Pulmonary Edema after Catastrophic Carbon Dioxide Embolism during Laparoscopic Ovarian Cystectomy

Carbon dioxide (CO₂) is used as a distention medium during laparoscopy because it is more tolerable to patients than other gases. However, the insufflation of several liters of CO₂ along with an extreme Trendelenburg position can cause the following: rapid absorption of a large amount of CO₂ into circulation, impeded compensatory ventilation with high intraabdominal pressure, and decreased cardiac index with increased systemic and pulmonary vascular resistance. These occurrences can result in a rare CO₂ embolism, a possibly fatal complication of laparoscopy.1

The CO₂ embolism usually occurs during the insufflation of CO₂ or immediately after the completion of CO₂ insufflation,2 but one should also consider the unexpected occurrence of a CO₂ gas embolism as late as a few hours after the patient has reached a stable cardiorespiratory status.3, 4 Sometimes, an air embolism is accompanied by pulmonary edema, which further complicates the management of the patient.5, 6 It is difficult to predict whether or not pulmonary edema will occur during the clinical course of a CO₂ embolism.

We report a case of a life-threatening CO₂ embolism with concomitant fulminant pulmonary edema that occurred approximately an hour after the completion of CO₂ insufflation.

A 29-year-old healthy woman who had previously undergone a cervical cerclage procedure for cervical incompetence was scheduled for laparoscopic gynecological surgery. The preoperative evaluation including chest PA was unremarkable (Fig. 1). Anesthesia was induced smoothly with tracheal intubation and was maintained with 1.5-2% enflurane, 3 L/min nitrous oxide, and 2 L/min O₂. The noninvasive blood pressure, electrocardiography (ECG), pulse oximetry, end tidal CO₂ (ETCO₂) and urinary output were monitored. The abdomen was insufflated with CO₂, and an intraabdominal pressure between 10 and 12 mmHg was maintained with the laparoscopic machine (CO₂-aqua-purator 1611: WISAP: Germany), followed by the lithotomy and a 20 degree Trendelenburg head-down position. The surgery and anesthesia were uneventful for one hour with an arterial blood pressure (BP) of 100 - 120/70 - 100 mmHg, a heart rate of 70 - 102 bpm, ETCO₂ of 30 - 40 mmHg, and SpO₂ of 100%.

Fig. 1
Chest PA taken before the surgery shows no evidence of a definite abnormality.

• Click for larger image

During irrigation and bleeding control, the ETCO₂ suddenly increased from 40 mmHg to 85 mmHg and then dropped to 13 mmHg. The BP decreased to 55/30 mmHg and a "mill-wheel" precordial murmur was heard on auscultation. A pulmonary CO₂ embolism was suspected. In response to these events, the inhalational anesthetics were discontinued, and the FiO₂ was changed to 100% while the CO₂ insufflation was stopped and the laparoscope was removed. At the same time, a 1 mg bolus of epinephrine was administered intravenously with a continuous infusion of 1 mg epinephrine and 4 mg norepinephrine in NS 100 mL while the IV fluid rate was increased. Despite these efforts, the BP and ETCO₂ were still low. Every 2 - 3 minutes, 1.0 mg of Atropine was administered, but the patient's heart rate decreased further and no beat was observed for approximately 15 to 30 seconds. Immediately, chest compression was initiated. After 2 - 3 minutes of compression, fine ventricular fibrillation was observed on the ECG, and defibrillator cardioversion (DC) at 200 J was delivered to the patient's anterior chest. The heart rate returned to sinus rhythm, but it was immediately converted to ventricular fibrillation. With another application of 200 J DC and 1% lidocaine 60 mg intravenously, the beat was converted to sinus tachycardia. However, the BP was still low at 55/28 mmHg.

The right subclavian vein was cannulated, and her hemodynamics improved gradually. Approximately 40 minutes after the incident, her BP was 130/80 mmHg, her heart rate was 135 bpm, her ETCO₂ was 30 mmHg, her SpO₂ was 100%, with her FiO₂ of 100%. At that time, chest radiography showed newly developing extensive bilateral consolidations in her lungs, suggesting pulmonary edema (Fig. 2). Twenty milligrams of Furosemide and 125 mg of methyl-prednisolone were administered intravenously, and her central venous pressure was maintained within 5 mmHg. In order to locate the hidden bleeding, the lower abdomen was incised and examined. But the bleeding was negligible.

Fig. 2
Chest AP taken 40 minutes after the CO2 embolism shows newly developing bilateral extensive consolidations in the lungs, indicating pulmonary edema.

• Click for larger image

A ventilator was applied to the patient in the intensive care unit, and supportive care was carried out to treat this case of pulmonary edema. Three hours after admission to the intensive care unit, the ventilator was disconnected and her trachea was extubated. She was alert, and her BP was 110/70 mmHg, her heart rate was 95 bpm, her respiration rate was 20/min, and her SpO₂ was 99% with a facemask. The next morning, the chest AP showed clear lungs (Fig. 3).

Fig. 3
Chest AP taken the morning after surgery shows clear lungs without an enlarged heart.

• Click for larger image

The initial sign in this case included an abrupt and transient increase in the ETCO₂ to 85 mmHg, which likely was related to the acute increase in the CO₂ level in the vascular system followed by exhalation from the respiratory system.7 Subsequently, there was a decrease in the ETCO₂ to 13 mmHg. In addition, there was a deterioration of the hemodynamic parameters with a decrease in the BP and heart rate, as well as a mill-wheel murmur consistent with the manifestation of a gas embolism. These manifestations can be explained by the increased CO₂ causing overcrowding in the right heart and the occluded right outflow tract of the heart, which was observed in the gas embolism.8 However, Geissler et al.9 have proposed that the primary mechanism for cardiac dysfunction after venous embolism is increased right ventricular afterload and arterial hypotension, possibly with subsequent right ventricular ischemia, rather than right ventricular outflow obstruction by an airlock. In this case, the presence of the CO₂ in the right ventricle was the primary etiology causing cardiac dysfunction. It is believed that the immediate termination of laparoscopic gas insufflation blocked any further CO₂ entry into the vascular system, and the immediate resuscitative measures were useful in preventing any devastating events.

The manner in which CO₂ entered the vessels during the laparoscopic surgery is unclear. When a CO₂ embolism is encountered, it normally occurs at the beginning of or immediately after the completion of CO₂ insufflation.2 In addition, the risk of an embolic episode decreases after the completion of the pneumoperitoneum because increased intra-abdominal pressure can cause the collapse of the injured vessel. In contrast, our case occurred more than an hour after the insufflation of CO₂.

There are several possible reasons for this embolism. First, the previous lower abdominal surgery might have caused peritoneal adhesions that could have ruptured during the forceful irrigation. This may have caused CO₂ to enter into the circulatory system through the injured vessels and then led to the CO₂ embolism.2 Another possibility is inadvertent use of the laparoscopic machine. In our case, irrigating solutions were introduced into the intraperitoneal cavity by CO₂ pressure; an unnoticed empty irrigation bottle or a CO₂ mixed irrigation solution could have allowed for the entry of CO₂ into the open venous sinuses under pressure. The coincidence of the occurrence of CO₂ embolism with the irrigation and bleeding control in this case support this possibility. Additionally, CO₂ might have become trapped in the portal system after transperitoneal absorption and later might have been released into the circulation.4 Finally, although direct injury to a vessel with a Veress needle might be possible, this possibility is less likely because the embolus occurred one hour after completing the pneumoperitoneum.

Although short-lived, the patient developed pulmonary edema in the operating room and required intensive therapy even after surgery in the intensive care unit. The cause of pulmonary edema associated with laparoscopic surgery, especially with CO₂, is unclear. However, there are some reports of noncardiogenic pulmonary edema as a consequence of an air embolism.5, 6 In these types of permeability edema, nonmechanical factors, such as the release of superoxide dismutase from leukocytes, are known to be involved.5, 10 Moreover, severe pulmonary edema after a venous air embolism is often associated with massive fluid loss.6 This is not believed to be the cause of this case of pulmonary edema even though much fluid volume had been administered during resuscitation. This administration of fluid might have helped to resuscitate the patient more rapidly.

In this case of pulmonary edema, steroid, mechanical ventilation with positive end expiratory pressure was required.5 To avoid aggravation of the edema, infused fluid should be carefully controlled. In this case we managed the patient with mechanical ventilation and fluid control under the guidance of central venous pressure. Luckily, patient oxygenation was improved when the edema was discovered, so we avoided the use of positive end expiratory pressure, which can cause cardiac depression during treatment.

Recovery time is relatively short after such events. CO₂, the causative agent in this case, is highly soluble compared with other gases including helium, argon, nitrogen, oxygen, and air.1 Furthermore, the bubbles of CO₂ gas rapidly traverse the vascular systems and disappear. These physiochemical characteristics of CO₂ may shorten the clinical course and contribute to a more rapid recovery.

According to this report, if a cardiac event occurs during laparoscopic surgery, one should consider a CO₂ embolism, even if the events occur after the completion of CO₂ insufflation. In order to prevent devastating events, preventive monitoring of the ETCO₂, SpO₂, and heart and lung sounds is essential. However, when it does occur, pulmonary edema should also be considered during patient management.