Supplementary Guidance for the Diagnosis of Death using Neurological Criteria (ECMO)

This supplementary guidance is endorsed by the Faculty of Intensive Care Medicine, the Intensive Care Society Standards Committee and the NHS England Severe Respiratory Failure Network for adults and children older than 16 years. It should be used in conjunction with the Academy of Medical Royal Colleges (AoMRC 2008) A Code of Practice for the Diagnosis and Confirmation of Death and the forms for the Diagnosis of Death using Neurological Criteria endorsed by the Faculty of Intensive Care Medicine, Intensive Care Society and the National Organ Donation Committee.

Forms available: https://www.ficm.ac.uk/standards-and-guidelines/access-standards-and-guidelines

Challenges of diagnosing death using neurological criteria on ECMO

Pharmacokinetic
It is known that ECMO circuits sequester drugs, both by adsorption and absorption, leading to the potential for altered kinetics, particularly clearance. It is also possible that some drugs including sedatives and potentially muscle relaxants may wash out of the membrane into the patient after cessation of administration, extending their effective half-life. Measures to exclude reversible causes of coma and apnoea may therefore be unreliable. Where there is concern that drugs may be contributing to unconsciousness, apnoea and loss of brainstem reflexes, specific drug levels should be performed, a reversal agent administered and/or train of four performed as deemed appropriate. If doubt persists, diagnosing death using neurological criteria should be abandoned and consideration given to undertaking ancillary investigations.

Apnoea Testing
Extracorporeal clearance of CO2 must be taken into account when performing apnoea testing on ECMO, and sweep gas flow titrated carefully to avoid large-scale, acute changes in cerebral PaCO2. In venovenous (VV) ECMO, the cerebral PaCO2 is identical to the PaCO2 at all peripheral arterial blood gas sampling sites. Modification of the PaCO2 through sweep gas changes, and subsequent analysis of the peripheral PaCO2 allows apnoea testing to proceed reliably. In peripheral venoarterial (VA) or hybrid VVA ECMO, the PaCO2 in the cerebral circulation may differ from the PaCO2 in some systemic arterial sampling sites due to the variable location of a mixing point between native pulmonary-cardiac blood flow and ECMO circuit blood flow in the aorta. Consequently, to provide absolute reassurance that blood gas measurements reflect the cerebral PaCO2, multiple sites (including post-membrane and systemic arterial sites furthest from the ECMO return flow) must be sampled and all PaCO2 levels taken into account when assessing the absence of neurological responses. The highest pH and lowest PaCO2 must be used.
The guidance below sets out a protocol for apnoea testing in patients on VV and VA-ECMO. Where brainstem death is suspected but the steps to diagnose using neurological criteria on ECMO cannot be reliably completed, ancillary testing is required to confirm the diagnosis. In this circumstance it is recommended that consultant neurology and/or neurointensive care advice is sought. It is important to appreciate that MRI cannot be undertaken on ECMO, however CT, CT angiography and electrophysiological tests (including somatosensory evoked potentials) are all possible.
With the exception of the pharmacokinetic and apnoea testing challenges above, testing for absence of brain-stem reflexes on ECMO is otherwise undertaken in accordance with standard practice.

Apnoea testing on ECMO

AoMRC (2008) Code of Practice apnoea test requires:

1. Starting apnoea test PaCO2 ≥ 6.0 kPa and starting pH <7.4 or [H+] >40 nmol/L.
   In patients with chronic CO2 retention, or those who have received intravenous bicarbonate,
   confirmation that the starting PaCO2 > 6.5 kPa and the pH < 7.4 / [H+] > 40 nmol/L.
2. There should be no spontaneous respiration within a minimum of 5 (five) minutes
   following disconnection from the ventilator.
3. Confirmation that the PaCO2 has increased from the starting level by more than 0.5 kPa.
4. Oxygenation and cardiovascular stability should be maintained through each apnoea test.
   These steps are possible on ECMO, although observation of spontaneous respiration may be
   challenging in a patient with absent native pulmonary function and total parenchymal consolidation.
   End-tidal CO2 may not be measurable in these circumstances although respiratory effort may be
   visualised.
   Apnoea testing on ECMO requires maintenance of the extracorporeal blood flow with a membrane
   sweep gas FiO2 100% in order to preserve systemic oxygenation. The sweep gas flow rate can then be
   manipulated to achieve the starting PaCO2 and pH requirements. During testing, it is important to
   titrate the sweep gas flow downwards in small decrements to prevent rapid changes in PaCO2 and
   potential precipitation of further neurological injury.

Steps in apnoea testing on ECMO

Testing can occur in any ECMO circuit configuration (VV, VA or hybrid VVA). For VV ECMO, systemic
arterial blood gases measurements at any site will be identical. For VA or hybrid VVA ECMO both right
upper limb and femoral blood samples need to be performed in order to assess variability in the
systemic blood gas measurements. All sites measured need to fulfil the criteria outlined below.
The following steps should be undertaken:

1. Ensure sweep gas FiO2 is 100%.
2. Temporarily increase the sweep gas flow rate to maximum and then reduce again to ‘sigh’ the
   extracorporeal membrane and eliminate condensation.
3. Adjust ECMO blood flow to achieve PaO2 > 10 kPa at all times at all sampling sites. Note that the
   blood flow may need to be increased above previously established baseline rates.
4. Reduce sweep gas flow rate by 0.5 L/minute every 5 minutes. Perform arterial blood gas analysis
   at each 5-minute point until the PaCO2 is ≥ 6.0 kPa and starting pH <7.4 or [H+] >40 nmol/L. In
   patients with an elevated bicarbonate, the PaCO2 may need to be further titrated to achieve the
   desired pH. Do not reduce the sweep gas flow rate below 0.5 L/min.
5. Suction the patient’s airway to ensure it is clear of obstruction, secretions or soiling.
6. Undertake a ventilator recruitment manoeuvre to optimise pulmonary gas exchange.
7. Disconnect the ventilator circuit and attach a Water’s (Mapleson C) circuit with inline ETCO2
   monitoring and valve adjusted to give approximately 10 cmH2O CPAP to the lungs.
8. Reduce the sweep gas flow rate further by 0.5 L/minute every 5 minutes and perform an arterial
   blood gas at each 5-minute point until all measured PaCO2 values have risen by at least 0.5kPa
   above the starting level. Do not reduce the sweep gas flow rate below 0.5L/min.
9. During the period of disconnection, observe for any respiratory effort – ETCO2, chest excursion
   and/or movement of the circuit reservoir bag. This must be for a minimum of 5 (five) minutes.
10. Abandon test if there is significant deoxygenation or cardiovascular instability, if respiratory
    effort is observed or if adequate rise in PaCO2 is not achieved.