A 42 year old man presented with a week-long history of increasing shortness of breath, cough  (productive of purulent sputum) and fevers on a background of significant chronic lung disease. He had a ten year history of interstitial lung disease and was on the waiting list for a lung transplant. He used oxygen at a rate of 2 litres per minute at home, 24 hours a day. His usual exercise tolerance of 200 metres had been significantly reduced for the past week. His regular medications included seretide and salbutamol inhalers, lansoprazole, azathioprine, prednisolone alendronate.

On arrival in hospital, he was alert and orientated. He had a patent airway, but was tachypnoeic (rate of 50/minute) using his respiratory accessory muscles and a tracheal tug was evident. An arterial blood gas revealed type two respiratory failure (pH 7.26; pO₂ 8.14, pCO₂ 7.52 on 15 liters/min of face mask oxygen). He was hypotensive (80/40mmHg) and tachycardic (130/minute, sinus rhythm). A pyrexia of 39.2°C was recorded. Blood results showed normal renal function, a slightly elevated white cell count of 14.

The patient was admitted to the high dependency for close monitoring in view of his history and presentation. He was commenced on treatment for a presumed infection (viral or bacterial) with oseltamivir, co-amoxiclav and clarithromycin and given three “pulsed” doses (750mg) of methylprednisolone. He remained stable for the next twelve hours.

Early the next morning, he became very hypoxic (oxygen saturations less than 50%), bradycardic (<35 beats per minute) and had a brief hypoxic respiratory arrest. He received 1 cycle of cardiopulmonary resuscitation and was intubated. There was subsequently a return of spontaneous circulation.

The next 24 hours involved a period of difficulty with ventilation. His peak airway pressures were very high, despite being paralysed and a low volume/high respiratory rate strategy being employed. He was discussed with a tertiary respiratory centre and it was decided that he should be transferred for insertion of a pumpless arteriovenous interventional lung assist (for extracorporeal carbon dioxide removal) as a bridge prior to lung transplantation. He had formal ultrasound measurement of his femoral arteries. His left common femoral artery was widely patent (AP and transverse diameter of 8-9mm throughout), but the right was only 4-5mm throughout.

In the meantime, his peak airway pressures were consistently between 35 and 40cmH2O, despite tidal volumes of 230ml, 3.8ml/kg). With a rate of 32-35 breaths per minute, his pH was  initially maintained above 7.2, with a pCO₂ of 9-11kPa. Over the course of the next few hours, this became increasingly difficult to achieve. His oxygen requirements did not escalate (an FiO₂ of 0.6 provided a pO₂ of 8-9kPa). When his pCO₂ increased to 15.4kPa and his pH dropped to 7.17, further adjustments were made and the PEEP decreased to 5cmH₂O from 10cmH₂O. His noradrenaline requirements were increasing and with the aid of the cardiac output monitoring, he was cautiously given fluid with a good response.

He was transferred to the centre in which a lung transplant could be performed within hours of the referral. A Novalung device was inserted and he underwent a bilateral lobar lung transplant several days later. He was in hospital for 6 weeks and made a very good long-term recovery. At six months, he was extremely well and was undertaking his activities of daily living completely normally with stable lung function. He even managed to complete an eight mile bike ride.

What is the rationale for extracorporeal lung assist?

Emma Fitzgerald

The case described here gives an example of some of the problems that we encounter with conventional ventilation. His presentation did not fit with the traditional definition of an acute lung injury [1], but his lung compliance was very poor and his lungs very fragile. The low volume strategies were important to not only reduce further damage, but to also prevent sequelae such as pneumothoraces which could have proven fatal with this degree of respiratory failure.

The balance of delivering oxygen and removing sufficient carbon dioxide without causing any further damage to the lungs is sometimes very hard to achieve. There is evidence from the ARDSnet [2] trial that large tidal volumes and high peak pressures are detrimental and actually contributed to the damage that the lung parenchymal tissue undergoes. It showed significant reduction in mortality by utilizing a low-volume ventilatory strategy based on predicted body weight (6 ml kg⁻¹ and peak pressures <30 cm H₂O vs 12 ml kg⁻¹ and peak pressures <50 cm H₂O). It suggested that large volumes are thought to damage any remaining areas of healthy lung  as it will cause over-inflation and that a more conservative ventilatory strategy is associated with a significantly lower level of circulating cytokines, the cause of biotrauma, and distant organ damage.

In recent years, extracorporeal lung assist has become a new therapeutic option in patients suffering from acute respiratory failure. The numbers of patients that have been treated remain relatively low, but are increasing [3]. Respiratory acidosis can become a serious problem during protective ventilation of severe respiratory failure and as in this case, conventional strategies do not always prove effective enough. A pumpless arteriovenous interventional lung assist for extracorporeal carbon dioxide removal has been used increasingly to control critical respiratory situations [4].There is evidence to suggest that the use of devices like the novalung^® will permit the reduction of tidal volumes to 3ml/kg and results in a decrease in severe hypercapnia without alveolar collapse or hypoxaemia and also a reduction in serum levels of IL-6 [5]. This will allow the removal of carbon dioxide without the damage caused by mechanical ventilation with high pressures.

Numerous studies have suggested that pumpless arteriovenous interventional lung assist devices might provide a sufficient rescue measure with severe acute respiratory distress syndrome and persistent hypoxia/hypercapnia. A study of ninety patients [6] in Germany showed a prompt and marked reduction in hypercapnia within two hours of using such a device and stated that it is not only an effective means of eliminating carbon dioxide, it improved oxygenation, had easy handling properties and was of low cost.

---

Lessons Learnt:

Use of a pumpless arteriovenous interventional lung assist device could be considered in patients who show signs of not tolerating the conventional ventilatory techniques with severe lung pathology. Early consideration could limit the damage from barotrauma. It must also be remembered that there are also significant implications of using this technique. A large bore femoral arterial line and a cvc line must be sited and in critically unwell patients this is not an insignificant risk.

---

References:

1. Mackay A, Al-Haddad M. Acute lung injury and acute respiratory distress syndrome. Contin Educ Anaesth Crit Care Pain (2009) 9 (5): 152-156.
2. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med 2000;342:1301-8.
3. Marc-Alexander von Mach, Joachim Kaes, Babatunde Omogbehin, Ingo Sagoschen, Jascha Wiechelt, Kristina Kaiser, Oliver Sauer, Ludwig Sacha Weilemann. An Update on Interventional Lung Assist Devices and Their Role in Acute Respiratory Distress Syndrome. Lung (2006) 184:169–175
4. Muller T et al Extracorporeal pumpless interventional lung assist in clinical practice: determinants of efficacy. Eur Respir J. 2009 Mar;33(3):551-8.
5. Bein T, Zimmermann M, Hergeth K, Ramming M, Rupprecht L, Schlitt HJ, Slutsky AS. Pumpless extracorporeal removal of carbon dioxide combined with ventilation using low volume and high positive end-expiratory pressure in a patient with severe ARDS. Anesthesia, 2009, 64: 195-8
6. Bein T, Weber F, Philipp A. A new pumpless extracorporeal interventional lung assist in critical hypoxemia/hypercapnia. Crit Care Med 2006 May;34(5):1372-7.