Outcomes of Prone Positioning in Mechanically Ventilated COVID-19 patients: A Prospective Observational Study

doi:10.21203/rs.3.rs-2747209/v1

Introduction

Prone positioning in mechanically ventilated patients with severe acute respiratory distress syndrome (ARDS) is associated with improved mortality. More data is needed to fully understand its utility in those with ARDS due to COVID-19.

Methods

We conducted a single center prospective observational study inclusive of 100 consecutive patients intubated for ARDS from COVID-19 admitted to the ICU from September 2020 to December 2020. Data was collected daily from time of intubation for 7 days along with 30-day outcomes.

Results

The study included a total of 53 patients proned and 47 non-proned during their hospitalization. Proned patients were 61.8 years old, and 56.6% men compared to 66.3 years old and 57.4% male in the non-proned group. Other baseline characteristics and treatments were similar between both groups other than proned patients having a higher BMI than non-proned patients (34.1 ± 7.5 vs 30.5 ± 7.4, p = 0.02), and lower initial P/F ratios (119.1 ± 54.5 vs 154.0 ± 92.7, p = 0.047). Proned patients required more neuromuscular blockade (OR 6.63, 95% CI 3.25–13.12, p < 0.0001) and higher sedation levels (2 sedatives: OR = 3.00, 95% CI = 1.77,5.08; ≥3 sedatives: OR = 7.13, 95% CI = 3.96,12.81) with similar ICU length of stays, ventilator days, newly initiated renal replacement therapy, and 30-day outcomes when compared to non-proned patients. Proned patients were re-intubated substantially less than the non-proned group (1.9% vs 19.1%, p = 0.006).

Conclusion

Proning mechanically ventilated COVID-19 patients was associated with more frequent use of neuromuscular blockade and sedation, and required significantly lower rates of re-intubation for respiratory failure when compared to non-proned patients.

COVID-19

Prone Positioning

ARDS

Hypoxemia

Intubation

The severity of pulmonary involvement in patients with Coronavirus Disease 2019 (COVID-19) is variable. As many as 42% of patients have been reported to have acute respiratory failure ^[1] with 10–20% having severe respiratory failure requiring mechanical ventilation ^[2]. Many of these patients with severe respiratory failure suffer from acute respiratory distress syndrome (ARDS) leading to implementation of a variety of interventions to reduce mortality, including prone positioning.

The landmark 2013 Proning Severe ARDS Patients (PROSEVA) trial demonstrating improved 28-day mortality in patients with severe ARDS ^[3], along with subsequent meta-analyses ^{[4, 5]} also showing improved mortality in prone positioning, led to the American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine clinical practice guidelines to strongly recommend prone positioning in severe ARDS^{[6, 7]}. Prone positioning improves mortality effectively by improving oxygenation and respiratory mechanics through a more homogenous distribution of ventilation and perfusion ^[7].

Throughout the COVID-19 pandemic, intensivists have widely performed prone positioning in patients with ARDS without fully knowing the utility of this intervention in this specific patient population. There has been a paucity of data with relatively few studies performed on the effects of prone positioning in intubated patients compared to non-intubated patients specifically with COVID-19 ^[8]. The studies that have been performed on prone positioning in intubated COVID-19 patients with ARDS have revealed mixed results on mortality^{[9, 10]} necessitating the need for further studies to determine the utility of this intervention.

Despite a lack of data, the Surviving Sepsis Campaign guidelines on the management of critically ill adults with COVID-19 suggests prone ventilation for 12–16 hours over no prone ventilation for mechanically ventilated adults with COVID-19 and moderate to severe (ARDS) ^[11]. Due to the paucity of data specifically in the COVID-19 patient population, this suggestion is a weak recommendation based off low-quality of evidence. We sought to better understand the utility of prone positioning in an intubated COVID-19 population and compare outcomes and associated complications to a non-proned patient population.

We performed a single center, prospective, observational study conducted at a large academic tertiary care center from September 2020 to December 2020. 100 consecutive patients 18 years of age or older admitted with COVID-19 pneumonia to the intensive care unit requiring invasive mechanical ventilation for ARDS were included. ARDS was defined by the classic Berlin Criteria ^[12] with patients having a partial pressure of arterial oxygen (PaO2) to fraction of inspired oxygen (FiO2) (P/F Ratio) of at least 300 or less. Patients were excluded from the study if intubation was performed for alternative causes to respiratory failure or were intubated at an outside facility greater than 24 hours prior to transfer. The study was approved by the local institutional review board.

Baseline characteristics were obtained via chart review of the patients’ electronic medical record along with daily data from the time of intubation for 7 days and 30-day outcomes. Daily data collection included variables prior to the onset of the proning session as well as 6 hours after the start of the proning session to assess for changes in patient stability. Patients were proned at the discretion of the critical care staff attending with most being proned within 6–12 hours of intubation. Prone positioning was performed by an experienced group of Intensive Care Unit (ICU) nurses and respiratory therapists with patients being proned for up to 16 hours. Subsequent episodes of prone positioning were performed based on patient response and were also at the discretion of the critical staff attending.

All complications were documented and noted by the patients care team. Nursing staff was educated on documentation of predefined complications including hypoxemia defined as a greater than 10% drop or required ventilator adjustment; hypotension defined as a greater than 20% drop in systolic blood pressure or increase in intravenous vasopressor support; cardiac arrythmia; endotracheal tube dislodgement; central line, arterial line, or foley catheter dislodgment; or cardiac arrest. Complications related to proning were defined as the occurrence of one or more complications during or immediately after the proning event. Predefined outcomes included 30-day mortality, ICU length of stay, complications related to proning, pre- and post-proning hemodynamic changes, rates of re-intubation for respiratory failure, need for renal replacement therapy, and need for tracheostomy.

For continuous variables, means and standard deviations were calculated and differences were compared using two-sampled t-tests or Wilcoxon rank sum tests for variables that did not meet t-test assumptions. For categorical variables, frequencies were compared using chi-square tests for association or Fisher’s Exact Test for variables with cell counts < 5. Repeated measures were modeled by random-intercept mixed effects ANOVA models with an autoregressive covariance structure and the Kenward-Roger approximation for degrees of freedom for continuous variables, generalized linear mixed models with binomial distribution and logit link for binary variables, or multinomial distribution with generalized logit link for multinomial variables with unstructured covariance structure and the Kenward-Roger approximation for degrees of freedom. Log transformations were applied to skewed continuous variables prior to repeated measures analysis (ABG-PO2, PO2:FiO2, and D-Dimer). Chi-square tests of association and two-sample t-tests were also used to test differences by 30-day survival.

The study population included a total of 53 patients proned and 47 patients not proned during their hospitalization. Proned patients were 61.8 years old, and 56.6% men compared to 66.3 years old and 57.4% men in the non-proned group. While proned patients had a higher body mass index (BMI) than non-proned patients (34.1 ± 7.5 vs 30.5 ± 7.4, p = 0.02), other baseline characteristics inclusive of race, comorbidities, mode of ventilation, and treatment with steroids, remdesivir, or convalescent plasma were similar between both groups (Table 1). Baseline vitals and inflammatory markers within 4 hours of intubation were overall similar between both groups other than the proned group having a lower initial P/F ratio (119.1 ± 54.5 vs 154.0 ± 92.7, p = 0.047), lower mean oxygen saturation (92.9% ± 5.5 vs 94.9% ±5.2, p = 0.04), and higher required Positive End Expiratory Pressure (PEEP) (13.2 ± 3.1 vs 10.3 ± 2.7, p < 0.001) and Fractional Inspired Oxygen (FiO2) (91.9% ±13.0 vs 84.7% ±19.5, p = .0.04). Proned patients were also intubated later into their ICU admission than the non-proned patients (5.1 days ± 5.4 vs 2.0 ± 2.8, p = 0.0006).

Table 1

Baseline Characteristics
	Total N (%)	Not proned N (%)	Proned N (%)	p-value
Baseline characteristics
Age (mean, SD)	64.0, 12.1	66.3, 11.6	61.8, 12.3	0.06
BMI (mean, SD)	32.5, 7.6	30.5, 7.4	34.1, 7.5	0.02
Sex
Female	43 (43.0)	20 (42.6)	23 (43.4)	0.93
Male	57 (57.0)	27 (57.4)	30 (56.6)
Race
White	56 (66.7)	22 (59.5)	34 (72.3)	0.43
African American	14 (16.7)	8 (21.6)	6 (12.8)
Other	14 (16.7)	7 (18.9)	7 (14.9)
Missing	16	10	6
Ethnicity
Not Hispanic or Latino	74 (85.1)	31 (79.5)	43 (89.6)	0.19
Hispanic or Latino	13 (14.9)	8 (20.5)	5 (10.4)
Missing	13	8	5
Active smoker
No	75 (92.6)	35 (97.2)	40 (88.9)	0.22
Yes	6 (7.4)	1 (2.8)	5 (11.1)
Missing	19	11	8
	Total N (%)	Not proned N (%)	Proned N (%)	p-value
Past medical history
Diabetes
No	53 (53.0)	24 (51.1)	29 (54.7)	0.71
Yes	47 (47.0)	23 (48.9)	24 (45.3)
Hypertension
No	27 (27.0)	11 (23.4)	16 (30.2)	0.45
Yes	73 (73.0)	36 (76.6)	37 (69.8)
Coronary artery disease
No	71 (71.0)	33 (70.2)	38 (71.7)	0.87
Yes	29 (29.0)	14 (29.8)	15 (28.3)
Chronic lung disease
No	80 (80.0)	38 (80.9)	42 (79.2)	0.84
Yes	20 (20.0)	9 (19.1)	11 (20.8)
Renal failure
No	82 (82.0)	37 (78.7)	45 (84.9)	0.42
Yes	18 (18.0)	10 (21.3)	8 (15.1)
Liver disease
No	96 (96.0)	45 (95.7)	51 (96.2)	1.00
Yes	4 (4.0)	2 (4.3)	2 (3.8)
Active cancer
No	94 (94.0)	43 (91.5)	51 (96.2)	0.42
Yes	6 (6.0)	4 (8.5)	2 (3.8)
Immunosuppression
No	91 (91.0)	45 (95.7)	46 (86.8)	0.17
Yes	9 (9.0)	2 (4.3)	7 (13.2)
None
No	90 (90.0)	45 (95.7)	45 (84.9)	0.10
Yes	10 (10.0)	2 (4.3)	8 (15.1)
CXR or CT prior to intubation
Normal	2 (2.0)	0 (0.0)	2 (3.8)	0.51
Bilateral opacities	77 (77.0)	35 (74.5)	42 (79.2)
Unilateral opacities	4 (4.0)	2 (4.3)	2 (3.8)
Not available	17 (17.0)	10 (21.3)	7 (13.2)
Patients with opacities:
Pleural effusion	6 (7.2)	5 (13.5)	1 (2.2)	0.08
Pneumothorax	1 (1.2)	0	1 (2.2)
Neither	76 (91.6)	32 (86.5)	44 (95.7)
Missing	17
ABG within 4 hours of intubation?
No	4 (4.0)	3 (6.4)	1 (1.9)	0.34
Yes	96 (96.0)	44 (93.6)	52 (98.1)
Mode of ventilation
Pressure A/C	12 (12.0)	3 (6.4)	9 (17.0)	0.43
Volume A/C (autoflow)	79 (79.0)	39 (83.0)	40 (75.5)
Volume A/C (no autoflow)	4 (4.0)	2 (4.3)	2 (3.8)
APRV	4 (4.0)	2 (4.3)	2 (3.8)
Other	1 (1.0)	1 (2.1)	0
Number of vasopressors
0	44 (44.0)	20 (42.6)	24 (45.3)	0.72
1	48 (48.0)	22 (46.8)	26 (49.1)
≥ 2	8 (8.0)	5 (10.6)	3 (5.7)
	Total N (%)	Not proned N (%)	Proned N (%)	p-value
Number of sedatives
0–1	6 (6.0)	6 (12.8)	0	0.02
2	84 (84.0)	36 (76.6)	48 (90.6)
3	10 (10.0)	5 (10.6)	5 (9.4)
Use of neuromuscular block
No	93 (93.0)	45 (95.7)	48 (90.6)	0.44
Yes	7 (7.0)	2 (4.3)	5 (9.4)
Received steroid
No	0	--	--	--
Yes	100 (100.0)	--	--	--
Received Remdesivir
No	15 (15.0)	10 (21.3)	5 (9.4)	0.10
Yes	85 (85.0)	37 (78.7)	48 (90.6)
Received convalescent plasma
No	55 (55.0)	29 (61.7)	26 (49.1)	0.20
Yes	45 (45.0)	18 (38.3)	27 (50.9)
Developed pressure ulcer during ICU stay
No	50 (50.0)	24 (51.1)	26 (49.1)	0.84
Yes	50 (50.0)	23 (48.9)	27 (50.9)
Baseline vitals within 4 hours of intubation	Mean (SD)	Mean (SD)	Mean (SD)	p-value
Systolic blood pressure	123.0, 30.9	122.2, 32.2	123.8, 29.9	0.80
Diastolic blood pressure	72.7, 20.3	71.8, 18.3	73.5, 22.1	0.67
Mean blood pressure	89.5, 22.1	88.6, 21.2	90.3, 23.0	0.71
Heart rate	90.9, 22.7	90.7, 25.7	91.2, 19.9	0.91
Respiratory rate	25.4, 5.6	25.5, 6.7	25.2, 4.5	0.79
Oxygen saturation	93.8, 5.4	94.9, 5.2	92.9, 5.5	0.04
< 88%	12 (12.0)	3 (6.4)	9 (17.0)	0.13
≥ 88%	88 (88.0)	44 (93.6)	44 (83.0)
ABG-PH	7.3, 0.1	7.3, 0.1	7.3, 0.1	0.95
ABG-PO2	121.3, 57.8	131.6, 67.9	112.5, 46.6	0.12
ABG-PCO2	55.3, 17.1	52.4, 14.0	57.7, 19.1	0.12
PO2:FiO2 ratio	135.1, 76.1	154.0, 92.7	119.1, 54.5	0.047
PEEP	11.8, 3.2	10.3, 2.7	13.2, 3.1	< 0.0001
FiO2	88.5, 16.7	84.7, 19.5	91.9, 13.0	0.04
Peak airway pressure	26.5, 8.5	25.9, 8.5	27.0, 8.5	0.53
Tidal volume (ml/kg)	6.90, 2.12	6.78, 1.82	7.01, 2.37	0.81
≤ 6.0 (N, %)	37 (37.0)	15 (32.0)	22 (41.5)	0.32
> 6.0 (N, %)	63 (63.0)	32 (68.1)	31 (58.5)
D-Dimer	9,834; 17,850	10,603; 16,416	9,241; 19,032	0.16
CRP	21.7, 65.7	16.8, 9.8	25.6, 87.9	0.49

Daily parameters assessed across the 7 days between both proned and non-proned patients were similar between both groups overall. However, proned patients had a lower pH (7.33; 95% CI 7.31,7.35 vs 7.36; 7.34,7.38 p = 0.03), higher pCO2 (53.95; 95% CI 51.18, 56.73 vs 49.11; 45.93, 52.28 p = 0.02), lower P/F ratio (log-scale) (4.81; 95% CI 4.73, 4.89 vs 5.00; 4.91, 5.09 p = 0.002), higher PEEP (11.31; 95% CI 10.71,11.92 vs 8.83; 8.15,9.50 p < 0.0001), higher FiO2 (67.59; 95% CI 63.48,71.70 vs 55.23; 50.56, 59.90 p = 0.001) and higher peak airway pressures (26.03, 95% CI 24.60, 27.47 vs 22.30; 20.68, 23.91 p = 0.0008) (Table 2). Proned patients required neuromuscular blockade over the 7-day study period more frequently than non-proned patients (OR 6.63, 95% CI 3.25–13.12, p < 0.0001). While proning was also associated with a significantly increased likelihood of higher sedation levels compared to non-proned patients (2 sedatives: OR 3.00, 95% CI 1.77,5.08, p < 0.0001; ≥3 sedatives: OR 7.13, 95% CI 3.96, 12.81, p < 0.0001), it was not associated a significant increased need for more vasopressor therapy (OR 1.28, 95% CI 0.95, 1.73, p = 0.11).

Table 2

Associations between Proning and Daily Parameters (as measured at 12pm each day) Across the 7-day Study Period
	Not proned LS Means (95% CI)	Proned LS Means (95% CI)	β (95% CI)	p-value
Systolic blood pressure	119.33 (115.68, 122.97)	118.52 (115.18, 121.87)	-0.80 (-5.75, 4.14)	0.75
Diastolic blood pressure	61.43 (59.26, 63.60)	61.51 (59.52, 63.51)	0.08 (-2.86, 3.03)	0.96
Mean blood pressure	80.83 (78.72, 82.94)	80.47 (78.53, 82.41)	-0.36 (-3.23, 2.50)	0.80
Heart rate	82.58 (77.83, 87.33)	82.62 (78.27, 86.96)	0.04 (-6.40, 6.48)	0.99
Respiratory rate	24.64 (23.48, 25.80)	25.37 (24.30, 26.44)	0.73 (-0.85, 2.31)	0.36
Oxygen saturation	94.26 (93.59, 94.93)	93.36 (92.75, 93.97)	-0.90 (-1.81, 0.003)	0.05
ABG-PH	7.36 (7.34, 7.38)	7.33 (7.31, 7.35)	-0.03 (-0.06, -0.003)	0.03
ABG-PO2 (log)	4.44 (4.38, 4.49)	4.42 (4.37, 4.47)	-0.02 (-0.09, 0.06)	0.67
ABG-PCO2	49.11 (45.93, 52.28)	53.95 (51.18, 56.73)	4.84 (0.63, 9.06)	0.02
PO2:FiO2 ratio (log)	5.00 (4.91, 5.09)	4.81 (4.73, 4.89)	-0.19 (-0.31, -0.07)	0.002
PEEP	8.83 (8.15, 9.50)	11.31 (10.70, 11.92)	2.48 (1.58, 3.39)	< 0.0001
FiO2	55.23 (50.56, 59.90)	67.59 (63.48, 71.70)	12.36 (6.14, 18.58)	0.0001
Peak airway pressure	22.30 (20.68, 23.91)	26.03 (24.60, 27.47)	3.74 (1.58, 5.90)	0.0008
Tidal volume (ml/kg)	7.31 (6.89, 7.73)	7.52 (7.15, 7.88)	0.21 (-0.35, 0.76)	0.47
D-Dimer (log)	8.09 (7.84, 8.34)	7.80 (7.57, 8.03)	-0.29 (-0.63, 0.05)	0.09
CRP	13.77 (11.56, 15.99)	15.20 (13.19, 17.21)	1.43 (-1.57, 4.42)	0.35
	OR (95% CI)	p-value
Number of vasopressors
1	ref
≥ 2	1.28 (0.95, 1.73)	0.11
Neuromuscular blocker
No	ref
Yes	6.63 (3.25, 13.12)	< 0.0001
Number of sedatives
0–1	ref
2	3.00 (1.77, 5.08)	< 0.0001
≥ 3	7.13 (3.96, 12.81)	< 0.0001

Daily parameters were also measured before and 6 hours after proning sessions to determine changes in patients’ clinical stability. Post-proning variables were largely unchanged other than a decrease in heart rate from 81.6 beats per minute to 73.7 beats per minute (Mean difference − 7.87; 95% CI -11.93, -3.81 p = 0.003), increase in PaO2 from 82.7 to 129.4 (Mean difference 36.88, 95% CI 15.73,58.02 p = 0.002) with corresponding increase in P/F ratio from 89.9 to 161.1 (Mean difference 60.58, 95% CI 34.98, 86.19 p = 0.0001) and decrease in FiO2 needs from 89.3 to 75.9 (Mean difference − 13.4, 95% CI -19.16,-7.65 p < 0.0001). Other parameters including blood pressure, PEEP, peak airway pressures, and tidal volumes were unchanged.

There was a total of 15 (28.3%) complications related to proning. Persistent hypoxemia was the most common with 10 (66.7%) occurrences in patients with a complication, followed by hypotension (4, 26.7%) and arrhythmia (3, 20%) (Table 3). No patients were self-extubated or suffered an accidental line removal related to proning throughout the study. Within the patients that were proned, having a lower mean PCO2 (50.0, 10.4 vs 60.8, 21 p = 0.02), lower mean PEEP (11.8, 3.0 vs 13.7, 3.0 p = 0.01), and longer proning session (13.0 hours, 2.4 vs 11.2, 2.6 p = 0.02) was associated with a significantly higher risk of proning complications (Table 4). Younger age (54.1, 12.9 vs 65.5, 10.4 p = 0.001) and higher PEEP (14.6, 3.4 vs 12.5, 2.7 p = 0.04) were associated with being alive at 30 days with shorter time to intubation from admission (3.5 days, 4.4 vs 5.8 days, 5.8 p = 0.08) being nearly significantly associated.

Table 3

Frequencies of complications among proned patients
	Total N (%)
Proning-related complications?
No	38 (71.7)
Yes	15 (28.3)
Complication type
Persistent hypoxia	10 (66.7)
Hypotension	4 (26.7)
Arrhythmia	3 (20.0)

Table 4

Associations between patient characteristics and outcomes (complications from proning and 30-day survival status) among proned patients.
		Proning complications			Alive at 30-days
	Total N (%)	No N (%)	Yes N (%)	p-value	No N (%)	Yes N (%)	p-value
Age (mean, SD)	61.8, 12.3	60.8, 13.5	64.4, 8.5	0.26	65.5, 10.4	54.1, 12.9	0.001
BMI (mean, SD)	34.1, 7.5	34.4, 7.6	33.3, 7.2	0.64	33.1, 7.7	36.2, 6.7	0.17
Sex
Female	23 (43.4)	16 (42.1)	7 (46.7)	0.76	15 (41.7)	8 (47.1)	0.71
Male	30 (56.6)	22 (57.9)	8 (53.3)		21 (58.3)	9 (52.9)
Number of vasopressors
0	24 (45.3)	18 (47.4)	6 (40.0)	0.54	16 (44.4)	8 (47.1)	0.71
1	26 (49.1)	17 (44.7)	9 (60.0)		17 (47.2)	9 (52.9)
≥ 2	3 (5.7)	3 (7.9)	0		3 (8.3)	0
ABG-PCO2 (mean, SD)	57.7, 19.1	60.8, 21.0	50.0, 10.4	0.02	59.7, 21.7	53.2, 10.4	0.15
PO2:FiO2 ratio (mean, SD)	119.1, 54.5	114.6, 46.7	130.2, 71.0	0.41	117.0, 58.8	123.8, 44.9	0.52
PEEP (mean, SD)	13.2, 3.1	13.7, 3.0	11.8, 3.0	0.01	12.5, 2.7	14.6, 3.4	0.04
FiO2 (mean, SD)	91.9, 13.0	93.7, 12.0	87.3, 14.9	0.12	91.9, 12.8	91.8, 13.8	0.92
Peak airway pressure	27.0, 8.5	26.7, 7.7	27.7, 10.8	0.71	26.8, 8.4	27.4, 9.2	0.83
Tidal volume (ml/kg) (mean, SD)	7.01, 2.37	7.1, 2.5	6.7, 2.2	0.52	6.8, 2.1	7.5, 2.9	0.39
≤ 6.0 (N, %)	22 (41.5)	15 (39.5)	7 (46.7)	0.63	16 (44.4)	6 (35.3)	0.53
> 6.0 (N, %)	31 (58.5)	23 (60.5)	8 (53.3)		20 (55.6)	11 (64.7)
Proned again after 48hr interruption
No	41 (77.4)	28 (73.7)	13 (86.7)	0.47	27 (75.0)	14 (82.4)	0.73
Yes	12 (22.6)	10 (26.3)	2 (13.3)		9 (25.0)	3 (17.3)
Received newly initiated dialysis
No	46 (86.8)	36 (94.7)	10 (66.7)	0.02	30 (83.3)	16 (94.2)	0.41
Yes	7 (13.2)	2 (5.3)	5 (33.3)		6 (16.7)	1 (5.9)
Alive on day 30
No	36 (67.9)	24 (63.2)	12 (80.0)	0.33	--	--	--
Yes	17 (32.1)	14 (36.8)	3 (20.0)
	Mean, SD	Mean, SD	Mean, SD		Mean, SD	Mean, SD
Total days proned	2.1, 2.3	3.8, 1.8	4.1, 1.7	0.55	4.1, 1.8	3.6, 1.7	0.37
Total hours proned	47.5, 24.9	44.9, 24.8	54.1, 24.7	0.23	49.4, 26.2	43.4, 22.2	0.42
Average hours per day proned	11.7, 2.6	11.2, 2.6	13.0, 2.4	0.02	11.7, 2.8	11.8, 2.3	0.87
ICU length of stay	17.0, 7.4	17.5, 7.4	15.8, 7.4	0.46	15.3, 6.2	20.6, 8.6	0.01
Total ventilator days (for all intubations)	12.7, 7.1	12.7, 7.5	12.6, 6.2	0.60	10.8, 5.8	16.6, 8.2	0.003
Days from admission to intubation	5.1, 5.4	4.8, 4.0	5.8, 8.1	0.61	5.8, 5.8	3.5, 4.4	0.08
Days from intubation to liberation	11.2, 3.4	10.7, 3.7	12.7, 2.5	0.45	--	--	--
Days from admission to death	16.7, 7.6	15.8, 5.3	18.4, 11.0	0.46	--	16.7, 7.6	--
Days from intubation to death	10.9, 5.5	10.5, 4.9	11.6, 6.7	0.60	--	10.9, 5.5	--

ICU length of stay was similar in the proned and non-proned groups (17 days ± 7.4 vs 15.8 ± 8.3, p = 0.42) as well as total ventilator days (12.7 days ± 7.1 vs 12.3 ± 7.3, p = 0.58) (Table 5). Proned patients were reintubated substantially less than the non-proned group (1.9% vs 19.1%, p = 0.006) while the need for tracheostomy (7.5% vs. 10.6%, p = 0.73) and newly initiated renal replacement therapy (13.2% vs 19.1%, p = 0.42) were similar for the proned and non-proned groups. No differences were observed between the proned and non-proned groups for the 30-day outcomes including: being alive at 30 days (32.1% vs 34.8%, p = 0.78), out of the ICU (70.6% vs 68.8%, p = 0.91), discharged from the hospital (64.7% vs 75%, p = 0.71), and being ventilator-free (70.6% vs 81.3%, p = 0.69).

Table 5

Outcome Variables for Proned vs. Non-proned Patients
	Total N (%)	Not proned N (%)	Proned N (%)	p-value
Re-intubated the same hospital stay for respiratory failure reasons
No	90 (90.0)	38 (80.9)	52 (98.1)	0.006
Yes	10 (10.0)	9 (19.1)	1 (1.9)
Received tracheostomy
No	91 (91.0)	42 (89.4)	49 (92.5)	0.73
Yes	9 (9.0)	5 (10.6)	4 (7.5)
Received newly initiated dialysis
No	84 (84.0)	38 (80.9)	46 (86.8)	0.42
Yes	16 (16.0)	9 (19.1)	7 (13.2)
Alive on day 30
No	66 (66.7)	30 (65.2)	36 (67.9)	0.78
Yes	33 (33.3)	16 (34.8)	17 (32.1)
Missing	1	1
Out of ICU on day 30
No	10 (30.3)	5 (31.3)	5 (29.4)	0.91
Yes	23 (69.7)	11 (68.8)	12 (70.6)
Is the patient discharged on day 30
No	10 (30.3)	4 (25.0)	6 (35.3)	0.71
Yes	23 (69.7)	12 (75.0)	11 (64.7)
Vent free
No	8 (24.2)	3 (18.8)	5 (28.4)	0.69
Yes	25 (75.8)	13 (81.3)	12 (70.6)
	Mean, SD	Mean, SD	Mean, SD	p-value
ICU length of stay	16.4, 7.8	15.8, 8.3	17.0, 7.4	0.42
Total ventilator days (for all intubations)	12.5, 7.2	12.3, 7.3	12.7, 7.1	0.58
Days from admission to intubation	3.6, 4.6	2.0, 2.8	5.1, 5.4	0.0006
Days from intubation to liberation	11.0, 6.1	10.8, 7.9	11.2, 3.4	0.55
Days from admission to death	15.7, 7.3	14.4, 6.8	16.7, 7.6	0.20
Days from intubation to death	11.6, 6.3	12.4, 7.2	10.9, 5.5	0.33

We prospectively sought to better understand the utility of prone positioning in an intubated COVID-19 population to compare outcomes and associated complications to a non-proned patient population. Our study showed that proning an intubated COVID-19 population is safe with minimal risk of complications, associated with more frequent use of neuromuscular blockade, reduced risk of re-intubation for respiratory failure, but was not associated with being alive on day 30, out of the ICU, or ventilator free when compared to a non-proned intubated COVID-19 patient population.

During our study period, the incidence of prone positioning was 53% with variable frequency described in other studies. Mathews et al had a similar incidence with 49% of patients proned ^[13], while other studies described incidences as low as 24%-40% ^{[9, 14]} with others as high as up to 60–70%^{[10, 15]}. Overall, the collective incidence of prone positioning was still significantly higher in these studies than pre-pandemic utilization as was described in the LUNG SAFE and APRONET trials with reported rates from 8–33% ^{[10, 16, 17]}. Increase in prone positioning for COVID-19 compared to previous trials may have been related to lack of evidence-based treatment alternatives for this patient population, especially early on in the pandemic. ICUs also experienced a larger proportion of patients with ARDS physiology and became more comfortable with proning critically ill patients.

While 28% of patients proned experienced a complication, all of these complications were persistent hypoxemia defined as a greater than 10% drop or required ventilator adjustment, hypotension defined as a greater than 20% drop in systolic blood pressure or increase in vasopressor support, or arrythmia. There were no complications involving dislodgement of the patients’ endotracheal tube, central line, arterial line, or foley catheter as well as no increase in development of pressure ulcers. Furthermore, none of the aforementioned complications led to cardiac arrest. The incidence of complications such as tube or line dislodgment, arrythmia, or cardiac arrest were similar to other studies who reported the incidence of complications ^[9], however the significance of persistent hypoxemia, hypotension, and arrhythmia has not been frequently reported in this population We also found a significant lower need for re-intubation in patients who underwent prone positioning which has been minimally described previously ^[9].

Prone positioning was associated with increased use of neuromuscular blockade when compared to non-proned patients and has also been described elsewhere ^{[10, 13]}. Neuromuscular blockade is used in patients with moderate to severe ARDS after being shown to improve ICU mortality, 28-day mortality, more free ventilator days, and less barotrauma ^[18]. It is possible that with the proned population in our study having lower P/F ratio’s, providers could have had a lower threshold to initiate neuromuscular blockade and explain the increased use in the proned group. However, the overall incidence of neuromuscular blocker use was low in both groups and was lower than described in other studies ^[10]. Lower overall use is likely related to the recent findings of the ROSE trial that showed no improvement in 90-day mortality ^[19] with another recent study in COVID-19 showing that neuromuscular blockade use was associated with no improvement in 90 day mortality or ventilator free days, but associated with a longer duration of ventilation and ICU length of stay ^[20].

While proning has been associated with improved mortality in patients with ARDS patients ^[3], its effect specifically in a COVID-19 intubated patient population has been mixed ^{[9, 10, 14]}. Our study showing no improved mortality was consistent with findings of Stilma et al ^[10]. There are a few variables that could explain the variability in findings between studies. The optimal duration of proning and its effect on mortality is not fully known. Recent studies have suggested that prolonged prone positioning of 12–16 hours in COVID-19 intubated patients with ARDS is helpful in refractory hypoxemic respiratory failure ^{[9, 21]} with another study showing prolonged prone positing for at least 24 hours was associated with improved mortality compared to intermittent prone positioning ^[22]. Our average length of prone positioning per day was 11.7 hours and may not have been long enough to show an improvement in mortality. However, Stilma et al had an average of 15 hours a day in patients that were proned and also failed to show a mortality benefit at 28-days ^[10]. The Surviving Sepsis Campaign guidelines on the management of critically ill adults with COVID-19 suggest prone ventilation for moderate to severe ARDS ^[11], but the landmark PROSEVA trial showed mortality benefit in patients with severe ARDS ^[3]. The non-proned patient population in our study had a slightly higher mean P/F ratio than 150 and would have not been included in the original PROSEVA trial. This discordance between P/F ratios in both groups could have affected the results of our study.

There remains an overall paucity of data comparing proning and not proning intubated COVID-19 with ARDS. A recently published systemic review and meta-analysis by Lee et al included only three non-randomized studies of intubated COVID-19 patients ^{[9, 10, 14]} with no randomized controlled trials comparing prone to supine positioning in this patient population ^[8]. Mathews et all also compared the effects of prone positioning within two days of ICU admission, however, was also in an observational, retrospective manor ^[13]. Other studies have included data on proning, but in the context of describing broader characteristics of patients with COVID-19 ARDS ^[15]. While our study contributes to the understanding of prone positioning in intubated COVID-19 patients, a randomized control trial is needed to fully understand the utility of intervention in this patient population.

Prone positioning in mechanically ventilated COVID-19 patients was associated with more frequent use of neuromuscular blockade and sedation, and significantly lower rates of re-intubation for respiratory failure when compared to non-proned patients. While mortality and 30-day survival were similar between both groups, larger, randomized trials are needed to better understand outcomes of proning ventilated COVID-19 patients.

Tables

The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.

The authors have no relevant financial or non-financial interests to disclose

All authors contributed to the study conception and design. Data collection was performed by Jonathan Chandler, Badr Jandali, Dale Smith, and Katie Joyce. Analysis and Tables were performed by Lynn Chollet-Hinton. The first draft of the manuscript was written by Jonathan Chandler and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

IRB approval was obtained from the University of Kansas Medical Center Institutional Review Board and the study was performed in line with the principles of the Declaration of Helsinki.

Calabrese, F., et al., Machine learning-based analysis of alveolar and vascular injury in SARS-CoV-2 acute respiratory failure. J Pathol, 2021. 254(2): p. 173–184.
Lauring, A.S., et al., Clinical severity of, and effectiveness of mRNA vaccines against, covid-19 from omicron, delta, and alpha SARS-CoV-2 variants in the United States: prospective observational study. Bmj, 2022. 376: p. e069761.
Guérin, C., et al., Prone positioning in severe acute respiratory distress syndrome. N Engl J Med, 2013. 368(23): p. 2159–68.
Munshi, L., et al., Prone Position for Acute Respiratory Distress Syndrome. A Systematic Review and Meta-Analysis. Ann Am Thorac Soc, 2017. 14(Supplement_4): p. S280-s288.
Beitler, J.R., et al., Prone positioning reduces mortality from acute respiratory distress syndrome in the low tidal volume era: a meta-analysis. Intensive Care Med, 2014. 40(3): p. 332–41.
Fan, E., et al., An Official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine Clinical Practice Guideline: Mechanical Ventilation in Adult Patients with Acute Respiratory Distress Syndrome. Am J Respir Crit Care Med, 2017. 195(9): p. 1253–1263.
Rampon, G.L., S.Q. Simpson, and R. Agrawal, Prone Positioning for Acute Hypoxemic Respiratory Failure and ARDS: A Review. Chest, 2022.
Lee, H.J., et al., Efficacy and safety of prone position in COVID-19 patients with respiratory failure: a systematic review and meta-analysis. Eur J Med Res, 2022. 27(1): p. 310.
Shelhamer, M.C., et al., Prone Positioning in Moderate to Severe Acute Respiratory Distress Syndrome Due to COVID-19: A Cohort Study and Analysis of Physiology. J Intensive Care Med, 2021. 36(2): p. 241–252.
Stilma, W., et al., Incidence and Practice of Early Prone Positioning in Invasively Ventilated COVID-19 Patients-Insights from the PRoVENT-COVID Observational Study. J Clin Med, 2021. 10(20).
Alhazzani, W., et al., Surviving Sepsis Campaign: Guidelines on the Management of Critically Ill Adults with Coronavirus Disease 2019 (COVID-19). Crit Care Med, 2020. 48(6): p. e440-e469.
Ranieri, V.M., et al., Acute respiratory distress syndrome: the Berlin Definition. Jama, 2012. 307(23): p. 2526–33.
Mathews, K.S., et al., Prone Positioning and Survival in Mechanically Ventilated Patients With Coronavirus Disease 2019-Related Respiratory Failure. Crit Care Med, 2021.
Chen, Y., et al., Prone positioning in intubated and mechanically ventilated patients with SARS-CoV-2. J Clin Anesth, 2021. 71: p. 110258.
Clinical characteristics and day-90 outcomes of 4244 critically ill adults with COVID-19: a prospective cohort study. Intensive Care Med, 2021. 47(1): p. 60–73.
Guérin, C., et al., A prospective international observational prevalence study on prone positioning of ARDS patients: the APRONET (ARDS Prone Position Network) study. Intensive Care Med, 2018. 44(1): p. 22–37.
Bellani, G., et al., Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. Jama, 2016. 315(8): p. 788–800.
Neto, A.S., et al., Neuromuscular blocking agents in patients with acute respiratory distress syndrome: a summary of the current evidence from three randomized controlled trials. Ann Intensive Care, 2012. 2(1): p. 33.
Moss, M., et al., Early Neuromuscular Blockade in the Acute Respiratory Distress Syndrome. N Engl J Med, 2019. 380(21): p. 1997–2008.
Li Bassi, G., et al., Early short course of neuromuscular blocking agents in patients with COVID-19 ARDS: a propensity score analysis. Crit Care, 2022. 26(1): p. 141.
in Coronavirus Disease 2019 (COVID-19) Treatment Guidelines. 2021, National Institutes of Health (US): Bethesda (MD).
Okin, D., et al., Prolonged Prone Position Ventilation Is Associated With Reduced Mortality in Intubated COVID-19 Patients. Chest, 2022.

No competing interests reported.

Outcomes of Prone Positioning in Mechanically Ventilated COVID-19 patients: A Prospective Observational Study

Status:

Version 1

Abstract

Introduction

Methods

Results

Conclusion

Introduction

Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Status:

Version 1