Early Rehabilitation Interventions and Physical Therapy in Adults Who Were Critically Ill With COVID-19 Pneumonia: A Retrospective Observational Study

Abstract

Objective

The primary objective of this observational study was to analyze the time to the first edge-of-bed (EOB) mobilization in adults who were critically ill with severe versus non-severe COVID-19 pneumonia. Secondary objectives included the description of early rehabilitation interventions and physical therapy delivery.

Methods

All adults with laboratory-confirmed COVID-19 requiring intensive care unit admission for ≥72 hours were included and divided according to their lowest PaO₂/FiO₂ ratio into severe (≤100 mmHg) or non-severe (>100 mmHg) COVID-19 pneumonia. Early rehabilitation interventions consisted of in-bed activities, EOB or out-of-bed mobilizations, standing, and walking. The Kaplan–Meier estimate and logistic regression were used to investigate the primary outcome time-to-EOB and factors associated with delayed mobilization.

Results

Among the 168 patients included in the study (mean age = 63 y [SD = 12 y]; Sequential Organ Failure Assessment = 11 [interquartile range = 9–14]), 77 (46%) were classified as non-severe, and 91 (54%) were classified as severe COVID-19 pneumonia. Median time-to-EOB was 3.9 days (95% CI = 2.3–5.5) with significant differences between subgroups (non-severe = 2.5 days [95% CI = 1.8–3.5]; severe = 7.2 days [95% CI = 5.7–8.8]). Extracorporeal membrane oxygenation use and high Sequential Organ Failure Assessment scores (adjusted effect = 13.7 days [95% CI = 10.1–17.4] and 0.3 days [95% CI = 0.1–0.6]) were significantly associated with delayed EOB mobilization. Physical therapy started within a median of 1.0 days (95% CI = 0.9–1.2) without subgroup differences.

Conclusion

This study shows that early rehabilitation and physical therapy within the recommended 72 hours during the COVID-19 pandemic could be maintained regardless of disease severity. In this cohort, the median time-to-EOB was fewer than 4 days, with disease severity and advanced organ support significantly delaying the time-to-EOB.

Impact

Early rehabilitation in the intensive care unit could be sustained in adults who are critically ill with COVID-19 pneumonia and can be implemented with existing protocols. Screening based on the PaO₂/FiO₂ ratio might reveal patients at risk and increased need for physical therapy.

Introduction

The coronavirus disease of 2019 (COVID-19) pandemic inflicted a substantial burden on health care systems around the world. Switzerland experienced a large second pandemic wave during the winter 2020−2021 and reported 3.6 million cases (as of May 16, 2022) in a population of 8.5 million.^1,2 Approximately 15% of patients hospitalized with COVID-19 were admitted to a Swiss intensive care unit (ICU).³

ICU procedures for treating patients with COVID-19 included lung-protective mechanical ventilation, deep sedation (sometimes with neuromuscular blockers), steroids, prone positioning, and complex organ support.⁴ However, these life-saving procedures are known to prolong ICU length of stay (LOS), which in turn is associated with numerous long-term sequelae, such as ICU-acquired weakness, cognitive and psychological impairments, and increased mortality.⁵ Specifically, ICU-acquired weakness is associated with extubation failure, increased hospital LOS, and functional impairments with a high dependency in activities of daily living that may lead to long-lasting physical impairments.^6–9

Previous studies have shown that early rehabilitation is safe,¹⁰ significantly reduces neuromuscular complications,¹¹ decreases ICU LOS, and improves functional outcomes at hospital discharge.^12–14 Higher mobilization levels, specifically sitting on the edge-of-bed (EOB), improve functional mobility and increase the likelihood of a discharge to home.¹⁵ In addition, sitting is associated with an increased cardiorespiratory training response.¹⁶ EOB mobilization is therefore an important milestone for patients’ functional recovery.¹⁷ Early rehabilitation is associated with important benefits in terms of health-related quality of life, physical functioning, and peripheral and respiratory muscle strength.^18,19 Thus, physical therapists have become an important part of the ICU team to integrate early rehabilitation measures in patients’ treatment plans.^20,21

We have recently demonstrated that early physical therapy within 48 hours and team-based EOB mobilization starting between 2 to 7 days (interquartile range [IQR]) are safe and feasible in adults who were critically ill and mechanically ventilated with a prolonged LOS.²² Switzerland reported one of the highest rates (33%) of active mobilization (defined as a mobilization level equal to or higher to sitting on the EOB) of mechanically ventilated patients prior to the pandemic²³ compared with Germany (24%)²⁴ or the United States (16%).²⁵ However, there is still an ongoing concern among physicians that early mobilization might be underused in our country.²⁶ Additionally, it remains unclear whether these evidence-based practices could have been sustained during the COVID-19 pandemic.²⁶ Recent data suggest that early mobilization of patients who were critically ill during the COVID-19 pandemic was poorly implemented.^27,28

We therefore designed a benchmark analysis to establish whether evidence-based practices such as early rehabilitation and physical therapist interventions could be sustained during the pandemic as per our previously reported non-COVID cohort.²² With this exploratory study we aimed to (1) analyze the time to the first EOB mobilization (time-to-EOB) after ICU admission as an important milestone of a patients’ trajectory; (2) describe early rehabilitation interventions (total number of patients and interventions) and the provision of physical therapist services (number, duration, and time to) in a mixed Swiss tertiary ICU during the first 3 waves of the COVID-19 pandemic; (3) draw comparisons between non-severe and severe COVID-19 pneumonia groups; and (4) explore the association of pre-specified explanatory variables with time-to-EOB.

Methods

Design and Setting

This retrospective, single-center, observational study was conducted in a mixed ICU of a large, academic tertiary center in Switzerland (Department of Intensive Care Medicine, Inselspital, Bern University Hospital) between March 1, 2020, and May 31, 2021. Adult patients (≥18 years of age) admitted to the ICU with a laboratory-confirmed diagnosis of COVID-19 pneumonia and an LOS >72 hours were included. Patients were included regardless of mechanical ventilation requirement. Patients who actively refused general consent (opt-out) and patients with non-pulmonary COVID-19 disease were excluded.

The research project was approved by the local ethics committee (Ethics Committee of Canton Bern, #2021-00613). Study reporting complies with the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines.²⁹

Definitions

Severity of COVID-19 Pneumonia

The severity of COVID-19 pneumonia was graded according to the lowest ratio of partial pressure of arterial oxygen to fraction of inspired oxygen (PaO₂/FiO₂ ratio, ie, Horowitz index) within the first 24 hours of ICU admission.³⁰ A PaO₂/FiO₂ ratio of 100 mmHg delineated severe COVID-19 pneumonia (PaO₂/FiO₂ ratio ≤ 100 mmHg) from non-severe COVID-19 pneumonia (PaO₂/FiO₂ ratio > 100 mmHg).³⁰

Early Rehabilitation Interventions and Physical Therapy

A consensus on the definition of early rehabilitation or mobilization interventions is currently lacking.³¹ The German Society for Anesthesiology and Intensive Care Medicine recommends to start protocolized early rehabilitation in collaboration with physical therapists within the first 72 hours.³² Accordingly, our center has adopted an interprofessional approach to early rehabilitation that includes critical care nurses and physical therapists. They closely collaborate with physicians to set goals and procedures for the patient’s overall course of recovery. Our usual therapy practices and interprofessional standards were previously described.^22,33 In brief, early rehabilitation interventions start as soon as possible based on the daily assessment by physical therapists on morning rounds and multidisciplinary team visits. The protocol entails a step-wise approach starting from in-bed activities such as cycling, advancing via a side-edge position (upright position with 1 leg on the floor),³⁴ to the EOB and out-of-bed. The protocol aims to ensure goal-specific adaptation and to gradually increase patient participation, mobilization level, and functional task complexity. Pre-pandemic physical therapist services were usually available from Monday to Friday from 8 am to 5 pm and for specifically selected patients on weekends from 9 am to 12 am. However, during the first 2 waves of the pandemic, physical therapists worked in 2 shifts from 8 am until 10 pm, 7 days per week. At that time, the main objectives of physical therapy were to assist with safe proning and implement early rehabilitation interventions.³⁴ The frequency of physical therapy sessions was between 1 to 3 times per day. It was at the therapist’s discretion to decide how often and intense a patient received therapy per day. These decisions were based on physical examinations and clinical reasoning.

For this benchmark analysis, we defined the time from ICU admission to early physical therapy as starting within 48 hours and a time-to-EOB <7 days as early. These definitions were based on our previously described non-COVID cohort and took into account the expected increased severity of patients with COVID-19 (accordingly, the previously reported 75% percentile for time-to-EOB was chosen).²² They served as a prospective objective to determine compliance with our standard of care.

Outcomes

The primary outcome of this study was to analyze the time-to-EOB from ICU admission in all patients with COVID-19 as well as in the subgroups of patients with severe and non-severe COVID-19 pneumonia. Sitting on the EOB was considered the most important milestone to achieve functional independence.¹⁵ Secondary outcomes were the total number of physical therapy sessions, the duration for each session (minutes spent with the patient), the time to the first physical therapy session, and the number of patients who received early rehabilitation interventions such as in-bed cycling, sitting in-bed, sitting side-edge³⁴, sitting EOB, sitting out-of-bed, standing, walking. We summarized total number of interventions, calculated the differences between the 2 defined groups (non-severe and severe), and reported the highest mobilization level (in-bed, EOB, out-of-bed). We further analyzed the association of a priori explanatory variables with the primary outcome time-to-EOB. Finally, to describe ICU outcomes, we collected the Medical Research Council sum score to assess the presence of ICU-acquired weakness (<48 points),³⁵ Glasgow Coma Scale, Richmond Agitation Sedation Scale, and survival status at ICU discharge.

Data Collection

Structured routine ICU, rehabilitation intervention, and physical therapy data were extracted from electronic databases (patient data management system from Centricity Critical Care Clinisoft, GE, Barrington, IL, USA; Phoenix CompuGroup Medical, Bern, Switzerland). Data were checked for their completeness and consistency using graphical illustrations and descriptive statistical analyses, whereby outliers or missing data were re-checked in the original sources. Baseline characteristics included patient demographics (age, sex, body mass index [BMI]), illness severity markers (Sequential Organ Failure Assessment [SOFA]), Acute Physiology and Chronic Health Evaluation II Score [APACHE II Score]) as well as the number of comorbidities (including the Charlson Comorbidity Index) at ICU admission. We further collected ICU LOS and interventions (intubation, tracheostomy, extracorporeal membrane oxygenation [ECMO], dialysis, proning, mechanical ventilation).

Statistical Analysis

Due to the exploratory nature of this analysis, we used a convenience sample and did not perform a prospective sample size calculation.

Continuous, normally distributed variables are described as means with SDs and non-normally distributed, ordinal data described as medians and IQRs. Severity-group comparisons for metric data were made using a pairwise, independent t test for ordinal data with the Mann–Whitney-U test and for categorical data with the Pearson χ-squared test or Fisher exact test. We hypothesized that patients with severe COVID-19 pneumonia mobilized substantially later to the EOB than patients with non-severe COVID-19 pneumonia.

The time-to-EOB for all and between non-severe and severe was evaluated with a Kaplan–Meier log-rank analysis between subgroups, censored at ICU discharge to account for non-mobilized patients and without censoring to estimate actually implemented time-to-EOB. The relation of the a priori–defined explanatory variables with time-to-EOB (continuous dependent variable in days) was determined with multiple linear regression. The variables were chosen on previous evidence and clinical reasoning and included BMI, SOFA, Charlson Comorbidity Index, oxygenation index, prone positioning, ECMO use, and intubation. The time to the first physical therapist sessions was analyzed with Kaplan–Meier without censoring using log-rank analysis between subgroups.

Analyses were performed using RStudio (2009-2020 RStudio, PBC, Version 1.3.1073, packages rio, psych, Hmisc, tab, survival, and ggplot2, Posit, Boston, MA, USA) and SPSS (IBM SPSS Statistics Version 25, IBM Corp, Armonk, NY, USA). We considered P < .05 as statistically significant.

Results

Study Participants

Between March 2020 and May 2021, a total of 301 patients were admitted to the mixed ICU with a laboratory-confirmed COVID-19 diagnosis. Of these, 168 (56%) met the inclusion criteria and were subsequently analyzed (Fig. 1). Baseline characteristics, ICU interventions, and outcomes for all participants and per severity group are reported in Table 1, whereby non-severe significantly differed from severe for illness severity, ICU interventions, and outcomes.

Primary Outcome

Overall, 107 (64%) patients sat at least once on the EOB (non-severe = 59 [77%]; severe = 48 [53%]). Censored, median time-to-EOB was 8.8 days (95% CI = 6.4–11.1) for all, 3.9 days (95% CI = 1.6–6.2) in the non-severe group, and 12.7 days (95% CI = 11.1–14.4) in the severe group (P < .001) (Fig. 2). Non-censored, median time-to-EOB was 3.9 days (95% CI = 2.3–5.5) for all and 2.5 days (95% CI = 1.8–3.5) and 7.2 days (95% CI = 5.7–8.8) for the non-severe group and severe group, respectively (P = .014).

Secondary Outcomes

Physical therapy, rehabilitation interventions, highest mobilization level, and time-to-first EOB are reported in Table 2. The total number of interventions per severity group are shown in Figure 3. A total of 166 (99%) patients received physical therapy within a median of 1.0 days (95% CI = 0.9–1.2): the non-severe group within a median of 1.3 days (95% CI = 0.86–1.6) and the severe group within a median of 1.0 days (95% CI = 1.0–1.1) (P = .637). Physical therapy was delivered within 48 hours for 135 (80%) patients. Patients had a median number of 10 physical therapist sessions (IQR = 6–19), whereby the number of sessions did not differ between groups (non-severe group = 9 sessions [IQR 5–19] vs severe group = 12 sessions [IQR 6–19]; P = .526) (Tab. 2).

Factors Contributing to Time-to-EOB

In the multiple linear regression model, SOFA score (plus 0.3 days per each additional point) and ECMO use (plus 13.7 days compared with no ECMO) were significantly associated with a delayed mobilization to the EOB (Tab. 3).

Discussion

With this single-center study in Switzerland, we report a high delivery and early start of rehabilitation and physical therapy in patients who were critically ill with COVID-19 pneumonia that was comparable with the standards we previously reported before the pandemic.²² Thus, we were able to perform this evidence-based practice during the COVID-19 pandemic despite a substantially sicker cohort (APACHE 30 vs 22) of patients with longer ICU LOS (10 vs 6 days).²² In addition, our patients were mobilized earlier compared with similar studies in patients with severe COVID-19, whereby McWilliams et al²⁸ mobilized their patients within a mean of 14 (SD = 7) days to the EOB and Rossi et al³⁶ within a median of 18 (IQR = 10–34) days.^28,36 This was also true for the sickest patients of our cohort with severe COVID-19 pneumonia.

In contrast, McWilliams et al²⁸ and Stutz et al³⁷ reported higher mobilization levels (such as standing or walking) at ICU discharge for patients with COVID-19 pneumonia compared with our study population. Reasons might include different case mixes (median APACHE II of 30 vs 17 in Stutz et al³⁷ vs 16 in McWilliams et al²⁸) and difference in health care systems and discharge criteria resulting in different LOS (median ICU LOS of 10 days vs 7 days in Stutz et al³⁷ vs mean of 22 days in McWilliams et al²⁸). During the COVID pandemic, we transferred patients back to referring centers, sometimes under sedation and mechanical ventilation as soon as stabilization had been reached. This might explain our comparatively shorter LOS together with lower Richmond Agitation Sedation Scale and Glasgow Coma Scale scores at ICU discharge. In addition, some studies reported that patients with a higher BMI are mobilized significantly later than patients with a lower or normal BMI.²⁸ We did not observe this in our cohort.

Not surprisingly, patients with severe COVID-19 pneumonia reached lower mobilization levels at ICU discharge with a delayed time-to-EOB compared with patients who had non-severe COVID-19 pneumonia. Major contributing factors for delayed mobilization were higher SOFA scores and enhanced level of organ support (ie, ECMO). Barriers for early mobilization might have been higher levels of sedation and increased use of neuromuscular blockade in the severe group.³⁸ This group of the sickest and complex patients is highly challenging and requires sufficient staffing and interprofessional management for the safe delivery of physical therapist interventions.³⁹

Delayed achievement of mobilization goals (eg, sitting at the EOB) is by itself a reason for poor physical outcomes such as ICU-acquired weakness.^7,9 In our population, almost 80% of patients with severe COVID-19 pneumonia had a confirmed ICU-acquired weakness at ICU discharge (Medical Research Council sum score < 48). Moonen et al⁴⁰ reported a similar high incidence (100%) of ICU-acquired weakness at ICU discharge after hospitalization for COVID-19 pneumonia. In another study by Wiertz et al,⁴¹ 72% of patients with COVID-19 had muscle weakness in all major muscle groups at admission to an inpatient rehabilitation care following ICU discharge. In addition, SARS-CoV-2 itself shows neurotropic propriety and affects the peripheral and central nervous system.⁴² Thus, there is a need for additional patient-directed, targeted rehabilitation for severe COVID-19 pneumonia after ICU discharge.

In our study, the observed median time to the first ICU physical therapy session was well within the recommended 72 hours³² and even earlier compared with our previously reported standards.²² It was also shorter than reported by Stutz el al,³⁹ who started physical therapy after a median of 4 days (IQR = 4–16) despite fewer intubated patients (40% vs 96%). An important factor contributing to the successful implementation of early physical therapy might have been the extension of physical therapist services, including coverage in the evenings. This was also observed by others.^28,43 In addition, we cannot exclude that it might have been the result of the ongoing “early mobility culture” that we established following our randomized controlled trial.²²

Due to the already established benefit of proning and rehabilitation, physical therapists in our unit primarily focused on these interventions during the pandemic.^12–14,44 However, in other parts of the world, physical therapists have been more involved in respiratory management than in delivering early mobilization therapies.⁴⁵ Higher caseloads, such as in the United Kingdom, could have led to different treatment priorities for physical therapists, such as the management of mechanical ventilation and respiratory therapy.^28,43 Although our therapists did perform respiratory interventions such as deep breathing exercises, secretion clearance, or assisted coughing, data on those interventions unfortunately were not routinely collected. More research about the efficacy of respiratory techniques would help physical therapists to choose the most beneficial treatment during pandemic peaks. In addition, future randomized trials are needed to investigate the effectiveness of early rehabilitation interventions to accelerate weaning and functional outcomes.

Limitations

This study provides real-world data from performed rehabilitation interventions and physical therapist services during the COVID-19 pandemic using prospective hypotheses and rigorous analysis. However, this study also has limitations. First, data from a single university hospital limits generalization to other hospital settings. Second, due to the retrospective nature of the study design, the availability of variables was limited to the available routine data. Accordingly, data on functional outcomes, reasons for missing rehabilitation intervention, or adverse events could not be retrieved. However, we previously reported a very low incidence of adverse events during physical therapy (<0.6%),²² and our safety protocols remained unchanged. Third, no conclusions about the efficacy of early rehabilitation in patients who were critically ill with COVID-19 pneumonia can be drawn due to the lack of a control group. Fourth, we decided to grade the severity of the COVID-19 pneumonia using a classification based on the Berlin definition for acute respiratory distress syndrome³⁰ and not according to the National Institutes of Health criteria, which would have classified all patients with COVID-19 who were admitted to the ICU as “severe.”⁴⁶ However, given that patients with a mild-to-moderate disease do not necessarily have to be admitted to the ICU, this classification would not have been appropriate for our study purposes. And, finally, as mentioned in the discussion, our LOS were comparatively short due to an increased inter-hospital as well as inter-ICU transfer of patients with COVID-19 during the pandemic. This was a result of regional bed shortages. In turn, we regionalized patients as soon as they stabilized to enable bed capacity.

Conclusion

Our study shows that early rehabilitation and physical therapy for adults who were critically ill with COVID-19 pneumonia was maintained during the pandemic peaks and started within 48 hours regardless of the severity of illness. The expansion of the physical therapy service, daily assessment, interprofessional management and a continuous culture of early mobilization may have contributed to the sustained practice of early mobilization. The median time-to-EOB was less than 4 days. However, time-to-first EOB was delayed in patients who had severe compared with patients who had non-severe COVID-19 pneumonia. Here, high SOFA scores and ECMO use were independently associated with delayed mobilization. Future studies should investigate rehabilitation pathways for this critically ill patient group to find safe and effective treatments.

Author Contributions

Concept/idea/research design: N. Pecorelli, S. Eggmann, M.-M. Jeitziner, Y.-A. Que, A.S. Messmer

Writing: N. Pecorelli, S. Eggmann, A.S. Messmer

Data collection: N. Pecorelli, S. Eggmann, M.-M. Jeitziner, Y.-A. Que, A.S. Messmer

Data analysis: N. Pecorelli, S. Eggmann, A.S. Messmer

Project management: S. Eggmann, A.S. Messmer

Providing participants: S. Eggmann, A.S. Messmer

Providing facilities/equipment: S. Eggmann, A.S. Messmer

Providing institutional liaisons: S. Eggmann, A.S. Messmer

Consultation (including review of manuscript before submitting): N. Pecorelli, S. Eggmann, M.-M. Jeitziner, Y.-A. Que, A.S. Messmer

Ethics Approval

The research project was approved by the Ethics Committee of Canton Bern, #2021-00613.

Funding

There are no funders to report for this study.

Disclosures

The authors completed the ICMJE Form for Disclosure of Potential Conflicts of Interest and reported no conflicts of interest.

Data Availability

The data supporting the findings of this study are available from the corresponding author on reasonable request.

References

Author notes