Hence, the question of whether PPCs can be reduced using sugammadex is raised.

Unlike anticholinesterases, which are traditional NMB reversal agents, sugammadex is the only reversal agent that can reverse deep and intense blocks. Moreover, various studies have confirmed that sugammadex can reduce the incidence of residual NMB [33–35]. Various meta-analyses performed for several randomized controlled studies comparing neostigmine and sugammadex reported that the incidence of respiratory events and residual curarization could be reduced with sugammadex [36–38]. However, it is difficult to use these studies as evidence that sugammadex reduces PPCs, because the PPCs in these studies were identified as one of the outcomes of various adverse events and not as the primary outcome. As an ideal condition of a randomized controlled trial (RCT) would minimize the overall incidence of PPCs and the difference between the two groups, the incidence of PPCs in the real world may be different from that in an RCT [39]. Therefore, it is important to investigate the relationship between reversal agents and PPCs through large-scale cohort studies.

A retrospective study published in 2014 with data from 1,444 patients reported that sugammadex could lower the risk of PPCs in high-risk patients with ASA PS ≥ III; however, this finding should be interpreted cautiously because the endpoints are mixed, including acute post-anesthetic care unit (PACU) complications, bronchospasm, airway intervention, cardiac arrhythmia, length of stay in the PACU, and pulmonary outcome within 7 days postoperatively [40]. The post-anaesthesia pulmonary complications after use of muscle relaxants (POPULAR) study published in 2019, supported by the European Society of Anaesthesiology and Intensive Care, is the first multicenter cohort study providing prospective data for PPCs and analyzing 22,803 patients to identify PPCs related to NMBAs [41]. The incidence of PPCs in that study was 7.8%. Moreover, it showed that the use of NMBAs was associated with an increase in PPCs, although the use of NMBA had a lesser effect on the risk of PPCs than did the type and duration of surgery or the patient’s preoperative pulmonary function. In contrast, monitoring of NMB, use of reversal agents, extubation at TOF ratio ≥ 0.9, and the use of sugammadex instead of neostigmine were not associated with a reduction in PPCs. Some of these results have been refuted in later studies [42,43]; PPCs were reduced with TOF ratio > 0.95 before tracheal extubation than with TOF ratio > 0.9 [42], and the use of sugammadex instead of neostigmine was associated with a decrease in PPCs [43]. In an exploratory analysis of the POPULAR data, applying a TOF ratio threshold of 0.95 before extubation instead of 0.9 showed a decrease in PPCs [42]. This is presumed to be due to the property of the AMG, which overestimates neuromuscular recovery. In the POPULAR study, 87% of the neuromuscular monitoring devices used were AMG, which requires “normalization” of the TOF ratio to obtain an accurate TOF ratio of 0.9. However, normalization is rarely performed in clinical practice. Additionally, NMB reversal was achieved only in less than half of the patients who received NMBAs, and the number of patients receiving sugammadex was < 2,000. Furthermore, only 20% of the patients had complete data that allowed for the matching of patients and comparison of the treatment methods. Therefore, a more appropriate conclusion might be that NMBAs can cause serious outcomes if managed improperly.

Until the sugammadex versus neostigmine for reversal of neuromuscular blockade and postoperative pulmonary complications (STRONGER) study was published in 2020, no large-scale multicenter study had focused on the relationship between reversal agents and PPCs [43]. Although this study has limitations in collecting data retrospectively, its hypothesis was that patients receiving sugammadex have fewer PPCs than those receiving neostigmine. PPCs were confirmed in 4.1% of the 45,712 patients. The incidence of PPCs was 3.5% in the sugammadex group and 4.8% in the neostigmine group, confirming that the use of sugammadex reduced the risk by 30%. Moreover, and pneumonia and respiratory failure were reduced by 47% and 55%, respectively. However, since the data collection period of 5 years was set retrospectively, and the definition of PPCs was established using the prescription codes of the International Classification of Diseases-9 and -10, the results should be interpreted considering these limitations compared to prospective studies.

However, unlike previous studies [40,43], in a retrospective registry study on the occurrence of PPCs related to the use of sugammadex and neostigmine published in 2021 [44], there was no significant difference in the incidence of PPCs between them. In this study, the absolute incidence of PPCs decreased over time (adjusted odds ratio, 0.91 [per year]), which may be a simple change due to the long data collection period (10 years). In addition to the difference between the use of neostigmine and sugammadex, because of the multiple quality improvement initiatives over time including enhanced recovery after surgery protocols, the use of objective criteria for ventilator-related pneumonia and of TOF monitoring were implemented in the hospital where this study was conducted. All these factors may have influenced the reduction of PPCs.

Among the studies published so far on the occurrence of PPCs with sugammadex and neostigmine, a recently published meta-analysis analyzed 14 RCTs and 1,478 patients [45]. In the main meta-analysis, the risk of overall PPCs was lower with sugammadex than with neostigmine. In the stratified subgroup analyses, sugammadex was associated with a reduced risk of respiratory failure as compared to neostigmine, but there was no statistical difference in the occurrence of pulmonary infection, atelectasis, or pneumothorax. This meta-analysis may have clinical heterogeneity due to differences in patient comorbidities and the type of surgery. Additionally, PPCs such as respiratory infections, atelectasis, and pneumothorax were included only in 1–3 studies. These limitations should also be clearly considered [45].

Despite the discussions in recent studies, it is difficult to determine the exact association between the choice of reversal agent and PPCs. Published retrospective studies were unable to control for several factors that affect PPCs. However, controlled situations do not accurately reflect the actual clinical environment. In a clinical environment, the individual patient characteristics and the condition according to the risk factors should be understood and considered.

As mentioned earlier, there are limited studies on the relationship between reversal agents and PPCs, and studies on certain specific populations and situations are lacking (Table 2).

Although several efforts are being made to reduce PPCs, it remains questionable whether the choice of the NMB reversal agent affects PPCs. To date, complete reversal of the NMB before extubation under neuromuscular function monitoring seems more important than choosing a reversal agent. Moreover, understanding and considering the characteristics of each patient and the surgery type are important to reduce PPCs. As mentioned in the beginning, because the strength of evidence regarding surgical or patient-related factors as risk factors for PPCs is greater than that for anesthesia-related factors, a multidisciplinary approach should be considered to reduce PPCs.