Alternative approaches to the problem of COVID-19 related respiratory failure

The case for a million bicycles

APRIL 1, 2020

Dhruv Joshi and Dileep Raman

In the face of the COVID-19 pandemic, in addition to lockdowns, curfews and economic upheaval, most planning centers and living rooms have been debating the shortage of ventilators and the need for critical care in the days to come. We have seen heartbreaking scenes from Italy, Iran and New York City where many lives have been lost and hospitals are full and ventilators are unavailable. It is important to delve on some of these issues and develop rational approaches and solutions, in the Indian context, to an impending surge of critically ill patients.    

Ventilators: Advantages and pitfalls

The modern ventilator is an advanced medical device that can help carry out the complex task of allowing a patient to survive respiratory illnesses in various settings. ICU specialist doctors, also known as intensivists, use it as a powerful tool and a vital ally to manage ICU patients. 

That said, intensivists understand more than anyone else, that the breathing machine or the ventilator can be a double edged sword. The prime function of the ventilator is to sustain breathing for a critically ill patient until the underlying disease process improves (either spontaneously or with treatment). The ventilator is not a cure or a treatment by itself. It merely buys time so that the therapies in place, or the body’s intrinsic healing mechanisms have a chance to act, allowing the body or the lungs to heal. In severe cases, the machine may have to deliver a concentrated amount of oxygen (up to 100% O2) in contrast to 21% oxygen that is present in ambient air.

When we breathe in normally, we create a negative pressure in our chest cavity using the muscles in our chest wall and the diaphragm to suck in air into the lungs. This is known as negative pressure ventilation. A ventilator on the other hand, pushes air into the lungs and thus a positive pressure ventilation (PPV) system is created. To be on a ventilator, the patient must also have a plastic breathing tube inserted into their mouth/airway to allow this to happen making this an invasive therapy. 

COVID-19 related lung injury is typically classified as a disease syndrome called Acute Respiratory Distress Syndrome or ARDS. The lining of the lung that participates in gas exchange is adversely affected in this disease. Contrast this with say, polio. In polio the underlying lung is often essentially normal but the mechanical pump, the muscles of breathing are weak. In muscular disorders simple ventilators, even those not using supplemental oxygen would suffice. However, in ARDS, where the lung lining is damaged, one would need supplemental oxygen in addition to mechanical ventilation to help a patient get sufficient oxygen. 

Severely ill COVID-19 patients will need ventilators, and India will still need to increase its ventilator capacity. What is also coming through from international data is that a high percentage of these patients will unfortunately succumb to their illness (1)(2). While mobilizing resources in terms of equipment and personnel, it would be prudent to consider easier strategies, which might allow us to save potentially hundreds of patients that do not meet the evidence based threshold for ventilator use. One such option that might give us the best “bang for buck” may come from large-scale production of high-flow devices and ensuring sufficient oxygen supply.

Primum non nocere or first, do no harm

While ventilators offer clear benefits, positive pressure ventilation (PPV) is also known to cause adverse effects in experimental and clinical models. The injury that it inevitably causes to the lungs is called Ventilator Induced Lung Injury (VILI). VILI is something that has been known for decades (3)(4). Intensivists spend years in training to learn and understand various modalities that minimize VILI and optimize the chance of lung recovery on a ventilator. In a disease such as ARDS, which afflicts a majority of the critically ill COVID-19 patients, the lungs are stiff and susceptible to injury from excessive pressures generated by ventilators. 

The prime goals of mechanical ventilation are safety, comfort and eventual liberation from the device. Given these objectives, when low cost ventilators with short development cycles reach the market (i.e. the patient), it raises concern. It is relatively easy to pressurize a breathing circuit but it takes years to perfect the safety and comfort systems that optimize patient outcomes (5). Similar concerns exist with using devices (‘splitters’) that enable ventilating more than one patient on a machine designed for ONLY one patient (6). Skill needed to administer advanced mechanical ventilation is also in short supply. If untested devices are used by untrained workers in experimental ways, the risks to patients could be significant. Once you combine this with the fact that COVID-19 patients who require mechanical ventilation are already at high risk of dying (2); the only conclusion is that these patients must be put on a validated and reliable ventilator and treated by trained personnel, as far as possible, to maximize the chance of recovery. 

Furthermore, ventilators require significant clinical effort by the nursing team to reduce the chances of causing harm. Ventilator Associated Pneumonia (VAP) is a common complication that occurs in about 30% of mechanically ventilated patients (7). Incidence of VAP is reduced by high quality nursing care and strict protocols, expertise that is not easy to impart to the vast majority of understaffed and underskilled ICUs in India. It would be unfortunate for large numbers of patients to be put on ventilators only for abnormally large numbers of them to succumb to VAP later. This story is not uncommon in India. 

‘So what are my options, Doc?’

In ideal situations, with unlimited ventilators and skilled hands there is no doubt that early invasive mechanical ventilation with high concentrations of oxygen and application of advanced techniques of ARDS care is the way to go. In the absence of this, the other option often used is a trial of non-invasive ventilation (NIV) that is achieved by way of using a mask to deliver air under pressure (2). While NIV is easier to use compared to ventilators, there is still a chance of harm to the patient and it is not easy to use in areas with reduced or unskilled personnel. NIV has not been found to be very useful in COVID-19 patients with lung injury and has been discouraged (8). Face masks are simple and commonplace but are not capable of delivering high flows or a therapeutic range of pressures to help the patient with their work of breathing.

Another attractive modality of delivering oxygen to critically ill patients that can be attempted prior to them requiring invasive ventilators is to deliver oxygen via High Flow Nasal Cannulas (HFNCs). HFNCs require large amounts of oxygen to be available. If supplies of oxygen can be maintained they are easier and cheaper to deploy and place on patients and have been recommended for use in COVID-19 patients with low oxygen levels (8).

Invasive Ventilation with a Mechanical Ventilator

Source: By Calleamanecer - Own work, CC BY-SA 3.0 

Non-Invasive Ventilation (NIV)

Source: By James Heilman, MD - Own work, CC BY-SA 4.0

High Flow Nasal Cannula

Source: http://rc.rcjournal.com/content/58/1/98

Face Mask

Source: By James Heilman, MD - Own work, CC BY-SA 4.0

Putting it all together

When dealing with respiratory failure due to hypoxemia (low oxygen levels in the blood) at a pandemic scale, there are four ways of delivering oxygen or respiratory assistance to a patient. These are analyzed in the table below. In order to evaluate all these therapies we made a few assumptions. Firstly, we decided to be patient centric with respect to risk and benefit. Secondly, we assumed that the treatment option should be easy to mass produce or source and that oxygen is plentiful. Lastly, we kept in mind the heterogeneous levels of skill in India; for example, untrained staff with a complex device are likely either use it ineffectively or worse cause inadvertent harm. 

Most COVID-19 patients who are admitted to hospitals have mild to moderate ARDS. HFNCs have the potential to treat a majority of these patients with a level of respiratory support that is evidence based and deployable. In an imperfect world this may well be a suitable solution for India. HFNCs are more comfortable to wear for patients, have greater benefit than a simple mask and have significantly lower risk when compared with NIV or invasive ventilation. At least one study has shown that it may even reduce the risk of mechanical ventilation altogether (9). These benefits cannot be ignored in India.

The other, less talked about type of ventilation

Areas where COVID-19 patients are treated need to be well ventilated so that the air is filtered of harmful and infectious aerosols. This is usually accomplished by artificial indoor ventilation that can clear the entire volume of the room at least 12 times per hour, the minimum requirement as per WHO guidelines (10). 

However, and quite alarmingly, most Indian ICUs only have a standard “split” air conditioner that does not have HEPA (high efficiency particulate air) filters. This is a disease transmission catastrophe waiting to happen.

The process of putting a patient on a ventilator generates infectious aerosols as does using NIV and HFNCs. It is estimated that India may not have enough personal protective equipment (PPE) for our medical workers. A lack of PPE coupled with recirculated and non filtered air would subject them to extraordinarily high amounts of aerosols, increasing their chances of getting infected. This is an underrecognized risk and compromises our finite healthcare resource.  

Source: Atkinson, et al., Natural Ventilation for Infection Control in Health-Care Settings, WHO Publication/Guidelines

The amount of aerosol generated from HFNCs is low (11) and not likely to be harmful if ventilation is adequate and protective measures are taken. Measures, such as putting a surgical mask on the HFNC while in use and making sure that the healthcare provider is wearing proper PPE should suffice. Negative pressure rooms that prevent aerosol dispersion are strongly recommended but are likely to be a pipe dream in India for now. Interestingly, the WHO (10) does suggest that in the right setting natural ventilation can have air exchange rates that meet or exceed the standards set for indoor air quality. 

Using natural ventilation in COVID-19 treatment units is pragmatic and safe when negative pressure rooms or sufficient artificial ventilation is not feasible. Given our temperate climate, India has the distinct advantage of being in a unique position to leverage this. 

Source: Atkinson, et al., Natural Ventilation for Infection Control in Health-Care Settings, WHO Publication/Guidelines

Keeping these facts in mind and putting them in the context of the Indian healthcare setting, the combination of HFNCs and naturally ventilated “open-air” ICUs is a pragmatic approach to:

• maximise benefit to the salvageable patients, and 
• optimize safety of healthcare workers.

Conclusions

We should redouble our efforts to manufacture reliable ventilators and upgrade our facilities over the next few months. We laud the ingenuity and industriousness of our engineers and the endeavors of government and private enterprises in trying to make this happen. Our premier hospitals with well qualified doctors, can treat patients as well and as safely as anywhere in the world. However, based on projections put forth by ICMR and other bodies on the projected rates of infection, these hospitals and their personnel are likely to exhaust their capacity. As such, even if India manages to source more ventilators, it is unlikely that we will have sufficient trained personnel and facilities that can use them safely. We require a simultaneous focus on more easily deployable and safe solutions that might address the need of the hour. One of the fundamental principles in medicine says, less is more. This may yet be true today.

______________

Dhruv Joshi and Dileep Raman are Pulmonary and Critical Care specialists who trained in the US and now live in Bengaluru. They run a company that delivers critical care services remotely. 

______________

REFERENCES

(1) Weiss, et al., Clinical course and mortality risk of severe COVID-19, The Lancet 

(2) Meng, et al., Intubation and Ventilation amid the COVID-19 Outbreak: Wuhan’s Experience, ASA Publications 

(3) Slutsky, et al., Ventilator-Induced Lung Injury, The New England Journal of Medicine 

(4) Loza, et al., Ventilator-Induced Lung Injury (VILI) in Acute Respiratory Distress Syndrome (ARDS): Volutrauma and Molecular Effects, NCBI

(5) The U.S. Tried to Build a New Fleet of Ventilators. The Mission Failed. The New York Times

(6) Joint Statement on Multiple Patients Per Ventilator, American Association for Respiratory Care

(7) Chastre, et al., Ventilator-associated Pneumonia, American Journal of Respiratory and Critical Care Medicine

(8) SARS CoV-2, Guidance Document, American Association for Respiratory Care

(9) Frat, et al., High-Flow Oxygen through Nasal Cannula in Acute Hypoxemic Respiratory Failure, NCBI

 (10) Atkinson, et al., Natural Ventilation for Infection Control in Health-Care Settings, WHO Publication/Guidelines

(11) Hui, et al., Exhaled air dispersion during high-flow nasal cannula therapy versus CPAP via different masks, European Respiratory Journal