Massive carbon dioxide embolism during pneumoperitoneum for laparoscopic adrenalectomy: A case report

Laparoscopy is deemed a “minimally” invasive procedure that allows for exploration through an endoscope of the peritoneal cavity after insufflation of carbon dioxide (CO2) [1]. It has been widely used during many types of surgery. There are a number of advantages compared to open procedure which includes: decreased risk of bleeding, infection, and postoperative pain6. However, there are also some complications associated with laparoscopic surgery. For example, the pneumoperitoneum may lead to decreased cardiac output and/or increased systemic vascular resistance. In addition, the Carbon dioxide pneumoperitoneum may cause respiratory acidosis and deep venous thrombosis [2].


Introduction
Laparoscopy is deemed a "minimally" invasive procedure that allows for exploration through an endoscope of the peritoneal cavity after insufflation of carbon dioxide (CO 2 ) [1]. It has been widely used during many types of surgery. There are a number of advantages compared to open procedure which includes: decreased risk of bleeding, infection, and postoperative pain6. However, there are also some complications associated with laparoscopic surgery. For example, the pneumoperitoneum may lead to decreased cardiac output and/or increased systemic vascular resistance. In addition, the Carbon dioxide pneumoperitoneum may cause respiratory acidosis and deep venous thrombosis [2].
Although CO 2 embolisms are rare, they are potentially fetal if they are not diagnosed or treated immediately. This is especially true when the CO 2 from the right heart flows into the left heart [3] Clinically, carbon dioxide embolism can present itself with systemic hypotension, dyspnea, cyanosis, tachycardia or bradycardia, arrhythmia or asystole, elevated pulmonary arterial pressure(PAP), elevated central venous pressure, hypoxemia and increased arterial partial pressure of carbon dioxide, increase or decrease end-tidal carbon dioxide tension [4]. There are case reports about cardiac arrest associated with CO2 embolism during laparoscopic surgery8. We are reporting this case of potentially fatal complication of massive CO2 embolism during laparoscope adrenalectomy at our institute. Even the patient had experienced a massive CO2 embolism during the surgery, he discharged without any complication by the twelfth day after the operation.

Case report
A 65-year-old man, suspected with right adrenal metastasis, was scheduled to undergo a laparoscopic right adrenalectomy. The patient presented with a history of hepatocellular carcinoma, BCLC stage A. He underwent a hepatic tumor excision 2 years ago, with repeated recurrence. The recurrence hepatic tumors were treated with RFA and TACE. However, following computed tomography scan 3 months ago found suspected right adrenal metastasis. The doctor suggested him to receive a laparoscopic right adrenalectomy. The patient is 167 cm tall and weighed 67 kg. His physical examination revealed clear breathing sounds to auscultation and regular heart rate with normal resting blood pressure (112/64 mmHg).
The patient's preoperative stats in the operation room were: blood pressure 141/69 mmHg; pulse 86/min; SaO 2 % on room air 99%. The patient was preoxygenated with 5L of oxygen via face mask, and underwent induction with 100 μg of fentanyl, 0.2 mg of glycopyrolate, 250 mg of Citosol, 30mg of propofol and 10mg of cisatracurium IV. Ventilation was set at a tidal volume of 550 ml at the rate of 10-12/min and with FiO 2 40-60%. Desflurane was administered to maintain a Minimum Alveolar Concentration (MAC) of 1.0 to 1.2 in combination with 0.5L of oxygen and 0.5L of air. After anesthesia, an arterial catheter was inserted into the patient to continuously monitor the patient's arterial pressure.
The patient was positioned in a left lateral recumbent position. An additional 50μg fentanyl was administered while the Veress needle was introduced. During the surgery, we began with a high flow of CO 2 intra-peritoneal insufflation to achieve an intra-abdominal pressure of 15 mmHg.
During the initial 75 minutes, the patient's vital signs were stable and the procedure was performed smoothly. However, when attempting to remove the adrenal gland and adrenal tumor which was tightly adhered to the IVC, the surgeon accidently nicked the large inferior vena cava which resulted in a large IVC tear of about 2×2 cm. The bleeding from the torn IVC was temporary compressed by the guaze. However, we suspect that a massive amount of CO 2 entered the vessel via the IVC broken hole. Taking into account the difficulty of performing exploratory laparotomy due to previous hepatic tumor excision related intra-abdominal adhesion, the surgeon tried to sow the torn IVC by the use of laparoscope and to control the bleeding via carbon dioxide (CO 2 ) insufflation pressure. However, we suspect that a massive amount of CO 2 entered the vessel via the IVC broken hole while the surgeon sow the torn IVC. They spent about 60 minutes on repairing the IVC tear. Blood pressure was stable during the IVC tear insult but dropped slightly after the torn IVC repaired. Systolic blood pressure during the insult was kept at 100~130 mmHg, pulse 70~90/min, airway pressure was maintained at 35mmHg which may be induced by the laparoscope and SPO2% was 100%. However, the end tidal CO 2 gradually elevated from 35 mmHg to 53 mmHg. The ventilator respiratory rate was then set higher to wash out CO 2 .
We then inserted TEE to evaluate the severity of the CO 2 embolism while the surgeon tried to fix the IVC tear. Through the TEE image, we noticed that the right atrium and right ventricle were totally white out and full of air ( Figure 1). About 20 minutes after IVC tear, the TEE image also revealed that the left heart also had scattered air embolisms ( Figure 2). We then inserted a CVC via the right internal jugular vein, to attempt to drain the CO 2 through the CVC, which failed. Sewing the fragile IVC tissue proved difficult, so we consulted cardiac surgeons who were then able to repair the torn IVC. The total massive CO 2 influx time was about 60 minutes with continuous 15 mmHg CO 2 insufflation pressure. And the total blood loss was about 1250 ml.
The drop-in blood pressure caused by the IVC tear was restored by low dose norepinephrine intermittent bolus (5mcg) and blood transfusion. In total, we infused 2u of pRBC, 2u of whole blood and 12 u of platelet. The blood gas 1hr after the insult showed respiratory acidosis with P H 7.01, PCO 2 129%, PO 2 75% (FiO 2 86%), SaO 2 83%. Pulmonary CO 2 embolism induced hypercapnia and hypoxemia was under impressed. We suspected pulmonary edema because of the presence of pink bubbled sputum. SPO 2 % was also gradually decreased to 93%. The blood gas 3 hours after the insult showed PH 7.1, PCO 2 82.9%, PO2 186% (FiO 2 85%), SaO 2 99%. Post-surgery, the patient was transferred to the ICU with mechanical ventilation to receive further care.
During the ICU admission, the ECG showed S1Q3 without ST-T change. Acute respiratory distress syndrome (ARDS) was also diagnosed via chest radiography and clinical presentation. Low dose norepinephrine and dopamine continuous infusion were administered for SIRS reaction but were then gradually tapered off in 2 days. Neurological examinations showed mild weakness with muscle power of 4+ in the patient's left limbs during the first day after the surgery but these symptoms dissipated by the second day after the operation. By the second day the patient was under a stable enough condition to be extubated and transferred out of the ICU into the general ward by the fifth day. He was discharged by the twelfth day after the surgery without any residual complication.

Discussion
Carbon dioxide is the most widely used insufflation gas. Most serious cases of CO 2 embolism reported in the literature occurs during the beginning of the procedure, usually due to the misplacement of the Veress needle either directly into a vein or parenchymal organ4. Lesser amounts of CO 2 may also enter circulation through openings in injured vessels, as seen in our case, which explains the late onset of the carbon dioxide embolisms [4].  Rapid entrainment of a large volume of gas can lead to the formation of large emboli, which may lodge in a large central vessel and potentially lead to cardiovascular collapse [5]. In our case, the surgeon was facing the dilemma of whether to transform into open surgery. Considering previous hepatic tumor excision related intra-abdominal adhesion, exploratory laparotomy to approach the bleeder would be also difficult [6]. Instead, we tried an adventurous method: we controlled the IVC bleeding via the carbon dioxide (CO 2 ) insufflation pressure, at the same time, the surgeon sow the torn IVC by the use of laparoscope. A necessary condition for the development of gas embolism is the presence of an open vein with a lower pressure in the vein than the surrounding pressure [7]. Nevertheless, it is because the little higher pressure of the CO 2 gas insufflation, the bleeding rate was not so fast to be life-threatening. But the pressure of CO 2 gas insufflation can't be too high to make fetal amount of CO 2 embolism.
Fortunately, our patient did not develop cardiovascular collapse even encounter this massive CO 2 embolism. It may because the slow entrainment of the CO 2 and the high solubility of CO 2 (20:1 to oxygen and 25:1 to nitrogen) [8]. Mayer, et al. described a mortality of 60% at a continuous intravenous CO 2 infusion rate of 1.2 mL/kg/min, which is equivalent to a rate of 72 mL/min for a 60 kg person. That volume is only 5% of the volume of carbon dioxide that may be infused into a vein, intentionally cannulated by a Veres needle, in one minute at a low-flow rate [9]. Our effort to keep the balance of IVC venous pressure and surrounding CO 2 insufflation pressure during the episode may prevent a massive fetal CO 2 embolism and a fetal bleeding rate.
About 20 minutes after IVC tear, the TEE image also revealed scattered air embolisms in the left heart. The foramen ovale didn't open on the TEE image. We suspect that the left heart gas emboli may have been caused from pulmonary capillary diffusion when large amounts of air entered the vena cava. Post-surgery, the patient developed neurologic signs of left limbs weakness but was able to rapidly recover in 2 days. The rapid improvement maybe attributed to the high solubility of CO 2 .
Treatment of CO 2 embolisms consists of the immediate cessation of insufflation and the release of the pneumoperitoneum, steep headdown and left lateral decubitus position. And the discontinuation of nitrous oxide and to ventilate with 100% oxygen. In addition, adequate hydration and the placement of a CVC to aspirate CO 2 is required. Our patient was already in the head-down and left lateral decubitus position as it was required for surgery. However, the insufflation of CO 2 didn't cease in our patient, because the surgeon needed to control the bleeding through applying pneumoperitoneum pressure.
In conclusion, our patient's prognosis was good despite having gone through a massive CO 2 embolism event. The patient's relatively stable vital signs and possibly intra-abdominal adhesion were the reasons why we did not transfer the laparoscope to exploratory laparotomy nor immediately cease CO 2 insufflation. In addition, the rapid improvement from the neurological symptoms maybe attribute to the high solubility of CO 2 . Hence, we hypothesize that the decision of whether to cease the pneumoperitoneum immediately when CO 2 embolisms occur, may depend on the patient's hemodynamics. In admission, it is important to find the balance between venous and CO 2 insufflation pressure to avoid massive bleeding or fetal amount of CO 2 embolisms. And that the neurological tolerance of CO 2 emboli may be much higher than that of other gases.