Intensive care rehabilitation outcomes in COVID-19 compared to other respiratory viruses: an observational evaluation

Introduction

The COVID-19 pandemic has resulted in severe repercussions on rehabilitation services. Intensive care unit (ICU) rehabilitation teams have struggled to meet the demand of COVID-19 due to the sheer volume of cases during the pandemic, as well as their complexity (1). In the United Kingdom, COVID-19 patients were admitted to ICUs for a median of 10 days, where 72% required invasive mechanical ventilation (IMV) for a median of 13 days (2). Consequently, a high incidence of intensive care acquired weakness (ICU-AW) has been reported, whereby 52% of patients undergoing IMV continue to experience significant weakness when discharged from the ICU (3, 4). Therefore, it is not surprising that critically ill COVID-19 patients return home at a reduced level of physical functioning, needing mobility aids (49%) or long-term rehabilitation (14%) (5). Overall, the pandemic has highlighted the fundamental need for rehabilitation in order to promote quality of life after ICU in severe COVID-19 cases (6).

As we enter the post-pandemic era, there is concern that the co-existence of COVID-19 and other well-established respiratory viruses (RVs) will place extreme pressure on healthcare systems. Epidemiological studies show that RVs such as influenza and respiratory syncytial virus (RSV) are responsible for between 16% and 49% of ICU admissions with lower respiratory tract infections (7, 8). Influenza occupies a significant portion of ICU beds and already puts pressure on the delivery of rehabilitation services during the winter period, consequently shaping healthcare services particularly during these months (9). Despite sharing many similarities, studies show that COVID-19 places a greater burden on ICUs compared to influenza. COVID-19 patients spend an additional four and five days in ICU and hospital respectively, are mechanically ventilated for twice as along, and have almost double the risk of requiring intubation (2, 10). This increases the likelihood of developing ICU-AW, demanding considerably more capacity for the provision of rehabilitation services.

During the COVID-19 pandemic we have seen a dramatic decrease in the incidence of RVs in ICUs, likely due to new-normal social distancing measures and flu vaccinations (11). We can only expect that the existence of COVID-19 will further complicate the forthcoming influenza seasons (12), however there is a need to better estimate the additional pressures ICU rehabilitation services will face in the winter months. The aim of this evaluation is to present the differences between mechanically ventilated COVID-19 and other RV patients, within the context of ICU rehabilitation and the achievement of basic milestones. This will help ICU and therapy managers better prepare for the combined pressures of COVID-19 and RVs that will constitute the new reality of a post-pandemic winter.

Methods

Design

A service evaluation was completed at two general, adult ICUs of an acute hospital in the United Kingdom (Nottingham University Hospitals NHS Trust). Data was collected for patients admitted to ICU with any RV, including COVID-19, between 1st December 2016 and 30th June 2020. Data was collected from patients’ medical and rehabilitation notes, as well as nursing charts. Patient anonymity was preserved by removing personal data and using a password protected database. Ethical approval was waived for this study as routine practice was not changed and patient confidentiality was maintained. This study was registered as a service evaluation within Nottingham University Hospitals NHS Trust (project ID number 20-598C).

Clinical setting

The extenuating circumstances of the pandemic called for significant structural changes within ICU services to accommodate the influx of COVID-19 admissions. Therefore, for this study, ICU was defined as any hospital space that provided specialist level two or three intensive care, including the use of operating theatre areas. During the pandemic, the ICU physiotherapy service hours were extended to a 12 hour 7-day service (8am–9pm), whereas previously this service would normally run between 8am–4pm, with one late-shift physiotherapist between 4pm–9pm. This change was possible due to staff redeployment from other physiotherapy specialties, increasing the capacity of the ICU physiotherapy team by approximately 40%. This was not the case for patients admitted with RVs before the COVID-19 pandemic, where rehabilitation services ran as normal.

Population

Adult patients admitted to ICU with COVID-19 and other RVs during the aforementioned time-frame were included in this study. More specifically, COVID-19 admissions were during March–June 2020, whereas other RV admissions were between December 2016 and February 2020. Other RVs included influenza (type A and B, H1N1), parainfluenza, human coronavirus, human metapneumovirus (HMPV) and RSV. Admissions for fungal or non-viral pneumonias and paediatric patients were excluded from this study.

Study variables

The rehabilitation variables studied were the number of days taken to achieve three rehabilitation milestones: sit on the edge of the bed (SOEOB), sit out of bed (SOOB) and stand. Level of mobility using the ICU mobility score (IMS) was collected at discharge from ICU. The number of physiotherapy sessions was also documented. Generic demographic data such as gender, age and comorbidities were collected. The independent variables included duration of ICU stay and the use of invasive or non-invasive ventilation, neuromuscular blockades (NMBAs), proning techniques, pre-oxygenation and tracheostomies. Pre-oxygenation refers to the use of a higher fraction of inspired oxygen prior to any physiotherapy manoeuvre, in anticipation of significant oxygen desaturation.

Statistical analysis

Continuous variables are presented as means and standard deviations, whereas categorical variables are described as percentages. Chi-squared tests were used for categorical variables and Student’s t-test for continuous variables, respecting the central limit theorem for sample sizes greater than 30 (13). A significance level of p <0.05 was used. The analyses completed included:

1.  Demographic data for COVID-19 and RVs.
2.  Demographics for COVID-19 and RVs between survivors and non-survivors.
3.  ICU outcomes and treatments for COVID-19 and RVs.
4.  Rehabilitation outcomes for COVID-19 and RVs for mechanically ventilated ICU survivors.

     *ICU = intensive care unit.

See Figure 1: Flowchart of data collection.

Results

A total of 175 patients were studied, where seven were excluded due to transfer to other hospital centres (Figure 1).

109 COVID-19 patients (65%) and 59 RV patients (35%) were studied. The distribution of all viruses for this study’s population is demonstrated in Figure 2.

See Figure 2: Distribution of respiratory viruses within study population (n = 168).

Fifteen percent more men were affected by COVID-19 compared to RVs (p = 0.054, Table 1). Common comorbidities shared between the two groups of viruses included hypertension, cardiovascular disease, musculoskeletal disorders and diabetes. The COVID-19 group had 16% more patients with raised body mass index (BMI) (p <0.001). A 12% increase in general respiratory conditions and chronic obstructive pulmonary disease (COPD) was observed in the RV group (p = 0.024, p = 0.007), as well as cancer (18%, p <0.001). RV patients with a history of cardiovascular disease and cancer had 18% (p = 0.054) and 34% (p = 0.019) more deaths in ICU respectively (Table 2).

Table 1: Demographic data between all COVID-19 and RV* patients (n = 168).

All results are presented in percentages (%) unless otherwise specified, with p values derived from Chi-squared tests.

* = RV (respiratory viruses).

** = Mean with standard deviation.

Table 2: Demographic data between survivors and deceased COVID-19 and RV* patients in ICU.

All results are presented in percentages (%) unless otherwise specified, with p values derived from Chi-squared tests.

* = RV (respiratory viruses).

An additional 8% (p = 0.337) of COVID-19 patients required level three care compared to RV patients (Table 3). COVID-19 patients were ventilated for an additional four days (p = 0.036), with 12% more NMBA use (p = 0.036), 16% more proning and 35% more pre-oxygenation during physical manoeuvring (p <0.001). Both groups had similar use of tracheostomies, which were performed at day 13 of ICU stay. The length of ICU stay was five days greater in COVID-19 patients than in the RV group (p = 0.194).

Table 3: ICU* outcomes and treatments used between COVID-19 and RV** patients who survived ICU.

All results are presented in percentages or median with interquartile ranges, with p values derived from Chi-squared tests and Student t-test.

* = ICU (intensive care unit); ** = RV (respiratory viruses); *** = IMV (invasive mechanical ventilation); **** = SBT (spontaneous breathing trial).

For both groups, a similar number of ICU physiotherapy sessions were received (p = 0.029) and rehabilitation began at day two of ICU stay (p = 0.496; Table 4). COVID-19 patients required an additional seven, one and four days to SOOB, SOEOB and stand respectively (p = 0.043, p = 0.614, p = 0.05; Figure 3). The COVID-19 group achieved a greater level of mobility (IMS grade five, for example transfer from bed to chair) compared to the RV group (p = 0.061).

Table 4: Rehabilitation milestone achievement and level of mobility between mechanically ventilated COVID-19 and RV* patients who survived ICU** admission.

All results are presented in median with interquartile ranges, with p values derived from Student t-test.

*RV (respiratory viruses).

**ICU (intensive care unit).

***IMS (Intensive care mobility scale).

****IMS 4 = standing.

*****IMS 5 = transfer from bed to chair by stepping.

See Figure 3: Rehabilitation milestone achievement between mechanically ventilated survivors with COVID-19 and RVs.

Discussion

This is the first evaluation that compares early rehabilitation milestones in ICU between COVID-19 and other RVs. Similar to other studies, male patients were more affected by RVs in the ICU, with a 15–16% increase seen in COVID-19 (2). Unlike gender, we found no significant difference in the ages of patients admitted to ICU with either virus, although our population had a lower average age compared to other studies (2, 10). Hypertension and cardiovascular disease were the most common comorbidities in COVID-19 and RV ICU admissions, however other studies show a higher prevalence of respiratory comorbidities in other viral pneumonias. Our findings are concurrent with comparative studies suggesting that COVID-19 patients experience an additional four to seven days of IMV, as well as a longer length of ICU stay by four to five days (2, 10, 14). Whilst our finding regarding length of ICU stay was statistically insignificant, it may suggest clinical significance with regards to the additional ICU bed costs for COVID-19 patients, as well as the need for more physiotherapy sessions and staff.

There are few studies that analyse rehabilitation trajectories in non-COVID-19 RVs to help us compare our findings. In this study, the achievement of basic rehabilitation outcomes, such as SOOB and standing, was delayed in COVID-19 patients by seven and four days, respectively. Mobilisation began on day 15 by SOEOB in COVID-19 patients. Likewise, other studies show that mechanically ventilated COVID-19 patients begin to mobilise within 15 days of ICU admission (3). The difference between the two groups may be explained by the greater use of complex ventilation strategies seen in severe COVID-19, such as proned ventilation, NMBAs and extended time of IMV. Previous non-COVID-19 studies associate the development of ICU-AW with risk factors such as IMV for more than five days, NMBAs and proning, however some of these relationships may be considered modest (15, 16). Pre­oxygenation was used twice as much in COVID-19 patients in this study, indicating that physical progression was limited by exertional hypoxia (17).

Furthermore, the significant weeks-worth delay for COVID-19 patients to SOOB may be explained by the limited physical space and mobility equipment during the pandemic. This has also been described in other hospitals, where limiting factors to SOOB included the lack of space, reduced nursing staff availability and altered staffing models (3). However, it is interesting to highlight the delay of rehabilitation milestone achievement within the context of physiotherapy staffing. The 40% increase in ICU physiotherapy team capacity enabled the provision of enough physiotherapy sessions to match a non-pandemic situation – 24 sessions for COVID-19 and 23 for RVs. However, despite these efforts, we continue to see a significant delay in milestone achievement. Therefore, we cannot argue that the delay was due to a lack of physiotherapy staff or number of treatments, thus emphasising the importance of well-staffed teams during high-peak respiratory virus seasons. From a physiological perspective, evidence that compares ICU-AW in COVID-19 to other generic ICU populations discusses the fact that COVID-19 uses angiotensive-converting enzyme two (ACE2) to enter the host cell (18). The expression of ACE2 onto skeletal muscle and nervous cells may lead to additional cytotoxic damage, thus increasing the risk of ICU-AW. However, there are no studies that study ACE2 and ICU-AW between COVID-19 and other RVs to support this statement.

Limitations

The main limitation of this study is the assumption that COVID-19 is a seasonal virus that will add more hospital pressures during the winter months. Epidemiological studies suggest that COVID-19 is temperature-sensitive and, therefore, seasonal, however further studies of full seasonal cycles of the virus are needed to confirm this (19). With regards to the methodology of this study, two outcome measures were not used due to lack of time – the incidence of ICU-AW (through muscle strength testing using the Medical Research Council scale) and the Chelsea Critical Care Physical Assessment Tool (20). Both outcome measures may have helped interpret our results better. Moreover, a survival bias has been acknowledged for this study, where we analysed rehabilitation milestone achievement only in mechanically ventilated survivors. Finally, the lack of patient premorbid status may have supported the interpretation of our findings, however this is a common limitation of ICU research where patients require unplanned and urgent care.

Future research

Future research should include multi-centre studies with larger sample sizes to help with the generalisability of our findings. Follow-up studies that consider rehabilitation outcomes after discharge from ICU, including home discharge and community follow-up would help describe the long-term outcomes between COVID-19 and other RVs. More recent data for COVID-19 patients should be collected, which will take into consideration the advances that have been made regarding its medical treatment and outcomes. Data should also be collected outside of a pandemic setting to interpret the results under ‘normal’ working conditions. Furthermore, predictive models for the co-existence of COVID-19 with other respiratory viruses should be linked to this study to help with the development of future ICU rehabilitation staff models.

In conclusion, COVID-19 results in patients requiring longer time-frames to achieve basic rehabilitation milestones when compared to other RVs. The provision of early ICU rehabilitation is key to prevent long-term effects of COVID-19, avoiding successive pressures on hospital and community rehabilitation services. ICU physiotherapy services require advanced planning of resources and staffing during the winter season to account for the added pressure of COVID-19, which is expected to continue occupying ICU bed-spaces despite the development of a vaccine.

Key points

• COVID-19 patients require longer time-frames to achieve basic rehabilitation milestones in the ICU when compared to other RVs.

• Early ICU rehabilitation is key to prevent long-term effects of COVID-19.

• ICU physiotherapy services require advanced planning of staffing and resources, in order to take into account the added pressure of COVID-19 during the winter season.

Funding

No funding was received for this study.

Conflict of interest

There is no conflict of interest associated with this study.

1 Agostini F, Mangone M, Ruiu P, Paolucci T, Santilli V, Bernetti A. Rehabilitation setting during and after COVID-19: An overview on recommendations. J Rehabil Med. 2021;53(1):jrm00141. Published 2021 Jan 5. https://doi.org/10.2340/16501977-2776.

2 Richards-Belle A, Orzechowska I, Gould D, Thomas K, Doidge J, Mouncey P, Christian M, Shankar-Hari M, Harrison D, Rowan K. COVID-19 in critical care: Epidemiology of the first epidemic wave across England, Wales and Northern Ireland [published correction appears in Intensive Care Med. 2021 Jun;47(6):731–732]. Intensive Care Med. 2020;46(11):2035-2047. https://doi.org/10.1007/s00134-020-06267-0.

3 McWilliams D, Weblin J, Hodson J, Veenith T, Whitehouse T, Snelson C. Rehabilitation levels in patients with COVID-19 admitted to intensive care requiring invasive ventilation. An observational study. Ann Am Thorac Soc. 2021;18(1):122–129. https://doi.org/10.1513/annalsats.202005-560oc.

4 Lad H, Saumur TM, Herridge MS, Santos C, Mathur S, Batt J, Gilbert, P. Intensive care unit-acquired weakness: Not just another muscle atrophying condition. Int J Mol Sci. 2020;21(21):7840. Published 2020 Oct 22. https://doi.org/10.3390/ijms21217840.

5 Musheyev B, Lara Borg, Janowicz R, Matarlo M, Boyle H, Singh G, Ende V, Babatsikos I, Hou W & Duong T. Functional status of mechanically ventilated COVID-19 survivors at ICU and hospital discharge. J Intensive Care. 2021;9(1):31. Published 2021 Mar 31. https://doi.org/10.1186/s40560-021-00542-y.

6 Demeco A, Marotta N, Barletta M, Pino I, Marinaro C, Petraroli A, Moggio L, Ammendolia A. Rehabilitation of patients post-COVID-19 infection: A literature review. J Int Med Res. 2020;48(8):300060520948382. https://doi.org/10.1177/0300060520948382.

7 Wiemken T, Peyrani P, Bryant K, Kelley R, Summersgill J, Arnold F, Carrico R, McKinney W, Jonsson C, Carrico K, Ramirez J. Incidence of respiratory viruses in patients with community-acquired pneumonia admitted to the intensive care unit: Results from the Severe Influenza Pneumonia Surveillance (SIPS) project. Eur J Clin Microbiol Infect Dis. 2013;32(5):705–710. https://doi.org/10.1007/s10096-012-1802-8.

8 Nguyen C, Kaku S, Tutera D, Kuschner WG, Barr J. Viral respiratory infections of adults in the Intensive Care Unit. J Intensive Care Med. 2016;31(7):427–441. https://doi.org/10.1177/0885066615585944.

9 Maltezou HC, Theodoridou K, Poland G. Influenza immunization and COVID-19. Vaccine. 2020;38(39):6078–6079. https://doi.org/10.1016/j.vaccine.2020.07.058.

10 Ludwig M, Jacob J, Basedow F, Andersohn F, Walker J. Clinical outcomes and characteristics of patients hospitalized for influenza or COVID-19 in Germany. Int J Infect Dis. 2021;103:316–322. https://doi.org/10.1016/j.ijid.2020.11.204.

11 Tang J, Bialasiewicz S, Dwyer D, Dilcher M, Tellier R, Taylor J, Hua H, Jennings L, Kok J, Levy A, Smith D, Barr I, Sullivan S. Where have all the viruses gone? Disappearance of seasonal respiratory viruses during the COVID-19 pandemic. J Med Virol. 2021;93(7):4099–4101. https://doi.org/10.1002/jmv.26964.

12 Wormser GP. COVID-19 versus seasonal influenza 2019–2020: U.S.A. Wien Klin Wochenschr. 2020;132(13–14):387–389. https://doi.org/10.1007/s00508-020-01685-y.

13 Lu HC, Fang GC, Wu YS. Estimating the frequency distributions of particulate matter and their metal elements in a temple. J Air Waste Manag Assoc. 2006;56(7):1033–1040. https://doi.org/10.1080/10473289.2006.10464508.

14 Gjurašin B, Santini M, Krajinovic V, Papic N, Atelj A, Kotarski V, Krznaric J, Vargovic M, Kutleša M. A retrospective comparison between influenza and COVID-19-associated ARDS in a Croatian tertiary care center. Wien Klin Wochenschr. 2021;133(7–8):406–411. https://doi.org/10.1007/s00508-020-01759-x.

15 Yang T, Li Z, Jiang L, Wang Y, Xi X. Risk factors for intensive care unit-acquired weakness: A systematic review and meta-analysis. Acta Neurol Scand. 2018;138(2):104–114. https://doi.org/10.1111/ane.12964.

16 Papazian L, Forel J, Gacouin A, Penot-Ragon C, Perrin G, Loundou A, Jaber S, Arnal J, Perez D, Seghboyan J, Constantin J, Courant P, Lefrant J, Guérin C, Prat G, Morange S, Roch A, ACURASYS Study Investigators. Neuromuscular blocking agents in acute respiratory distress syndrome: A systematic review and meta-analysis of randomized controlled trials. Crit Care [Internet]. 2013 Mar 11 [cited 2021 May 20];17(2):1–10. Available from: https://link.springer.com/articles/10.1186/cc12557.

17 Bhasin A, Bregger M, Kluk M, Park P, Feinglass J, Barsuk J. Exertional hypoxia in patients without resting hypoxia is an early predictor of moderate to severe COVID-19. Intern Emerg Med. 2021;16(8):2097–2103. https://doi.org/10.1007/s11739-021-02708-w.

18 Qin ES, Hough CL, Andrews J, Bunnell AE. Intensive care unit-acquired weakness and the COVID-19 pandemic: A clinical review. PM R. 2022;14(2):227–238. https://doi.org/10.1002/pmrj.12757.

19 Byun W, Heo S, Jo G, Kim J, Kim S, Lee S, Park H, Baek J. Is coronavirus disease (COVID-19) seasonal? A critical analysis of empirical and epidemiological studies at global and local scales. Environ Res. 2021;196:110972. https://doi.org/10.1016/j.envres.2021.110972.

20 Corner E, Wood H, Englebretsen C, Thomas A, Grant R, Nikoletou D, Soni N. The Chelsea critical care physical assessment tool (CPAx): Validation of an innovative new tool to measure physical morbidity in the general adult critical care population; an observational proof-of-concept pilot study. Physiotherapy. 2013;99(1):33–41. https://doi.org/10.1016/j.physio.2012.01.003.