Weiner et al.45, through a randomized control trial, studied the effect of specific inspiratory muscle training with incentive spirometry (IS) started two weeks before surgery and maintained until three months after. Compared to a control group that did not receive any respiratory training, the experimental group had significant improvements in FEV1, FVC, and MIP before surgery, which became even more pronounced at the end of the study. In this study, despite the apparent benefits of IMT, the weight of preoperative and postoperative training periods to the results was not ascertained.

Sekine et al.46 obtained similar results in a preoperative study with COPD candidates for pulmonary surgery. Apart from the IS training, candidates were also advised to walk 5000steps/day for two weeks prior to surgery. They were instructed about diaphragmatic breathing exercises, pursed-lip breathing, huffing and coughing for 15min after nebulization with bronchodilation, five times a day. At baseline, the FEV1 in the intervention group was significantly lower than in the control group composed by historical data (1.8±0.5L vs. 2.0±0.45L, respectively; P=0.04), and despite that the intervention group had a significant LOS reduction (21.0±6.8 vs. 29.0±9.O P=0.0003) and a lower rate of PPC (54.5% vs. 70%, respectively; P=0.202). Despite the limitation inherent to the lack of a true control group, the results support the importance of physiotherapy intervention programs, particularly those involving physical exercise to prevent PPC. This was later reinforced by a study in severe COPD patients,47 who did not meet the inclusion criteria for lung surgery due to their reduced maximal oxygen consumption (VO2max). Patients underwent a four-week program of peripheral muscle training on a cycle ergometer with an initial load of 50% of their maximum work load on CPET, and a weekly load increase of 10% until reaching 80%. Patients were also encouraged to practice breathing exercises and IS twice a day. Despite the absence of lung function improvements, the aerobic capacity was significantly enhanced after training, fitting the inclusion criteria for lung surgery. Interestingly, the initial VO2max of 13.5ml/kg/min was enhanced to 16.3ml/kg/min after training, being 15ml/kg/min the accepted VO2max cut-off for pulmonary resection. However, the lack of a control group weakens the conclusions in this study regarding the true effect of the intervention.

More recently, Pehlivan et al.48 in an RCT examined the effect of a short-time intensive program in patients awaiting pulmonary surgery. The program started one week before surgery and was continued during the hospital stay, which was similar to others performed in the previous studies but with the advantage of assessing a larger range of variables and having a real control group. All patients were assessed before surgery for heart rate, performance in the six-minute walk test, and pulmonary function, including, FVC, FEV1, DLCO, oxygen arterial pressure, and carbon dioxide arterial pressure. The experimental group showed significant improvements in all parameters compared with the control group. Some patients from the intervention group were submitted to a ventilation/perfusion ratio (V/Q) scan, which was repeated one week after intensive training, and showed advantageous changes in the perfusion distribution. Indeed, the V/Q suffered a reduction around the tumour side and an improvement in the contra-lateral lung, probably modulated by the physiotherapy program. Although the modus operandi was unknown, the authors speculated that the exercise-training program might have induced a hypoxic constriction of the pulmonary arterial branches around the tumour, and a significant improvement in V/Q in the healthy lung, which would explain the better gas-exchange after the program. Regarding PPC and LOS, both were also significantly reduced in the intervention group.

According to Pouwels et al.49 preoperative exercise therapy (PET) might be an important component to reduce PPC, mortality and LOS; and to improve physical fitness and quality of life. Curiously, the authors found two studies indicating a decreased incidence of persistent air leak, although it was not possible to assume a relationship with PET.

Based on the above, and similarly to what Bartels et al.50 recommend for lung-volume reduction surgery, multidisciplinary interventions in the preoperative period should also be focused on smoking cessation and healthy lifestyle advice, pharmacological optimization, nutritional counselling, stress management, bronchial hygiene and an exercise training program.

Many studies have focused on physiotherapy treatment in the perioperative period. Some have attempted to analyse the importance of physiotherapy in patients undergoing pulmonary resection, but apparently, all have methodological flaws, such as the absence of control group. Indeed, regarding this subject, due to ethical reasons, it is difficult to conduct a study without any kind of approach or recommendation to patients who do not receive treatment. The study of Reeve et al.51 has been frequently classified in the literature as the only true randomized controlled trial (RCT) in which the control group did not receive treatment for respiratory physiotherapy. In that study, the authors analysed the effectiveness of a postoperative physiotherapy program on PPC and LOS in patients submitted to different types of lung resection by posterolateral thoracotomy. They have concluded that routinely respiratory physiotherapy is not necessary in such patients. Nevertheless, this conclusion must be carefully analysed, because in this study the control group was not blinded towards physiotherapy, since before surgery the patients received a document explaining the need for breathing and coughing exercises, early deambulation, mobilization, and shoulder exercises. Moreover, although patients did not formally receive any physiotherapy intervention, they were encouraged to carry out exercises that they had learned before. Additionally, the exercise program in the treatment group consisted only of 20 repetitions of breathing exercises with three seconds apnoea in the end of inspiration, two forced expirations and coughing, walking 15m, mobilization exercises, active or active-assisted shoulder mobilization, five repetitions of each exercise, shoulder flexion, horizontal abduction and adduction. These exercises lasted only 15min, which can be considered the minimum treatment as the author stated. At baseline the groups were similar except for body mass index (BMI), which was higher in the treatment group with an average of 29kg/m2, and 45% presented a value ≥30kg/m2, while in the control group the average was 26.5kg/m2 and only 15% presented a value ≥30kg/m2. In addition, all patients included in this sample showed good pulmonary function. Besides being widely accepted as a hallmark study, the above-mentioned methodological flaws, such as not controlling for BMI differences between groups at baseline, as well as the lack of recording of the frequency of autonomous training in both groups, seriously compromise the conclusions. Indeed, another study in this topic has been frequently criticized due to the lack of a real control group, which was composed by historical data.52 Briefly, the study was conducted in a hospital unit that did not routinely perform respiratory physiotherapy interventions until 2002, but since then an intensive postoperative physiotherapy program has been implemented for all patients. For that reason, the control group encompassed the patients submitted to lung surgery prior to 2002. This physiotherapy program intervention consisted of maximal inspiratory manoeuvres, coughing and mobilization exercises of the upper limbs, and supervised exercise on cycle-ergometer or treadmill. The effectiveness of the program was evaluated for outcomes such PPC, LOS and hospital costs. Regarding the preoperative assessed parameters, the groups were similar at baseline. The physiotherapy intervention did not bring any improvements in mortality rates or pneumonia when compared to controls, although there was a trend to a reduction in the physiotherapy group. Reinforcing this tendency, the physiotherapy group evidenced a significant reduction in the prevalence of atelectasis compared to the historical group.52 Responding to methodological criticisms of the Varela et al. study, Novoa et al.53 used the same data but with a different approach (i.e., through a propensity matched analysis), and clearly showed that the intensive physiotherapy program significantly reduced the patients’ morbidity. This conclusion was based on 359 pairs of patients (before vs. after 2002) using a logistic regression where the dependent variable was the PPC, assessed by radiological analysis, leucocytosis and body temperature, the independent variables being age, BMI, FEV1% and ppoFEV1, the Charlson index, and the clinical T status. This quasi-experimental study supported by a solid analysis, weakens the criticisms made of the study of Varela et al.52 and reinforces the importance of a regular and intensive physiotherapy program to avoid PPC, enhancing the recovery process of these patients.

Despite the scientific evidence supporting the physiotherapy programs performed routinely in surgical patients, uncertainties still remain regarding the advantages of external respiratory devices, such as IS, or PEP Mask, among others, to the traditional chest physiotherapy interventions. For instance, it is common to associate the IS device with the chest physiotherapy program; however, this practice was questioned by Gosselink et al.54 in a RCT that assessed patients submitted to lobectomy or to oesophageal resections. For each surgery the sample was divided into a control group, which received physiotherapy, and an intervention group that received both physiotherapy and the volumetric IS device. Patients started the program preoperatively, and continued after surgery until hospital discharge. As expected, there was a decline in pulmonary function in all patients immediately after surgery, but without differences between the groups in their recovery rate and in pulmonary complications. Additionally, Agostini et al.55 reviewed the efficacy of the routine use of IS devices in patients submitted to thoracic surgery and corroborated the findings of Gosselink et al.54 concluding that physiotherapy with or without IS is equally effective in reducing postoperative pulmonary complications and enhancing pulmonary function. Agostini et al.55 also state that IS cannot replace nor significantly improve the physiotherapist's work. Thus, a well-organized and regular physiotherapy program seems to be the most effective approach for recovery and PPC prevention. Later, Agostini et al.56 performed a RCT to investigate the effect of IS compared with deep breathing exercises in patients submitted to lung resections by thoracotomy in terms pulmonary function recover, PPC development and LOS, and also adjusted according to the patients’ risk stratification and concluded that IS is not more effective in pulmonary function improvement or in the PPC reduction. However, it is possible to observe a tendency with no statistical significance of more favourable result for PPC frequency in the high risk patients.

Regarding the use of other external devices associated with physiotherapy treatment, the results reported in literature do not allow safe conclusions regarding its advantage/costs. Indeed, Orman et al.57 conducted a systematic review of the effectiveness of using simple devices that allow breathing exercises with positive expiratory pressure (PEP) in postoperative abdominal and thoracic surgery. Although there are many devices available to achieve this goal, it is important to note that due to the scarcity of literature, this review only includes the blow-bottle system and the PEP-mask®. Moreover, from a methodological point of view and considering the PEDro scale, the low quality of these studies must be highlighted. From a conceptual point of view, it must be outlined that the positive expiratory pressure (PEP) can be used either to increase or to reduce lung volume, depending on the respiratory technique applied, and none of the included studies referred to the breathing pattern or the pressure applied. Consequently, well-designed studies about this issue are required in order to clarify the advantage of using these external devices in surgical patients. It would also be interesting to find out the extent to which the association of non-invasive ventilation (NIV) with a conventional physiotherapy program could be an asset to reduce PPC in patients undergoing pulmonary resection. Freynet et al.58, in a review article including five studies, reported that all were unanimous in demonstrating the effectiveness of NIV therapy in association with physiotherapy, either to treat or to prevent PPC after lung resection.

It is widely accepted that postoperative pain is highly restrictive to patients, particularly after posterolateral thoracotomy. Thus, pain control is necessary not only for the patients’ comfort, but also, to assure better cooperation with physiotherapy treatment, and therefore to achieve faster improvements in their pulmonary function. It is worth noting that analgesic drugs as opioids may have a detrimental effect on pulmonary function and, therefore, it is now recommended to use analgesia in the form of ropivacaine and sufentanil via epidural catheter rather than intravenous morphine administration, with significant improvements in pulmonary function recovery after lung surgery.59 Therefore, patients with epidural pain control had more rapid improvement in lung volume without increased airway resistance, reducing the likelihood of pulmonary complications such as atelectasis and pneumonia. Regarding the disadvantages of analgesia via epidural catheter, including hypotension, urinary retention, respiratory muscle weakness, and neurological injury risk, other analgesic alternatives have been used, such as thoracic paravertebral block, which proved to be a good and safe alternative for thoracotomy pain control. In a recent meta-analysis comparing thoracic epidural analgesia with thoracic paravertebral block, differences in terms of pain control efficacy and opioid consumption were not found between the two procedures. However, patients submitted to thoracic paravertebral block revealed a lower incidence of urinary retention and hypotension comparatively to those submitted to epidural analgesia.60

The use of analgesia by application of transcutaneous electrical nerve stimulation (TENS) can be an asset as a technical adjunct to control severe pain. This technique was effective in reducing the administration of opioid analgesics, and significantly increased FVC, positively influencing the cooperation of the patients with the physiotherapy treatment and improving the extent of chest and shoulder mobilization.61 Since thoracotomy leads to shoulder functional limitations, with loss of shoulder mobility, strength, and pain, Reeve et al.24 carried out a study in which all participants received an explanatory document about the importance of early mobilization and deambulation, changing position in bed regularly, as well as the importance of performing upper limb and breathing exercises. The intervention group was submitted to a daily physiotherapy program, including active-assisted shoulder mobilization. The program finished at hospital discharge, but patients were advised to continue the program by themselves. They also received a diary to record the exercises performed during the following three months. Patients were evaluated before surgery, on the day of hospital discharge, and one and three months after discharge. Pain, shoulder function, shoulder mobility and strength, and quality of life were evaluated. At hospital discharge the intervention group had substantially less pain, both at the shoulder and the chest. The same results were observed one month and three months after discharge. Shoulder function was also significantly better in this group, but with no significant differences in mobility, strength and quality of life. Oddly enough, this was the only article found that tested a shoulder treatment protocol after pulmonary resection.

After hospital discharge, patients still present dyspnoea, pain, fatigue, and changes in lung function, and they still are highly limited in their daily activities, with significant loss of their quality of life. Considering these issues, Cesario et al.62 conducted a study with a physiotherapy program after-discharge. Unfortunately, this study loses strength in the evidence, since patients in the intervention group were enrolled by convenience, and therefore, the study was not a RCT. The intensive physiotherapy program lasted one month, five days/week, with session duration of 3h. The program included a supervised incremental cycle ergometer exercise until the patient reach 30min of continuous exercise at 70–80% of the load achieved in CPET at baseline. Exercises to strengthen the abdominal muscles, upper and lower limb training, inspiratory muscle training were included, as well as educational sessions on lung pathophysiology, pharmacology, nutrition, relaxation and stress management, principles of energy conservation, and breathing control. Since this program demanded a prolonged hospitalization of one month, it is interesting to note that only 25 out of the 211 eligible patients accepted to participate in the program, all of them being the ones with worse clinical status at baseline. The remaining 186 patients agreed to take part in the control group. Despite the better status of the control group at baseline, one month after hospital discharge there was an inversion in the condition of both groups, with the intervention group showing higher improvements in functional parameters in contrast to the impaired functionality observed in control group. Even with methodological limitations, it is clear that the physiotherapy program accelerated functional recovery, with improvement in dyspnoea and exercise tolerance, without compromising the adjuvant oncologic treatment.

After pulmonary resection patients are often submitted to additionally adjuvant therapies, such as chemotherapy and radiotherapy. Unfortunately, these therapies induce permanent changes in these patients. During the treatment period, fatigue is one of the most limiting factors and causes a significant negative impact on quality of life. It is not yet clear whether this symptomatology is directly related to the malignancy, to a side effect of these therapies, or if it arises from a combination of both.63 Thus, like observations in other types of cancer/diseases with the same kind of limitations, such as COPD and heart failure, it is possible that an exercise training program applied to lung cancer patients might also be effective in controlling symptoms and improving work capacity.

Indeed, the application of an exercise program seems to have a positive impact on a wide variety of neoplasms, as described by McMillan et al.64, although lung cancer patients are not included in any of these studies. Regarding lung cancer, it might be expected that a positive impact of physical training on these patients would be seen because pulmonary surgical resection, in parallel with other factors, such as COPD and cardiac overload, tends to result in a decreased exercise tolerance. According to Nagamatsu et al.65, long-term recovery of lung function and exercise capacities seem to have different timings, because pulmonary function recovers faster as the pain gets progressively less severe. Lung function seems to be regained in the first three months, while the recovery of exercise tolerance is slower, reaching a maximum after one year. Considering this subject, there are some doubtful results in literature presented by studies with several methodological flaws.66 Nevertheless, there are studies supporting the beneficial effects of exercise training with robust methodology such as the one from Spruit et al.30 These authors conducted a non-randomized pilot study of a multidisciplinary rehabilitation program of eight weeks in patients undergoing surgical treatment and complementary cancer therapy. The program was focused on general exercise training following the recommendations of the programs preconized for COPD patients.67 All patients presented at baseline a restrictive lung function, with abnormalities in diffusion and exercise intolerance. After eight weeks, they showed no improvement in lung function, but in contrast showed significant improvements in exercise capacity. Worth noting are the late start of the exercise program, which was at least three months after completing intensive lung cancer treatment, and the absence of specific respiratory exercises on this training program. Consequently, improvements in the pulmonary function after such long time following surgery were not expected.