Surgery in and for COPD

Surgery for COPD

During the past few decades multiple surgical interventions have been suggested to improve symptoms in patients with COPD [32]. These include bullectomy, lung volume reduction surgery and lung transplantation (see Patient section).


Bullectomy appears to be of benefit in highly selected patients [33], resulting in short-term improvements in airflow obstruction, lung volumes, hypoxaemia and hypercapnia, exercise capacity, dyspnoea, and health-related quality of life.

Surgical mortality ranges from 0-22.5% [33]. Long-term follow-up data are more limited with 1/3-1/2 of patients maintaining benefits for ~5 yrs [33].

Patient selection
Based on the presumption that improvement is dependent on relief of compressed normal lung, most investigators have attempted to identify optimal surgical candidates on the basis of pulmonary function and radiographic features, as enumerated in table 2 [34].

Table 2. - Factors associated with favourable or unfavourable outcome in classical bullectomy

Parameter Favourable Unfavourable
Clinical Rapid progressive dyspnoea despite maximal medical therapy
Older age
Comorbid illness
Cardiac disease
Pulmonary hypertension
>10% weight loss
Frequent respiratory infections
Chronic bronchitis.
Physiological Normal FVC or slightly reduced
FEV1 >40% pred
Little bronchoreversibility
High trapped lung volume
Normal or near normal DL,CO
Normal Pa,O2 and Pa,CO2
FEV1 <35% pred
Low trapped gas volume
Decreased DL,CO
CXR Bulla >1/3 hemithorax Vanishing lung syndrome
Poorly defined bullae
CT Large and localised bulla with vascular crowding and normal pulmonary parenchyma around bulla Multiple ill-defined bullae in underlying lung
Angiography Vascular crowding with preserved distal vascular branching Vague bullae; disrupted vasculature elsewhere
Isotope scan Well-localised matching defect with normal uptake and washout for underlying lung Absence of target zones, poor washout in remaining of lung

CXR: chest radiography; CT: computed tomography; FVC: forced vital capacity; FEV1: forced expiratory volume in one second; DL,CO: carbon monoxide diffusing capacity of the lung; Pa,O2; arterial oxygen tension; Pa,CO2; arterial carbon dioxide tension. Modified from [34].

Lung Volume Reduction Surgery

LVRS results in short-term improvements in spirometry, lung volumes, exercise tolerance, dyspnoea and health-related quality of life, and potentially long-term improvement in survival [35].

LVRS using resection of lung tissue has significantly better results than laser treatment of the lung [36].

Bilateral LVRS shows greater improvements compared to unilateral LVRS among similar patients [37].

Patient selection
Selection criteria for LVRS remain controversial [27].

A systematic review proposed the following features, as determined by expert opinion, to be associated with better outcomes: smoking-related emphysema, heterogeneous emphysema with surgically accessible "target" areas, bilateral surgery, good general fitness/condition and thoracic hyperinflation [38]. The National Emphysema Treatment Trial (NETT) suggests that upper lobe predominance of emphysema on high resolution computed tomography of the chest and low post rehabilitation exercise capacity measured while breathing 30% inspiratory oxygen fraction on a cycle ergometry are predictive of the best chance of post-surgical improvement [35].

Table 3 enumerates potential criteria to identify patients more likely to experience benefit after LVRS. Figure 3 proposes an algorithm delineating expected improvements of bilateral LVRS compared to medical therapy based on the NETT Research Group experience [35]. Potential outcomes of such a selection algorithm are enumerated in table 4.

Table 3. - Factors associated with favourable or unfavourable outcome in lung volume reduction surgery

Parameter Favourable Unfavourable
Clinical Age <75 yrs
Clinical picture consistent with emphysema#
Not actively smoking (>3-6 months) #
Severe dyspnea despite maximal medical treatment including pulmonary rehabilitation#
Requiring <20 mg prednisone·day-1
Age > 75-80 yrs.
Comorbid illness which would increase surgical mortality
Clinically significant coronary artery disease
Pulmonary hypertension (PA systolic >45, PA mean >35 mmHg)
Severe obesity or cachexia
Surgical constraints
    Previous thoracic procedure#
    Chest wall deformity#
Physiological FEV1 after bronchodilator <45% pred
    RV >150%
    TLC >100% pred
Pa,O2 > 6 kPa (45 mmHg)
Pa,CO2 < 8 kPa (60 mmHg)
Postrehabilitation 6-min walk >140 m
Low post-rehabilitation maximal achieved cycle ergometry watts#
FEV1 <20% pred and DL,CO <20% pred#
Decreased inspiratory conductance
Radiographical High-resolution computed tomography confirming severe emphysema, ideally with upper lobe predominance# Homogeneous emphysema and FEV1 <20% pred#
Non-upper lobe predominant emphysema and high post-rehabilitation cycle ergometry maximal achieved wattage#

PA: alveolar pressure; RV: residual volume; TLC: total lung capacity; Pa,O2: arterial oxygen tension; Pa,CO2: arterial carbon dioxide tension; FEV1: forced expiratory volume in one second; DL,CO: carbon dioxide diffusing capacity of the lung. #: confirmed recommendations using NETT data [35, 39] or expert opinion [40]. Modified from [34].

Click to view larger image.

Fig. 3. - Diagnostic algorithm based on the National Emphysema Treatment Trial (NETT) [35, 39]. The groups refer to the outcomes detailed in table 4. LVRS: lung volume reduction surgery; FEV1: forced expiratory volume in one second; DL,CO: carbon dioxide diffusing capacity of the lung.

Table 4. - Results of bilateral lung volume reduction surgery compared to medical therapy in patients with severe emphysema

Patients 90-Day mortality p-value Total mortality Risk ratio# p-value
  LVRS Medical Therapy LVRS Medical Therapy
Group A 48/608 (7.9) 8/610 (1.3) <0.001 42/70 30/70 1.82 0.06
Group B 4/139 (2.9) 5/151 (3.3) 1.00 26/139 51/151 0.47 0.005
Group C 6/206 (2.9) 2/213 (0.9) 0.17 34/206 39/213 0.98 0.70
Group D 7/84 (8.3) 0/65 (0) 0.02 28/84 26/65 0.81 0.49
Group E 11/109 (10.1) 1/111 (0.9) 0.003 27/109 14/111 2.06 0.02

Patients Improvement in Exercise Capacity Improvement in Health-related Quality of Life+
  LVRS Medical Therapy OR p-value LVRS Medical therapy OR p-value
Group A 4/58 (7) 1/48 (2) 3.48 0.37 6/58 (10) 0/48 (0) 0.03
Group B 25/84 (30) 0/92 (0) <0.001 40/84 (48) 9/92 (10) 8.38 <0.001
Group C 17/115 (15) 4/138 (3) 5.81 0.001 47/115 (41) 15/138 (11) 5.67 <0.001
Group D 6/49 (12) 3/41 (7) 1.77 0.50 18/49 (37) 3/41 (7) 7.35 0.001
Group E 2/65 (3) 2/59 (3) 0.90 1.00 10/65 (15) 7/59 (12) 1.35 0.61

Data are presented as n/n (%). Groups A-E refer to the patients as defined in fig. 3. OR: odds ratio. #: risk ratio for total mortality in surgically versus medically treated patients during a mean follow-up of 29.2 months; : increase in the maximal workload >10 W from the patient’s postrehabilitation baseline value (24 months after randomisation); +: decrease in the score on the St George’s Respiratory Questionnaire >8 points (on a 100-point scale) from the patient’s postrehabilitation baseline score (24 months after randomisation). Modified from [35].

The mean improvement from baseline in FEV1 ranges from 5-96%, although 20-50% of patients show little spirometric improvement after LVRS [27, 35].

Lung volume
Lung volume changes include a mean decrease in total lung capacity from baseline varying from 1-23% and residual volume ranging from 3-46% [27].

Exercise tolerance
Improved timed walk distance, ranging a mean of 7-103% [27], and an increase in maximal work load [35], oxygen uptake and minute ventilation [27] have been reported after LVRS.

Dyspnoea and health-related quality of life
Improved dyspnoea and health-related quality of life has been reported after LVRS (table 5).

Table 5. - Health-related quality of life and survival following bilateral lung volume reduction surgery (LVRS)

First author
Operation Subjects
Duration of
Cohort studies with n³50, consecutive enrollment and full accountability of patients at follow-up
YUSEN[41] Sternotomy 200 Yes Patient satisfaction 6 months 184 12 4 Good-to-excellent satisfaction in most patients
YUSEN [41] Sternotomy 193 Yes Patient satisfaction 3 yrs 159 25 8 Good-to-excellent satisfaction in most patients
YUSEN [41] Sternotomy 144 Yes Patient satisfaction 5 yrs 82§ 41 13 Good-to-excellent satisfaction in most patients
YUSEN [41] Sternotomy 119 Yes MRC dyspnoea scale 6 months 109 7 3 Decreased dyspnoea
YUSEN [41] Sternotomy 112 Yes MRC dyspnoea scale 3 yrs 92 15 5 Decreased dyspnoea
YUSEN [41] Sternotomy 66 Yes MRC dyspnoea scale 5 yrs 35 23 7 Decreased dyspnoea
YUSEN [41] Sternotomy 159 Yes SF-36 6 months 150 6 3 Improved physical functioning and physical component summary scores
YUSEN [41] Sternotomy 152 Yes SF-36 3 yrs 122¶ 21 8 Improved physical functioning and physical component summary scores
YUSEN [41] Sternotomy 111 Yes SF-36 5 yrs 67 31+ 7 Improved physical functioning and physical component summary scores
COOPER [42] Sternotomy 150 Yes NHP 6 months 108 7 35 Improved physical mobility, energy, emotional reaction and sleep scale scores
BRENNER [43] VATS 145 Not required MRC dyspnoea 3 months 130 6 9 Decreased dyspnoea
BRENNER [43] VATS 145 Not required MRC dyspnoea 6 months 84 6 55 Decreased dyspnoea
LEYENSON [44] Sternotomy 50 Yes SIP 3 months 42 0 8 Improved psychosocial, physical and total SIP scores
First author
Operation Subjects
Duration of
Randomised controlled trials: bilateral LVRS compared to medical therapy
GEDDES [45] Sternotomy, VATS 24/24 Yes SF-36 12 months 13/19 5/3 6/2 Improved total score in LVRS cohort; worsened score in medical therapy cohort
CRINER [46] Sternotomy 18#/18 Yes SIP 3 months 11/15 3/0 4/3 Improved psychosocial, physical and overall SIP scores in LVRS cohort; no change in medical therapy cohort
GOLDSTEIN [47] VATS, sternotomy 28/27 Yes CRQ 12 months 28/27 4/1 1/1 Improved dyspnea, fatigue, emotional function, mastery
NETT [35] Sternotomy
608ƒ/610 Yes SF-36
29.2 months 608/610 157/160 0/0## Improved SGRQ, UCSD SOBQ, QWB

HRQOL: health-related quality of life; VATS: video-assisted thoracoscopic surgery; SF-36: medical outcomes survey short-form 36; NHP: Nottingham health profile; SIP: symptom impact profile; MRC: modified medical research council dyspnoea scale; CRQ: Chronic Respiratory Questionnaire; SGRQ: St. George’s Respiratory Questionnaire; QWB: Quality of Well-Being; UCSD SOBQ: University of California, San Diego Shortness of Breath Questionnaire. #: one of the 19 patients randomised to LVRS that died during preoperative rehabilitation period is excluded from the analysis; ¶: one patient excluded due to lung transplantation; +: eight patients excluded due to lung transplantation; §: six patients excluded due to lung transplantation; ƒ: 406 median sternotomy/174 VATS, 28 patients refused LVRS; ##: patients missing data had imputed scores. Modified from [48].

Mortality following LVRS varies greatly among centres. The NETT Research Group documented a 90-day surgical mortality of 7.9% in all randomised patients, compared to 1.3% in a comparable medically treated arm [35]; much of this mortality was accounted for by high-risk patients [39] in whom the 90-day surgical mortality was 28.6%, as compared to 0% in the respective medical arm [35]. In nonhigh risk patients, the 90-day surgical mortality was 5.2%, as compared to 1.5% in the medically treated patients [35].

In the NETT study, baseline patient characteristics were found to predict long-term mortality risks. In patients with upper lobe predominant emphysema on high-resolution computed tomography and a low postrehabilitation, maximal-achieved cycle ergometry work load, there was an improved long-term (mean follow-up 29 months) survival in patients undergoing bilateral LVRS compared to those treated with medical therapy (risk ratio 0.47, p=0.005). Early higher mortality in patients treated surgically was compensated for by lower mortality risk in LVRS patients during long-term follow-up [35]. In patients with nonupper lobe predominant emphysema and a higher post-rehabilitation cycle ergometry work load, surgically treated patients experienced a higher mortality than comparable, medically treated patients (risk ratio 2.06, p=0.02) [35] (table 4). The other two sub-groups experienced no mortality difference with LVRS [35] (table 4).

Long-term results
Few studies have reported long-term results, but they suggest widely varying long-term morbidity and mortality among centres, return of spirometric function and lung volumes towards preoperative baseline and worsening dyspnoea over time [49]. There appears to be slower loss of 6-min walk distance after LVRS than of other functional measures [40].

Lung transplantation

Lung transplantation should be considered in selected patients with advanced COPD. COPD is the most common indication for lung transplantation (UNOS on-line data base). The choice of single lung transplantation (SLT) or bilateral lung transplantation (BLT) for COPD remains controversial [50, 51].

Lung transplantation results in improved pulmonary function, exercise capacity and quality of life. However, its effect on survival remains controversial.

Patient selection
In selecting candidates, several issues must be considered, including the patient’s pulmonary disability, projected survival without transplantation, comorbid conditions and patient preferences. To optimise results of transplantation, the procedure must be carefully timed such that transplantation is performed when the patient is neither "too healthy" nor "too ill" [52]. Selection criteria for COPD patients are shown in tables 6 and 7 [52].

Table 6. - General selection guidelines for candidate selection for lung transplantation in COPD patients

Relative contraindications Age limits
    Heart-lung transplants ~55 yrs
    Double lung transplant ~60 yrs
    Single lung transplant ~65 yrs
Symptomatic osteoporosis
Oral corticosteroids >20 mg·day-1 prednisone
Psychosocial problems
Requirement for invasive mechanical ventilation
Colonisation with fungi or atypical mycobacteria
Absolute contraindications Severe musculoskeletal disease affecting the thorax
Substance addiction within previous 6 months
Dysfunction of extrathoracic organ, particularly renal dysfunction
HIV infection
Active malignancy within 2 yrs except basal or squamous cell carcinoma of skin
Hepatitis B antigen positivity
Hepatitis C with biopsy-proven evidence of liver disease

HIV: human immunodeficiency virus.

Table 7. - Disease-specific guidelines for candidate selection for lung transplantation in COPD patients.

FEV1 £25% pred (without reversibility) and/or
Resting, room air Pa,CO2 >7.3 kPa (55 mmHg) and/or
Elevated Pa,CO2 with progressive deterioration requiring long-term oxygen therapy
Elevated pulmonary artery pressure with progressive deterioration.

FEV1: forced expiratory volume in one second; Pa,CO2: arterial carbon dioxide tension. Modified from [52].

Pulmonary function
Following SLT for COPD, FEV1 is expected to rise to ~50% of the predicted normal value and FVC to ~70% of the predicted normal value [50, 51, 53].

Following BLT, FEV1 increases to 78-85% and FVC to 66-92% of the predicted normal values [53, 54].

Exercise capacity
Despite the differential improvements in spirometry, peak exercise capacity is similar between SLT and BLT [55].

Quality of life
Quality of life following lung transplantation improves dramatically, in particular for those patients who do not develop chronic rejection [56].

Only a minority of patients return to full-time work [57].

Average actuarial survival following lung transplantation for recipients with COPD is 81.7, 61.9 and 43.4% at 1, 3 and 5 yrs (UNOS online data base).

Compared to patients with other cardiopulmonary diseases, patients with emphysema exhibit the best overall survival after transplantation [57-59].

Data on whether transplantation actually confers a survival advantage compared to the natural history of the disease are conflicting [50, 60, 61].

By 5 yrs following lung transplantation, the prevalence of chronic allograft rejection (obliterative bronchiolitis), the leading cause of long-term morbidity and mortality, is as high as 50-70% among survivors [62].

Last Reviewed: February 2015