Secondary Logo

Journal Logo


Reexpansion Pulmonary Edema After Thoracoscopic Mediastinal Tumor Resection

Yanagidate, Fumi MD*†,; Dohi, Shuji MD, PhD*,; Hamaya, Yoshihiro MD, PhD*,; Tsujito, Tatsuya MD*

Author Information
doi: 10.1097/00000539-200106000-00011
  • Free

Reexpansion pulmonary edema (RPE) is a rare complication associated with the treatment of collapsed lung caused by pneumothorax, pleural effusion, and atelectasis. Rapid reexpansion of the lung after its prolonged collapse predisposes the condition to RPE (1). Can the lung reexpansion after one-lung ventilation cause edema to form in the nondependent lung? This case could indicate that an answer to this question is affirmative in a compromised patient.

Case Report

A 78-yr-old man (149 cm, 54 kg) was scheduled for left thoracoscopic mediastinal tumor resection and right lower lobectomy for squamous cell carcinoma. A chest roentgenogram revealed pulmonary emphysema in both lung fields but no atelectasis of the left lung by the tumor (Fig. 1A). He had a history of hypertension and was an exsmoker, having smoked one-half pack of cigarettes per day for 50 yr. Because of bronchial asthma, he had been receiving theophylline and procaterol hydrochloride for 1 yr; he had an asthma attack during bronchoscopic examination 1 mo before the operation. His preoperative pulmonary function studies revealed a forced vital capacity of 1.87 L (66%) and a forced expiratory volume in 1 s of 1.36 L (73%). Analysis of arterial blood gases and arterial pH (pHa) while he breathed room air were Pao2 97 mm Hg, Paco2 42 mm Hg, pHa 7.40, and base excess (BE) 1.8 mEq/L. Blood pressure was 155/81 mm Hg, and heart rate was 80 bpm. The data of other laboratory tests were unremarkable.

Figure 1A
Figure 1A:
A chest roentgenogram obtained several weeks before surgery showed pulmonary emphysema in both lungs but no atelectasis of the left lung caused by the tumor. 1B. An intraoperative chest roentgenogram showed diffuse alveolar infiltrates over the left lung, with a clear right lung field.

After diazepam 5 mg was given orally for 90 min, an epidural catheter (18-gauge) was placed into the epidural space through the T4-5 intervertebral space. General anesthesia was induced with 90 mg propofol IV supplemented with a 50 μg fentanyl after 100 mg IV hydrocortisone. The trachea and left main bronchus were intubated with a 37F, left-sided, double-lumen endobronchial tube (Bronchocath; National Catheter, New York, NY). The appropriate placement of the endobronchial tube was confirmed with chest auscultation of breath sounds and fiberoptic bronchoscopic examination. Anesthesia was maintained with thoracic epidural anesthesia by using 1.5% lidocaine, propofol, nitrous oxide, and intermittent IV fentanyl. Vecuronium was injected IV to maintain muscle paralysis and facilitate controlled mechanical ventilation. Thoracoscopic operation for the left mediastinal tumor was performed under right-lung ventilation; the bronchial lumen of the double-lumen tube was disconnected from the anesthesia machine with the patient in the right lateral position. During this period, analysis of arterial blood gas tensions and pHa revealed Pao2 146 mm Hg, Paco2 37 mm Hg, and pHa 7.398, BE −2.0 mEq/L with inspired oxygen of 100%. Removal of the mediastinal tumor was completed uneventfully 90 min later, during which time the left lung was deflated and opened to room air. The blood loss was approximately 100 mL during the procedure. Before closure of the thorax, the left lung was reinflated to examine any air leakage, and both lungs were expanded with manual positive-pressure ventilation with 20 cm H2O of peak inspiratory pressure. Blood pressure and heart rate did not change remarkably. When his position was changed to the left lateral position for the right lower lobectomy, oxyhemoglobin saturation was decreased from 99% to 96% abruptly, with an Fio2 of 0.5. A chest roentgenogram showed diffuse alveolar infiltrates over the entire left lung, with clear right lung fields (Fig. 1B). There were frothy pinky sputa, not mucous plugs, apparent by fiberoptic bronchoscopy, but we could not collect enough of the edema fluid for measuring albumin concentration. Moist rales were heard over the left lung field. Right lower lobectomy was postponed. The endobronchial tube was replaced by a regular tracheal tube and positive-pressure ventilation with a 5 cm H2O endotracheal suctioning. Furosemide and dopamine were administered IV. The patient received 2000 mL of crystalloid solution; urine output was 800 mL for 5 h. A chest roentgenogram improved markedly 2 h later, and analysis of arterial blood gases showed pHa 7.376, Pao2 171 mm Hg, Paco2 38 mm Hg, and BE −2.7 mEq/L with an Fio2 of 0.5. The patient was transferred to the intensive care unit, where his trachea was extubated on the following day and he recovered uneventfully.

Two months later, the patient received a right lower lobectomy for squamous cell carcinoma. He was not treated with theophylline and procaterol hydrochloride because of no recent asthma attack. During one-lung ventilation, both his lungs were gently expanded every 30 min with manual positive-pressure ventilation. The operation was performed uneventfully.


Many factors can cause the development of RPE during perioperative periods. The duration and severity of lung collapse and speed of reexpansion are important (2). In this case, there was no lung compression by the mediastinal tumor before surgery, and the duration of the lung collapse was 90 minutes for one-lung ventilation. Other reports described an acute form of RPE associated with lung reexpansion after several hours (3,4). Matsumiya et al. (5) reported that a two-hour collapse of the lung during one-lung ventilation for mediastinal tumor removal could contribute to the formation of RPE. However, in their patient the tumor compressed the lung field more than 80 days before surgery. The pathogenesis of RPE is unknown, although several potential mechanisms have been suggested, including increased permeability of pulmonary capillaries, a decrease in perivascular pressure, application of a high negative pressure, decreased surfactant, or decreased lymph flow (1,6,7).

Because one-lung ventilation may change the partitioning of blood flow between the nondependent and dependent lungs (8), sympathectomy of the pulmonary vasculature by thoracic epidural anesthesia in our patient might have affected the sympathetic control of the pulmonary circulation; this might have been responsible for deleterious changes in pulmonary hemodynamics of the left lung. RPE occurs mainly because of the increased permeability of the collapsed lung when the lung is inflated (9). The timing was atypical for RPE in our patient. Although there is a report indicating that dyspnea and hypoxemia developed a few hours after reexpansion of a collapsed lung (10), without measuring the protein concentration of the edema fluid, we may not exclude what the other factors (such as pulmonary hemodynamic changes associated with the lung inflation) with oxygen would contribute in our patient. Because of bronchial asthma, medication of theophylline and procaterol could affect both the development and resolution of RPE. Because theophylline (which has surprisingly few data on pulmonary circulation), an adenosine-receptor antagonist, attenuates hypoxic-induced changes in pulmonary vascular resistance (11), it is possible that during one-lung ventilation the blood flow to the collapsed left lung might not be decreased, and thus might predispose the left lung to develop pulmonary edema when it is reexpanded with a high volume of oxygen. In addition, procaterol, a β-adrenergic agonist, might be useful as a treatment to hasten the resolution of alveolar edema (12), and thus the rapid recovery of arterial blood gas tension of oxygen in our patient could be have been caused by procaterol. These factors may have contributed to one atypical form of RPE observed in our patient.

Endoscopic thoracotomy has become a very popular surgical procedure for lung and mediastinal surgery. This report suggests that pulmonary edema can develop after reexpansion of a collapsed lung for thoracoscopy that lasts 90 minutes.


1. Mahajan VK, Simon M, Huber GL. Reexpansion pulmonary edema. Chest 1979; 75: 192–4.
2. Mahfood S, Hix WR, Aaron BL, et al. Reexpansion pulmonary edema. Ann Thorac Surg 1988; 45: 340–5.
3. Ravin CE, Dahmash NS. Re-expansion pulmonary edema [letter]. Chest 1980; 77: 709–10.
4. Sherman S, Ravikrishnan KP. Unilateral pulmonary edema following re-expansion of pneumothorax of brief duration [letter]. Chest 1980; 77: 714.
5. Matsumiya N, Dohi S, Kimura T, Naito H. Reexpansion pulmonary edema after mediastinal tumor removal. Anesth Analg 1991; 73: 646–8.
6. Trapnell DH, Thurston JG. Unilateral pulmonary oedema after pleural aspiration. Lancet 1970; 1: 1367–9.
7. Pavlin J, Cheney FW Jr. Unilateral pulmonary edema in rabbits after reexpansion of collapsed lung. J Appl Physiol 1979; 46: 31–5.
8. Benumof JL. One-lung ventilation and hypoxic pulmonary vasoconstriction: implications for anesthetic management. Anesth Analg 1985; 64: 821–33.
9. Sprung CL, Loewenherz JW, Baier H, Hauser MJ. Evidence for increased permeability in reexpansion pulmonary edema. Am J Med 1981; 71: 497–500.
10. Iqbal M, Multz AS, Rossoff LJ, et al. Reexpansion pulmonary edema after VATS successfully treated with continuous positive airway pressure. Ann Thorac Surg 2000; 70: 669–71.
11. Konduri GG, Mital S, Gervasio CT, et al. Purine nucleotides contribute to pulmonary vasodilation caused by birth-related stimuli in the ovine fetus. Am J Physiol 1997; 272: H2377–84.
12. Sakuma T, Folkesson HG, Suzuki S, et al. Beta-adrenergic agonist stimulated alveolar fluid clearance in ex vivo human and rat lungs. Am J Respir Crit Care Med 1997; 155: 506–12.
© 2001 International Anesthesia Research Society