Safe doubling of ventilator capacity: a last resort proposal for last resorts

The best way to ventilate two patients on a single ventilator is simply not to do it: The Authors

In light of the COVID-19 pandemic, this common-sense approach was recently clarified in a SCCM-ASA-AARC-AACN-ASPF-CHEST consensus statement on the Society of Critical Care Medicine (SCCM) website [1]: ‘We recommend that clinicians do not attempt to ventilate more than one patient with a single ventilator while any clinically proven, safe, and reliable therapy remains available (ie, in a dire, temporary emergency)’ [1].

The current situation in several European countries and states in the USA is a ‘dire emergency’. Physicians have been, or may be, asked to make difficult choices in the face of ventilator shortages [2]. Nevertheless, if you are faced with a decision to ventilate two patients at once, or deny care to one, we believe we can propose the next best way.

The SCCM recommendation [1] addresses a series of popular Internet concepts with multiple patients breathing in-parallel [3, 4]. In-parallel is a critical point, as inspiration and expiration all take place at the same time, so there is thus no change to respiratory rate (RR) and tidal volume or driving pressure are adjusted for the number of patients. All of these add risk over over-/under-ventilating patients and causing harm [1].

Instead, we recommend a multiplex in-series breathing approach to double (2-for-1) the patients on a ventilator (Fig. 1). In-series breathing means only 1 circuit volume (split between patients) is active at a time, but each patient’s inspiratory effort is singular. This approach addresses the limitations of shared, in-parallel breathing in the SCCM statement, where Table 1 addresses in detail each consensus statement concern [1].

Fig. 1

Top: Schematic of a proposed in-series breathing circuit for two patients using an active inspiration valve. Bottom: Resulting ventilation waveforms and active (filled in) and in-active (not filed) inspiratory and expiratory circuit lines at any given 2-s period for 2 × 4-s breaths, one by each patient. The ventilator will display the given patient data in each breath. Patients are colour-coded for clarity and show how end-expiration of one patient overlaps inspiration and initial expiration of a second patient, although using different parts of the circuit

Table 1 Specific reasons and multiplex series approach mitigation: The SCCM/ASA/AARC/AACN/ASPF/CHEST consensus statement [1] strongly recommends against parallel breathing and ventilation of patients; this poses a wide range of valid criticisms. We address these criticisms in terms of the serial ventilation approach/concept presented VC volume-controlled, PC pressure-controlled ventilation, PEEP positive end-expiratory pressure, PIP peak inspiratory pressure, Vt tidal volume

Instead of the same RR and higher tidal volume or driving pressure, in-series breathing doubles the RR and keeps the other ventilator settings the same. One patient breathes in, while the other breathes out. With typical I:E ratios around 1:3, there is also shared expiration time when neither is breathing in (Fig. 1). As with other proposals [3], driving pressure can be modified by added resistors in the inspiratory circuit, and PEEP can be customised with in-line expiratory PEEP valves (Table 1).

This in-series approach ensures breath-by-breath ventilation parameters of each patient are displayed to ensure monitoring and safety are maintained. Staff can thus be assured more (or less) compliant patients are not over- (under-) ventilated, both of which could result in harm if ventilated in-parallel, where monitoring individual patients is not possible. Finally, one-way expiratory valves and filters prevent rebreathing and cross-contamination.

Clinically, we would suggest pressure-control modes, where driving pressures are easily customised per-patient with resistors, and are more commonly used, currently. This choice allows customised PEEP and driving pressure for each patient effectively as if they were ventilated separately.

This setup requires an active valve to switch between patients (Fig. 1), comprising a pressure sensor at the single end of a y-splitter on the inspiration circuit, and two active valves at the outlet. It uses measured inspiratory pressure to switch the inspiratory circuit from one patient to the other after inspiration (as pressure drops). The active y-splitter valve thus allows flow down only one inspiratory path per ventilator-supplied breath (at 2xRR). The components, sensors, and computation are low-cost and easily 3D printable by hospital bioengineers or others.

However, nothing is perfect. This approach is not suited for spontaneous, triggered breathing, which cannot be synchronised nor limited. In addition, it cannot be used if I:E < 1, or if a patient’s respiratory mechanics are such they will receive inadequate minute ventilation within the time for each allocated breath.