Enhancement of Pulmonary Function and Reduction of Complications Through EIT-Guided Yoga Breathing Exercise After Esophagectomy

Introduction

Esophageal cancer is one of the most aggressive cancers and is the sixth leading cause of cancer death worldwide [1]. Globally, 0.54 million deaths and 0.6 million new cases of esophageal cancer were identified in 2020 [2]. According to a retrospective analysis of 8403 patients in 19 countries, the 30-day mortality rate is 2.0%, and the 90-day mortality rate is 4.2% after esophagectomy [3]. Across a 17-year period, the overall hospital death rate was 11% at a tertiary referral center. The leading causes of death after esophageal resection are postoperative pulmonary complications (PPCs) (45.5%) and progression of malignant disease (21.5%). Anastomotic leakage accounts for 9% of deaths [4]. Major postoperative overall complications and PPCs appeared to be independent risk factors in the 5-year overall survival (OS) and the 5-year disease-free survival (DFS) after esophagectomy [5]. PPCs and mortality rate after esophagectomy remain high, and the potential for relief has not been adequately studied.

Breathing exercises (eg, kapalabhati) are first introduced as a non-pharmacological therapy in patients with bronchial asthma. The immediate increase in FEV1 is achieved after 10 min of kapalabhati practice [6]. Before elective major open upper-abdominal surgery, the addition of a 30-min physiotherapy session, including education and breathing exercise training, reduced incidence rates of PPCs and hospital-acquired pneumonia by half [7]. In addition to effectively treating asthma, breathing exercises also have some positive effects on quality of life, hyperventilation symptoms, and lung function after 4 to 6 months [8]. Preoperative exercise was effective in reducing postoperative complications and length of hospital stay in patients with lung cancer. However, for patients with esophageal cancer, preoperative exercise was not effective in reducing length of hospital stay [9].

Electrical impedance tomography (EIT) is a real-time monitoring tool for continuously evaluating pulmonary ventilation at the bedside for critically ill patients. The regional differences between inspiratory and expiratory lung physiology facilitate the characterization of ARDS phenotypes and personalized ventilation settings [10]. The application of EIT has been extended to estimate lung collapse and overdistension, pneumothorax detection, monitor the effects of endotracheal aspiration on pulmonary volumes, ventilatory desynchrony, and pulmonary perfusion [11,12]. EIT is also to PET for measuring and tracking changes of relative lung perfusion [13]. However, among clinicians, there is still a lack of knowledge regarding the technical principles of EIT and potential applications in PPCs patients.

The purposes of this study were to determine whether EIT-guided breathing exercise after esophagectomy reduces the incidence of PPCs, and to evaluate the effect on length of hospital stay.

Material and Methods

PATIENTS:

The prospective study was conducted in the anesthesia intensive care unit (AICU) of the Second Hospital of Anhui Medical University. This randomized clinical trial was approved by the hospital’s ethics committee (Ref#YX2021-033 approved on 12/04/2021), and then registered in the Chinese Clinical Trial Registry (ChiCTR2100045449, registered on 29/11/2021). A total of 66 patients were included from 17/05/2021 to 29/03/2022. Informed consent was obtained from all individual participants. Inclusion criteria were patients who were eligible to undergo surgical resection for squamous cell carcinoma or adenocarcinoma of the middle or lower third of the esophagus, ability to provide written informed consent, ability to undergo one of the investigated surgical procedures, and forced expiratory volume in 1 s higher than 65% of the predicted value. All patients who did not meet all the inclusion criteria were excluded. The other exclusion criteria were the common contraindications for surgery related to patient status, disease extension, or operative technique, and those who could not tolerate the breathing training due to pathological fracture, pulmonary edema, pneumothorax, wound infection, bleeding, or hemoptysis, as well as heart, kidney, brain, and other organ dysfunction, or serious organic diseases, severe asthma, and a history of lung surgery. Patients were randomly allocated in 2 groups (Control and Yoga Breathing Exercise) with a computer-generated randomization sequence using http://www.randomization.com. Allocation was concealed in opaque sealed envelopes that were opened by the primary investigator just before performing the breathing training.

PERIOPERATIVE MANAGEMENT:

All patients received standard general anesthesia and postoperative analgesia. Before anesthesia, all patients were moved to the pre-anesthesia room for radial artery and internal jugular vein intubation. Ultrasound-guided thoracic paravertebral nerve block (PVNB) and erector spinae plane block (ESPB) were performed by an experienced anesthesiologist who was blinded to the trial. By using a convex array probe (Sonosite Micromaxx, Bothell, WA, USA) with a frequency of 2–5 MHz, unilateral blocks were completed with the diffusion of 0.33% ropivacaine (including dexmedetomidine 1 μg/kg).

A standardized anesthesia protocol was conducted to all patients. Combined intravenous and inhalation anesthesia was maintained by continuously infusing with dexmedetomidine, propofol, cisatracurium, and remifentanil, as well as sevoflurane inhalation. Cardiovascular variables were standardized and carefully controlled in a similar way in each treatment group, both intraoperatively and postoperatively, with single-lung ventilation during thoracotomy. A volume-controlled model patient-controlled intravenous analgesia (PCIA) device was used as analgesia complement after surgery. All surgical procedures were completed by the same surgical team, who were experienced in minimally invasive McKeown esophagectomy. All surgeons had over 5 years of experience performing esophagectomy. All surgeons and nurses in the operating room were blinded to the intervention.

STUDY PROTOCOL:

After the esophagectomy, patients with an endotracheal tube were immediately transferred to the Anesthesia Intensive Care Unit (AICU). Following extubation, patients received standard care with multiple perioperative management. Patients randomized to the intervention group also received yoga breathing exercise under the guidance of EIT. Complete breathing exercise, also known as chest-abdominal breathing, is a breathing method that combines chest breathing with abdominal breathing. Thoracic breathing, also known as costal breathing, is a method of breathing in which the ribs are lifted or moved down with the contraction or relaxation of intercostal muscles, and the chest is also expanded or smoothed. Abdominal breathing, also known as diaphragmatic breathing, is a method of breathing that makes the abdominal organs move down or up with the contraction or relaxation of the diaphragm, and the abdomen rises and falls. Patients in the Control group received neither breathing exercise training nor education.

While undergoing breathing training, the patients lay down, relaxed, and put their hands on the chest or the upper edge of the umbilicus, then exhaled through the mouth and felt the rib cage fall, always contracting the abdomen in the process of inhalation and exhalation. In abdominal breathing, patients inhaled through the nose, and slowly expanded the abdomen when inhaling for 5 s, then slowly exhaled through the mouth to contract the abdomen. Each session was alternated 10 times. The frequency of the repetition was every 15 min within the first hour, then every 30 min from 1–6 h, and 1 h after 6 h in the AICU. A senior nurse was responsible for breathing education and training. She reminded patients to follow the instructions to perform breathing exercise at each session and kept the records.

EIT MEASUREMENTS:

EIT measurements were obtained with a Swisstom BB2 device (Swisstom, Landquart, Switzerland). During the intervention, a 32-electrode belt was placed on chest between the fifth and sixth intercostal space. Computed tomographic axial slice imaging of the thorax facilitated schematic representation of electrical current pathways through thorax. One pair of electrodes injects electrical current while the remaining electrodes read voltages produced as a result of the distribution of current density inside the thorax. The injection pair of electrodes is alternated sequentially, and 1 image is generated after a full cycle. Functional image reconstruction by EIT uses a color scale – the lighter the blue, the greater the regional ventilation. Of note, this color scale is not universal. All tests were conducted by a physician who received standard training and was blinded to the randomization and interventions for all participants.

OUTCOME MEASURES:

The primary outcome was the incidence of PPCs in the hospital. Assessors masked to the group allocation assessed patients prospectively and daily until discharge. Participants were screened using a standardized validated diagnostic tool consisting of 8 symptomatic and diagnostic criteria [14]. PPC was diagnosed when 1 or more of these diagnoses were present at any time on each postoperative day.

The secondary outcomes included the time to the first pulmonary complication; the early recovery outcomes indicators (the time of first flatus and ambulate) and the postoperative length of hospital stay (LOS); the EIT parameters center of ventilation (CoV), dependent silent spaces (DSS), non-dependent silent fFspaces (NSS) assessed at 1, 6, and 12 h after tracheal tube extubation in the AICU. Blood gas analysis was performed at the same time. Time to the first pulmonary complication was defined as the time from extubation to the first pulmonary complication, with higher values indicating better pulmonary aeration.

STATISTICAL ANALYSIS:

Based on the preliminary trial, the incidence of pulmonary complications in the Control group and EIT group were 80% and 40%, respectively. The sample size was calculated to be 29 patients in each group, with a significance level of 5% in the two-sided test and a detection power of 85%, which allows for a 10% loss to follow-up. We recruited 32 patients in each group to account for random errors and additional comparisons.

Continuous variables are represented as a mean±standard deviation (SD) or median (interquartile ranges, IQR). The Shapiro-Wilk test was used to test for normality. The comparisons between the 2 groups were conducted with t tests or Wilcoxon rank sum tests, as appropriate. Categorical variables are represented as number (n) and percentage (%), compared with the chi-square test or Fisher’s exact test. Relative risks (RR) with 95% CIs (confidence intervals) were estimated with the log-binomial model for the incidence of pulmonary complications. Time to the first pulmonary complications was compared using the log-rank test. Linear mixed models were used to analyze the difference between groups in changes in EIT parameters and blood gas indexes, followed by pairwise comparisons with Bonferroni correction adjusted for age and sex. Data analyses were conducted with SAS version 9.4 (SAS Institute, Inc.). Two-sided P values less than 0.05 were considered statistically significant.

Results

From 17/05/2021 to 29/03/2022, a total of 149 patients received thoracoscopic radical esophagectomy for esophageal cancer. Then, 66 patients were randomly assigned to receive thoracoabdominal breathing training or not. Sixty-one patients (31 in the Control group and 30 in the YB group) were included in the analysis, as 1 patient did not complete the assessment of EIT parameters (Figure 1). The median age of participants in the Control group was 68.0 (IQR: 62.0, 71.0) years and 70.0 (IQR: 68.0, 72.0) years in the YB group, with no significant difference between groups (P=0.149). There were no significant differences between the 2 groups in baseline and surgical characteristics, including preoperative comorbidities, surgery duration, and time in the AICU (Table 1).

In this trial, 44 patients experienced pulmonary complications after esophagectomy – 17 (36.7%) in the YB group and 27 (87.1%) in the Control group (RR, 0.42 (95% CI: 0.26, 0.69). The most common pulmonary complication was pleural effusion, with an incidence of 30% in the YB group and 74.2% in the Control group, with RR of 0.40 (95% CI: 0.23, 0.73). For pulmonary atelectasis, the RR was 0.16 (95% CI: 0.05, 0.47) with a 10% incidence rate in the YB group and 64.4% in the Control group (Figure 2).

Time to first pulmonary complication was significantly longer in the YB group than in the Control group (hazard ratio, HR, 0.43; 95% CI 0.21 to 0.87; P=0.019) (Figure 3), and patients in the YB group performed better in early recovery. The median postoperative hospital stay was significantly longer in the YB group (median, 8.0, IQR: 8.0, 9.0) than in the Control group (median, 19.0, IQR: 11.0, 28.0), with a P value less than 0.001. Patients in the YB group had shorter time to first flatus, ambulation, and the duration of closed thoracic drainage (P<0.05), and these were shorter than in the Control group (P<0.001) (Table 2).

For the EIT parameters, there was no statistically significant difference in the parameters scores before thoracoabdominal breathing training (P>0.05). The mixed model showed that the EIT parameters scores over time were significantly different in the 2 groups (Pinteraction <0.001). During the follow-up, a significant increase was observed in all parameters scores in both groups (Ptime <0.001). Patients in the YB group had significantly higher scores in CoV, DSS, and NSS than in the Control group after training for 5 h. After 11 h of training, the parameters scores of patients in the YB group were significantly increased, while there was no change in the Control group (Table 3).

The effect of time and thoracoabdominal breathing training on blood gas indexes is shown in Table 4, using linear mixed models. During the training period, there was no significant difference in changes in blood gas indexes between groups (Pinteraction >0.05). We further compared the hemogram and biochemical indexes between groups, and found that C-reactive protein, aspartate transaminase, and creatinine was higher in the Control group than in the YB group, while preoperative indexes were similar in the 2 groups (Table 5).

Discussion

In this trial using EIT, postoperative yoga breathing exercise (YB) reduced the incidence of PPCs after esophagectomy, including pulmonary atelectasis, pulmonary infection, and pleural effusion. The significant increases of DSS and NSS may explain the immediate effect of YB on lung function. Finally, the LOS was remarkably shortened by YB under EIT guidance.

More than 50% of the world’s esophageal cancer patients live in China, and esophageal squamous cell carcinoma (ESCC) is the main pathological type [15]. Esophagectomy is the main curative treatment for ESCC, but 29% of patients are readmitted at 30 and 90 days after major cancer surgery, and they have substantially poorer outcomes and higher medication costs. Risk of readmission is most strongly associated with length of stay and discharge destination [16]. PPCs have been considered as an independent predictor of shorter long-term survival in patients undergoing resection of ESCCs [17]. In a prospective cohort study of 1055 consecutive patients within 7 days after elective non-thoracic surgery, mean lengths of stay (LOS) were substantially longer in patients with PPCs (27.9 days versus 4.5 days) [18]. Even mild PPCs are associated with increased early postoperative mortality, ICU admission, and LOS (ICU and hospital). Therefore, to improve perioperative outcomes, mild frequent PPCs (eg, atelectasis and prolonged oxygen therapy need) are increasingly used [19]. In our study, 61 patients older than age 65 years received esophagectomy and their results were analyzed. The 30-day readmission rate was 26.2% independent of interventions. Use of yoga breathing exercise in the AICU significantly reduced the incidence of PPCs, hospitalization costs, and postoperative stay.

The risk factors for pulmonary complications include older age, operation duration, and proximal tumor location based on logistic regression analysis, and older age and greater blood loss are predictive of mortality [20]. Therefore, interventions to reduce risk for PPCs span throughout the perioperative period, including the evidence for lung-specific strategies, anesthetic and analgesic techniques, surgical techniques, and perioperative care [21]. Therefore, compared with open esophagectomy, surgical modification such as hybrid minimally invasive esophagectomy has can reduce the incidences of perioperative major complications, specifically PPCs, although this surgical modification failed to promote overall and disease-free survival over a period of 3 years [22]. Inappropriate management of intraoperative fluid is also associated with higher morbidity, mortality, cost, and length of stay [23]. In healthy patients undergoing general surgery, the combination of intraoperative low tidal volume ventilation (LTV) and positive end-expiratory pressure (PEEP), and intermittent recruitment maneuvers (IRM) are associated with significant improvement of clinical pulmonary outcomes and reduction in LOS. Therefore, a protective ventilation strategy can consist of LTV, PEEP, and IRM [24].

Breathing exercises have been widely used as a non-pharmacological therapy to effectively treat asthma. In 441 patients undergoing elective major open upper-abdominal surgery, a 30-min preoperative physiotherapy session (including education and breathing exercise training) strikingly reduced the incidence of PPCs (27% vs 12%) within 14 postoperative days, in particular for hospital-acquired pneumonia, but there was no a statistically significant reduction in LOS [7]. Yogic breathing (YB, also called Pranayama) is a collection of techniques to regulate breathing voluntarily. YB induces a valid relaxation response via the vagal and parasympathetic stimulation. YB can also stimulate salivary expression of nerve growth factor [25]. For patients with chronic respiratory disease, yoga practice (2 days a week for 8 weeks, a total of 16 sessions) also improved dyspnea, sleep quality, and fatigue [26]. Therefore, according to a systematic analysis of 113 articles, there are systemic benefits of yoga therapy, including the cardiovascular, respiratory, and musculoskeletal systems, along with the psychological impact and effect on quality of life in patients with chronic health conditions [27]. Similarly, the preoperative application of inspiratory muscle training significantly improves respiratory (muscle) function, and reduces the risk of PPCs in the early period after cardiothoracic or upper-abdominal surgery, but there was no effect on LOS [28], perhaps due to the repetition and strength of breathing exercises. We successfully used YB in patients after esophagectomy under EIT guidance. In addition to the decreased PPCs incidence, such as pulmonary atelectasis, pulmonary infection, pneumothorax, and pleural effusion, YB reduced levels of CRP, AST, and creatinine in the AICU.

Electrical impedance tomography (EIT) is very helpful in establishing the hypoxemia etiology at bedside via combining regional ventilation and perfusion information [29]. For critically ill patients or patients after liver transplantation, bedside EIT can be used as a real-time monitoring tool for evaluating the early phase distribution of pulmonary ventilation continuously [30]. Several clinical applications of EIT have been developed as clinical tools for estimation of lung collapse and overdistension, pneumothorax detection, and to monitor the effects of endotracheal aspiration on pulmonary volumes, ventilatory desynchrony, and pulmonary perfusion [11]. EIT and PET should agree in measuring and tracking changes in lung perfusion. An advantage of EIT is that it can be used to measure pulmonary perfusion at the bedside [13]. EIT can also detect regional differences between inspiratory and expiratory lung physiology. EIT allows in-depth characterization of ARDS phenotypes and can guide personalized ventilation settings [10]. Therefore, EIT can be applied to generate diagnoses and impact clinical decision-making and therapy planning [31]. Compared with a single constant PEEP, EIT-guided individualized PEEP from the beginning of durasuturing to extubation reduces the incidence of PPCs within 1 week after a craniotomy, and also reduces the length of ICU and hospital stays [32].

In our study, the improvements of lung function were immediately displayed by EIT during AICU, including CoV, DSS, and NSS. Finally, the total incidence of PPCs was comprehensively reduced.

In this trial, use of EIT increased the efficacy of postoperative breath exercise in preventing postoperative complications. However, our study has some limitations. First, the sample size was rather small, with mid-term follow-up, and it is hard to perform sub-group analysis according to the breathing exercise frequency or times. Second, the PPCs rate in our intervention group, although halved, was still 36.7%, higher than 31.4% [33]. Third, pre-surgical pulmonary function was not evaluated. Fourth, intraoperative characteristics, in particular the anesthetics consumption, may affect the respiratory muscle recovery. Fifth, postoperative pulmonary function was not assessed, and other postoperative complications were not recorded, such as cardiac complications, which may be risk factors for PPCs. Considering the high mortality associated with PPCs, more needs to be done to prevent PPCs in these high-risk patients, beyond preoperative physiotherapy education and postoperative ambulation.

Conclusions

We found that the postoperative breathing exercise effectively reduced the incidence of pulmonary complications and LOS after esophagectomy. EIT is a convenient way to evaluate pulmonary function in real time. Therefore, postoperative breathing exercise should be applied in the perioperative management for esophageal cancer patients.