|Home | About | Journals | Submit | Contact Us | Français|
Puncture or laceration of the pulmonary, intercostal, or peripheral vessels is an uncommon, but potentially life-threatening, complication of thoracentesis, which has been reported to result in severe haemothorax in 75% of patients. Damage to these vessels typically requires surgical intervention or intra-arterial embolisation. We report the successful non-surgical management of an unusual case of pulmonary artery perforation without concomitant haemothorax in an 82-year-old man who underwent thoracentesis.
Thoracentesis is generally considered to be a safe procedure; however, complications are not infrequent.1 Perforation of the pulmonary artery is a potentially life-threatening complication of thoracentesis, which has been reported to result in severe haemothorax in 75% of patients.2 Damage to the pulmonary artery typically requires surgical intervention or intra-arterial embolisation.1,2 We herein report the non-surgical management of an unusual case of pulmonary artery perforation without concomitant haemothorax in an 82-year-old man who underwent thoracentesis.
An 82-year-old man with chronic obstructive pulmonary disease (COPD), hypertension, and congestive heart failure was admitted to the intensive care unit because of urosepsis with septic shock (blood pressure, 85/40 mmHg). The patient’s haemodynamic status stabilised after adequate crystalloid fluid hydration, and the underlying infection was controlled with antibiotic therapy. On the fifth day after admission, respiratory distress was observed (respiratory rate, 30 per minute), and bilateral, large pleural effusions were detected by chest X-ray.
Thoracentesis was performed with the patient in the left decubitus position by a junior resident following a standard procedure under direct supervision of an attending cardiologist. After injection of 2% xylocaine, an 8-Fr pig-tail catheter was inserted into the right pleural space under sonographic guidance. Non-pulsatile, fresh blood immediately drained from the catheter, and the catheter was immediately clamped. However, when the catheter was unclamped 5 h later, non-pulsatile blood was again observed and it was clear that the catheter was unobstructed. A right haemothorax was suspected, although no tachycardia or hypotension was present. We performed a fine-needle diagnostic thoracentesis which produced a small amount of yellowish, transudative fluid; however, no blood was produced. Thus, vessel perforation without pericatheter leakage of blood was possible. Chest X-ray revealed inappropriate positioning of the pig-tail catheter and a large, residual pleural effusion (Fig. 1A). Contrast-enhanced multidetector computed tomography (MDCT) of the chest revealed the tip of the pig-tail catheter was within the bronchopulmonary artery of the right anterior basal segment (Fig. 1B). A large pleural effusion and an ipsilateral pneumothorax, but no pulmonary haematoma or haemothorax were noted. We performed a repeat thoracentesis with a second pig-tail catheter in the right pleural space (Fig. 1C), and 720 ml of yellowish, transudative pleural effusion was drained.
Analysis of the blood obtained from the first pig-tail catheter was consistent with that of an arterial blood sample, except for a lower PaO2 in the effusion. We consulted a thoracic surgeon and an interventional radiologist, both of whom recommended close observation. We considered arranging a pulmonary angiogram and removing the catheter; however, it was believed conservative management was more appropriate given the risk that surgical intervention may have been required if the catheter was removed. Daily radiographs were performed and, on the fourth day, we observed that the first pig-tail catheter had spontaneously dislodged from the bronchopulmonary artery (Fig. 1D). The following day, we carefully removed the catheter without subsequent haemothorax formation. For personal reasons, the patient was referred to another hospital on the seventh day after removing the catheter. His haemodynamic status was stable before discharge.
The major complications of thoracentesis include pneumothorax, haemothorax, haemoperitoneum, spleen or liver puncture, and sheered-off catheter.1 Although puncture of the pulmonary arteries is rare, haemothorax occurs in approximately 75% of patients in whom the intercostal, pulmonary, or peripheral vessels have been punctured or lacerated, and may be life-threatening.3 Measures suggested to reduce the rate of complications during thoracentesis include performing the procedure under direct ultrasonographic guidance and limiting the procedure to well-trained and experienced physicians.1,3 In our patient, despite the use of ultrasonographic guidance and supervision by an attending cardiologist, perforation of the bronchopulmonary artery still occurred.
The unusual feature of this case, absence of haematoma or haemothorax, was likely because the diameter of the pig-tail catheter was similar to that of the punctured artery, thus, no blood leaked from the perforation. Although there is pulmonary parenchyma between the pulmonary artery and pleura, when major artery trauma occurs during thoracentesis, it usually results in haemothorax. A pulmonary haematoma is typically evident on radiographic studies in cases of lung parenchyma or pulmonary perforation; yet, despite the malpositioned catheter, no evidence of pulmonary haematoma or haemothorax was ever present. Though infusion of contrast material through the catheter to verify position is frequently necessary, we had sufficient evidence by chest CT regarding the position of the catheter, and is was not necessary to subject the patient to the risk of an angiogram.
The first choice for management of a damaged vessel is surgical repair; however, many patients cannot tolerate the stress of operative intervention. This complication in our patient was difficult to manage because removal of the pig-tail catheter could have resulted in massive haemothorax, whereas leaving the catheter in situ could have resulted in pulmonary emboli; thus, various treatment options were considered.
Unlike the situation with our patient, most cases of damage to the pulmonary artery result in severe haemothorax. Coil occlusion of a damaged pulmonary artery appears to be a valuable strategy to prevent fatal haemorrhage,4 and coil embolisation has also been used to treat haemothorax successfully.5 In a review of the literature, and the experience at their institution of catheter-induced pulmonary artery perforation, Sirivella et al.2 reported that surgical intervention was effective in controlling haemorrhage, but did not reduce the number of deaths. The authors suggested conservative treatment as an option for haemodynamically stable patients without haemothorax. Given our patient’s stable haemodynamic status without hypoxaemia, and the risks associated with an invasive surgical procedure, we chose conservative treatment with intensive monitoring. Fortunately, the catheter dislodged spontaneously without the occurrence of haemothorax. The absence of haemothorax after dislodgment of the catheter was likely the result of thrombus occlusion caused by the prolonged presence of the catheter.
This case high-lights the possibility that severe complications may occur during thoracentesis. Our experience supports the conclusion that emergency surgical intervention may not be necessary in managing pulmonary artery perforation in patients without haemothorax.