The Association Between Preoperative Inspiratory Muscle Training
Variables and Postoperative Pulmonary Complications in Subjects
With Esophageal Cancer
Meike C Overbeek, Elja AE Reijneveld, Karin Valkenet, Edwin J van Adrichem, Jaap J Dronkers,
Jelle P Ruurda, and Cindy Veenhof
BACKGROUND: Preoperative inspiratory muscle training (IMT) is frequently used in patients
waiting for major surgery to improve respiratory muscle function and to reduce the risk of postoperative pulmonary complications (PPCs). Currently, the mechanism of action of IMT in reducing
PPCs is still unclear. Therefore, we investigated the associations between preoperative IMT variables
and the occurrence of PPCs in patients with esophageal cancer. METHODS: A multi-center cohort
study was conducted in subjects scheduled for esophagectomy, who followed IMT as part of a
prehabilitation program. IMT variables included maximum inspiratory pressure (PImax) before
and after IMT and IMT intensity variables including training load, frequency, and duration.
Associations between PImax and IMT intensity variables and PPCs were analyzed using independent samples t tests and logistic regression analyses, corrected for age and pulmonary comorbidities and stratified for the occurrence of anastomotic leakages. RESULTS: Eighty-seven
subjects were included (69 males; mean age 66.7 6 7.3 y). A higher PImax (odds ratio 1.016, P 5
.07) or increase in PImax during IMT (odds ratio 1.020, P 5 .066) was not associated with a
reduced risk of PPCs after esophagectomy. Intensity variables of IMT were also not associated
(P ranging from .16 to .95) with PPCs after esophagectomy. Analyses stratified for the occurrence of anastomotic leakages showed no associations between IMT variables and PPCs.
CONCLUSIONS: This study shows that an improvement in preoperative inspiratory muscle
strength during IMT and training intensity of IMT were not associated with a reduced risk on
PPCs after esophagectomy. Further research is needed to investigate other possible factors explaining the mechanism of action of preoperative IMT in patients undergoing major surgery, such as the
awareness of patients related to respiratory muscle function and a diaphragmatic breathing pattern.
Key words: inspiratory muscle training; postoperative pulmonary complications; cancer; surgery; physiotherapy; respiratory muscles. [Respir Care 2024;69(3):290–297. © 2024 Daedalus Enterprises]
Introduction
Despite advances in perioperative care in the last decades, the risk of postoperative pulmonary complications
(PPCs) after a major thoracic or abdominal surgery
remains high.1-3 The development of PPCs seems to be
related to a dysfunction of the diaphragm, the most important muscle used in inspiration, leading to a decreased
inspiratory capacity after surgery.4,5 Preoperative inspiratory muscle training (IMT), aimed at improving inspiratory muscle strength and endurance, can lead to an increase
of the inspiratory capacity and a better-quality deep breathing after surgery.6,7 Therefore, preoperative IMT can be
used to reduce the risk of PPCs after a major surgery.8,9 In
290
patients undergoing cardiac and upper abdominal surgery,
preoperative IMT has been shown to reduce the incidence
of PPCs.4,7,9-12 In patients undergoing esophagectomy,
PPCs are very common (27–57%).13-17 However, the effectiveness of IMT to reduce PPCs in patients undergoing
esophagectomy varies between studies.18-20
Based on the current evidence, IMT seems to result in
significantly improved inspiratory muscle strength after
training.10,18-20 Nevertheless, in previous studies, no clear
association between an improvement of the inspiratory
muscle strength and a reduced risk of PPCs after esophagectomy has been demonstrated.18-22 Therefore, the possible mechanism of action of IMT in reducing PPCs after
esophagectomy is still unclear.
RESPIRATORY CARE MARCH 2024 VOL 69 NO 3
PREOPERATIVE IMT IN ESOPHAGEAL CANCER
An improvement of the inspiratory muscle strength
and the effectiveness of IMT may also be related to the
training intensity of the performed IMT. Previous studies
indicate that a higher training intensity and an increase in
SEE THE RELATED EDITORIAL ON PAGE 382
training intensity during IMT seem to be associated with
a reduced risk of PPCs.18,20 There is no information
available on the influence of the training frequency and
duration of IMT on the risk of PPCs. To create more
insight into the possible mechanism of action of IMT, the
associations between inspiratory muscle strength and
training intensity and the risk of PPCs need to be investigated further. Therefore, the aim of this study was to
determine associations between preoperative maximum
inspiratory pressure (PImax) and intensity variables of
preoperative IMT with the occurrence of PPCs in subjects undergoing esophagectomy.
QUICK LOOK
Current knowledge
Preoperative inspiratory muscle training (IMT) seems to
reduce the risk of postoperative pulmonary complications
(PPCs) after major surgery. However, in previous studies,
no clear associations have been demonstrated between an
improvement of the inspiratory muscle strength, training
intensity of IMT, and a reduction of the risk of PPCs in
patients with esophageal cancer.
What this paper contributes to our knowledge
This study shows that a higher inspiratory muscle
strength or increase of inspiratory muscle strength was
not associated with a reduced risk of PPCs in subjects
after esophagectomy. Training intensity of IMT was
also not associated with the risk of PPCs after esophagectomy. These findings address the need for a better
understanding and possibly an alternative rationale for
IMT before major surgery.
Methods
Study Design
Current analyses were part of the preoperative intervention to improve outcomes in esophageal cancer patients
after resection (PRIOR) study, a multi-center, observational cohort study evaluating the implementation of prehabilitation to improve (inspiratory) muscle function, general
fitness, and nutritional status for patients with esophageal
cancer.
Participants and Procedures
Patients in the University Medical Center Utrecht,
University Medical Center Groningen, University Medical
Center Utrecht, University Medical Center Groningen,
Gelre Hospital, Isala Hospital, and Twente Hospital
Group were asked to participate in the PRIOR study from
March 2018–January 2020. The inclusion criteria were (1)
Mss Overbeek and Reijneveld and Dr Dronkers are affiliated with
Research Centre for Healthy and Sustainable Living, Research Group
Innovation of Movement Care, HU University of Applied Sciences
Utrecht, Utrecht, the Netherlands. Drs Valkenet and Veenhof are affiliated
with Research Centre for Healthy and Sustainable Living, Research
Group Innovation of Movement Care, HU University of Applied Sciences
Utrecht, Utrecht, the Netherlands; and Department of Rehabilitation,
Physiotherapy Science and Sport, University Medical Center Utrecht,
Brain Centre, Utrecht, Netherlands. Dr van Adrichem is affiliated with
School of Nursing, Hanze University of Applied Sciences, Groningen, the
Netherlands. Dr Ruurda is affiliated with Department of Surgery, University
Medical Center Utrecht, Utrecht, the Netherlands.
The authors have disclosed no conflicts of interest.
RESPIRATORY CARE MARCH 2024 VOL 69 NO 3
diagnosis of esophageal cancer and (2) scheduled for curative treatment consisting of neoadjuvant chemoradiotherapy and esophagectomy. There were no exclusion criteria.
Subjects followed a prehabilitation program consisting of
physical training and nutritional support as part of the
standard curative treatment pathway. Curative treatment
started with a 5-week schedule of chemoradiotherapy.
After completing chemoradiotherapy, subjects followed a
6–8 week training program consisting of overall physical
training and IMT under the supervision of a physiotherapist, combined with nutritional support by a dietitian.
After diagnosis and before the start of the medical treatment, subjects were informed about the study and asked to
participate by the physiotherapist. All subjects enrolled in
the study signed an informed consent for the use of their
treatment data for research. This study protocol was
approved by the medical ethics committee of the
University Medical Center Utrecht (protocol number 17–
533/C).
This work was supported by the foundation “Vrienden Integrale
Oncologische Zorg.”
Mss Overbeek and Reijneveld contributed equally to this study.
Supplementary material related to this paper is available at http://www.
rcjournal.com.
Correspondence: Elja AE Reijneveld MSc, HU University of Applied
Sciences, Heidelberglaan 7, 3584, CS, Utrecht, the Netherlands. E-mail:
elja.reijneveld@hu.nl.
DOI: 10.4187/respcare.11199
291
PREOPERATIVE IMT IN ESOPHAGEAL CANCER
Inspiratory Muscle Training
IMT was performed using an inspiratory threshold-loading POWERbreathe Medic Plus device (POWERbreathe,
Southam, United Kingdom). The starting level of the training sessions was based on the PImax at baseline. The resistance on the threshold ranged from 0–10, corresponding to
9–78 cm H2O (Figure S1; see related supplementary materials at http://www.rcjournal.com). The training consisted
of high-load training starting with 60% of the PImax in the
first week and 80% of the PImax from the second week.
Perceived exertion was rated using a Borg scale from 0
(no exertion) to 10 (maximal exertion). If the Borg scale
was < 5, resistance was increased by 5% of the measured PImax. Six series of 6 repetitions were performed per training session. When subjects reached the maximum resistance of 78 cm H2O, the number of repetitions per series was increased until an exertion of 5 on the Borg scale was achieved. Between each series, a resting period was scheduled. The first resting period was 60 s, and it was shortened to, respectively, 45, 30, 15, and 5 s after each subsequent series. Subjects performed the training twice a week under supervision of a physiotherapist and once a week independently at home. After each training session, the training load of the threshold trainer and the Borg score were recorded in a training log by the physiotherapist or the subject. The mean Borg score of all training sessions within a subject was calculated representing the average exertion of the subject. Measurements The PImax was measured with the respiratory pressure meter (Micro Medical RPM, PT Medical, Leek, the Netherlands)23 before and after the training period. Measurements were performed on a chair without armrests, with the subject holding the mouth pressure gauge in one hand and the other arm hanging next to the body or lying loose on the leg.24 A nose clip was placed on the subject’s nose, and after a maximum exhalation, the subject closed their lips tightly around the mouthpiece of the mouth pressure gauge. The subject inhaled as forcefully as possible for a minimum of 2 s and was encouraged by the physiotherapist. The test was repeated at least 5 times with a pause of at least half a minute. The highest measured negative pressure in cm H2O was noted. The test-retest reliability of the PImax in healthy subjects showed high reliability with an intraclass correlation coefficient of 0.78–0.8725 and a high reliability (r ¼ 0.97) in subjects with COPD.26 Intensity variables of the IMT included training load (from the first and last session, progress in training load, and mean training load), training frequency, and training duration. To determine the training load of each training 292 session, the recorded resistance on the device was converted to the corresponding training load in cm H2O (Figure S1 of online supplement, see related supplementary materials at http://www.rcjournal.com). The training load from each training session was also calculated as percentage of the PImax at baseline, and the mean training load in cm H2O of all training sessions within a subject was calculated. Progress in training load was calculated by subtracting the training load of the first IMT session from the last IMT session. The mean training frequency per week was calculated by dividing the total number of IMT sessions by the number of training weeks. Training duration included the total training period in weeks. Data on postoperative complications were obtained from the Dutch Upper Gastrointestinal Cancer Audit,27 including the occurrence of anastomotic leakage and PPCs. PPCs included pneumonia (diagnosed in the presence of new lung infiltrate, based on imaging, plus at least 2 of the 3 clinical signs: [1] fever, [2] purulent sputum, and [3] leukocytosis or leukopenia),28 pleural effusion requiring drainage, pneumothorax requiring treatment, mucus plug atelectasis requiring bronchoscopy, respiratory failure requiring re-intubation, acute aspiration, ARDS, and/or persistent air leakage requiring chest drainage.29 At the presence of one of these complications, a PPC was registered. The outcome measure in this study was the occurrence of PPCs (yes/no). Demographic and medical data were collected from the medical record and included sex, age, body mass index (BMI), pulmonary comorbidity, tumor location, tumor type, surgery procedure, and the American Society of Anesthesiologists physical status classification level. Statistical Analysis Statistical analyses were performed with IBM SPSS Statistics version 26 (IBM, Armonk, New York). Descriptive statistics were performed on the demographic and medical data. A histogram, Q-Q plot, and ShapiroWilk test were used to check whether demographic and medical data were normally distributed.30 In case of normal distribution, variables were described as mean and SD and in case of a skewed distribution as median and interquartile range. The independent sample t test (in case of normal distribution), Mann-Whitney U test (in case of nonnormal distribution), or chi-square test was used to determine differences between subjects with and without PPCs in demographic and medical data and in PImax (at baseline, follow-up, and change in PImax between baseline and followup) and IMT intensity variables. To determine progression during the IMT, a paired-sample t test or Wilcoxon signedrank test was used to test differences within the groups between PImax and training load at baseline and at the last training session. RESPIRATORY CARE MARCH 2024 VOL 69 NO 3 PREOPERATIVE IMT IN ESOPHAGEAL CANCER Table 1. Subjects included in PRIOR study 248 Excluded 161 Metastatic disease: 37 Wait and see approach/chemo: 22 Gastric resection: 5 Emergency surgery: 3 Too frail, no measurements: 8 Too frail, no surgery: 6 Declined measurements: 6 No IMT performed: 8 Died: 4 Other/unknown: 3 Missing IMT logbooks: 59 Baseline Demographic and Medical Characteristics of Subjects All (N ¼ 87) No PPCs (n ¼ 58) PPCs (n ¼ 29) P Male 69 (79.3) 43 (74.1) 26 (89.7) .09 Female 18 (20.7) 15 (25.9) 3 (10.3) 66.7 (7.3) 67.1 (6.9) 66.2 (8.3) .60 .92 Sex Age, y Total Age, y < 60 14 (16.1) 8 (13.8) 6 (20.7) 60–69 38 (43.7) 27 (46.6) 11 (37.9) 70–79 35 (40.2) 23 (39.7) 12 (41.4) 26.1 (3.7) 25.9 (3.8) 26.4 (3.5) .53 19 (21.8) 11 (19.0) 8 (27.6) .36 .51 BMI, kg/m2 Total Comorbidity Subjects included in the current study 87 Pulmonary comorbidity Tumor location Fig. 1. Flow chart. PRIOR ¼ preoperative intervention to improve outcomes in esophageal cancer patients after resection; IMT ¼ inspiratory muscle training. Logistic regression analyses were used to assess the association of PImax and IMT intensity variables with the occurrence of PPCs, corrected for age and pulmonary comorbidities.31,32 To investigate a possible interaction of the occurrence of anastomotic leakage on the association between PImax and IMT variables with PPCs, analyses were stratified for subjects with and without an anastomotic leakage. Odds ratios and 95% CIs were determined. The analyses were considered statistically significant if the P value was < .05. Results Between March 2018–December 2020, 248 subjects were enrolled in the PRIOR study. Of these subjects, 102 dropped out because they did not undergo surgery, measurements were stopped, subjects did not perform the IMT, or other reasons (Fig. 1). In addition, 59 subjects did not return an IMT log. Therefore, data from 87 subjects were analyzed, of which 69 (79.3%) were males and 18 (20.7%) females (Table 1). The mean age was 66.7 (SD 7.3) y, and the mean BMI was 26.1 (SD 3.7). PPCs were diagnosed in 29 (33.3%) of 87 subjects (Table 2). None of the demographic and medical data were significantly (P < .05) different between subjects with PPCs and subjects without PPCs. Demographic and medical data are presented in Table 1. PImax and IMT Variables The PImax and IMT variables in the total group and in subjects with and without PPCs are described in Table 3. The RESPIRATORY CARE MARCH 2024 VOL 69 NO 3 Intrathoracic, middle part 10 (11.5) 7 (12.1) 3 (10.3) Intrathoracic, distal part 72 (82.8) 47 (81.0) 25 (86.2) Esophagus-stomach transition 5 (5.7) 4 (6.9) 1 (3.4) Adenocarcinoma 76 (87.4) 49 (84.5) 27 (93.1) Squamous cell carcinoma 11 (12.6) 9 (15.5) 2 (6.9) Tumor type .25 Surgery procedure Transhiatal 3 (3.4) 2 (3.1) 1 (4.5) Transthoracic 84 (96.6) 63 (96.9) 21 (95.5) > .99
ASA physical status
Normal healthy patient
4 (4.6)
2 (3.4)
2 (6.9)
Mild systemic disease
51 (58.6)
33 (56.9)
18 (62.1)
Severe systemic disease
30 (34.5)
21 (36.2)
9 (31.0)
Constant life-threatening illness
1 (1.1)
1 (1.7)
Unknown
Anastomotic leakage
1 (1.1)
1 (1.7)
16 (18.4)
9 (15.5)
7 (24.1)
.78
.33
Data are presented as n (%) or mean (SD).
PPCs ¼ postoperative pulmonary complications
BMI ¼ body mass index
ASA ¼ American Society of Anesthesiologists
Table 2.
Postoperative Pulmonary Complications
Pneumonia
Pleural effusion requiring drainage
Pneumothorax requiring treatment
Mucus plug atelectasis requiring bronchoscopy
Respiratory failure requiring re-intubation
Persistent air leakage requiring chest drainage
22 (75.9)
8 (27.6)
2 (6.9)
2 (6.9)
5 (17.2)
5 (17.2)
Data are presented as n (%).
mean PImax increased from 77.6 (SD 28.8) cm H2O to 101.7
(SD 33.0) cm H2O in the total group. In the group without
PPCs, the mean PImax increased from 76.7 (SD 27.9) cm
H2O to 96.4 (SD 32.4) cm H2O (P < .001) and from 79.5 (SD 31.2) cm H2O to 112.1 (SD 32.4) cm H2O in the group with PPCs (P < .001). No significant differences between the groups were found in the PImax values (Table 3). 293 PREOPERATIVE IMT IN ESOPHAGEAL CANCER Table 3. Association of the Preoperative Maximum Inspiratory Pressure and Intensity Variables of the Inspiratory Muscle Training With Postoperative Pulmonary Complications Corrected for Age and Pulmonary Comorbidity Total PImax, cm H2O Baseline Follow-up Difference baseline-follow–up Training load, cm H2O First IMT Last IMT Difference first-last IMT Mean training load Training load related to PImax, % First IMT Last IMT Difference first-last IMT Mean training load Training parameters Training frequency per wk Total number of training sessions Total training period in wk All subjects, n Mean (SD) No PPCs, n Mean (SD) PPCs, n 79 72 65 77.6 (28.8) 101.7 (33.0) 22.8 (25.6) 53 48 44 76.7 (27.9) 96.4 (32.4) 18.8 (24.4) 26 24 21 79.5 (31.2) .68 1.002 (0.984–1.019) .85 112.1 (32.4) .058 1.016 (0.998–1.033) .07 31.3 (26.7) .064 1.020 (0.999-1.043) .066 86 86 86 85 40.3 (16.7) 56.3 (18.0) 16.0 (14.7) 50.0 (17.4) 57 57 57 56 40.5 (17.2) 55.1 (18.1) 14.6 (15.0) 49.6 (17.1) 29 29 29 29 39.9 (15.9) 58.7 (17.8) 18.7 (13.8) 50.9 (18.4) .88 .39 .22 .76 0.996 (0.968–1.023) 1.011 (0.985–1.037) 1.023 (0.991–1.057) 1.003 (0.977–1.029) .75 .40 .16 .84 78 78 78 77 54.3 (15.0) 81.1 (35.6) 26.8 (32.6) 69.9 (22.0) 52 52 52 51 54.8 (14.8) 81.0 (37.5) 26.2 (34.0) 70.2 (21.7) 26 26 26 26 53.2 (15.6) 81.2 (32.2) 28.0 (30.2) 69.5 (22.9) .65 .98 .81 .90 0.993 (0.961–1.025) 1.002 (0.988–1.015) 1.003 (0.989–1.018) 1.001 (0.979–1.023) .66 .83 .65 .95 84 86 85 3.0 (2.0) 20.6 (12.1) 7.6 (3.7) 55 57 56 2.9 (1.3) 19.7 (11.7) 7.2 (3.4) 29 29 29 3.2 (2.9) .46 22.5 (12.8) .30 8.4 (4.1) .14 Mean (SD) P OR (95% CI) P 1.091 (0.871–1.366) .45 1.018 (0.981–1.057) .34 1.092 (0.964–1.238) .17 *The number of subjects varies between the analyses because of missing values in the measurements. PPCs ¼ postoperative pulmonary complications OR ¼ odds ratio PImax ¼ maximum inspiratory pressure IMT ¼ inspiratory muscle training The IMT was performed at an average Borg of 4.4 (SD 1.2) in the total group. In the group without PPCs, the training load increased from 40.5 (SD 17.2) cm H2O at baseline to 55.1 (SD 18.1) cm H2O at the last training session (P < .001) and in the group with PPCs from 39.9 (SD 15.9) cm H2O to 58.7 (SD 17.8) cm H2O (P < .001). The training load as percentage from PImax at baseline increased from 54.8% (SD 14.8) to 81.0% (SD 37.5) in the group without PPCs (P < .001) and from 53.2% (SD 15.6) to 81.2% (SD 32.2) in the group with PPCs (P