Low-flow extracorporeal carbon dioxide removal to enable reduced airway pressures in a ventilated patient with a persistent air leak

16 May 2014 User Reports

Pr. Gilles Capellier and Dr. Loïc Barrot
Critical Care and Emergency Department, Centre Hospitalier Régional Universitaire Hôpital Jean Minjoz (Besançon)

This report describes the case of a 45 year old male who was admitted to a peripheral hospital in respiratory distress, where he was diagnosed with a tuberculosis infection, pulmonary embolism (PE), and pneumothorax. The patient had low body weight, with a BMI of 16, and a past medical history of COPD, depression, and poor living conditions. A regime of antibiotics was started for the tuberculosis infection, and anticoagulation was initiated to address the PE. After two days, he was transferred to the ICU where he was intubated due to low oxygen saturation and tachypnea.

Case Description

After 16 days at the peripheral hospital, the patient was transferred to our critical care unit due to his complex respiratory problem and worsening hemodynamic status. On arrival, he was neurologically sedated and paralyzed, on mechanical ventilation, and receiving norepinephrine at 1.6 mg/hr. The primary challenge was extensive subcutaneous emphysema from head to abdomen and respiratory failure with bilateral wheezing and crackles on both sides. Our examination confirmed mycobacterium pneumonia with necrosis and bullae in both lungs. Consequent to a major bronchial air leak related to the bullae, a right side chest tube was inserted, but this did not result in any improvement in respiratory status. Over the next two days, the air leak did not improve and ventilation using volume assist control could not be implemented safely, requiring high airway pressures of 35–40 mmHg. While the oxygen levels were normal, the patient was also hypercapnic, with a pH of 7.23 and a PaCO2 of 15.53 kPa (116 mmHg). His respiratory status after two days in our unit is summarized in the first column of data in Table 1.

The primary concern was that if the airway pressures and respiratory rate could not be reduced, the air leak would not resolve or might even worsen. At this point, alternative measures were discussed including both extracorporeal membrane oxygenation (ECMO) and low-flow extracorporeal carbon dioxide removal (ECCO2R) with the Hemolung Respiratory Assist System (RAS). The patient was not hypoxemic, and jugular cannulation would have been difficult because of very severe subcutaneous emphysema of the neck, so ECMO was ruled out in favor of the Hemolung RAS, which could be used with a single, dual-lumen catheter in the femoral vein. It was also thought that improving pH and decreasing PaCO2 with a less invasive therapy could improve the patient’s hemodynamic status and ability to oxygenate. This would further enable a decrease in tidal volume and an increase in PEEP to levels which could achieve a recruitment effect in the lung.

Therapy with the Hemolung RAS was initiated on the patient’s third day in our unit. Using the Hemolung 15.5 Fr femoral catheter, the system ran at a blood flow of 450 mL/min which removed CO2 at a rate of 105–110 mL/min (measured and displayed by the Hemolung controller). After initiation of ECCO2R, the patient’s respiratory rate was reduced from 30 to 14 which reduced peak airway pressure from 37 to 27 cmH2O. The goal of Hemolung therapy was to decrease the airway pressures, and this goal was achieved. In addition, hypercapnia was reduced and pH normalized.

Paralytics and sedation were weaned over the next two days, according to patient comfort, in an effort to limit the airway leak. After the patient regained consciousness, the ventilation mode was switched to pressure support. Therapy continued uneventfully for over a week, but despite all efforts, the airway leak did not improve, nor did the underlying tuberculosis infection. On day 10, atelectasis of the left lung occurred, after which conditions continued to worsen. On day 14, it was decided to limit treatment and the Hemolung RAS was discontinued. The patient ultimately did not survive.

Table 1: Ventilatory status and arterial blood gases during first 6 days of Hemolung RAS therapy
Time Relative to Hemolung Start -1 Day +1 Hour Day 2 Day 3 Day 4 Day 5 Day 6
FiO2 (%) 65 70 65 55 50 65 75
Respiratory Rate 30 14 14 18 38 37 30
Tidal Volume (mL) 340 310 320 340 270 190 190
Minute Ventilation (L) 10.2 4.4 4.3 6.0 9.8 7.4 6.0
PEEP (cmH2O) 6 5 5 8 10 7 10
Peak Airway Pressure (cmH2O) 38 27 28 32
Arterial Blood Gases
pH 7.23 7.30 7.36 7.47 7.39 7.42 7.33
PaCO2 (mmHg) 116 112 101 74 88 81 101
Bicarbonates (mmol/L) 48 54 55 53 52 51 52
PaO2 (mmHg) 76 59 81 59 73 58 72
SaO2 (%) 94.4 88.8 95.8 89.7 92.9 93.3 91.1


Discussion with Dr. Capellier

Q: Do you feel that there were other alternatives which may have resulted in a better outcome?

A: No. This patient was very ill with a complex respiratory problem and a refractory tuberculosis infection in his lungs. Use of lowflow ECCO2R was successful in allowing us to reduce the airway pressures, tidal volumes and respiratory rate to give this patient the best opportunity to recover. It also allowed the patient to regain consciousness and to communicate, which we hoped would help to resolve the air leak. ECMO was considered, but the patient was not hypoxemic so this level of invasiveness was not necessary, and jugular cannulation would have been risky due to severe subcutaneous emphysema in his neck. Unfortunately we were unable to save this patient, but we feel that we were able to give him an improved opportunity for survival with the Hemolung RAS.

Q: Were there any complications associated with using low-flow ECCO2R?

A: No, there were no complications from low-flow ECCO2R with this patient. Anticoagulation was maintained at the same level already initiated for the pulmonary embolism. We were able to run continuously on the same Hemolung cartridge and catheter for the full duration of therapy (14 days). Interestingly, we did not need to provide any units of platelets for thrombocytopenia, which in our experience can be associated with extracorporeal therapy and with the antibiotics we were using to fight his infections. The platelet count dropped over the first few days of Hemolung therapy from 500,000 to 230,000/μL, as is typical with extracorporeal therapy, but remained above 100,000/μL.

Q: What has been the perception of the Hemolung RAS in your unit?

A: We have found it to be easy to implement. We work as a team with our cardiovascular surgeons and perfusionists to implement ECMO. Although we keep our team informed when we use the Hemolung RAS, we are able to implement and operate the Hemolung on our own.

About the Author

Pr. Capellier is a Professor in the Critical Care and Emergency Department at Centre Hospitalier Régional Universitaire Hôpital Jean Minjoz (Besançon), and is head of the medical ICU. He received no compensation in association with this case report and has no conflicts of interest to disclose. Pr. Capellier can be reached via email at samu@chu-besancon.fr.

About the Hemolung RAS

The Hemolung RAS from ALung Technologies provides Respiratory Dialysis®, a simple, minimally-invasive form of extracorporeal carbon dioxide removal (ECCO2R). The system utilizes patented technology to provide highly efficient CO2 removal at dialysis-like blood flow rates which are achieved through a single 15.5 Fr venous catheter. For more information, please visit http://www.alung.com/products/hemolung-ras/

View the PDFpdf_icon