Introduction
Background
Pneumothorax is defined as the presence of air or gas in the pleural cavity. Primary spontaneous pneumothorax (PSP) occurs in people without underlying lung disease and in the absence of an inciting event. Many patients whose condition is labeled as primary spontaneous pneumothorax have subclinical lung disease.
Secondary spontaneous pneumothorax (SSP) occurs in people with a wide variety of parenchymal lung diseases. Iatrogenic pneumothorax results from incursion into the pleural space secondary to diagnostic or therapeutic medical intervention. Traumatic pneumothorax results from injury, typically blunt or penetrating trauma. Tension pneumothorax develops when air is trapped in the pleural cavity under positive pressure.1
This article identifies several areas of new information in the medical literature: 1) studies comparing aspiration and tube drainage for treatment of primary spontaneous pneumothorax, 2) long term follow-up of surgical treatment of pneumothorax, 3) assessment of the impact of pleurodesis on transplantation outcomes in patients with lymphangiomyomatosis, 4) demonstrated utility of ultrasonography in the bedside diagnosis of iatrogenic pneumothorax, and 5) inability of ultrasonography to distinguish between intrapulmonary bullae and pneumothorax.
Pathophysiology
The inner surface of the thoracic cage (parietal pleura) is contiguous with the outer surface of the lung (visceral pleura); this space contains a small amount of lubricating fluid and is normally under negative pressure compared to the alveoli. Determinants of pleural pressure are the opposing recoil forces of the lung and chest wall.
Primary spontaneous pneumothorax (PSP) is typically observed in tall young people without parenchymal lung disease and is thought to be related to increased shear forces in the apex. PSP is associated with the presence of apical pleural blebs lying under the visceral pleura, but the exact anatomic site of air leakage is often uncertain. Fluorescein-enhanced autofluorescence thoracoscopy (FEAT), a novel method to examine the site of air leak in PSP, shows FEAT-positive lesions that are normal when viewed under normal white-light thoracoscopy.2
Blebs and bullae (sometimes called emphysematous-like changes or ELCs) are related to the occurrence of primary spontaneous pneumothorax. Thoracic computerized tomography (CT) of patients with PSP shows ipsilateral ELC in 89% and contralateral changes in 80% compared to a rate of 20% among control subjects matched for age and smoking. Nonsmokers with PSP had CT ELC abnormalities of 80% compared with a rate of 0% among nonsmoker controls without PSP.1
While patients with PSP do not have overt parenchymal disease, more than 90% of them are smokers. The relative risk of PSP increases as the number of cigarettes smoked per day increases. This incremental risk with increasing number of cigarettes smoked per day is much more pronounced in female smokers.
Lung inflammation and oxidative stress are hypothesized to be important to the pathogenesis of PSP.3 Current smokers, at increased risk for PSP, have increased numbers of inflammatory cells in the small airways. Bronchoalveolar lavage studies in patients with PSP associated the degree of inflammation with the extent of ELCs. One hypothesis is that ELCs result from degradation of lung tissue due to imbalances of enzymes and antioxidants released by innate immune cells.4 In one study, erythrocyte superoxide dismutase activity was significantly lower and plasma malondialdehyde levels higher in patients with PSP than in normal control subjects.
A growing body of evidence indicates that genetic factors may be important in the pathogenesis of many cases of primary spontaneous pneumothorax. Familial clustering of this condition has been reported. Genetic disorders that have been linked to primary spontaneous pneumothorax include Marfan syndrome, homocystinuria, and Birt-Hogg-Dube (BHD) syndrome.
Birt-Hogg-Dube syndrome is an autosomal dominant disorder that is characterized by benign skin tumors (hair follicle hamartomas), renal and colon cancer, and spontaneous pneumothorax. The spontaneous pneumothorax occurs in about 22% of patients with this syndrome. The gene responsible for this syndrome has been identified and is a tumor suppressor gene located on chromosome 17p11.2. The gene encoding folliculin has been identified and is thought to be the etiology of Birt-Hogg-Dube syndrome. Multiple mutations have been found, and phenotypic variation is recognized. In a recent study, 8 patients without skin or renal involvement had lung cysts and spontaneous pneumothorax. A germ line mutation to this gene has been found in 5 patients. Genetic testing is now available.5
Secondary spontaneous pneumothoraces (SSP) occur in the presence of lung disease, primarily in the presence of chronic obstructive pulmonary disease (COPD). Other diseases that may be present when SSPs occur includetuberculosis, sarcoidosis, cystic fibrosis, malignancy, and idiopathic pulmonary fibrosis.
Pneumocystis jiroveci pneumonia (previously known as Pneumocystis carinii pneumonia [PCP]) was a common cause of secondary spontaneous pneumothorax in patients with AIDS during the last decade. With the advent of highly active antiretroviral therapy (HAART) and widespread use of trimethoprim-sulfamethoxazole prophylaxis, the incidence of PCP and associated SSP has significantly declined.
PCP is now primarily seen in patients who are noncompliant with HIV therapy or trimethoprim-sulfamethoxazole prophylaxis or those taking inhaled pentamidine for PCP prophylaxis (probably related to nonuniform distribution of the medication aerosol). PCP in other immunocompromised patients is seen only when trimethoprim-sulfamethoxazole prophylaxis is withdrawn prematurely. For practical purposes, if the immunocompromised patient has been taking trimethoprim-sulfamethoxazole prophylaxis reliably, PCP is reasonably excluded from the differential diagnosis.
Iatrogenic pneumothorax is a complication of medical or surgical procedures. It most commonly results from transthoracic needle aspiration. Other procedures commonly causing iatrogenic pneumothorax are therapeutic thoracentesis, pleural biopsy, central venous catheter insertion, transbronchial biopsy, positive pressure mechanical ventilation, and inadvertent intubation of the right mainstem bronchus. Therapeutic thoracentesis is complicated by pneumothorax 30% of the time when performed by inexperienced operators in contrast to only 4% of the time when performed by experienced clinicians.
The routine use of ultrasonography during diagnostic thoracentesis is associated with lower rates of pneumothorax (4.9% vs 10.3%) and need for tube thoracostomy (0.7% vs 4.1%). Similarly, in patients who are mechanically ventilated, thoracentesis guided by bedside ultrasonography without radiology support results in a relatively lower rate of pneumothorax.
Traumatic pneumothoraces can result from both penetrating and nonpenetrating lung injuries. Complications include hemopneumothorax and bronchopleural fistula. Traumatic pneumothoraces can create a 1-way valve in the pleural space (only letting in air without escape) and can lead to a tension pneumothorax.
Tension pneumothorax typically occurs in the intensive care setting in patients who are ventilated. With air trapping in the pleural space, positive pressure rises. This pressure compresses the mediastinum, decreasing venous return to the heart and reducing cardiac output. In addition, owing to ipsilateral lung collapse and contralateral lung compression, gas exchange is compromised, leading to hypoxemia.
Frequency
United States
For men, the age-adjusted incidence of primary spontaneous pneumothorax (PSP) is 7.4 cases per 100,000 persons per year. For women, age-adjusted incidence is 1.2 cases per 100,000 persons per year.6
For men, the age-adjusted incidence of secondary spontaneous pneumothorax (SSP) is 6.3 cases per 100,000 persons per year; for women, age-adjusted incidence is 2.0 cases per 100,000 persons per year. In patients with COPD, the incidence is 26 cases per 100,000 patients per year.7
Traumatic pneumothoraces occur more frequently than spontaneous pneumothoraces, and the rate is increasing.
Mortality/Morbidity
Recurrences usually strike within the first 6 months to 3 years. The 5-year recurrence rate is 28% for primary spontaneous pneumothorax (PSP) and 43% for secondary spontaneous pneumothorax (SSP).
- Recurrences are more common among patients who smoke, patients with COPD, and patients with AIDS. Predictors of recurrence include pulmonary fibrosis, younger age, and increased height-to-weight ratio. Bullous lesions found on CT scan or at thoracoscopy are not predictive of recurrence. In a retrospective study of 182 consecutive patients with a newly diagnosed first episode of pneumothorax, a higher rate of recurrence was noted in taller patients, thin patients, and patients with SSP. Patients who underwent pleurodesis had cumulative rates of recurrence 13%, 16%, and 27% at 6 months, 1 year, and 3 years, respectively, compared to 26%, 33%, and 50%, respectively, in patients treated with chest tube drainage. The use of tetracycline or gentamicin did not have any significant impact on the recurrence rate.
- Complications include hypoxemic respiratory failure, respiratory or cardiac arrest, hemopneumothorax, and bronchopulmonary fistula.
- PSP is typically benign and often resolves without medical attention.
- While the risk of mortality with PSP is low, a higher risk of mortality with SSP exists. In particular, patients with COPD are at great risk, with a 3.5-fold increase in relative mortality.
- Studies indicate a mortality rate of 1-17% in patients with COPD and an SSP. One study indicated that 5% of patients with COPD died before a chest tube was placed.
- Patients with AIDS also have a high inpatient mortality rate of 25% and a median survival of 3 months after the pneumothorax. These data derive from the pre-HAART therapy era (see above).
Sex
For primary spontaneous pneumothorax (PSP), the male-to-female ratio of age-adjusted rates is 6.2:1. For a secondary spontaneous pneumothorax (SSP), the male-to-female ratio of age-adjusted rates is 3.2:1.
Age
- Primary spontaneous pneumothoraces (PSPs) occur in people aged 20-30 years. Peak incidence is in the early twenties, and PSP is rarely observed in people older than 40 years.
- Secondary spontaneous pneumothoraces (SSPs) occur more frequently in patients aged 60-65 years.
Clinical
History
Most episodes of spontaneous pneumothorax (SP) occur at rest. By definition, spontaneous pneumothorax is not associated with trauma or stress.
- Acute onset of chest pain and shortness of breath were present in all patients in one series. Typically, both symptoms are present in 64% of patients.
- Acute onset of chest pain - Severe and/or stabbing pain, radiating to ipsilateral shoulder and increasing with inspiration (pleuritic)
- Sudden shortness of breath
- Anxiety, cough, and vague presenting symptoms (eg, general malaise, fatigue) are less commonly observed.
- Dyspnea tends to be more severe with secondary spontaneous pneumothoraces (SSPs) because of decreased lung reserve.
- Bilateral pneumothorax - Primary bilateral spontaneous pneumothorax (PBSP) was significantly more common in patients with lower BMI and among smokers.8
Physical
- General appearance
- Diaphoretic
- Splinting chest wall to relieve pleuritic pain
- Cyanotic (with tension pneumothoraces)
- Vital signs
- Tachypnea
- Tachycardia (most common finding) - If faster than 135 beats per minute (bpm), tension pneumothorax is likely
- Pulsus paradoxus
- Hypotension (often with tension pneumothorax)
- Asymmetric lung expansion - Mediastinal and tracheal shift to the contralateral side with a large tension pneumothorax
- Distant or absent breath sounds
- Hyperresonance on percussion
- Decreased tactile fremitus
- Cardiovascular - Jugular venous distension (tension pneumothorax)
- Neurologic - Altered mental status
- If patients who are mechanically ventilated are difficult to ventilate during resuscitation, high peak airway pressures are a clue to an impending pneumothorax. A tension pneumothorax causes progressive difficulty with ventilation as the normal lung is compressed. On volume-control ventilation, this is indicated by marked increase in both peak and plateau pressures, with relatively preserved peak and plateau pressure difference. On pressure control ventilation, tension pneumothorax causes sudden drop in tidal volume. However, these observations are neither sensitive nor specific for making the diagnosis of pneumothorax or ruling out the possibility of pneumothorax.
Causes
- Risks factors for primary spontaneous pneumothorax (PSP)
- Smoking
- Of patients with PSP, 91% reportedly are smokers or were smokers.
- The risk of PSP is related to the intensity of smoking, with 102-times higher incidence rates in males who smoke heavily (ie, >22 cigarettes/d), compared to a 7-fold increase in males who smoke lightly (1-12 cigarettes/d).
- Tall, thin stature in a healthy person
- Marfan syndrome
- Pregnancy
- A 10-year retrospective series of 250 SP cases found 5 pregnant women, suggesting that pregnancy is an unrecognized risk factor.9
- The cases were all managed successfully with simple aspiration or vacuum-assisted thoracostomy (VATS), and no harm occurred to mother or fetus.9
- Familial pneumothorax
- Familial spontaneous pneumothorax has been described as an autosomal dominant inheritance with incomplete penetrance. One family study reported 9 cases of SP among 54 members ascertained.
- A review of the literature summarized 61 reports of familial spontaneous pneumothorax among 22 families. Up to 10% patients with SP report a positive family history.10
- Diseases and conditions associated with secondary spontaneous pneumothorax
- Chronic obstructive lung disease
- Asthma
- HIV/AIDS with Pneumocystis jiroveci (PCP) infection: 77% of AIDS patients with SP had thin-walled cavities, cysts, and pneumothorax from PCP infection.11
- Necrotizing pneumonia
- Bronchogenic carcinoma
- Metastatic malignancy
- Pneumothorax in a patient with malignancy should prompt a look for metastatic disease. Many different types of malignancies have been implicated, especially sarcomas, but also genitourinary cancers and primary lung cancer.
- Chemotherapeutic agents and, at times, the response of cancer to the agent can induce SP.12
- Tuberculosis
- Cystic fibrosis (CF)
- Up to 18.9% of patients with CF have been reported to have SP and have a high incidence of recurrence on the same side after conservative management (50%) or intercostal drainage (55.2%). The average annual incidence is 1 per 167 each year (Batten 1982).
- The risk of SP increases with cepacia or pseudomonas infections, allergic bronchopulmonary aspergillosis (ABPA), and lower lung function.13 Pleurodesis increases the risk of bleeding associated with lung transplantation but is not an absolute contraindication.
- Inhalational and intravenous drug use such as marijuana and cocaine has been implicated as etiology of SP as well.14
- Interstitial lung diseases associated with connective tissue diseases
- Ankylosing spondylitis (AS) when apical fibrosis is present: The typically low incidence of SP in patients with AS (0.29%) increases 45-fold to 13% when apical fibrotic disease exists.15
- Idiopathic pulmonary fibrosis
- Sarcoidosis
- Lymphangioleiomyomatosis (LAM)
- SP may be the presenting sign of LAM, a disease characterized by thin-walled cysts in women of childbearing age.
- Respiratory failure may lead to a need for lung transplantation, and prior pleurodesis is no longer an absolute contraindication for lung transplantation.
- Langerhans cell histiocytosis
- Acute respiratory distress syndrome (ARDS) and positive pressure ventilation in ICU: High peak airway pressures can translate into barotrauma in up to 3% of patients on a ventilator and up to 5% of patients with ARDS.16
- Severe acute respiratory syndrome (SARS): 1.7% of patients with this acute viral syndrome developed SP.17
- Thoracic endometriosis
- In a case series of 229 patients, catamenial pneumothorax caused by thoracic endometriosis was localized to the visceral pleura in 52% of patients and to the diaphragm in 39% of patients.
- Treatment is almost always surgical; most cases present during or shortly after menses, and the SP is usually right-sided. It occurs mostly in women aged 30-40 years.
- The risk of thoracic endometriosis can not be predicted from the site of peritoneal lesions.18
- Causes of iatrogenic pneumothorax
- Transthoracic needle aspiration biopsy of pulmonary nodules
- Transbronchial biopsy
- Thoracentesis
- Central venous catheter insertion
- Intercostal nerve block
- Tracheostomy
- Cardiopulmonary resuscitation
- Positive pressure ventilation and ARDS in the ICU: High peak airway pressures can translate into barotrauma-associated pneumothorax in up to 3% of patients on a ventilator and up to 5% of patients with ARDS.16
- Acupuncture
- Nasogastric feeding tube placement
- Causes of traumatic pneumothorax
- Trauma - Penetrating and nonpenetrating injury
- Rib fracture
- High-risk occupation (eg, diving, flying)
Differential Diagnoses
Esophageal Spasm
Myocardial Ischemia
Pericarditis, Acute
Pleurodynia
Pulmonary Embolism
Other Problems to Be Considered
Large bulla can simulate pneumothorax on chest radiographs. CT scan may be required to clarify the diagnosis.
Occasionally, skin folds, the scapula, and bed sheets can mimic the pleural line, falsely suggesting pneumothorax on the chest radiograph.
Workup
Laboratory Studies
- Arterial blood gas - In patients with severe underlying lung disease and in those with persistent respiratory distress despite treatment
- Hypoxemia occurs with increased alveolar-arterial oxygen tension gradient.
- Hypoxemia tends to be more severe in patients with secondary spontaneous pneumothoraces.
Imaging Studies
- Chest radiograph (confirms pneumothorax)
- A linear shadow of visceral pleura with lack of lung markings peripheral to the shadow may be observed, indicating collapsed lung.
- In supine patients, deep sulcus sign with radiolucency along costophrenic sulcus may help to identify occult pneumothorax.
- Mediastinal shift toward the contralateral lung may also be apparent.
- Small pleural effusions commonly are present and increase in size if the pneumothorax does not reexpand.
- Airway or parenchymal abnormalities in the contralateral lung suggest causes of secondary pneumothorax. Evaluation of the parenchyma in the collapsed lung is less reliable.
- For more information, see eMedicine Radiology article Pneumothorax.
- Method to estimate the fractional size of pneumothorax
- Calculate the ratio of the transverse radius of the pneumothorax (cubed) to the transverse radius of the hemithorax (cubed).
- To express the pneumothorax size as a percentage, multiply the fractional size by 100.
- The cut-point distinguishing small and large pneumothoraces varies somewhat among professional societies and experts. The British Thoracic Society uses 2 cm as the cut-off,19 the American College of Chest Physicians uses 3 cm as the cut-point,20 and the Light Index uses 15% of the thoracic volume on the posterior-anterior film as the cut point.21
- Lateral decubitus film
- Confirmation of a suspected pneumothorax that is not readily observed on standard supine anteroposterior (AP) radiograph can be demonstrated by obtaining a lateral decubitus film with the involved hemithorax positioned uppermost.
- Rib films
- Rib films are indicated if the patient has localized rib pain.
- Rib films are also indicated when evaluating the patient's posttrauma status if nothing is found on the main posteroanterior (PA) and lateral films.
- CT scan
- CT scan is not recommended for routine use but can help to accomplish the following:
- Distinguish between a large bulla and a pneumothorax
- Indicate underlying emphysema or emphysemalike changes
- Determine the exact size of the pneumothorax, especially if it is small
- Confirm the diagnosis of pneumothorax in patients with head trauma who are mechanically ventilated
- CT is widely used in actual clinical practice to assess the possibility of associated concurrent pulmonary disease because of the inherent superiority of CT scans to visualize the details of lung parenchyma and pleura.
- Ultrasonography
- Ultrasonography is increasingly used in the acute care setting as a readily available bedside tool, especially in ICU and emergency departments.
- Traumatic pneumothorax in the ICU setting can be followed accurately and early (initial 24 hours) with ultrasonography alone for resolution of the lesion.
- Lung sliding is the terminology for normal pleural movement in patients without pneumothorax.22 One study showed absent lung sliding from an anterior approach indicated pneumothorax (n =9) with 81% sensitivity and 100% specificity.
- Ultrasonography has high sensitivity (95.65%), specificity (100%), and diagnostic effectiveness (98.91%) for pneumothorax when using CT as the criterion standard. The sensitivity drops in the ICU, especially in patients with acute respiratory distress syndrome (ARDS).23
- Ultrasonography cannot be used to discriminate between a COPD-associated bleb and pneumothorax.24