Optimizing Ventilation Strategies to Reduce Pulmonary Complications in Frail Patients: A Comparative Study

Introduction

Postoperative pulmonary complications (PPCs) can influence prognosis, prolonging hospital stays and significantly increasing mortality and long-term disability rates [1–3]. PPCs are among the most common serious complications after cardiovascular events [4], and are closely related to the process of surgery, anesthesia methods, and the patient’s underlying health. Because pulmonary complications tend to exacerbate the condition and result in delayed postoperative recovery and excessive use of medical resources, implementing effective preventive and management strategies is particularly critical in high-risk patients. Therefore, reducing the incidence of PPCs is important for improving patient prognosis and minimizing adverse surgical outcomes [5].

Among all the risk factors that can affect prognosis, frailty is an independent factor that is especially common among elderly patients, and it is characterized by multi-system functional decline and reduced ability to respond to external stressors [6]. Additionally, frail people tend to be at higher risk for PPCs, especially for respiratory complications [7,8]. It was shown by previous studies [9,10] that frailty is associated with PPCs, as frail patients tend to have longer recovery times and higher risk of lung complications than non-frail patients due to reduced physiological reserves. Furthermore, frail patients tend to have respiratory dysfunction, highlighting the importance of prevention and management of postoperative pulmonary complications.

In recent years, individualized ventilation strategies have gradually shown potential in the improvement of postoperative respiratory function [11,12]. The aim of the individualized ventilation strategy was to titrate the PEEP and obtain driving pressure with the lowest level, which can minimize the risk of over-ventilation and lung injury [13–15]. Previous studies [16,17] indicated that this strategy improves function in postoperative lungs by alleviating injury related with the ventilation and decreasing the occurrence of PPCs. However, its use in frail patients requires further investigation.

Therefore, this prospective observational study included 61 frail patients undergoing non-cardiac surgery and aimed to compare postoperative pulmonary outcomes between driving pressure-guided individualized ventilation and standard positive end-expiratory pressure (PEEP) ventilation.

Material and Methods

ETHICAL APPROVAL AND STUDY DESIGN:

This study was approved by the Ethics Committee of the First People’s Hospital of Xiangshan County, China (Approval No: XYYJ-2-23-008). All participants were fully informed of the study details and provided written informed consent.

A prospective observational study was conducted at the First People’s Hospital of Xiangshan County between January and December 2024. Frail patients undergoing non-cardiac surgery (NCS) requiring tracheal intubation and general anesthesia were enrolled. Despite the relatively limited sample size, a preliminary sample size calculation was performed to ensure statistical validity for the analysis of primary outcomes.

Sample size estimation was conducted using a two-sided Z-test for 2 independent proportions, with α=0.05 and power=90%. According to previous literature, the incidence of PPCs was assumed to be 50% in the control group and 11% in the intervention group [18,19]. The calculation was performed using PASS 2023 (NCSS, LLC. Kaysville, Utah, USA). The estimated required total sample size was 53. Since our study enrolled 61 patients, the sample size was considered sufficient and statistically justified.

PATIENT SELECTION:

Frail patients scheduled for elective non-cardiac surgery under general anesthesia at the First People’s Hospital of Xiangshan County between January and December 2024 were screened for eligibility. Inclusion criteria were: age ≥65 years, elective surgery requiring tracheal intubation and postoperative hospitalization, and ability to undergo frailty assessment prior to surgery. Exclusion criteria included: ASA physical status IV or higher, severe pulmonary disease, inability to complete frailty scoring, or withdrawal of informed consent.

Frailty was assessed using the Edmonton Frail Scale (EFS), which evaluates physical, cognitive, and social domains. Only patients with an EFS score ≥5 were included in the study. Based on the EFS score, patients were further categorized into mild, moderate, or severe frailty subgroups for descriptive analysis.

FRAILTY ASSESSMENT:

Frailty was assessed preoperatively using the Edmonton Frailty Scale (EFS), a validated and widely used clinical tool for evaluating frailty in elderly patients. The EFS includes 9 components: cognitive function, general health status, functional independence, social support, medication use, nutrition, mood, continence, and functional performance. Each item is scored, with a maximum total score of 17 [20]. Patients with a score ≥5 were classified as frail and included in the study. Frailty severity was further categorized as mild (5–6), moderate (7–8), and severe (≥9). The assessment was conducted by a trained staff member during the pre-anesthesia clinic visit.

INCLUSION AND EXCLUSION CRITERIA:

Frailty assessment, patient grouping, and informed consent were all completed during the pre-anesthesia visit. Inclusion criteria were: (1) American Society of Anesthesiologists (ASA) physical status I–III, and (2) preoperative Edmonton Frailty Scale (EFS) score ≥5. Frailty severity was categorized according to the total EFS score as follows: mild frailty (5–6 points), moderate frailty (7–8 points), and severe frailty (≥9 points) [6]. The exclusion criteria were: (1) Under age 18 years; (2) Visual or hearing impairments, cognitive disorders, or mental illnesses that could hinder the understanding of the scale; (3) Underwent preoperative adjuvant chemotherapy or radiotherapy; (4) History of severe allergic reactions to general anesthesia agents; and (5) Severe, non-improvable respiratory diseases, defined per Global Initiative for Chronic Obstructive Lung Disease (GOLD) and American Thoracic Society/European Respiratory Society (ATS/ERS) criteria [21,22]:

ASSIGNMENT AND BLINDING PROCEDURES:

Patients were assessed for frailty using EFS by a fixed evaluator (score ≥5), and eligible patients were referred to the research coordinator. Patients were then assigned into 2 groups (individualized PEEP group and standard PEEP group) based on a pre-established group allocation list developed prior to the study. The assignment was communicated to the attending anesthesiologist upon the patient’s entry into the operating room, and ventilation strategy was applied accordingly. All data were collected by an independent evaluation team blinded to the group assignments. The group allocations were not disclosed to either patients or assessors throughout the study.

ANESTHESIA MANAGEMENT:

All the patients in the study were fasted according to the standard preoperative protocols. General anesthesia was performed by board-certified attending anesthesiologists with at least 5 years of independent clinical experience. Upon entering the operating room, peripheral venous access was established, followed by the application of standard monitoring, including electrocardiogram, non-invasive blood pressure, and pulse oximetry.

ANESTHESIA INDUCTION:

Anesthesia induction used midazolam (Jiangsu Nhwa Pharmaceutical Co., Ltd., China) 0.03 mg/kg, etomidate (Jiangsu Nhwa Pharmaceutical Co., Ltd., China) 0.2–0.6 mg/kg, sufentanil (Yichang Renfu Pharmaceutical Co., Ltd., China) 0.4 μg/kg, and rocuronium (Emeishan Tonghui Pharmaceutical Co., Ltd., China) 0.6 mg/kg. Intubation was performed when the patient became unconscious after mask ventilation. Mechanical ventilation was initiated, and end-expiratory carbon dioxide (P_etCO₂) was monitored throughout the surgery. Radial artery puncture was performed for monitoring, and central venous catheterization was used for intraoperative fluid administration.

MAINTENANCE OF ANESTHESIA:

During the surgery, a combined intravenous and inhaled anesthesia was maintained with sevoflurane (Jiangsu Hengrui Medicine Co., Ltd., China) 1–2%, propofol (Xi’an Libang Pharmaceutical Co., Ltd., China) infusion 3–4 mg/kg/h, and remifentanil (Yichang Renfu Pharmaceutical Co., Ltd., China) 0.05–0.1 μg/kg/min. A bolus dose of sufentanil 0.2 μg/kg was administered before skin incision, and intermittent boluses of rocuronium were given. The depth of anesthesia was shown by the bispectral index (BIS), and the value was kept at 40–60. Blood pressure and heart rate fluctuations were maintained at 20%. Both groups were managed with the same fluid strategy, and postoperative pain management was consistent between groups.

VENTILATION PARAMETER SETTINGS:

All patients received volume-controlled ventilation, with the tidal volume (VT) of 8 ml/kg predicted body weight. Body weight was calculated using standard formulas:

In the control group (Group N), PEEP was fixed at 5 cmH₂O throughout the surgery. In the individualized ventilation group (ΔP Group), PEEP was titrated to minimize the driving pressure (ΔP=Pplat - PEEP). After anesthesia induction and intubation, the initial PEEP was set at 1 cmH₂O and increased in 1 cmH₂O increments following every 12-breath cycle. If the peak airway pressure (Paw) exceeded 35 cmH₂O, further PEEP increases were stopped. The optimal PEEP level was defined as the one resulting in the lowest driving pressure, measured during the final 12-breath cycle. The selected PEEP was then maintained throughout the surgery.

The inspiratory-to-expiratory ratio was set at 1: 2. When necessary, an inspiratory pause of 10% was applied. Fraction of inspired oxygen (FiO₂) was maintained at 50%, with a fresh gas flow of 2 L/min. Respiratory rate was adjusted to keep end-tidal CO₂ (PetCO₂) at 35–45 mmHg. Individualized PEEP titration based on driving pressure was conducted with full consideration of anesthetic management requirements associated with specific surgical procedures or positions (eg, one-lung ventilation, prone positioning). Ventilation parameters were adjusted accordingly to ensure patient safety and protocol consistency, while still adhering to protective ventilation principles.

EQUIPMENT USED:

General anesthesia was conducted using the Mindray WATO EX-55 anesthesia system (Shenzhen Mindray Bio-Medical Electronics Co., Ltd., China). Monitoring of vital signs was performed using the Mindray BeneVision N17 patient monitor, and anesthesia depth was assessed via the BIS Complete Monitoring System (Covidien, USA). Arterial blood gas analyses were carried out using the ABL800 FLEX blood gas analyzer (Radiometer Medical ApS, Denmark). All equipment was regularly calibrated and operated strictly following the manufacturers’ instructions.

MEASUREMENT OF OUTCOMES:

The primary outcome was the incidence of PPCs during the first postoperative week. PPCs were defined according to the 2015 European Perioperative Clinical Outcomes (EPCO) criteria and included pneumonia, respiratory failure, atelectasis, pleural effusion, and bronchospasm [23]. Only patients who were extubated within 24 hours postoperatively were included in the PPCs analysis to minimize confounding due to prolonged mechanical ventilation.

Secondary outcomes included perioperative respiratory parameters: tidal volume (VT), plateau pressure (Pplat), positive end-expiratory pressure (PEEP), and dynamic compliance (Cdyn), recorded at T0 (after intubation), T1 (after 60 minutes of ventilation), and T2 (end of surgery). Arterial blood gas measurements were collected at T0, T1, T2, and 24 hours after surgery (T3), and oxygenation index (OI) was calculated at each time point.

All patients were transferred to the intensive care unit (ICU) according to preoperative planning. For those who remained intubated, ICU ventilation settings – including PEEP – were maintained according to the intraoperative group allocation.

STATISTICAL ANALYSIS:

Statistical analyses were performed using SPSS version 26.0 (IBM Corp., USA), and graphs were generated with GraphPad Prism 9.0 (GraphPad Software, Inc., USA). The normality of continuous variables was assessed using the Kolmogorov-Smirnov test. For normally distributed variables, values are presented as mean±standard deviation, and comparisons were conducted using t-tests and repeated-measures ANOVA. For non-normally distributed variables, rank-sum tests were employed. Categorical variables were compared using chi-square tests or Fisher’s exact test as appropriate. Repeated-measures ANOVA was used to account for within-subject variability across time points (T0, T1, T2). All analyses were two-sided, and a P value <0.05 was considered statistically significant.

All data were independently collected by trained clinical researchers, and cross-validated by a second investigator to ensure data reliability and objectivity.

Results

BASELINE CHARACTERISTICS:

From January to December 2024, 76 frail patients were screened based on the Edmonton EFS among the patients with surgery records at Wenzhou Medical University Affiliated Xiangshan Hospital. However, 5 of the patients were excluded due to preoperative surgery cancellation, which occurred before anesthesia or study enrollment, along with 10 patients without tracheal general anesthesia; therefore, 61 patients were finally included in this analysis. The study consisted of 2 groups: the individualized strategy group (31 patients) and the traditional strategy group (30 patients).

We found no obvious differences in sex, age, ASA classification, body mass index (BMI), weight, ideal body weight, frailty severity level, preoperative pulmonary complications, tidal volume at each time point, anesthesia duration, or ICU transfer rate (P>0.05) (Table 1).

INTRAOPERATIVE VENTILATION PARAMETERS:

At T0 (immediately after intubation), significant differences were found in Pplat, Paw, and Cdyn (P<0.01). However, at T1 (1 hour after PEEP titration), there were no significant differences in Pplat, Paw, or Cdyn (P>0.05). At T2 (end of surgery), significant differences were found in Paw and Cdyn, but not in Pplat (P>0.05) (Table 2).

ARTERIAL BLOOD GAS ANALYSIS:

Analysis of arterial blood gas showed significant differences in PaO2 at T0 (P<0.05), but no significant differences were observed in PaO2 at T1, T2, or 24 hours after surgery (T3) (P>0.05). The oxygenation index (OI) did not significantly vary during the whole process (T0, T1, T2, T3) (P>0.05) (Table 3).

REPEATED-MEASURES ANALYSIS:

Repeated-measures analysis showed significant effects of PaO2, OI, and Cdyn over time (P<0.01), as did PaO2 and Cdyn (P<0.05). However, differences were not found in interaction effects among all 3 indicators (P>0.05) (Table 4, Figure 1).

POSTOPERATIVE PULMONARY COMPLICATIONS:

In the ΔP group, pulmonary complications did not occur within 7 days postoperatively, while 7 cases (23.3%) occurred in the traditional ventilation strategy group (N group), and the difference was statistically significant (P<0.01) (Table 5, Figure 2).

Discussion

This study found that driving pressure-guided individualized ventilation significantly reduced the incidence of postoperative pulmonary complications (PPCs) in frail patients undergoing non-cardiac surgery compared with standard PEEP ventilation. The findings of this study were consistent with Zhang et al [24], in which the individualized ventilation strategies were demonstrated to reduce PPCs in patients undergoing abdominal surgery. Similarly, our study supports the protective effect of this approach. Notably, our focus on frail patients undergoing non-cardiac surgery offers new clinical evidence regarding the effectiveness of individualized ventilation in this high-risk population. It was also found that significant differences occurred in Pplat, Paw, Cdyn, and PaO2 immediately after intubation (T0) between the 2 groups. This could be attributed to the study design, with a fixed PEEP of 5 cmH2O in the control group, and another group began with a PEEP of 1 cmH2O and titrated it upward based on ΔP, resulting in significant statistical differences without clinical significance. However, by T2, Paw and Cdyn had statistical differences, indicating the improvements in pulmonary function due to the ventilation strategy, which reduced Paw with the increasing of Cdyn. Importantly, the maximum individualized PEEP level applied in our study was 12 cmH2O, without peak airway pressure exceeding 35 cmH2O in patients throughout the procedure. Additionally, it was within the safety threshold supported by Campos et al [11], who reported that individualized intraoperative PEEP up to 12 cmH2O exhibited no connection with the increase in the incidence of postoperative pulmonary complications when protective ventilation strategies were maintained.

Except for the ventilation parameters, operative factors such as surgical approach and patient positioning can influence intraoperative lung compliance. In this study, laparoscopic, thoracoscopic, and prone-position procedures were equally distributed between groups, thereby reducing the potential bias.

Notably, no significant differences were found in PaO2, OI, and Cdyn at any time point, and the interaction effects was similar to those of previous studies [25–27], which could be attributed to differences in the sample size of this study. However, intra-group comparisons revealed that dynamic lung compliance in the experimental group improved over time, which was consistent with most previous studies [18,28,29], although it differed from the results of Mauri et al [30].

Preoperative frailty assessments are critical for the management of high-risk patients, particularly in surgical settings. EFS is widely used to assess multiple physiological, psychological, and social functions and provide a comprehensive measure of frailty [31]. In this study, the EFS was used to identify frail patients, and it was found that frailty played an important role in the incidence of PPCs, which corresponded to prior research [32,33]. Furthermore, in this study, frailty severity was stratified into mild, moderate, and severe levels based on EFS scores. As shown in Table 1, the distributions of severity levels were comparable between the groups, indicating that frailty severity was unlikely to confound the observed PPCs differences. Frail patients were more likely to develop respiratory complications postoperatively due to reduced physiological reserves and impaired stress responses, such as pneumonia and acute respiratory distress syndrome (ARDS). In the control group, 7 patients developed PPCs; among them, 4 had moderate frailty (EFS 7–8) and 3 had severe frailty (EFS ≥9), and no patients had mild frailty (EFS 5–6). Although formal statistical analysis was not conducted due to the limited sample size, this distribution may suggest a potential trend toward more frequent PPCs in patients with higher frailty severity.

In recent years, individualized ventilation strategies showed considerable potential in reducing PPCs. Unlike traditional ventilation approaches, individualized strategies can adjust ventilation parameters based on the requirement of patient’s specific physiologic, which can minimize lung injury, which is necessary for those with impaired pulmonary function and risk of over-ventilation or alveolar damage with conventional approaches [11]. An individualized strategy was employed depending on the EFS results, adjusting airway pressure and ventilation parameters to minimize the risk of lung injury. This strategy can significantly reduce the incidence and accelerate treatment in frail patients [34].

This study’s clinical significance is that it provides a new approach for perioperative management of frail patients. By combining the EFS frailty assessments with individualized ventilation strategies, the high-risk patients were more precisely identified, effectively reducing the incidence of PPCs.

However, our study has some limitations. Firstly, individual variations could have resulted in biased results [31], although the EFS is widely used for frailty assessment. Additionally, other potential influencing factors include intraoperative medication and anesthesia methods. Lastly, the sample size was relatively small, and the results are generalizable to limited populations. These limitations, including the single-center design, relatively small sample size, and other potential confounders, suggest the importance of further multicenter studies with larger populations to validate our findings. In addition, although comprehensive spirometry was not feasible for all participants due to frailty and cognitive limitations, exclusion of severe irreversible lung disease was ensured through standardized clinical and imaging-based assessments. Furthermore, the individualized ventilation strategy required intraoperative PEEP titration to achieve the lowest driving pressure, which can vary depending on clinical judgment and ventilator responsiveness. This process can be relatively complex and time-consuming, potentially limiting its feasibility in routine clinical practice.

Conclusions

The findings from this study indicate that, compared with traditional perioperative ventilation strategies, driving pressure-guided individualized PEEP titration significantly reduced PPCs and improved lung compliance in frail patients undergoing non-cardiac surgery. This strategy may help optimize intraoperative ventilation in similar high-risk surgical patients.