Sugammadex is associated with shorter hospital length of stay after open lobectomy for lung cancer: a retrospective observational study

Lung resection surgery is associated with a high incidence of postoperative pulmonary complications (PPCs), including prolonged air leak, atelectasis, pneumonia, empyema, and acute respiratory distress syndrome [1]. PPCs are associated with prolonged hospitalizations, admissions to the intensive care unit (ICU), and hospital re-admissions, as well as increased morbidity, mortality, and healthcare expenditure [2‐6]. Residual neuromuscular blockade (NMB) after emergence from anesthesia can lead to pulmonary complications including hypoxia, pulmonary collapse, and acute respiratory failure [7, 8]. In addition, complete pharmacological reversal improves pulmonary outcomes by reducing the likelihood of residual neuromuscular blockade after anesthesia [9].

Anticholinesterases such as neostigmine or pyridostigmine are commonly used to reverse NMB during general anesthesia. Despite the use of relatively short-acting agents (e.g., rocuronium), which decrease the incidence of residual NMB, the incidence of residual NMB is reported to 82% at 6 min post administration of neostigmine [10]. And it has various cholinergic side effects including bradycardia, hypotension, bronchoconstriction, and airway secretion [11, 12]. Sugammadex is modified gamma-cyclodextrin compound that selectively binds to steroidal non-depolarizing neuromuscular blocker by forming a 1:1 complex. In contrast to anticholinesterase, it shows rapid and reliable neuromuscular block reversal [13]. It reduces postoperative complications and improves patient outcomes [14].

Hospital length of stay (LOS) is considered an important quality metric for recovery from surgery and anesthesia [15]. The length of time patients spend in hospital is a good representation of the amount of hospital resources utilized, such as bed utilization, staffing, and equipment [16]. To date, few studies have evaluated the relationship between neuromuscular block reversal with sugammadex and patient outcomes including LOS with controversial results in different surgeries [12, 17‐19].

We hypothesized that there would be a difference in the LOS due to PPC when the NMB reversed with sugammadex compared pyridostigmine after open lobectomy for lung cancer. We investigated the relationship between reversal agent and LOS/postoperative complication rate/overall survival, and identified risk factors associated with prolonged LOS in these patients.

During the study period, 266 patients underwent open lobectomy for lung cancer at our institution. Of those patients, 9 were excluded due to cisatracurium use, ICU transfer after surgery, or missing medical records. A total of 257 patients were enrolled; 127 patients received pyridostigmine (pyridostigmine group) and 130 patients received sugammadex (sugammadex group) for NMB reversal. After propensity score matching, 127 patients in each group were included in the final analyses (Fig. 1). Demographic and clinical characteristics at baseline are summarized in Table 1 and were comparable between the two groups.

The collected data included 4 thoracic surgeons and 5 anesthesiologists (excluding residents). The initial dose of rocuronium was 0.8–1.0 mg/kg. Pyridostigmine was 0.1–0.2 mg/kg and sugammadex was 1.5–2 mg/kg for reversal of NMB. The use of sugammadex continued after it was decided at the anesthesiology faculty meeting, which led to more intensive NMB during surgery. The degree of NMB during anesthesia was monitored using the MechanoSensor™ DatexOhmeda GE Healthcare NMT-EMG (Helsinki, Finland). However, it was excluded from the analysis due to inconsistent medical records. A double lumen endotracheal tube was used for one-lung ventilation during surgery, and the surgical approach for the affected area was a conventional posterior lateral thoracotomy in the lateral position. In the following cases, an anesthesiologist discussed with a thoracic surgeon to determine whether to extubate: Difficulty breathing before surgery, hypoxemia (blood oxygen saturation < 90%) frequently occurring during one-lung ventilation, or excessive fluid administration (≥ 30 ml/kg).

Table 2 shows intra- and postoperative outcomes of matched cohorts in both pyridostigmine and sugammadex groups. Median hospital LOS was significantly shorter in the sugammadex group than in the pyridostigmine group (10.0 days (IQR 8.0–15.0 days) vs. 12.0 days (IQR 9.5–16.5 days) (p = 0.005). The incidence of atelectasis (18.1 vs. 29.9%, p = 0.040) was lower and Epidural PCA (70.9 vs. 86.6%, p = 0.004) was less used in the sugammadex group than in the pyridostigmine group. However, no differences were found regarding other postoperative complications reviewed between the two reversal agents.

Intra- and postoperative outcomes of matched cohorts in patients with and without prolonged LOS are shown in Table 3. Median LOS was 1 day shorter in the sugammadex group than in the pyridostigmine group (9.0 days [IQR 7.0–10.0 days] vs. 10.0 days [IQR 9.0–12.0 days], p = 0.002) in patients without prolonged LOS, whereas it did not differ between the two groups in patients with prolonged LOS (19.0 days [IQR 16.0–23.0 days] vs. 19.0 days [IQR 16.0–26.0 days], p = 0.537). And, PACU stay time was significantly shorter when reversed with sugammadex than with pyridostigmine (60.0 vs. 70.0 min, p = 0.007) in patients with prolonged LOS. Among postoperative complications, the incidence of atelectasis was significantly lower in the sugammadex group than in the pyridostigmine group (11.8 vs. 25.3%, p = 0.034) in patients without prolonged LOS, but did not differ between the groups (35.3 vs. 38.6%, p = 0.947) in patients with a prolonged LOS.

LOS was also analyzed based on the discharge rate to determine the proportion of discharged patients on each postoperative day during hospitalization (Fig. 2). Discharge rate was significantly higher in the sugammadex group throughout the admission period (p = 0.025, Fig. 2a). Sugammadex also facilitated patient discharge compared to pyridostigmine in patients without prolonged LOS (p = 0.0083, Fig. 2b), but not in patients with prolonged LOS (p = 0.41, Fig. 2c).

All variables that were significant in univariate analyses were included in multivariate analyses to identify covariates associated with LOS in the propensity matched total study cohort (Table 4). In multivariate analyses, demographic predictors included four variables (age ≥ 65 years, male sex, DM, COPD). COPD and male sex showed higher risk for prolongation of LOS. Age ≥ 65 years was associated with a 9% increase in LOS. By contrast, DM reduced the LOS by 18%. The only intraoperative predictor was sugammadex, which reduced LOS by 12%. Among the postoperative factors significant in univariate analyses, dyspnea, atelectasis, pneumonia, and air leak > 5 days remained significant for increasing the risk for prolonged LOS. Pneumonia and air leak > 5 days showed a higher risk for extended LOS. Dyspnea and atelectasis were also associated with increased hospitalization.

The Kaplan-Meier curves of overall survival in the propensity-matched cohort are illustrated in Fig. 3. The estimated 1-year survival rates were 93.4% (89.1 to 97.9%) in the pyridostigmine group and 93.7% (89.6 to 98.0%) in the sugammadex group. There were no significant differences between the two groups in overall survival with an unadjusted hazard ratio for death at 1 year of 0.967; 95% CI, 0.0.363 to 2.577 (p = 0.947).

In this retrospective study, we observed a median 2.0 day decrease in time from surgery to discharge, a significantly lower postoperative complication rate (e.g., atelectasis), and a similar mortality over 1 year after the surgery when reversed with sugammadex compared to pyridostigmine. Moreover, sugammadex was the only intraoperative predictor associated with reduced LOS.

Hospital LOS was decreased, as was the incidence of adverse postoperative outcomes (e.g., atelectasis) in patients reversed with sugammadex for NMB, in line with previous studies that reported that the use of sugammadex is associated with 20% shorter LOS with reduced postoperative adverse outcomes after major abdominal surgery [12] and is associated with a 0.6 day shorter hospital LOS and a lower postoperative complication rate after laparoscopic gastric cancer surgery [18]. Residual NMB occurs in approximately 20–60% of surgical patients [23] and is associated with an increased incidence of PPCs (e.g., hypoxemia and atelectasis) [8]. On the other hand, sugammadex has been shown to reduce the incidence of residual NMB upon arrival in the PACU compared to other classic NMB reversal agents [24]. Therefore, sugammadex may decrease hospital LOS through its improved muscle relaxant reversal, leading to a reduction in PPCs and early patient discharge.

By contrast, Ledowski et al. [17] observed that overall hospital LOS after surgery did not differ between patients treated with sugammadex or acetylcholinesterase inhibitors; the cause of this discrepancy remains unclear. It has been shown that postoperative residual NMB and associated adverse PPCs are more common in elderly patients than in younger patients [25]. Moreover, thoracic and abdominal procedures that reduce lung volume are associated with increased risk of developing atelectasis and postoperative complications [26]. Ledowski et al. [17] studied relatively young patients (mean age ~ 50 years) who underwent surgical procedures including orthopedic, plastic, general, and others, whereas we studied elderly patients (mean age ~ 66 years) who underwent open lung surgery. Thus, different ages of surgical populations and type of surgery may be responsible for differences between the studies. Indeed, Ledowski et al. [17] demonstrated that NMB reversal with sugammadex significantly improves postoperative pulmonary outcomes compared to neostigmine, particularly in elderly patients.

It has previously been reported that poorly controlled acute postoperative pain is a risk factor associated with respiratory complications [27], and that postoperative pain may lead to the development of atelectasis because it can interfere with the normal activity of respiratory muscles and forced respiratory effort [28]. On the other hand, other studies have shown that epidural analgesia provides better postoperative pain control than systemic opioid administration in abdominal or open thoracotomy surgery [29, 30]. In the present study, epidural patient-controlled analgesia was used less commonly in patients reversed with sugammadex than those reversed with pyridostigmine. The postoperative pain is expected to be more severe in the sugammadex group, but the hospital LOS was short and PPCs occurred less. These findings are in accordance with those of a recent study that showed that patients reversed with sugammadex had fewer postoperative complications and a shorter LOS despite more severe postoperative pain compared to those with neostigmine in patients who had undergone laparoscopic gastric cancer surgery [18].

Several studies have reviewed patients undergoing pulmonary resection for lung cancer and identified risk factors for prolonged hospital LOS [1, 20, 28, 31‐34]. Some risk factors identified previously include older age [20, 32, 33], male sex [20], ASA physical status score [20, 32], insulin-dependent diabetes [20], renal dysfunction [20], percentage predicted FEV1 [20, 33], surgeon [33], smoking [20], COPD [32], and postoperative complications (e.g., pneumonia [32, 33], unplanned reintubation [32, 34], or prolonged ventilation [32, 34]). In this study, we confirmed the important risk factors for morbidity and LOS after lung resection (Table 4). In fact, the use of sugammadex is becoming increasingly common for NMB reversal, particularly in the elderly, with the advantage that it can reverse profound NMB, although reversal agent options are currently limited by price.

Overall survival 1 year after surgery did not differ between the two reversal agents (Fig. 3). Death after lung cancer surgery may be attributable to surgery-related major complications and to cancer progression. Although sugammadex decreases the incidence of PPCs and shortened hospital LOS in the present study, this agent is unlikely to significantly reduce surgery-related major complications. In addition, we found that once a patient reached a medically stable state and was discharged, the mortality after 1 year was not different across the type of reversal agent, suggesting that an advantage for sugammadex does not extend to the long term.

This study had several limitations. First, this is a small number, single center, retrospective study that is not free from bias in selection and treatment. Not all covariates were controlled, although the demographics and clinical characteristics were balanced by propensity score matching. Second, some fundamental intra- or postoperative covariates associated with respiratory complications were not collected. The degree of pain after surgery is considered an important factor associated with respiratory complications [27]. The severity of pain and opioid consumption were not assessed in this study. In addition, the total dose of neuromuscular blocking agent administered and depth of NMB at the time of reversal were not included in our analyses. Use of single or repeated doses [35] and depth of NMB at the time of reversal [36] are important factors that affect recovery after NMB. Third, diagnosis of atelectasis was entirely dependent on plain chest radiographs, although only relatively obvious cases of atelectasis seen on plain radiography were included. However, this technique for the diagnosis of postoperative lung collapse is less sensitive than computed tomography, which was not routinely performed after open lobectomy in our hospital. Finally, the extent of surgery are strongly related to patient outcomes. However, the cohorts analyzed were a highly selective group that underwent open lobectomy for lung cancer.

Compared with pyridostigmine, NMB reversal with sugammadex after open lung lobectomy for lung cancer was associated with a shorter hospital stay and a lower PPC, but with a similar mortality after 1 year. In particular, the effects of sugammadex on LOS was obvious in patients without prolonged LOS. Our data suggest that sugammadex is a preferable agent for NMB reversal than cholinesterase inhibitors in this patient population. However, further prospective, randomized, controlled, and sufficiently powered studies on larger patient populations are required.