Pneumothorax in patients with COPD and emphysema receiving home chronic non-invasive ventilation: is it the emphysema phenotype or ventilator setting?

Abstract

We describe three patients with chronic obstructive pulmonary disease (COPD) and emphysema who developed a pneumothorax while receiving chronic home non-invasive ventilation (NIV). These cases raise the question whether the high alveolar pressures given by NIV may have contributed to the development of their pneumothorax by barotrauma. Pneumothorax in patients with COPD receiving NIV is uncommon, the pressures in our patients with COPD who developed pneumothorax were not extremely high and time to development of pneumothorax was relatively long after the initiation of NIV. Further, in our patients, the CT scan showed paraseptal emphysema, a known risk factor for pneumothorax. This suggests that COPD/emphysema phenotype is probably a more important factor for indicating pneumothorax risk than ventilator settings. Better phenotyping of patients with COPD in whom benefits of NIV can be expected at minimal risk of serious side-effects is needed to inform our patients properly and bring the field of chronic NIV in COPD forward.

Background

This article describes three cases of patients with chronic obstructive pulmonary disease (COPD) and emphysema who developed a pneumothorax while receiving chronic home non-invasive ventilation (NIV). Our described cases are part of a growing group of patients with COPD and emphysema who are on chronic home NIV. These are frail patients with a decreased cardiopulmonary and physical reserve. As a consequence, pneumothorax in these patients leads to high morbidity and mortality. Emphysema is a widely known risk factor of pneumothorax in patients with COPD, caused by rupture of bullae or blebs. Our described cases raise the question whether the relatively high NIV pressures that are used in patients with COPD contribute to pneumothorax risk due to barotrauma.

Case presentation

Patient A was a female with COPD GOLD IV/D and severe emphysema due to alpha-1 antitrypsin deficiency. Pulmonary function tests showed forced expiratory volume in 1 s (FEV₁) 0.51 L (17% predicted) and forced vital capacity (FVC) 1.46 L (37% predicted). CT of the thorax showed severe paraseptal emphysema (figure 1A). In her late 50s, NIV was initiated because of mild hypercapnia (partial pressure of carbon dioxide (pCO₂) 6.0 kPa) and symptoms of nocturnal hypoventilation. NIV was initiated and titrated during a tertiary pulmonary rehabilitation programme. She was ventilated with bilevel positive airway pressure with an inspiratory positive airway pressure (IPAP) of 18 cmH₂O, expiratory positive airway pressure (EPAP) of 4 cmH₂O and backup respiratory rate (BURR) of 10 cycles/min. She received continuous oxygen at 3 L/min. She was accepted for lung transplantation. Eight months after the start of NIV, the patient developed a large left-sided pneumothorax, which was successfully treated by drainage. Afterwards, NIV was restarted at the same settings. Three months later, she experienced increased dyspnoea and was therefore not able to tolerate the NIV. A couple of weeks after this, she appeared to have a recurrent left-sided pneumothorax and again a chest tube was placed. After 7 days, she was discharged from the hospital because the pneumothorax had resolved. NIV was stopped because of the recurrent pneumothorax. Two weeks later, the patient received a bilateral lung transplantation.

Figure 1.

Slides of CT of the thorax of our three cases with chronic obstructive pulmonary disease/emphysema. Slides with most severe emphysema per patient are shown. (A) Patient A: paraseptal emphysema predominantly in right middle lobe and left lower lobe. (B) Patient B: paraseptal and centrilobular emphysema predominantly in right upper lobe and right lower lobe. (C) Patient C: centrilobular and paraseptal emphysema predominantly in upper lobes. Slides of patients A and B are made before start of non-invasive ventilation (NIV), while slide of patient C is made after start of NIV during hospitalisation due to pneumothorax.

Patient B, a woman in her late 50s, started NIV for chronic respiratory failure (pCO₂ 7.4 kPa) due to COPD GOLD IV/D, also during a tertiary pulmonary rehabilitation programme. CT of the thorax showed severe paraseptal and centrilobular emphysema (figure 1B). Pulmonary function tests showed FEV₁ 0.46 L (19% predicted) and FVC 1.67 L (59% predicted). She was initiated on NIV because of chronic respiratory failure, with an IPAP 26 cmH₂O, EPAP 8 cmH₂O and BURR of 16 cycles/min. NIV had good effect; gas exchange improved and she experienced reduced symptoms of fatigue. Five years later, the patient was hospitalised because of pneumothorax on the right side that was treated by a small pleural catheter. NIV was stopped before hospital admission by the patient. A CT of the thorax showed occlusion of right lower lobe bronchi by sputum. A high amount of sputum was removed by bronchoscopy. The pneumothorax did not resolve and 3 days later, a thicker drain was placed. After a couple of days, the drain could be removed. However, there was increased respiratory failure (pCO₂ 11.2 kPa) and the patient was not able to tolerate the NIV. Active treatment was stopped in agreement with the patient and she died after a couple of days.

Patient C is a patient in his 60s, with COPD GOLD IV/B (FEV₁ 0.9 L (27% predicted), FVC 3.24 L (75% predicted)) and recently diagnosed obstructive sleep apnoea (apnoea/hypopnoea index 69 per hour). CT of the thorax showed severe centrilobular and paraseptal emphysema (figure 1C). He suffered from dyspnoea, impaired sleep quality and exhaustion. Because his blood gas analysis showed chronic hypercapnia (pCO₂ 8.3 kPa), he was initiated on NIV. NIV was titrated in a hospital setting to IPAP of 27 cmH₂O, EPAP of 8 cmH₂O and BURR of 10 cycles/min, with added oxygen at 2 L/min. Four months later, the patient was hospitalised because of COPD exacerbation due to an upper airway infection. X-ray of the thorax showed a small pneumothorax on the right side, and therefore a conservative approach was chosen. However, the patient experienced increased dyspnoea and worsened respiratory rate. Therefore, a surgical drain was inserted and symptoms improved. Afterwards, NIV was restarted. Two months later, the patient was again admitted because of exacerbation of COPD and a pneumothorax on the right side. The pneumothorax was again approached conservatively because it was a small pneumothorax; NIV was continued and COPD exacerbation treatment was given. The patient recovered clinically but a small pneumothorax remained. After 1 month, the patient was admitted for the third time because of COPD exacerbation. No pneumothorax was seen on X-ray of the thorax. During a couple of days, NIV treatment was extended to daytime and night-time because of increased respiratory failure. He recovered and NIV could be tapered to only night-time. The patient went to a rehabilitation clinic for a couple of weeks because of his decreased physical condition. There he recovered further and went home.

Discussion

The aim of chronic NIV in patients with COPD is to relieve symptoms and improve health-related quality of life by reducing hypercapnia caused by alveolar hypoventilation. Alveolar hypoventilation in patients with COPD is induced by decreased inspiratory muscle capacity on one hand (eg, diaphragm dysfunction) and increased load on the respiratory system on the other hand. To achieve effects on health-related outcomes, high inspiratory pressures aiming at a maximal reduction in CO₂ are necessary, called high-intensity NIV (HI-NIV).^1–3 In contrast to studies that used inadequate inspiratory pressures, HI-NIV has shown beneficial results on health-related quality of life, sleep quality, lung function, exercise tolerance and gas exchange.⁴ Known side-effects of NIV are dynamic hyperinflation because of high backup respiratory rate, discomfort of the patient (eg, due to mask leaks or mask decubitus) and aerophagia.

To our knowledge, in literature, not much is known about barotrauma being a side-effect of HI-NIV in patients with emphysema. One case report describes a patient in his 70s with end-stage COPD who developed pneumothorax while receiving chronic home NIV.⁵ Ventilator settings in this case are not described. In patients with cystic fibrosis receiving home chronic NIV, a case report published by Haworth and colleagues describes the development of pneumothorax in three patients. These patients were on the transplant waiting list, indicating the severity of their underlying lung disease that contributed to the development of their pneumothorax.⁶ Cho and colleagues performed a retrospective study that analysed the risk of pneumothorax in 176 patients with Duchenne muscular dystrophy who were admitted for chronic home NIV. Sixteen (9.0%) of these patients developed pneumothorax. Most patients with pneumothorax showed underlying pulmonary abnormalities, had low body mass index and were young boys—all known risk factors for pneumothorax.⁷ Further, ventilator settings were relatively low, aimed at IPAP of 20 cmH₂O. It is therefore not known whether NIV contributed to the development of pneumothorax in these patients.

In our hospital (University Medical Centre Groningen), pneumothorax in patients with COPD receiving NIV is uncommon. During the last decennium, we identified three patients with COPD (out of 389 patients with COPD initiated on chronic home NIV in our centre) who developed pneumothorax while receiving HI-NIV. Despite of this low incidence, the high burden of pneumothorax in our patients emphasises the need to investigate risk factors for pneumothorax. Reasonably, the high alveolar pressures caused by NIV in these patients may have contributed to the development of their pneumothorax by rupture of blebs. However, IPAP pressures in the presented cases were 18, 26 and 27 cmH₂O, respectively, which are pressures that are quite common in our population with COPD on NIV. For comparison, in the total group of patients with COPD initiated on NIV in our centre in the last 10 years, the mean IPAP was 23±4 cmH₂O with a BURR 13±3 cycles/min. Furthermore, time to development of pneumothorax was relatively long after the initiation of NIV. Additionally, in our patients, the CT of the thorax showed paraseptal emphysema being an important risk factor for pneumothorax. This suggests that COPD/emphysema phenotype is probably a more important factor for indicating pneumothorax risk than ventilator settings.

Currently, we do not consider emphysema/COPD phenotypes while preselecting patients for home chronic NIV, while both risk for side-effects (as a pneumothorax) as well as clinical effect and benefit may depend on this. Better phenotyping of patients with COPD and selecting patients in whom benefits of NIV can be expected at minimal risk of serious side-effects are needed to inform our patients properly and bring the field of chronic NIV in COPD forward.

Learning points.

• Patients with chronic obstructive pulmonary disease (COPD) and emphysema receiving home non-invasive ventilation (NIV) are often frail patients, in which pneumothorax leads to high morbidity and mortality.

• High inspiratory pressures are necessary for patients with COPD who need NIV, and may contribute to their already increased pneumothorax risk due to emphysema.

• COPD/emphysema phenotype is probably a more important factor for indicating pneumothorax risk than ventilator settings.

• Better phenotyping of patients with COPD and selecting patients in whom benefits of NIV can be expected at minimal risk of serious side-effects are needed.

Ethics statements

Patient consent for publication

Obtained.