Pneumothorax and Barotrauma
Definition and History
Pneumothorax is defined as air in the pleural space because of a break in the visceral or parietal pleura. The term pneumothorax was used for the first time in 1803.1 Laennec2 gave the first clinical description of pneumothorax in 1819; however, the first chest radiograph demonstrating this entity was not published until 1901.3 Definitive therapy became available with the advent of tube thoracostomy in 1876.4 It was thought to be always associated with tuberculosis until 1932 when “spontaneous pneumothorax in the apparently healthy” was first described.5
Spontaneous pneumothorax can be classified as primary or secondary. Primary spontaneous pneumothorax (PSP) occurs in patients with no apparent underlying lung disease. Secondary spontaneous pneumothorax (SSP) occurs in association with a known underlying lung disease such as chronic obstructive pulmonary disease (COPD).6 Nonspontaneous causes of pneumothorax include traumatic and iatrogenic (Table 47.1).
Table 47.1
Classification of Pneumothorax
From Grundy S, Bentley A, Tschopp JM: Primary spontaneous pneumothorax: A diffuse disease of the pleura. Respiration 2012;83:185-189.
Incidence
The incidence of PSP in men varies geographically, from 7.4 per 100,000 population per year in the United States to 37 per 100,000 population per year in the United Kingdom. In women, the incidence is substantially less, ranging from 1.2 per 100,000 population per year in the United States to 15.4 per 100,000 population per year in the United Kingdom.7 The reason for these differences is not known. In a study that evaluated 1199 patients with pneumothorax that included 865 male patients and 334 female patients, 60.3% of the pneumothoraces were spontaneous, 33.6% were traumatic, and 6.1% were iatrogenic.8
The estimates of the incidence of recurrent PSP range from 20% to more than 50% with most recurrences occurring within the first year.9
Pathophysiology
Emphysema-like Changes
PSP occurs in patients with no previously known lung disease. It is important to note, however, that this does not mean there is no underlying pathologic process. A finding of abnormal pleura is very common in PSP if looked for carefully.10 Abnormalities seen in PSP are summarized in Table 47.2 and include blebs and bullae, which are otherwise known as emphysema-like changes (ELCs). These areas of weakness of the visceral pleura are prone to rupture, allowing air to leak into the pleural space. Abnormalities can be visualized radiographically with high-resolution computed tomography (CT) scans and macroscopically at thoracoscopy.10,11 High-resolution CT imaging reveals these defects in approximately 80% of PSP patients. The literature is mixed as to whether the presence of or extent of ELCs is directly related to the risk of recurrence. Some case series suggest that there is no association,12,13 although others suggest the presence of contralateral blebs/bullae is a risk factor for future pneumothorax.14,15 The only clear conclusion that can be drawn is that there is a direct association between the presence of ELCs and the occurrence (but not necessarily recurrence) of a pneumothorax (Table 47.2 and Fig. 47.1).
Table 47.2
Pathologic Changes Associated with Primary Spontaneous Pneumothorax (PSP)
From Grundy S, Bentley A, Tschopp JM: Primary spontaneous pneumothorax: A diffuse disease of the pleura. Respiration 2012;83:185-189.
Pleural Porosity
ELCs are not the sole cause of PSP. Air leak has been described in areas where no ELCs are seen, leading to the concept known as “pleural porosity.”16,17 When evaluated with fluorescein-enhanced autofluorescence, areas of high-grade abnormality in the visceral pleura were frequently visualized separate from any area of abnormality seen with white light at thoracoscopy in patients with PSP. High-grade abnormalities were not seen in control patients.18 Areas of fluorescein leak (i.e., areas of potential air leak) were visualized in only a small proportion of PSP, but these areas were noted to be distinct from areas of ELC. When studied with electron microscopy, the linings of some resected areas of ELC have been shown to be almost completely absent of mesothelial cells and have abnormal pores present19 (Fig. 47.2).
Distal Airway Inflammation
Pathologic findings suggest an inflammatory cause to the formation of ELCs. Chronic distal airway inflammation with lymphocyte and macrophage infiltration alongside fibrotic changes and compensatory emphysema can be seen microscopically in areas of lung tissue from patients with PSP.20 In a different study, the presence of respiratory bronchiolitis (RB) was seen in close to 90% of patients with PSP who underwent surgical resection in a different study.21 However, all these patients in this study were smokers, and smoking is a recognized cause of RB. It has been proposed that distal airway inflammation associated with PSP leads to obstructive gas trapping and consequent increases in distal airway pressure, which possibly causes air leak into the pleural space.22
Smoking
Cigarette smoking is a significant risk factor for PSP. It is thought to be due to the consequences of airway inflammation leading to airway obstruction with a check valve phenomenon, causing air trapping and development of pneumothorax.22 The lifetime risk is 12% in smokers compared to 0.1% in nonsmokers. Risk is also directly related to the amount of cigarette smoking. Compared to nonsmokers, the relative risk of PSP in men was seven times higher in light smokers (1-12 cigarettes per day), 21 times higher in moderate smokers (13-22 cigarettes per day), and 102 times higher in heavy smokers (>22 cigarettes per day). For women, the relative risk was 4, 14, and 68 times higher in light, moderate, and heavy smokers, respectively.23 Cessation of smoking appears to reduce the risk of recurrence,9 and continued smoking increases the risk of recurrence.24
RB, a form of airway inflammation associated with cigarette smoking, may contribute to the development and recurrence of PSP. In a study with 115 patients with PSP who underwent video-assisted thoracoscopic surgery (VATS), pneumothorax recurrence rates were higher in patients with extensive rather than nonextensive RB for both nonoperative and postoperative pneumothorax.25
Genetics
Familial inheritance of pneumothorax describing the clustering of PSP in certain families has been published. Autosomal dominant, autosomal recessive, polygenic, and X-linked recessive inheritance mechanisms have been proposed.26–28 Birt-Hogg-Dubé (BHD) is an autosomal dominant cancer disorder that predisposes patients to benign skin tumors and renal cancer. It is associated with pleuropulmonary blebs and cysts that lead to PSP.29 In one study of 198 patients with this syndrome, 48 patients (24%) had a history of pneumothorax.30 The gene responsible for this familial cancer syndrome (FLCN) has been mapped to chromosome 17p11.2.31,32 Other mutations of FLCN have been associated with spontaneous pneumothorax and bullous lung disease in the absence of Birt-Hogg-Dubé syndrome.33
Patients with Marfan syndrome are tall, and pneumothorax is a common pulmonary complication. Marfan syndrome is caused by the mutation in FBN1 gene on chromosome 15. This gene is responsible for the formation of 10- to 12-nm microfibrils in the extracellular matrix of connective tissue. It is hypothesized that familial spontaneous pneumothorax is caused by a connective tissue disorder that exhibits mendelian inheritance and FBN1 has been postulated as the causative gene.34
Classification
Spontaneous Pneumothorax
Primary Spontaneous Pneumothorax
PSP is classically seen in previously healthy young men with an asthenic body habitus. The incidence of PSP rises with increasing height among adults of both sexes, more so in men. For those 76 inches or taller, the rate was 200 per 100,000 person-years.35 It is hypothesized that individuals with tall stature and low body mass index combined with smoking are predisposed to develop ELCs owing to the pressure gradient between the lung base and the apex, resulting in increased alveolar distending pressures at the apex.36 Smoking, as previously described, greatly increases the risk of PSP. Smoking increases the relative risk of developing spontaneous pneumothorax about ninefold in women and 22-fold in men, and there is a statistically significant dose-response relationship between smoking and spontaneous pneumothorax.37
Secondary Spontaneous Pneumothorax
SSP has been described in a large variety of diseases including COPD with emphysema, cystic fibrosis (CF), tuberculosis, lung cancer, human immunodeficiency virus (HIV)-associated Pneumocystis jiroveci pneumonia, followed by more rare but “typical” disorders such as lymphangioleiomyomatosis (LAM) and histiocytosis X (Box 47.1). Because lung function in these patients is already compromised, SSP often presents as a potentially life-threatening disease requiring immediate action, as opposed to PSP, which is more of a nuisance than a dangerous condition. The general incidence is almost similar to that of PSP.38
Chronic Obstructive Pulmonary Disease
COPD is the most common cause of SSP, with nearly 70% of SSP attributed to COPD.39 The peak incidence of SSP from COPD typically occurs later in life averaging 60 to 65 years of age.40 The clinical presentation of pneumothorax in COPD is often atypical—pain may be absent, anxiety and breathlessness may predominate and be out of proportion to the collapsed lung, and the classic sign of hyperresonance may not be helpful because of the underlying emphysema. The air leak in these patients is usually large, and the tissues are slow to heal, so it is weeks before the tubes can be taken out.41
Pneumothorax in Drug Abusers
When the peripheral veins of chronic abusers of drugs become obliterated because of a sclerotic or infectious process, the individual may attempt to use larger veins in the groin or neck. Attempted subclavian or supraclavicular (“pocket shot”) injection of drugs in the street setting has led to unilateral or bilateral pneumothoraces.42–44 Douglas and Levison45 found that the incidence of pneumothoraces is equal in both sexes and that it is less of a problem in teenagers and in addicts older than 40 years of age. It was also noted that although most drug users describe using small (21- or 22-gauge) needles, a large, complete, or tension pneumothorax usually develops.
Pneumothorax in HIV-Infected Patients
Pneumothorax is an uncommon but potentially fatal complication of HIV infection. The first report of spontaneous pneumothorax in patients with acquired immunodeficiency syndrome (AIDS) was in 1984.46 With the diagnosis of AIDS, a patient’s risk of sustaining a nontraumatic pneumothorax increases to 450 times that of the general population.47 It has since been described in a generalized HIV-infected population.48,49 Pneumothorax complicated 1.2% of all hospital admissions in a cohort of 599 HIV-infected patients followed over 3 years in a prospective observational study. There was also an associated increase in in-hospital mortality rate (31% versus 6%) for patients without pneumothorax.48
A high incidence (2-9%) of pneumothorax has been reported in patients with AIDS and Pneumocystis carinii pneumonia (PCP).50–52 Pneumocystis carinii, which was thought to be a protozoan, has been renamed as Pneumocystis jiroveci and is now classified as an archiascomycetous fungus.53 Causes of pneumothorax in HIV-infected individuals include P. jiroveci54–57 along with other infectious agents such as Mycobacterium tuberculosis, M. avium intracellulare, pulmonary cytomegalovirus, Pneumococcus organisms,54 or pulmonary toxoplasmosis.54 Pneumothorax has also been described in HIV-infected individuals from Kaposi sarcoma.58
The cause of pneumothorax in patients with PCP is unclear. Several investigators believe that extensive tissue invasion within the alveolar interstitium in severe PCP is an important factor in causing necrosis and subsequent pneumothorax. Several observations highlight this point. The most common sites of tissue invasion with PCP are the alveolar septa, pleurae, and vasculature.59 Tissue invasion could cause necrosis as a result of direct tissue injury by toxins from Pneumocystis,59 infarction from vascular compromise,60,61 or as a result of the host inflammatory response.62
The administration of aerosolized pentamidine has been implicated in the pathogenesis of cavitation, cyst formation, and pneumothorax,50,63,64 but the biologic basis for this relationship is unknown. No direct toxic action of pentamidine on the lungs has been described, so an indirect effect may be present. Cavitation due to PCP may occur primarily in the upper lobes and periphery because aerosolized pentamidine is preferentially delivered to the proximal parenchyma of the lower lobes. Inadequate deposition of pentamidine in the periphery of the lung could allow a chronic, low-grade infection with Pneumocystis to persist, leading to peripheral lung destruction and pneumatocele formation. Increased survival time of AIDS patients due to prophylaxis could allow for development of these lesions.50
Several other risk factors for AIDS-related pneumothorax have been identified. In addition to previous or active infection of P. jiroveci and aerosolized pentamidine, cigarette smoking and the presence of pneumatoceles on chest radiograph are risk factors.65 The association between cigarette smoking and AIDS-related pneumothorax could be explained by subclinical obstructive disease preventing adequate deposition of aerosolized pentamidine in the lung periphery, resulting in subpleural Pneumocystis infection.65 Pulmonary tuberculosis also appears to increase the risk of pneumothorax in AIDS.66
Catamenial Pneumothorax
In most cases, catamenial pneumothorax is related to pelvic or thoracic endometriosis.67,68 Catamenial pneumothorax occurs typically within 24 to 72 hours after onset of menstruation. It is often recurrent and more common than previously thought. Two mechanisms have been described for pneumothorax related to endometriosis. The most common is the movement of endometrial implants to the diaphragm, preferentially to the right side because of the recognized peritoneal circulation up from the pelvis to the right side. These implants then create channels or “holes” through the diaphragm that allow the implants or air to move into the chest. The second and much less frequent cause of endometrial implants causing pneumothorax in the chest is through the venous implants that lodge into the lung itself.69 Clinical manifestations of thoracic endometriosis include chest pain, dyspnea, and hemoptysis. Treatment for the prevention of recurrence is indicated after a first episode of catamenial pneumothorax because recurrences are frequent.38
Cystic Fibrosis
SSP occurs in approximately 6% of all patients with CF and this number increases to 16% to 20% among those who survive to age 18.70,71 SSP from CF is usually due to rupture of apical subpleural cysts. The risk of pneumothorax is inversely proportional to the forced expiratory volume in 1 second (FEV1). Other factors associated with an increased risk of pneumothorax include infection with Pseudomonas aeruginosa, Burkholderia cepacia complex, or Aspergillus species. A previous history of massive hemoptysis also increases risk.
Nonspontaneous Pneumothorax
Traumatic Pneumothorax
Pneumothorax ranks second to rib fractures as the most common sign of chest trauma. It occurs in up to 50% of chest trauma victims.72 Most are caused by a penetrating injury, but closed chest trauma causing alveolar rupture from thoracic compression, fracture of a bronchus, and esophageal rupture have also been reported.73,74 Traumatic pneumothorax can be classified as open, closed, tension, or hemopneumothorax. A tension pneumothorax should be managed immediately by decompression with a large-bore needle usually in the second anterior interspace in the midclavicular line. Open pneumothorax should have a moist sterile gauze pack placed over the open wound, followed by a chest tube. Hemopneumothorax (20% of trauma patients) requires insertion of a large-bore (28-36F) chest tube.38
Occult pneumothorax may be present in half of blunt abdominal trauma patients, many of which are undetected by chest radiograph.75–79 CT of the chest should therefore always be performed in these patients. Most surgeons and emergency physicians will place a chest tube in occult and nonoccult pneumothoraces. Studies suggest, however, that clinically stable patients and those who do not have an enlarging pneumothorax may be treated conservatively, ultimately requiring chest tube placement in about 10% of cases.80
Traumatic Iatrogenic Pneumothorax
Iatrogenic pneumothorax occurs most often following transthoracic needle biopsy (24%), subclavian vein catheterization (22%), thoracentesis (20%), transbronchial lung biopsy (10%), pleural biopsy (8%), and positive-pressure ventilation (7%).81 Diagnosis of iatrogenic pneumothorax is often delayed. Small and asymptomatic iatrogenic pneumothorax, however, often do not need any treatment and resolve spontaneously. In larger or symptomatic pneumothoraces, simple manual aspiration or placement of a small catheter or chest tube attached to a Heimlich valve is usually sufficient.82 Larger tubes may be necessary in patients with emphysema or when the patient is placed on a mechanical ventilator.
Pulmonary Barotrauma During Mechanical Ventilation
Pulmonary barotrauma (PBT) refers to alveolar rupture due to elevated transalveolar pressure (the alveolar pressure minus the pressure in the adjacent interstitial space). PBT was previously estimated to range between 3.8% and 41.7% of patients undergoing mechanical ventilation.83 The rate in actuality may be lower because low tidal volume ventilation is becoming more common. Consequences of barotrauma include pneumothorax, pneumomediastinum, pneumoperitoneum, and subcutaneous emphysema.
Positive-pressure ventilation increases transalveolar pressure, which can cause alveolar rupture.84 Alveolar rupture allows air from the alveolus to enter the pulmonary interstitium where it can dissect along the perivascular sheaths toward the mediastinum. This can lead to pneumothorax, pneumomediastinum, pneumoperitoneum, or subcutaneous emphysema85,86 (Fig. 47.3). Bronchopleural fistula, tension pneumothorax, tension lung cyst, and subpleural air cyst have also been reported but are less common.87
In a multicenter prospective cohort study of 5183 mechanically ventilated patients, the incidence of PBT was 3%.88 Asthma, chronic interstitial lung disease, and acute respiratory distress syndrome (ARDS) were identified as independent risk factors for barotrauma. Other studies have also demonstrated that acute lung injury (ALI) and ARDS are independent risk factors for PBT.89,90 Elevated peak and plateau pressures have been identified as risk factors.91,92
Neither open lung strategies using high levels of positive end-expiratory pressure (PEEP) nor recruitment maneuvers have been shown to increase the risk of barotrauma.93,94
Clinical Presentation
The clinical presentation of PBT can vary. With pneumothorax, patients may complain of dyspnea or chest pain. Physical findings can include tachycardia, tachypnea, hypertension, or oxyhemoglobin saturation accompanied by unilateral reduction of breath sounds. If a tension pneumothorax develops, there may be hypotension and tracheal deviation. Patients with pneumomediastinum may complain of dyspnea and chest or neck pain. Other findings include tachycardia, tachypnea, and hypertension. A crunching sound may be heard during auscultation. Rarely, hypotension from decreased venous return and cardiac output may occur if tension pneumomediastinum develops.95 Pneumoperitoneum may manifest itself as abdominal pain. Other physical findings include abdominal distention, tenderness, and tympany. Rarely, abdominal compartment syndrome may develop if the pneumoperitoneum progresses to a tension pneumoperitoneum.96 Subcutaneous emphysema generally presents as painless soft tissue swelling. It typically appears in the upper chest, neck, and face. Compression of the affected areas can reveal crepitus. A rare consequence of severe subcutaneous emphysema is compartment syndrome.97
Diagnosis
The diagnosis of pneumothorax is suspected when a patient presents with the symptoms and signs described earlier, then confirmed with a portable chest radiograph. An upright chest radiograph has the highest diagnostic yield for pneumothorax, although diagnosis in the intensive care unit (ICU) may be difficult as most patients are semirecumbent or supine.98 In a fully upright chest radiograph, a pneumothorax appears as a radiolucent collection between the visceral and parietal pleurae in the superior portion of the chest. In contrast, when a patient is supine, free air collects in the anterior chest, displacing the costophrenic angle inferiorly, often creating a “deep sulcus” sign. The deep sulcus sign refers to a unilateral increase in the apparent size of the costophrenic angle (Fig. 47.4).
Bedside ultrasound is being used more readily in the ICU to rapidly diagnose pneumothorax. Utilizing the M-mode, the absence of “lung sliding” is indicative of the presence of a pneumothorax99 (Video 47.1).
Pneumomediastinum frequently coexists with pneumothorax. It is usually diagnosed with a portable chest radiograph. It typically appears as radiolucent streaks in the mediastinum (Fig. 47.5).
Pneumoperitoneum is diagnosed with a chest radiograph less than one third of the time. A suspected pneumoperitoneum is best evaluated by chest CT.100 The patient should remain in position for 5 to 10 minutes before the radiograph is taken. This allows time for air to collect in a sufficient volume to be detected radiographically. Pneumoperitoneum may be identified on a supine abdominal radiograph (see Fig. 47.5). Free air accumulates anteriorly when the patient is supine and on a chest or abdominal radiograph may present in several ways. Gas appearing on both sides of the bowel wall is referred to as Rigler’s sign. Gas outlining the peritoneal cavity is known as the football sign. Gas outlining the medial umbilical folds is called an inverted V sign. Gas may also outline the falciform ligament or localize in the right upper quadrant.101
Subcutaneous emphysema is often found by identifying crepitus during physical examination. On chest radiograph of areas of tissue swelling, it can appear as radiolucent streaks throughout the subcutaneous tissue and muscle (see Fig. 47.5).
Prevention
To prevent barotrauma, it is generally recommended that plateau airway pressure be maintained at or below 35 cm H2O. Plateau pressure is the most indicative of the alveolar pressure and therefore is the measure of greatest concern for the prevention of PBT. Lower plateau airway pressures have been associated with a lower incidence of PBT. A threshold pressure appears to exist at 35 cm H2O, above which there is a higher incidence of barotrauma. A meta-analysis of 14 clinical trials demonstrated a strong relationship between PBT and a plateau airway pressure greater than 35 cm H2O or a static compliance less than 30 mL per cm H2O.102 There have not been direct comparisons between management targeting a plateau airway pressure or peak airway pressure. Peak airway pressure is likely a less reliable predictor of PBT given the conflicting data.103–109
Management
The best treatment for PBT is early recognition, and immediate attempts should be made to reduce plateau airway pressure.83 This may require lowering the tidal volume or PEEP, as well as increasing sedation, administering neuromuscular blockade, or advancing treatment of the underlying condition. In cases of pneumothorax while on a mechanical ventilator, there is no high-quality evidence that supports routine insertion of chest tubes for all patients. However, more than 30% of pneumothoraces in mechanically ventilated patients progress to tension pneumothoraces, indicating that these patients must be monitored closely. Treatment for mechanically ventilated patients who develop a pneumomediastinum, pneumoperitoneum, or subcutaneous emphysema is generally supportive unless there is evidence of tension pneumomediastinum or compartment syndrome from pneumoperitoneum or subcutaneous emphysema.95
Prognosis
PBT appears to be associated with increased mortality rate, even though barotrauma is not a direct cause of death in most patients. In a multicenter prospective cohort study, patients with barotrauma had a significantly higher mortality rate (51% versus 39%), a longer length of ICU stay (median 9 versus 7 days), and a longer duration of mechanical ventilation (median 6 versus 4 days) than patients without barotrauma.88 Mortality rate may be related to the severity of the PBT. In one retrospective cohort study of 1700 mechanically ventilated patients, the mortality rate approached 100% when PBT caused a large (>500 mL per breath) bronchopleural fistula.109 High-frequency jet ventilation is FDA (Food and Drug Administration) approved for the management of large bronchopleural fistulas, but this may be outweighed in some patients by increased plateau airway pressure (alveolar pressure), decreased oxygenation, or worse hypercapnia.110
Pneumothorax After Fiberoptic Bronchoscopy and Needle Biopsy of the Lung
Multiple literature reviews have documented the relative safety of fiberoptic bronchoscopy (FOB) with transbronchial biopsy. One review of more than 9000 such procedures found that the rate of pneumothorax was 1.9%.111 An immediate postbronchoscopic chest radiograph rarely provides clinically useful information, and in FOB without transbronchial biopsy an immediate postbronchoscopy radiograph is not necessary.112,113 Another study in 2006 concluded that in asymptomatic patients, routine radiograph after transbronchial biopsy is not necessary.114 It was determined that certain patient populations should have routine radiographs performed after FOB with transbronchial biopsy: comatose or mentally retarded patients, patients receiving positive-pressure ventilation, patients with severe respiratory compromise as a result of disease or surgery, patients with bullous disease, patients who complain of chest pain, and outpatients. Pneumothorax after bronchoalveolar lavage without biopsy is extremely rare. The complication of pneumothorax after transbronchial needle aspiration is also low.115
Pneumothorax is the most common complication of needle aspiration or biopsy of the lung. It has been reported to occur in 17% to 26.6% of patients.116–119 The chest tube insertion rate is much lower, ranging from 1% to 14.2%.116–119 Risk factors for the development of biopsy-related pneumothorax include the presence of COPD, the absence of a history of ipsilateral surgery, small lesion size, a long needle path, and repeated pleural puncture.116–121 Enlarging or symptomatic pneumothorax can be managed by manual aspiration or placement of a small-caliber chest tube.120
Delayed pneumothorax after percutaneous fine-needle aspiration has been reported. A study by Choi and colleagues122 reported on their series of 458 patients who had undergone transthoracic needle biopsy. A follow-up chest radiograph was obtained immediately and at 3, 8, and 24 hours after the biopsy procedure. A pneumothorax that developed after 3 hours was defined as delayed pneumothorax. Pneumothorax developed in 100 of the 458 patients (21.8%), and delayed pneumothorax developed in 15 patients (3.3%). Female gender and absence of emphysematous changes correlated with an increased rate of delayed pneumothorax.
Pneumothorax After Thoracentesis
According to a 1998 National Center for Health Statistics study,123 physicians perform an estimated 173,000 thoracenteses annually in the United States. Iatrogenic pneumothoraces resulting from thoracentesis increase morbidity rate, mortality rate, and length of hospitalization. Previous reports indicated chest tube insertion may be required in up to 50% of cases with a mean duration of placement of approximately 4 days.124,125 Gordon and colleagues126 performed a systematic review and meta-analysis of 24 studies reporting pneumothorax rates after thoracentesis involving 6605 thoracenteses. The overall pneumothorax rate was 6%. In cases in which pneumothorax developed, 34.1% required chest tube placement. Statistically significant risk factors for developing thoracentesis included performing thoracentesis as a therapeutic procedure as opposed to as a diagnostic procedure; the presence of cough, dyspnea, or chest pain during the procedure; and witnessing the aspiration of air during the procedure.126 Although not statistically significant, other possible predictors included the need for two or more needle insertions and concurrent mechanical ventilation.126 Ultrasonography guidance,127–129 more experienced operators,130 and fewer needle passes conferred lower complication rates,131 which paralleled findings from central venous catheter insertion studies. Various mechanisms may explain the pneumothoraces that occur after thoracentesis: the lung may be punctured at the time of needle entry or after the fluid has been withdrawn, or a small amount of air may be drawn into the chest during aspiration or along the needle track if high negative intrapleural pressure develops.132
Pneumothorax Resulting from Nasogastric Feeding Tubes
Small-bore Silastic feeding tubes are being used with increasing frequency for short- and long-term enteral hyperalimentation. The first reported case of pneumothorax as a complication of passing a narrow-bore feeding tube was in 1978.133 This once rare complication has now become more common.134–136 Narrow-bore feeding tubes are particularly likely to give rise to pneumothorax because of the tube’s small diameter (2.7 mm), self-lubricating properties, and wire stylet. These factors allow undetected entry of the tube into the tracheobronchial tree, perforation of pulmonary tissue, and lodging in the pleural cavity.137 Other factors that increase the risk of a misplaced feeding tube include the presence of an endotracheal or tracheostomy tube (these may increase pulmonary passage of the tube by preventing glottis closure and perhaps by inhibiting swallowing), altered mental status, denervation of airways, esophageal stricture, enlargement of the heart, and neuromuscular weakness.138 The clinical signs commonly used to determine correct placement of the feeding tube may be misleading. Normally, to confirm the correct placement of a feeding tube in the stomach, a small amount of air is injected. This produces a characteristic gurgle in the left upper quadrant of the abdomen, but a “pseudoconfirmatory gurgle” with a feeding tube in the chest has been reported.139 Aspiration of large amounts of fluid through the tube is also taken to be a test of correct placement into the stomach, but delayed aspiration of a large quantity of undigested feeding solution from the pleural space, mistaken for gastric contents, has been reported.140
Pneumothorax After Percutaneous Dilational Tracheostomy
Percutaneous dilational tracheostomy (PDT) was first described in 1985 by Ciaglia and colleagues.141 A case series described subcutaneous emphysema and pneumothorax as complications after percutaneous tracheostomy in a series of 326 cases.142 Their review of the literature showed that the incidence of subcutaneous emphysema was 1.4% and that of pneumothorax was 0.8%. Findings associated with pneumothorax included difficult PDT and the use of a fenestrated cannula.
