Incentive Spirometry for Prevention of Postoperative Pulmonary Complications After Thoracic Surgery

IS Devices and Theory

Atelectasis management involves removal of retained secretions and stretching the lung tissue sufficiently to achieve parenchymal re-expansion.⁹ The “ideal” deep-breathing technique to re-expand collapsed alveoli was originally described by Bartlett et al,⁶ who reported that a long, slow inspiration with an inspiratory hold for several seconds is required to achieve optimal re-expansion. IS was designed to allow patients to perform this breathing technique more accurately and enthusiastically by providing them with visual feedback of their inspiratory effort.

There are 2 types of IS devices: flow-oriented and volume-oriented. Flow-oriented IS devices typically consist of 3 interconnected columns, each containing a lightweight plastic float that acts as a marker. The columns are connected to a mouthpiece through which the patient inhales. The deep-breathing exercise with this device involves the patient attempting to lift the float, through inspiratory flow, to a certain point in the columns for a certain amount of time. The volume-oriented device consists of a mouthpiece connected to a chamber with a visible scale. The deep-breathing exercise with this device involves the patient attempting to lift a marker as high as possible. Clinical practice guidelines recommend the volume-oriented devices to be used postoperatively because they are considered to impose lower work of breathing, pain, and fatigue.^10,11

Outcomes in Thoracic Surgery

In 2009, Agostini and Singh¹² reviewed 3 randomized controlled trials (RCTs), 1 cross-sectional study with historical control, and 1 systematic review comparing the effects of postoperative physiotherapy and IS in subjects who underwent thoracic surgery. The authors reported that there is little evidence of benefit of IS after thoracic surgery. They concluded that studies looking into the clinical effectiveness of IS after thoracic surgery were sparse in number, not adequately powered, and incomparable due to different outcome measures studied. They identified a need for more adequately powered studies with well-defined outcome measures to better evaluate the treatment benefit of IS in thoracic surgery patients. Since the publication of this review, there have been 4 further studies^13-16 with 8,166 subjects investigating the clinical effectiveness of IS in the postoperative management of thoracic surgery patients. Three were RCTs, and one was a population-based observational study (Table 1).

Liu et al¹³ performed a retrospective, population-based observational study utilizing Taiwan’s Longitudinal Health Insurance Database for Catastrophic Illness Patients. They identified 7,549 subjects with lung cancer who underwent surgical lung resection by video-assisted thoracoscopic surgery (VATS) or thoracotomy between 2000 and 2008. Subjects in the experimental arm were prescribed IS for 7 d after admission, and the control group included subjects who were not prescribed IS postoperatively. The experimental group had a lower prevalence of asthma and a higher proportion of subjects who had received neoadjuvant therapy (P < .05). They found that subjects in the experimental group had lower hospitalization costs in both the VATS and thoracotomy cohorts (P < .05). Subjects who underwent VATS and received postoperative IS had a lower incidence of pneumonia (3.09% vs 5.46%, P < .05), but a statistically significant reduction in pneumonia incidence was not observed in the cohort of subjects who underwent thoracotomy (5.56% vs 6.19%, P > .05). The authors concluded that there may be selective benefit conferred by IS in patients undergoing VATS with the reduced hospitalization cost and reduced incidence of pneumonia.

Malik et al¹⁴ performed an RCT with 387 subjects undergoing lung resection in Canada between 2014 and 2017. They investigated whether any incremental benefit was conferred by the addition of IS to standard physiotherapy. They confirmed similar patient demographics in both groups. Subjects in the experimental arm received IS in addition to standard physiotherapy, whereas subjects in the control arm received only standard physiotherapy. Subjects in both groups had a similar incidence of PPCs (12.3% vs 13.0%, P = .88). Examining individual complications also revealed no significant differences: pneumonia (P = .21), atelectasis (P > .99), requirement for bronchoscopy (P = .64), and requirement for mechanical ventilation (P = .69). The addition of IS also had no impact on length of hospital stay, with both groups showing a median length of stay of 4 d (P = .34). In a subgroup analysis of subjects undergoing minimally invasive or open thoracic surgery, the authors observed similar findings. The incidence of PPCs were reduced in the subgroup of subjects with COPD receiving IS, although this did not reach statistical significance (9.3% vs 16.4%, P = .39). The authors concluded that their findings did not support the addition of IS to the postoperative care of patients who underwent pulmonary resection. However, it is worth emphasizing that the authors were attempting to show an incremental benefit above physiotherapy, and despite this challenge, there was suggestion of benefit for subjects with COPD.

Gunay et al¹⁵ performed an RCT with 50 subjects who underwent thoracoscopy and pleural biopsy for pleural effusion in Turkey between 2013 and 2014. They compared the outcomes of respiratory physiotherapy with IS alone in the intervention group with the outcomes of physiotherapist-guided physiotherapy in the control group. The authors reported that subjects undergoing physiotherapy had a significantly greater improvement (mean ± SD) in FEV₁ after surgery (0.35 ± 0.2 L vs 0.2 ± 0.17 L, P < .001). However, improvements in FVC (P = .39), (P = .51) and (P = .45) were not significantly different between the groups. There was, however, a significantly reduced length of hospital stay (mean ± SD) in subjects receiving physiotherapy (8.63 ± 3.86 vs 10.12 ± 3.34 d, P < .001). The authors concluded that exercises performed under the supervision of a physiotherapist were more effective than IS alone in the postoperative recovery period; however, the authors did not comment on PPCs.

Agostini et al¹⁶ performed an RCT with 180 subjects undergoing thoracotomy and lung resection in the United Kingdom between 2008 and 2010. They aimed to investigate the effectiveness of IS compared with a control group who performed thoracic expansion with deep breathing exercises. There was a notable increase in age (P < .01) and American Society of Anaesthesiologists (ASA) grade (P = .04) for subjects in the control group. The incidence of PPCs was similar between the groups (12.5% vs 15.0% in the control group, P = .80). The authors also performed subgroup analysis on a group of high-risk patients who included subjects over the age 75 y, ASA grade ≥ 3, subjects with COPD, current/recent ex-smokers, or body mass index ≥ 30 kg/m². In this high-risk group, they found a reduced incidence of PPCs than in the experimental group (14% vs 23%, P = .41) but this did not reach statistical significance due to the small number of subjects. The difference in frequency of PPC, although nonsignificant, was relatively large in subjects with COPD at 19% and current smokers/ex-smokers of ≤ 6 weeks at 15%, with 95% CIs that indicated improved outcomes in those subjects receiving IS. The authors found no significant difference in postoperative lung function, frequency of PPCs, or length of stay between the 2 groups and concluded that their data did not support the hypothesis that IS improves recovery of lung function or reduces the incidence of PPCs. However, they identified that it is difficult to rule out the benefit of IS in high-risk patients.

Outcomes of IS in Other Surgical Specialties

IS use is not restricted to thoracic surgery, and its use is widely documented in other surgical specialties. Odor et al¹⁷ conducted a systematic review and meta-analysis of 95 RCTs with 18,062 subjects that looked at perioperative interventions including IS for prevention of PPCs in noncardiac surgery patients. They concluded that there is evidence of moderate quality that IS offers no benefit in preventing PPCs compared to standard medical care (risk ratio 1.06, 95% CI 0.85–1.34).

A Cochrane review from 2012 investigated the use of IS compared to physical therapy, positive-pressure breathing techniques (including CPAP, bi-level positive airway pressure, intermittent positive-pressure breathing, and active cycle of breathing techniques), or preoperative patient education for preventing PPCs after coronary artery bypass graft surgery.¹⁸ That review included 7 studies with 592 subjects. The authors found no difference between the groups in the rates of PPCs. Subjects treated with IS had worse pulmonary function and arterial oxygenation compared with the positive-pressure breathing group. The authors concluded that there was no evidence that IS reduced PPCs or improved pulmonary function after coronary artery bypass graft surgery.

A Cochrane review published in 2014 investigated the use of IS for preventing PPCs after upper abdominal surgery.¹⁹ The authors included 12 studies with 1,834 subjects. Four studies (152 subjects) compared IS with no respiratory treatment and reported no statistically significant difference between the 2 groups for clinical complications (relative risk 0.59, 95% CI 0.30–1.18). Two studies (194 subjects) compared IS with deep breathing exercise and found no statistically significant differences between the 2 groups in the meta‐analysis for respiratory failure (relative risk 0.67, 95% CI 0.04–10.50). Two other studies (964 subjects) compared IS and other chest physiotherapy and found that there was no statistically significant difference between the 2 groups in terms of the risk of developing PPCs. The authors concluded there was no evidence that IS is effective in preventing PPCs.

The other area of surgery in which IS is commonly reported is bariatric surgery.²⁰ Pantel et al²⁰ conducted an RCT comparing IS with no IS after bariatric surgery. They found no significant effect of IS in preventing PPCs. However, they identified a low adherence rate for IS use, with IS use at about 4 times per day on postoperative day 1 and 10 times per day on postoperative day 2 compared to the prescribed 10 times per hour.

Should We Persevere With IS in Postoperative Patients?

Considering these data together, there is little evidence supporting the widespread routine use of IS in all patients after thoracic surgery. However, several studies hint at a treatment benefit conferred by IS in high-risk patients, such as those with COPD.^14,16,21 Despite these generally neutral results, IS remains widely used, which suggests that thoracic surgical teams have some confidence in their ability to encourage patients to undertake breathing exercises. It is possible that trials struggle to demonstrate a benefit because subjects within these trials receive enhanced physiotherapy compared to what is normally available outside of a trial setting, perhaps due to the Hawthorne effect.²²

It is worth noting that there are significant limitations within trials on the use of IS. For example, there is a lack of standardization on the prescription and use of IS between studies, difficulties in achieving blinding with the use of IS devices, lack of appropriate control comparisons, and an inability to isolate IS effects due to co-intervention.⁸ These trials may also be underpowered to show differences for the higher-risk groups, which has been suggested in several trials.^11,17 It is also widely recognized that there is significant heterogeneity in the definition of PPCs in the published literature and that this term is used loosely.²³ It is therefore likely that the definition of a chest infection differs significantly between the studies.

Furthermore, in a study that examined patient use of IS in postoperative subjects, 26.2% were not using the device correctly, and 38.1% denied using the device at all.²⁴ In a similar study that surveyed care providers on their thoughts on IS use in clinical practice, 86% of respondents believed adherence was poor due to patients forgetting to use IS devices, not using them correctly, and not using them frequently enough.²⁵ Although this may be true, the clinical reality is that providing more patient training and encouraging more frequent use will come at a cost, in addition to the cost of the devices, which may become difficult to justify in the context of numerous studies questioning the clinical benefit of IS. Eltorai et al²⁶ evaluated the financial impact of implementing IS using prospective questionnaires to evaluate the time spent by health professionals doing IS-related activities and national survey work load data. The authors estimated that the total annual cost of implementing postoperative IS to be $1.04 billion (95% CI $949.4 million to $1.13 billion). One strategy that may be useful in improving adherence and technique is educating patients in the use of the device preoperatively rather than introducing them to the device after surgery, when they are potentially in pain and may not be able to concentrate as easily.⁷

It should be considered that these studies attempt to demonstrate equivalence or superiority of IS over physiotherapy. However, these should not be considered as independent entities, and it is likely that the maximum clinical benefit could in fact come from the use of IS as an adjunct to physiotherapy. In support of this suggestion, IS was included as a component of the multidisciplinary I-COUGH patient care program, which was demonstrated to result in almost halving the incidence of postoperative pneumonia.²⁷ It should also be considered that correct use of IS may be particularly beneficial for patients after discharge when motivated patients will continue to perform IS at home when physiotherapy is no longer available. The stay following thoracic surgical procedures is typically on the order of a few days, so the true benefit of IS in reducing complications may in fact occur after discharge and may explain why it has been difficult to demonstrate benefit in studies typically focusing on in-hospital outcomes.

Summary

For patients undergoing thoracic surgery, there is limited evidence of clinical benefit in widespread routine use of IS. There is a suggestion of potential benefit for patients at higher risk for the development of PPCs, such as those with COPD, which needs to be investigated further. Despite various studies questioning the efficacy of IS, it continues to be relatively widely used. This may reflect belief that when used effectively, IS can act as a significant motivator to patients to continue performing breathing exercises and contribute to reducing the incidence of PPCs. However, this remains difficult to prove in clinical trials. It may be that, with appropriate preoperative training, and its use as a component of a multidisciplinary package of care, IS can be an effective intervention with benefits lasting long beyond the hospital stay.