Indications, contributing factors, and short-term outcomes of pneumonectomy: an 8-year retrospective study in a resource-limited setting

Abstract

Background

Pneumonectomy, the surgical removal of an entire lung, was first performed in 1933 by Evarts A. Graham for lung carcinoma. Today, pneumonectomy is primarily indicated for non-small cell lung cancer (NSCLC) worldwide. However, it carries a higher risk of morbidity and mortality compared to less extensive lung resections.

Objectives

This study aims to investigate the indications for pneumonectomy and its short-term outcomes among patients who underwent the procedure between 2016 and 2023.

Methods and materials

A retrospective study was conducted on 112 patients who underwent pneumonectomy. Data were analyzed using the Statistical Package for the Social Sciences (SPSS) version 25. Univariate, bivariate, and multivariate analyses were performed to address the study’s objectives. The Pearson chi-square test, crude odds ratio (COR), and adjusted odds ratio (AOR) were calculated for categorical variables where appropriate. To identify factors associated with pneumonectomy, a binary logistic regression model was used, and odds ratios with 95% confidence intervals were calculated. A p-value of < 0.25 was used as a threshold for including variables in the multivariate analysis, while a p-value of < 0.05 was considered statistically significant in the final model.

Results

The majority of patients were aged between 20 and 30 years, with 54 patients (48.2%) in this age group, and a male-to-female ratio of 1.6:1. The primary indication for pneumonectomy was inflammatory conditions, accounting for 62.5% (70 patients), followed by carcinoid tumors at 32.1% (36 patients). Lung cancer was the diagnosis in 5.4% (6 patients) of cases. Factors significantly associated with pneumonectomy included a history of tuberculosis treatment (AOR 40.3; 95% CI: 3.01, 540.2), intraoperative blood loss > 500 mL (AOR 9.07; 95% CI: 1.04, 79.13), and surgical duration > 180 min (AOR 37.9; 95% CI: 1.82, 792.3). The morbidity rate was 25%, while the mortality rate was 5.4%.

Conclusion and recommendations

In our setting, pneumonectomy is most commonly performed for inflammatory conditions. Significant factors associated with the procedure include a history of tuberculosis treatment, intraoperative blood loss > 500 mL, and surgical duration > 180 min. To reduce the need for pneumonectomy, strategies should focus on tuberculosis prevention, screening, and proper patient evaluation and diagnosis before treatment, to prevent extensive lung damage that often necessitates this procedure.

Introduction

Background

In 1933, Evarts A. Graham performed the first pneumonectomy on a 48-year-old obstetrician with lung carcinoma, who went on to have an extended survival. The first extrapleural pneumonectomy was described by Sarot in 1949; initially used to treat tuberculosis empyema, it later became more commonly associated with the treatment of mesothelioma [1].

Pneumonectomy remains a crucial surgical option for anatomically resectable non-small cell lung cancer (NSCLC) when lesser resections would be insufficient to clear the tumor. In cases of inflammatory lung conditions, particularly those linked to tuberculosis, pneumonectomy is often associated with significant morbidity and mortality. However, there is limited evidence identifying the key risk factors for poor outcomes in these patients. Preoperative assessment is vital to identify high-risk patients, and providing adequate preoperative and perioperative coverage with anti-tuberculosis drugs may improve outcomes1-5)

Pneumonectomy is a procedure used to treat a variety of nonmalignant lung diseases. Despite the challenges associated with performing a pneumonectomy for inflammatory diseases, the cure rates for multidrug-resistant tuberculosis (MDR-TB), non-tuberculous mycobacterial (MOTT) infections, and fungal diseases are excellent. However, pneumonectomy for trauma is associated with a very high mortality rate, and every effort should be made to avoid it if possible. Pneumonectomy for other benign conditions is uncommon. Low albumin levels have been shown to increase the risk of postoperative complications in patients undergoing surgical resection for a destroyed lung. Nonetheless, pneumonectomy should not be avoided as a treatment option in eligible patients with a destroyed lung [6, 7].

Several factors have been identified as significant risks for BPF development, including perioperative transfusion, respiratory infection at the time of presentation, neoadjuvant therapy, right-sided pneumonectomy, manual bronchial closure, prolonged postoperative hospitalization, and the need for mechanical ventilation. Implementing a stump coverage policy has been shown to reduce the incidence of BPF, although mechanical stapling is superior to manual closure. Early recognition of potential risk factors for fistula development is crucial to improving patient outcomes [8,9,10].

The approach to bronchial stump closure is a topic of debate in the literature. Some studies advocate for reinforcement with a pericardial flap to prevent post-pneumonectomy bronchopleural fistula (BPF), while others have found no significant impact of this technique on BPF occurrence. Additionally, certain studies have shown that right-sided pneumonectomy is associated with a higher incidence of BPF, regardless of the closure technique used [11, 12].

Literature review

Pneumonectomy refers to the surgical removal of an entire lung, a procedure first performed by Evarts A. Graham in 1933 to treat lung cancer. The main indications for pneumonectomy are categorized into malignant and non-malignant lung conditions. This surgery is typically executed through a posterolateral thoracotomy incision, which provides optimal access [1,2,3].

Initially, pneumonectomy was employed for severe lung diseases, including suppurative conditions. Over time, it became a standard approach for lung cancer until it was found that lobectomies offered similar survival rates with fewer complications. Nowadays, pneumonectomy is mainly reserved for cases where cancer necessitates complete lung removal for a cure or for certain severe suppurative issues [30].

Three variations of pneumonectomy are used based on the disease type: intrapleural, intrapericardial, and extrapleural. Intrapleural pneumonectomy is performed for bronchogenic carcinoma without chest wall or pericardial involvement. Intrapericardial pneumonectomy is used when there is pericardial invasion or when ligation of pulmonary veins is required. Extrapleural pneumonectomy is reserved for malignant mesothelioma that is resectable and part of a multi-modality treatment plan [8, 27, 30, 31, 33].

Pneumonectomy for benign diseases is also high-risk and is usually considered for conditions like tuberculosis-destroyed lung or life-threatening pulmonary hemorrhage due to suppurative disease. While it can be curative, the procedure is associated with significant morbidity and mortality, especially in cases of active hemorrhage or when performed under emergency conditions [6, 27, 28, 32].

Despite the risks, pneumonectomy can be an effective treatment for non-small cell lung cancer when other resections are insufficient. However, it carries high morbidity and mortality rates, with pneumonectomy having the highest mortality among lung cancer resections. The need for pneumonectomy is often dictated by the extent of the malignancy, though intraoperative findings may also influence this decision [1].

Completion pneumonectomy, typically done for recurrent cancer or complications from previous resections, is associated with higher morbidity and mortality rates. It is generally performed in specific cases where its benefits outweigh the risks [29, 33].

Successful pneumonectomy depends on mastering techniques such as positive pressure ventilation, managing the pleural space, and precise anatomic dissection. Postoperative complications can include cardiovascular issues like arrhythmias, with atrial fibrillation being particularly common. Cardiopulmonary complications are frequent, with arrhythmias affecting about 25% of patients and heart failure or herniation potentially occurring [30].

Cardiopulmonary complications developing after pneumonectomy are described. Arrhythmias are seen in about 25% of patients undergoing pneumonectomy, and supraventricular tachyarrhythmias are common. Atrial fibrillation/flutter is the most common and usually occurs in the first three days following surgery. Postpneumonectomy heart failure shares a common cause with postpneumonectomy edema. Cardiac herniation usually occurs within the first 24 h after surgery, but it has been reported up to 6 months following pneumonectomy. The condition presents with an abrupt drop in blood pressure and hemodynamic collapse with subsequent torsion is a rare but fatal complication after pneumonectomy. Early recognition and prophylaxis are always crucial. Post-pneumonectomy AF, with remarkable incidence, risk and independent predictors including age > 60 years and LAd ≥ 35 mm, was mostly solitary and possibly secondary to stump and non-stump PV triggers. POAF, along with infection and hemorrhage, was a RF for perioperative death [3, 13, 23, 26].

Pulmonary complications such as pneumonia, atelectasis, and respiratory failure are common and can significantly impact patient outcomes. Postpneumonectomy empyema, which may include a fistula, is a serious complication that requires comprehensive management including drainage and antibiotic therapy. Bronchopleural fistulas, occurring in 1.5–4.5% of cases, are severe and may be mitigated by using carina closure techniques [3, 19].

Chylothorax, though rare, can complicate pneumonectomy, and its management is debated, with conservative approaches like controlled drainage and nutritional support being employed in some cases [17].

Preoperative assessment is crucial to identify patients at higher risk for complications and to guide treatment decisions. Pneumonectomy is associated with a high incidence of major complications, and factors such as age, physical status, and lung function are closely linked to these outcomes, with a 30-day mortality rate of less than 8% [1, 4, 14].

Statement of the problem

Pneumonectomy has the highest mortality rate among lung resections, with limited literature distinguishing predictors of in-hospital mortality from early post-discharge mortality. Major complications associated with the procedure include treated cardiac arrhythmias, unplanned intensive care admissions, additional surgeries, inotrope usage, and 30-day mortality. A thorough preoperative physiological assessment is essential for identifying patients at increased risk and enables informed decisions about the most appropriate therapeutic approach for treating lung cancer [1,2,3,4].

The altered postoperative physiology following pneumonectomy can have severe implications, particularly in patients receiving adjuvant therapy for advanced lung cancer. These changes can lead to cardiovascular and respiratory complications, significantly increasing both morbidity and mortality [13]. Pulmonary edema complicating lung resection was first described by Gibbon and Gibbon in 1942. It is now recognized as noncardiogenic pulmonary edema, which can manifest as acute lung injury (ALI), acute respiratory distress syndrome (ARDS), or postpneumonectomy pulmonary edema (PPE). The primary mechanisms include increased pulmonary perfusion, endothelial damage, and disruption of the lymphatic system. Treatment depends on the physiological condition of the remaining lung after pneumonectomy, and effective prevention requires close collaboration between the medical and surgical teams [14].

Bronchopleural fistula, arrhythmia, cardiac herniation, pulmonary complications, and respiratory failure are common complications associated with pneumonectomy [34]. This study aims to evaluate the indications for pneumonectomy and to investigate the rates of major complications and in-hospital mortality in our setting.

Objectives of the study

The objective of this study is to identify the indications for pneumonectomy and assess its short-term outcomes for patients who underwent the procedure between 2016 and 2023. Additionally, the study aims to stimulate further research on this topic in the future.

Methods and material

Study design

An 8-year retrospective study was conducted through a review of patient records for those who underwent pneumonectomy during the specified period.

Sample size determination

During the study period, a total of 1,207 patients underwent surgery in the thoracic surgery operating room for various thoracic conditions. Out of these, 116 patients underwent pneumonectomy. Of these 116 cases, 112 met the eligibility criteria and were included in the study sample.

Preoperative preparation and selection for surgery

Patients with thoracic conditions are referred to Tikur Anbessa Specialized Hospital (TASH) from across the country, either as emergencies or elective cases. Emergency referrals typically involve patients presenting with massive hemoptysis who require urgent stabilization and intervention. Elective cases are evaluated at the outpatient department, placed on a waiting list, and usually admitted within three months based on their position on the list. Surgery is scheduled based on imaging and bronchoscopy findings.

Preoperative assessment includes reviewing CT scan results to determine the extent of lung damage, which influences the scope of lung resection. Bronchoscopy is performed only for a few patients due to limited expertise, specifically for those with centrally located tumors requiring biopsy.

Patients undergo basic laboratory tests and physical performance assessments, such as stair climbing. Patients who can climb 5 stairs are considered candidates for pneumonectomy, while those able to climb at least 2 flights of stairs are eligible for less extensive lung resections. Spirometry and other pulmonary tests are not routinely performed in our setup. For patients older than 45 years, echocardiography is conducted as part of cardiac assessment. Patients with significant structural or functional cardiac abnormalities are excluded from surgery.

Surgical techniques

Posteriolateral thoracotomy is the standard surgical approach for lung resections. Prophylactic antibiotics are administered after the induction of anesthesia. Extrapleural pneumonectomy is performed for patients with significant inflammatory adhesions between the parietal and visceral pleura, which complicate access to the hilum. In other cases, intrapleural pneumonectomy is performed.

The main pulmonary artery is double ligated with silk 0 stitches and sutured with prolene 4 − 0. The superior and inferior pulmonary artery is separately double ligated with silk 0 and running suture applied with 4 − 0 prolene. The bronchial stump is typically closed with hand-sewn non-absorbable sutures in two layers, and sometimes covered with tissue depending on the stump’s condition and the surgeon’s preference. Staplers are not usually used both for the vessels and bronchial stump as it is not available in our set up. A chest tube is always placed at the end of the procedure, kept clamped and released intermittently to monitor for ongoing bleeding, and is removed on the first postoperative day.

Study variables

Outcome variables

• Improvement.

• Complications.

• Death.

Exposure variables

• Socio-demographic Characteristics.

• Preoperative Parameters/Clinical Parameters.

• Intraoperative Parameters.

• Postoperative Parameters.

Data collection procedures

Data were collected from patients’ medical records using a data abstraction checklist developed by trained data collectors. The checklist included all relevant variables, such as socio-demographic characteristics, preoperative details, intraoperative specifics, and postoperative outcomes.

Quality assurance

To ensure the appropriateness of the checklist, 10 medical charts were randomly selected for a pretest. Revisions to the data collection tool were made based on the results of this pretest. Data collectors were supervised by the principal investigator throughout the data collection process.

Data management and analysis

The collected data were cleaned and entered into the Statistical Package for the Social Sciences (SPSS) version 25 for management and analysis. Both univariate and bivariate analyses were conducted, and multivariate analysis was performed to address the study objectives. A binary logistic regression model was primarily used to examine associations between study variables and to predict population parameters.

Chi-square tests, Crude Odds Ratios (COR), and Adjusted Odds Ratios (AOR) with 95% Confidence Intervals (CI) were used to compare socio-demographic, clinical, and operative parameters, following the assumptions for each test. Statistical significance was set at a p-value of < 0.05. Results were summarized and presented using tables and charts.

Results

Socio-demographic characteristics of the study subjects

A total of 112 patients were included in the study. The majority (42.8%) were aged between 21 and 30 years. The mean age of the study subjects was 36.66 years (SD ± 14.78). Pneumonectomy was predominantly performed on males, accounting for 62.5% of the cases (Table 1).

Table 1 Socio-demographic characteristics of the study subjects

Clinical characteristics of the study subjects

Approximately 80.4% (90) of patients were admitted on an elective basis, while 19.6% [22] were admitted as emergencies but underwent elective surgery after stabilization. In 67.9% (76) of the patients, the duration of symptoms was 6 weeks or less, whereas 32.1% (36) had symptoms lasting more than 6 weeks. Additionally, 92.9% (104) of the patients had no history of smoking, while 7.1% [8] had a history of cigarette smoking.

Furthermore, 75.0% (84) of the patients had a history of tuberculosis (TB) treatment, compared to 25.0% [28] who had no such history. The association between TB and pneumonectomy was statistically significant, with a p-value of < 0.01. For 12.5% [14] of patients, the pneumonectomy was unplanned preoperatively, whereas 87.5% (98) had planned pneumonectomy. Of the pneumonectomies performed, 55.4% (62) were on the left side, while 44.6% (50) were on the right side (Table 2).

Table 2 Clinical characteristics of the study subjects

Among the presenting symptoms, cough was the most common, affecting 89.29% (100) of patients. Shortness of breath (SOB) was the presenting symptom in 8.93% [10] of patients, while 7.14% [8] presented with excessive foul-smelling sputum, and chest pain was the presenting symptom in 17.86% [20] of patients. Hemoptysis was reported by 80.34% (90) of patients. Additionally, 3.57% [4] of patients presented with weight loss, and 5.36% [6] had other symptoms (Fig. 1).

Fig. 1

Presenting symptoms of the study subjects. N.B the total symptom may not = 112(n) since one patient may have more than one symptoms

Among the 7.1% [8] of patients with comorbid illnesses, 1.8% [2] had asthma, 1.8% [2] had cardiac illness, and 3.6% [4] had diabetes mellitus (DM) (Fig. 2).

Fig. 2

Comorbidities

The findings indicate that 62.5% (70) of the patients underwent surgery for inflammatory conditions, including post-TB complications such as Aspergilloma, bronchiectasis, lung fibrosis, bleeding from cavitary lesions, and extensive bullous diseases. Carcinoid tumor was the surgical indication for 32.1% (36) of the patients (Fig. 3), while non-small cell lung cancer was diagnosed in 5.4% [6] of the patients (Fig. 4).

Fig. 3

Removed destroyed lung which was obstructed by large endobroncheal mass

Fig. 4

Surgical indication

Intra and postoperative characteristics of the study subjects

The results showed that 91.1% (102) of the procedures were intrapleural, 8.9% [10] were extrapleural, and 5.4% [6] were completion pneumonectomies.

Regarding intraoperative factors, 82.1% (92) of patients had an estimated blood loss greater than 500 ml, with a p-value of < 0.01. Additionally, 87.5% (98) of patients had an operative time exceeding 180 min, with a p-value of < 0.01. Intraoperative incidents occurred in 3.6% [4] of patients. Postoperatively, 12.5% [14] of patients required inotropes, and 1.8% [2] required positive ventilation. Most patients were discharged from the hospital within 7 days, while 23.2% [26] stayed longer than 7 days (Table 3).

Table 3 Intra and postoperative parameters

Outcomes of pneumonectomy

Postoperative complications developed in 25.0% [28] of patients. Among these, arrhythmia and myocardial infarction occurred in 14.3% [4] of patients, postoperative bleeding in 14.3% [4], empyema in 14.3% [4], pneumonia in 50.0% [14], and wound infection in 7.1% [2] of patients.

The majority of patients, 92.9% (104), were discharged with improvement, while 5.4% [6] died, and 1.8% [2]as readmitted within one month. Among the deaths, the causes were sepsis in 33.3% [2], pulmonary thromboembolism (PTE) in 33.3% [2], and sudden death in 33.3% [2] of patients.

Factors associated with pneumonectomy

The final model identified several morbidities significantly associated with pneumonectomy (p < 0.05). These factors include a history of tuberculosis (TB) treatment, estimated intraoperative blood loss (EBL) greater than 500 ml, and a surgical duration exceeding 180 min.

Patients with a history of TB treatment were significantly more likely to undergo pneumonectomy, with an Adjusted Odds Ratio (AOR) of 40.3 (95% Confidence Interval [CI]: 3.01, 540.2). Two explanations may account for this association: First, patients who were misdiagnosed with TB but actually had endobronchial tumors might develop obstructive bronchiectasis and permanent lung damage, eventually requiring pneumonectomy after completing anti-TB treatment and referral following complications. Second, patients with actual TB who did not receive adequate medical treatment may develop complications and extensive lung damage, leading to pneumonectomy.

Patients with an EBL greater than 500 ml were associated pneumonectomy, with an AOR of 9.07 (95% CI: 1.04, 79.13). Additionally, a surgical duration exceeding 180 min was associated with pneumonectomy, with an AOR of 37.9 (95% CI: 1.82, 792.3) (Table 4).

Table 4 Factors Associated with pneumonectomy

Discussion

Historically, the earliest successful pneumonectomies were performed for suppurative lung diseases. Today, pneumonectomy is primarily reserved for patients with malignancies requiring complete resection for cure and for selected cases of suppurative lung disease. It is typically considered for advanced disease, particularly when the tumor is located in the main stem bronchus or extends across major fissures. Non-small cell lung cancer (NSCLC) is the most common malignancy necessitating pneumonectomy, though it can also be indicated for mesotheliomas and extensive tumors from adjacent structures such as the thymus [1, 5, 6, 30].

Among non-malignant conditions, inflammatory lung diseases are significant indications for pneumonectomy. These include post-TB bronchiectasis, Aspergilloma with massive hemoptysis, lung fibrosis, extensive bullous disease, and cavitary lesions. It is crucial to optimize these patients before surgery through medical treatment of underlying infections, lung function optimization, and nutritional support [5, 29, 30].

In our study, which included 112 patients undergoing pneumonectomy, the median age was 35 years with a male predominance of 58.3%. The most common presenting symptom was cough, reported in 89.2% of patients, followed by hemoptysis in 80.34%. A history of TB treatment was prevalent in 75% of the patients. Pneumonectomy was primarily indicated for inflammatory conditions, particularly post-TB complications, which accounted for 62.5% of the cases.

A 1979 retrospective study from Nigeria University Hospital noted that pneumonectomy was primarily performed for life-threatening hemorrhage following TB-destroyed lungs. A study from the University Of Natal Medical School in South Africa (2000) found that 72% of pneumonectomies were associated with TB. In contrast, a prospective study from the UK (2009) and a retrospective study from Al Hussein Teaching Hospital in Baghdad (2022) reported that pneumonectomy was more commonly indicated for lung cancer [1, 3, 5, 28].

This study demonstrates that pneumonectomy is significantly associated with several factors: a history of previous tuberculosis (TB) treatment (75%), greater intraoperative blood loss (87.5%), and a longer duration of surgery (> 180 min) (87.5%). The postoperative morbidity rate in this study was 25%, and the hospital mortality rate was 5.4%.

The complications observed in our study included pneumonia (12.5%), arrhythmia (3.6%), postoperative bleeding (3.6%), empyema (3.6%), and wound infection (1.8%). These findings are relatively consistent with the literature. A study from the University of Natal Medical School (2000) reported a morbidity rate of 23%, with higher incidences of empyema (14.8%) and bleeding (1.9%), and lower rates of bronchopleural fistula (BPF) (1.9%) and wound sepsis (0.6%). Conversely, studies from Baghdad and the UK reported higher complication rates, including arrhythmia and atelectasis (40%) in Baghdad and a morbidity rate of 19.9% in the UK [1, 3, 5, 28].

Pneumonectomy is known for having the highest mortality rate among lung resections. In our study, the mortality rate was 5.4%, which is comparable to the 5.4% reported in the UK but higher than the 1.2% reported in South Africa and 3% in Baghdad. The mortality rate in our study, despite being associated with a higher proportion of benign indications, aligns closely with rates observed in studies involving malignant indications [1, 3, 5, 25, 28, 30].

Limitations

The main limitations of this research is its retrospective nature, which made tracing patient records challenging and resulted in the exclusion of some patients from the study. Additionally, absence of long term patient follow up, important patient information such as the American Society of Anesthesiologists (ASA) classification, performance status, absence of spirometer, and bronchial closure techniques was not recorded, limiting our ability to analyze factors associated with the outcomes of pneumonectomy.

Conclusion and recommendations

Conclusion

At our center, pneumonectomy is predominantly performed for benign inflammatory conditions such as Aspergilloma, bronchiectasis, and lung fibrosis, which are frequently associated with tuberculosis (TB). This contrasts with studies from developed countries where pneumonectomy is more commonly indicated for malignant conditions. Despite the predominance of benign indications, the morbidity and mortality rates observed in our study are comparable to those reported in well-resourced settings.

Recommendations

Given the association between pneumonectomy and a history of TB treatment—either due to lung damage from untreated TB or from delayed diagnosis resulting in extensive lung damage—several recommendations are proposed:

• 1. Enhanced TB Prevention and Management: Effective TB prevention strategies should be implemented at all health facility levels. Proper patient evaluation and timely, accurate diagnosis are crucial to prevent unnecessary lung damage and avoid extensive resections.

• 2. Improved Diagnostic Accuracy: Efforts should be made to enhance diagnostic accuracy to distinguish between TB and other conditions that may require surgical intervention. This can help in managing patients with less invasive procedures when possible.

• 3. Optimized Patient Care: Patients with a history of TB should be carefully monitored and managed to prevent progression to conditions that necessitate pneumonectomy. Early intervention and appropriate treatment should be prioritized to minimize the need for major lung resections.