To our best knowledge, this study is the first to describe and investigate the prognosis of patients managed in a temporary ICU. Beyond the medical value, this study demonstrates a specific episode of our recent sanitary history. It describes the adaptation of a French ICU team to a massive pandemic to increase ICU beds by threefold, without affecting patient care.
Patients initially admitted to the temporary ICU did not present an increased mortality rate compared with those initially admitted to the classical ICU.
Nevertheless, our results can roughly be generalized to all Temporary ICUs: there has been many different ways to manage Temporary ICUs. We have described here our way to make a temporary ICU work.
Ventilation strategies might have differed between ICU categories and, consequently, MV duration, without sufficient power to relate this effect.
First, as the crisis evolved and the number of infected patients increased, the number of ICU beds decreased, and the demand for resuscitation beds increased. A shift in skills occurred with the implementation of noninvasive ventilation techniques and high-flow oxygen therapy in the classical ward. Although 27% of patients admitted to the classical ICU were not ventilated, only 3% of those admitted to the temporary ICU were not under MV.
Second, ventilator availability in both ICUs differed. Ferré et al. used a transport ventilator without identifying consequences regarding the outcome (13). Because of the lack of intensive care specialized ventilators at the beginning of the crisis, the temporary ICU in our department was equipped with either older model or anaesthesia ventilators. Our team has previously shown that anaesthesia ventilators can be used in these circumstances for 72-hour prolonged ventilation in ICU patients. However, in this preliminary observation, 2 of 16 COVID-19 patients required a ventilator shift for high plateau pressure or hypercapnia (14). Other teams have encountered greater difficulties: Bottoroli reported increased mortality in COVID-19 patients exclusively ventilated with anaesthesia ventilators and reported several incidents, mostly airway obstruction and sudden ventilator failure (15). This finding might be related to the fresh gas flow ranging from 80% of minute ventilation in patients receiving halogenates to 100% of minute ventilation in patients without inhaled aesthetics, while we used 150% of minute ventilation and no inhaled aesthetics (16). In our cohort, no patient had been exclusively ventilated with anaesthesia machines. These devices were considered rescue devices, and they were replaced by ICU ventilators as soon as possible. In ARDS patients with difficult ventilation, less efficient ventilators may have made it more challenging to adjust ventilator parameters and monitor ventilation pressures and thus lung compliance. Consequently, we cannot exclude more frequent ventilator maladjustments and/or deeper sedation or even the use of neuromuscular blockers in some cases.
Although the temporary ICUs were set in common rooms, we did not observe multiresistant bacterial transmission. The COVID-19 patient isolation protocol and strict adherence to hospital hygiene rules helped achieve this outcome.
We did not measure potential side effects that may have crucial consequences on the survivor’s life (17): rate of delirium and posttraumatic stress (18) and occurrence of pressure sores or neurological lesions linked to the absence of specific beds or lower quality prone position material. This analysis would require a dedicated prospective study.
During the COVID-19 first wave, our team doubled its capacity to receive patients requiring intensive care because of the creation of temporary ICUs. In our cohort, 62.7% of patients were initially managed in temporary ICUs. The temporary ICU allowed access to intensive care for all patients requiring it immediately. This benefit could not be modelled in our study. However, during the first pandemic wave, on a national level, population access to ICUs was correlated with the global mortality observed (19,20). Beyond the care provided to each patient, ensuring universal access to diagnostics and treatment during such situations is crucial for a society to survive a health crisis (21).
The mortality observed in our cohort and associated risk factors are consistent with the main studies published during the related period (1,2,22,23).
Temporary ICUs are an operating model developed in a health crisis context and should not represent a long-term strategy to increase regular ICU capacities. Although no loss of opportunity occurred related to managing patients presenting respiratory distress linked to SARS-CoV-2 infection during the first wave, this model certainly has collateral effects that are more difficult to measure.
The potential psychological impacts of the development of temporary ICUs for all the actors in this episode are not established:
Patients were already under significant stress, inherent to acute respiratory distress, and increased by the uncertainty of their fate in the context of an unknown and feared new infectious emerging disease. This situation could have been worsened by admission to an unusually configured ICU ward, which is a common room. The few observant patients admitted to the temporary ICU could observe patients under MV and/or in the prone position, such as patients in MV weaning phases. This situation could have favoured the occurrence of posttraumatic stress states.
Visits from relatives were not allowed in any ward of the hospital. The only exception made was for end of life, where relatives could come and accompany the patient in his or her final moments. Although the health care workers were careful to arrange screens around the patient bed, the temporary ICU configuration in a common room was not appropriate for privacy and relative recollection and made their experience more challenging to endure.
Finally, the primary work of most of the temporary ICU care workers was anaesthesiology before the sanitary crisis. This was a personal choice for the external reinforcements but not for the recovery room and operating theatre staff. Therefore, they had to adapt to an anxiety-provoking situation, which could have worsened the psychological experience of the health care workers that was already challenging in the context of sanitary crisis (24–27).
The development of temporary ICUs requires surgical deprogramming, the consequences of which are difficult to measure at this time. In November 2020, the French hospital federation published the first global and national estimate of the impact of the epidemic on non-COVID-19 activity (28). Over the period from March to August 2020, compared with 2019, a 58% decrease occurred in total hospitalization surgery activity and an 80% decrease in outpatient surgery activity, with significant temporal and geographical disparities, affecting both public and private structures. From March to May 2020, compared with 2019, overall decreases were observed of 87% in diagnostic colonoscopies, 99% in cataract surgery, and 80% in kidney transplants. Worse, urgent medical conditions were affected, with a decrease of 17% in the number of strokes diagnosed and 24% in diagnosed myocardial infarctions, with no clear catch-up during the summer of 2020. According to the cancer league, 93,000 cancers were not diagnosed in 2020 because of the epidemic (29). Several studies worldwide indirectly confirmed the harmful consequences of changing access to patient care: global excess mortality during a wave of epidemics (30–32), increase in the occurrence of out-of-hospital cardiac arrest and a worsened outcome (6,33–35), a delay in myocardial infarction management and a worsened outcome (36–38), and a decrease in transplant activity (39–41).
The medicalization of these new units was ensured by the redeployment of 2,500 doctors, 80% of whom were anaesthesiologists, and 715 residents, 80% of whom were taking the anaesthesia and intensive care course. This unprecedented reinforcement clearly illustrates the value of a dual ability in anaesthesiology and intensive care.