Pulmonary Hemodynamics and Ventilation in Patients With COVID-19-Related Respiratory Failure and ARDS

Abstract

Background:

It has been suggested that COVID-19-associated severe respiratory failure (CARDS) might differ from usual acute respiratory distress syndrome (ARDS) due to failing autoregulation of pulmonary vessels and higher shunt. We sought to investigate pulmonary hemodynamics and ventilation properties in patients with CARDS compared to patients with ARDS of pulmonary origin.

Methods:

This was a retrospective analysis of prospectively collected data from consecutive adults with laboratory-confirmed severe acute respiratory syndrome coronavirus 2 patients treated in our ICU in 04/2020 and a comparison of the data to matched controls with ARDS due to respiratory infections treated in our ICU from 01/2014 to 08/2019 for whom pulmonary artery catheter data were available.

Results:

CARDS patients (n = 10) had ventilation characteristics similar to those of ARDS (n = 10) patients. Nevertheless, mechanical power applied by ventilation was significantly higher in CARDS patients (23.4 ± 8.9 J/min) than in ARDS (15.9 ± 4.3 J/min; P < 0.05). COVID-19 patients had similar pulmonary artery pressure but significantly lower pulmonary vascular resistance, as cardiac output was higher in CARDS vs. ARDS patients (P < 0.05). Shunt fraction and dead space were similar in CARDS compared to ARDS (P > 0.05) and were correlated with hypoxemia in both groups. The arteriovenous pCO₂ difference (▵pCO₂) was elevated (CARDS 5.5 ± 2.8 mmHg vs. ARDS 4.7 ± 1.1 mmHg; P > 0.05), as was the P_(v-a)CO₂/C_(a-v)O₂ ratio (CARDS mean 2.2 ± 1.5 vs. ARDS 1.7 ± 0.8; P > 0.05).

Conclusions:

Respiratory failure in COVID-19 patients seems to differ only slightly from ARDS regarding ventilation characteristics and pulmonary hemodynamics. Our data indicate microcirculatory dysfunction. More data need to be collected to assure these findings and gain more pathophysiological insights into COVID-19 and respiratory failure.

Keywords

acute respiratory distress syndrome mechanical ventilation novel coronavirus disease 2019 COVID-19 pulmonary artery catheter

Introduction

Severe acute respiratory syndrome-coronavirus 2 (SARS-CoV-2) causes a disease termed COVID-19. Symptoms range from mild upper airway symptoms to acute respiratory distress syndrome (ARDS). There are hints that some characteristics of respiratory failure in COVID-19 patients are different from acute respiratory distress syndrome, as we know it from other causes, e.g. viral or bacterial infection. We and others observed relatively normal respiratory mechanics¹ despite poor oxygenation. It has been suggested that in COVID-19-related ARDS (CARDS), this is in part due to defective pulmonary vasoregulation so that hypoxic pulmonary vasoconstriction does not occur, leading to ventilation/perfusion mismatch and consequently may lead to profound hypoxemia.² These findings and hypotheses led to the question of whether our present understanding of ARDS with regard to pulmonary hemodynamics and right heart failure can be transferred to CARDS.

Previous reports suggest that COVID-19 induces the massive release of cytokines, including interleukin 6 (IL-6), and hence, it has been suggested to treat COVID-19 by IL-6 blockade. Increased pulmonary vascular resistance is a consequence of various factors, including hypercarbia, acidosis, inflammation, vascular endothelial proliferation³ and hypoxic pulmonary vasoconstriction. There is ample evidence that perivascular inflammation triggered by various mediators, including IL-6, contributes to pulmonary vascular disease resulting in pulmonary hypertension. Pulmonary hypertension due to increased pulmonary vascular resistance is a common finding in ARDS.^4,5 The exact prevalence of pulmonary hypertension in ARDS is not known^4,6 and varies considerably across studies. These differences may result from general differences between cohorts, cardiovascular comorbidities, etiology of ARDS and therapeutic interventions (e.g., permissive hypercapnia). ARDS is normally characterized by low-pressure pulmonary edema, shunt-related hypoxemia and reduced aerated lung volume, causing severely impaired lung compliance. It is widely accepted that ARDS patients with elevated pulmonary pressure are at risk of dying from right heart failure.⁷

In the present study, we compared CARDS patients to ARDS patients with regard to hemodynamic and derived physiological parameters to elucidate whether there are distinct features in CARDS that separate CARDS and ARDS with regard to ventilation and pulmonary hemodynamics.

Methods

This observational case-control study describes data of extended hemodynamic monitoring via pulmonary artery catheterization of 10 consecutive patients with laboratory-confirmed COVID-19-related respiratory failure (CARDS) treated in our tertiary-care ICU in April 2020. Patients were mainly transferred for the possibility of ECMO support or for further management from hospitals under the surge of COVID-19 in near-by France. Furthermore, we gathered the characteristics of mechanical ventilation, arterial blood gas and laboratory findings in these patients. The characteristics of ventilation and pulmonary hemodynamics were 1:1 compared to patients with ARDS due to pneumonia (n = 10). Groups were matched according to disease severity using the simplified acute physiology score II (SAPS II), P_aO₂/F_iO₂ ratio, age, sex and body mass index (BMI).

The data of the ARDS group was collected from an electronic patient recording system. Measurements were part of routine intensive care therapy.

Ventilation was in pressure-controlled mode in all patients. Generally, we targeted a protective ventilation approach, aiming at driving pressures below 15 cmH₂O. The mean arterial pressure target is mainly 60-65 mmHg if physiological aims are reached. These include capillary refill time (i.e. warm periphery), urinary output (≥0.5 mL/kg/h) or lactate levels (≤2.0 mmol/L). Norepinephrine was the vasopressor of choice in patients with pH ≥ 7.25. Inotrope use was initiated if S_cvO₂ was below 65% despite adequate hemoglobin levels. Nutrition in both groups was done according to the same standards.

Measurements and Data Recording

Extended hemodynamic monitoring was established synchronized with collection of blood gas analysis. Briefly, after inserting a sheath catheter, a 6F balloon-tipped Swan-Ganz catheter (Edwards Lifesciences Corp., Irvine, CA, USA) was advanced to the pulmonary artery until a curve characteristic of pulmonary artery occlusion pressure (PAOP) pressure could be derived. Calibration of the hydraulic monitoring system with zero-level at mid-thoracic line was done prior to each measurement. PAOP was measured at end-expiration (preferably at the beginning of QRS in the ECG). Values were averaged over 3 respiratory cycles. For the historical cohort of ARDS patients, no graphs were available to assess pulmonary hemodynamics.

Arterial and venous blood gases with parameters such as partial pressures of oxygen (pO₂) and carbon dioxide (pCO₂), hemoglobin, lactate, glucose, and pH were obtained via analysis on an ABL800 Blood Gas Analyzer (Radiometer, Copenhagen, Denmark).

The characteristics of ventilation correspond to the moment of data collection. Laboratory parameters have been collected on the same day as pulmonary artery catheterization. Protocols for hemodynamic measurement were identical for both cohorts.

Calculation of Hemodynamic and Physiological Parameters

Cardiac output (CO) was measured in thermodilution technique via the pulmonary arterial catheters. Patients were placed in the same position as during pulmonary pressure measurements. Three different measurements were accepted if there was a stable baseline and a regular curve. Further, the measurements were accepted, if the difference between cardiac output measurements was not greater than 0.5 L/min under stable conditions. In CARDS patients there was no use of positive inotrope substances, in ARDS patients 3 patients received dobutamine. The cardiac index (CI) relates CO to body surface area, calculated according to the Du Bois formula. The following variables were calculated: stroke volume (SV = CO/HR, mL); pulmonary pulse pressure (PP = sPAP-dPAP, mmHg); transpulmonary pressure gradient (TPG = mPAP-PAWP, mmHg); diastolic pressure gradient (DPG = dPAP-PAWP, mmHg); pulmonary vascular resistance (PVR = TPG/CO Wood units: mmHg*min/L, or mmHg*s/mL); pulmonary arterial compliance (C_PA = SV/PP, mL/mmHg); and time constant (RC time, τ = C_PA × PVR, s).

Furthermore, we calculated the ratio of the venous–arterial CO₂ gap to arterial-venous oxygen content difference (P_(v-a)CO₂/C_(a-v)O₂) as (P_vCO₂ − P_aCO₂)/(C_aO₂ − C_vO₂) with C_aO₂ = (1.34 mL/g × S_aO₂ × Hb) + (0.003 mL/(dL*mmHg) × P_aO₂) and C_vO₂ = (1.34 mL/g × S_vO₂ × Hb) + (0.003 mL/(dL*mmHg) × P_vO₂). DO₂ was calculated as CO × C_aO₂ and VO₂ as CO × (C_aO₂ − C_vO₂). The P_(v-a)CO₂ gap was calculated as P_vCO₂ − P_aCO₂. The calculation of the estimated dead space fraction was performed using a rearranged alveolar gas equation V_D/V_T = 1 − [(0.86 mmHg × VCO_2est)/(VE×P_aCO₂)], where VCO_2est is the estimated CO₂ production calculated gender specific from the Harris Benedict equation.⁸ Minute ventilation (VE) is obtained from the ventilator rate times expired tidal volume and P_aO₂ and P_aCO₂ from arterial gas analysis. The ventilatory ratio (VR), as a measure of ventilator efficiency, was calculated as (VE × P_aCO₂) divided by (100 mL/min × predicted body weight (PBW; kg) × 40 mmHg (expected P_aCO₂)).⁹ Predicted body weight was calculated according to the ARDSNet PBW calculator.¹⁰ The pulmonary shunt fraction was calculated as Q_S/Q_T = (C_cO₂-C_aO₂)/(C_cO₂-C_vO₂). The alveolar-arterial O₂ gradient (A-aDO₂) was calculated as [F_iO₂ × (barometric pressure – 47 mmHg) – P_aCO₂/0.81] - [(P_aO₂ + age in years/4*mmHg) + 4]. Barometric pressure was recorded on the day of pulmonary artery catheterization.

Mechanical power applied by ventilation was calculated by multiplying each component of the equation of motion by the minute ventilation using the following equation: mechanical power = 0.098 × tidal volumes × respiratory rate × (Ppeak – 0.5 × driving pressure) in J/min.

Statistical Analysis

Data are reported as median (range) or median (range; mean ± SD), if not otherwise indicated. Statistical analysis was performed with GraphPad Prism 5.02 (GraphPad Software, Inc., La Jolla, CA). An unpaired t-test was used to test for differences between the groups, as all groups showed a Gaussian distribution in the Kolmogorov-Smirnov test. P-values <0.05 were considered significant.

Results

Baseline Demographics and Laboratory Findings

All patients in the CARDS group received mechanical ventilation as a result of SARS-CoV-2 infection with pneumonia and respiratory failure. These patients were matched to a historical cohort of ARDS patients treated in our ICU. The SAPS II score was 42 in the COVID-19 group (n = 10) and 41 in the ARDS group (n = 10). Patients in both groups had mild to severe ARDS, according to the Berlin classification of ARDS.¹¹ P_aO₂/F_iO₂ did not differ between the 2 groups (118 vs. 120 in median; P > 0.05). The median age in CARDS patients was 62 years (range 59-71) compared to 71 in ARDS patients (range 45-85), P > 0.05. Nine out of 10 CARDS and 8 out of 10 ARDS patients were male. Three of 10 COVID-19 patients had obesity I°, 3 had normal BMI and 4 patients had light overweight; however, there was no difference in ARDS patients. There was a statistically significant difference in body temperature during hemodynamic measurement. While ARDS patients had no fever (median 37.4°C), every CARDS patient showed a body temperature above 38.1°C (median 38.5°C).

The clinical characteristics and laboratory findings of the patients are shown in Table 1 . In CARDS patients, CRP and interleukin-6 were elevated. The median procalcitonin level was 0.69 ng/mL (0.09-276 ng/mL). CRP in ARDS patients was slightly lower than in CARDS patients (104 vs. 234, P > 0.05). In CARDS, D-dimer levels were elevated at 3.2 mg/L (range 0.9-19.4 mg/L). Thrombocyte levels were in the normal range. Renal function (creatinine 1.13 mg/dL) and liver function (bilirubin 1.15 mg/dL) were slightly decreased in both groups.

Table 1.

Baseline Characteristics and Laboratory Findings of CARDS Patients (N = 10) Compared to Patients With ARDS Due to Pulmonary Infections (N = 10) With Invasive Hemodynamic Measurement (Interleukin 6 and PCT Were Not Part of the Laboratory Routine for Patients With Usual ARDS at the Time of Evaluation).

Laboratory findings	CARDS; N = 10 median (range; mean ± SD)	ARDS; N = 10 median (range; mean ± SD)	P value
Cause of ARDS	SARS-CoV2	7/10 CAP; 3/10 HAP
SAPS	42 (29-54; 42 ± 8)	41 (25-51; 38 ± 8)	>0.05
P_aO₂/FiO₂	120 (57-171; 114 ± 42)	118 (73-259; 142 ± 65)	>0.05
Age [years]	62 (59-71; 62.6 ± 5.1)	71 (45-85; 69 ± 12.2)	>0.05
Male [N;%]	9/10; 90%	8/10; 80
BMI [kg/m²]	28.5 (24-33; 28 ± 3)	27 (22-56; 32 ± 11.4)	>0.05
Norepinephrine	4/10; 40%	6/10; 60%
Hemoglobin [g/dL]	9.4 (7.1-14; 10.0 ± 2.2)	9.5 (7.1-12-2; 9.8 ± 1.8)	>0.05
Leukocytes [G/L]	9.45 (6.1-24.1; 11 ± 5)	12.6 (1.7-31.4; 14 ± 8)	>0.05
Thrombocytes [G/L]	186 (71-320; 190 ± 64)	137 (18-455; 165 ± 129)	>0.05
CRP [mg/L]	234 (94-399; 233 ± 115)	104 (35-447; 152 ± 117)	>0.05
Creatinin [mg/dL]	1.1 (0.7-3.2; 1.3 ± 0.7)	1.9 (0.5-4.6; 2.1 ± 1.3)	>0.05
Bilirubin [mg/dL]	1.2 (0.3-7; 1.6 ± 2)	0.6 (0.2-1.5; 0.7 ± 0.4)	>0.05
Interleukin-6 [pg/mL]	471 (11-2397; 828 ± 940)	n.a.
PCT [ng/mL]	0.7 (0.09-276; 28 ± 87)	n.a.
D-Dimer [mg/L]	3.2 (0.85-19.35; 5.7 ± 5.4)	n.a.

Abbreviations: BMI, body mass index; CARDS, COVID-19 associated ARDS; CAP, community acquired pneumonia; HAP, hospital acquired pneumonia; SAPS, implified acute physiology score; CRP, C-reactive protein; NT-proBNP, N-terminal pro-B-type natriuretic peptide; PCT, procalcitonin.

Characteristics of Ventilation and Blood Gas Analysis in COVID-19 and Comparison to ARDS Patients

All patients received ventilation in a pressure-controlled mode. The plateau pressure was 25 cmH₂O (range 20-32) in CARDS and 23 cmH₂O (range 20-28) in ARDS patients. PEEP and ΔP did not differ significantly between the 2 groups (CARDS PEEP: 10 cmH₂O; ΔP 14 cmH₂O vs. ARDS PEEP 12 cmH₂O; ΔP 18 cmH₂O; P > 0.05). Tidal volumes resulting from these pressure characteristics were 9 mL/kg in CARDS vs. 8 mL/kg in ARDS. None of the CARDS patients had tidal volumes less than 400 mL (median 613 mL; range 415 – 994 mL). Higher tidal volume in CARDS patients was a result of better dynamic compliance in these patients. Dynamic lung compliance was 42 mL/cmH₂O (range 27-77 mL/cmH₂O) in CARDS patients compared to 37 mL/cmH₂O (range 23-67 mL/cmH₂O) in ARDS patients (P > 0.05). Only in 4 COVID-19 patients dynamic lung compliance was less than 40 mL/cmH₂O. Fifty percent of the patients needed a PEEP equal to or higher than 10 cmH₂O to maintain adequate oxygenation. Gas exchange was different between groups CARDS and ARDS, with higher P_aCO₂ in the CARDS group (52 mmHg vs. 41 mmHg; P > 0.05), causing mild respiratory acidosis in CARDS (pH 7.33 vs. 7.46; P < 0.05) despite higher minute ventilation volume. The mechanical power applied by ventilation in CARDS patients was significantly higher than that in ARDS patients (P < 0.05). Table 2 shows the characteristics of ventilation in CARDS patients compared to patients with usual ARDS.

Table 2.

Mechanical Ventilation Characteristics of CARDS Patients (n = 10) Compared to Mechanical Ventilation Characteristics of Patients With ARDS (n =10).^a

Patient characteristics	CARDS; N = 10 median (range; mean ± SD)	ARDS; N = 10 median (range; mean ± SD)	P value
Body temperature [°C]	38.5 (38.1-39.3; 38.6 ± 0.3)	37.4 (36.2-38.2; 37.3 ± 0.7)	<0.05
Compliance [mL/cmH₂O]	42 (27-77; 44 ± 17)	37 (23-67; 39 ± 13)	>0.05
PEEP [cmH₂O]	10 (8-14; 10 ± 2.3)	10 (5-14; 9 ± 3)	>0.05
ΔP [P_plat-PEEP]	14 (12-22; 16 ± 3.7)	15 (8-18; 15 ± 3.2)	>0.05
Respiratory rate [/min]	18 (11-36; 21 ± 8)	18 (16-22; 19 ± 2)	>0.05
Vt [mL/kg]	9 (5-14; 9 ± 2.4)	8 (6-10; 8 ± 1.1)	<0.05
VE [L/min]	13.0 (6.7-20.3; 13.4 ± 4.2)	10.5 (7.5-13.2; 10.1 ± 2)	<0.05
Mechanical power [J/min]	20.8 (12.2-38.3; 23.4 ± 8.9)	14.1 (11.8-23.9; 15.9 ± 4.3)	<0.05
P_aO₂ [mmHg]	77 (57-97; 75 ± 14)	75 (67-115; 83 ± 16.5)	>0.05
P_aCO₂ [mmHg]	52 (33-74; 51 ± 12)	41 (26-73; 43 ± 12)	>0.05
pH	7.33 (7.29-7.43; 7.33 ± 0.05)	7.42 (7.32-7.53; 7.43 ± 0.1)	<0.05

Abbreviations: CARDS, Covid19-related acute respiratory distress syndrome; PEEP, positive end expiratory pressure; ΔP, difference between plateau pressure and PEEP; Vt, tidal volume; MV, minute volume; Compliance in mL/cmH2O.

^a P values <0.05 were considered significant.

Hemodynamic Monitoring in CARDS Vs. ARDS Patients

We compared pulmonary hemodynamics as assessed by pulmonary artery catheters in CARDS compared to ARDS patients ( Table 3 ). The median time between the diagnosis of respiratory failure and PAC insertion was 13 days in CARDS patients. ARDS patients received PAC insertion a median of 7 days from diagnosis. The time between intubation and PAC was 11 days (1 – 21 days) in CARDS patients and 3 (0-9 days) in ARDS patients (P < 0.05). This delay was mainly because most CARDS patients were diagnosed with respiratory failure, intubated, and transferred to our ICU from other hospitals.

Table 3.

Invasive Hemodynamics Acquired by Pulmonary Artery Catheterization of CARDS Patients (N = 10) Compared to ARDS of Other Causes. (N = 10) Differences in Time From Intubation to PAC Is Due to the Fact That Most Patients With CARDS Were Referred to Our Center.^a

Variable	CARDS; N = 10 median (range; mean ± SD)	ARDS; N = 10 median (range; mean ± SD)	P-value
Days from intubation to PAC	11 (1-21; 10 ± 6)	3 (10-9; 3 ± 2.9)	<0.05
P_aO₂/F_iO₂ ratio	120 (57-171; 114 ± 42)	134 (73-259; 149 ± 62.1)	>0.05
PAPs [mmHg]	53 (23-62; 47 ± 13.3)	46 (27-61; 44 ± 9)	>0.05
PAPd [mmHg]	24 (11-31; 23 ± 6.1)	25 (15-34; 25 ± 5.5)	>0.05
PAPm [mmHg]	36 (16-44; 32 ± 8)	34 (20-46; 33 ± 6.8)	>0.05
PCWP [mmHg]	18 (8-26; 17 ± 5.9)	16 (9-28; 17 ± 6.8)	>0.05
PVR [dynscm^-5]	133 (13-329; 139 ± 77)	248 (101-385; 242 ± 89.2)	<0.05
WU*m² [mmHg/L/min/m²]	1.7 (0.2-4.1; 1.7 ± 1)	3.1 (1.3-4.8; 3 ± 1.1)	<0.05
DPG [mmHg]	7 (1-10; 6 ± 5)	8 (3-19; 7 ± 4.6)	>0.05
TPG [mmHg]	15 (8-26; 15 ± 5.1)	15 (8-25; 15 ± 4.2)	>0.05
C_PA [mL/mmHg]	4.1 (2-11; 4.6 ± 2.4)	2.8 (2-6.1; 3.3 ± 1.3)	>0.05
C_{PA x} PVR (τ) in seconds	7.3 (0.9-11.9; 6.9 ± 3.1)	9.0 (3.3-16.6; 9.1 ± 3.3)	>0.05
SVR [mmHgminml^-1]	783 (515-1,171; 801 ± 168)	1,374 (539-1,985; 1,240 ± 481)	<0.05
CO [L/min]	7.8 (6.4-11.8; 8.1 ± 1.7)	5 (1.9-8.5; 5.4 ± 1.5)	<0.05
CI [L/min/m²]	3.9 (2.7-6.4; 4.1 ± 1.1)	2.7 (1.7-3.8; 2.7 ± 0.7)	<0.05
SV [mL]	95 (60-126; 95 ± 19)	50 (36-117; 61 ± 23.4)	<0.05
NT-proBNP [pg/mL]	365 (70-77,275; 336 ± 23,082)	2671 (264-50,403; 8,996 ± 14,462)	>0.05

Abbreviations: CARDS, COVID-19-related acute respiratory distress syndrome; SAPS, simplified acute physiological score; PAPs, systolic pulmonary arterial pressure; PAPd, diastolic pulmonary arterial pressure; PAPm, median pulmonary arterial pressure; PCWP, pulmonary capillary wedge pressure; PVR, pulmonary vascular resistance; DPG, diastolic pulmonary gradient; TPG, transpulmonary pressure gradient; SVR, ystemic vascular resistance; CO, cardiac output; CI, cardiac index; C_PA, pulmonary artery compliance.

^a P values <0.05 were considered significant.

Pulmonary arterial hypertension was common in both CARDS and ARDS patients. Nine out of 10 CARDS and ARDS patients had elevated mPAP values above 25 mmHg. Systolic and diastolic pulmonary arterial pressure did not differ between groups (P > 0.05). Both groups had partial post-capillary pulmonary hypertension with pulmonary artery occlusion pressures (PAOP) higher than 15 mmHg. ARDS patients had combined pre- and post-capillary pulmonary hypertension (PVR > 3 WU, DPG ≥ 7), while CARDS patients showed more isolated post-capillary pulmonary hypertension (median PVR ≤ 3 WU, median DPG = 7 mmHg). Only one patient in the CARDS group showed DPG ≥ 7 mmHg and PVR > 3 WU meeting the criteria of pre- and post-capillary pulmonary hypertension.¹² PEEP was not correlated with PVR in either group (P > 0.05). The pulmonary vascular compliance C_PA as part of right ventricular afterload showed no significant difference, being 4.1 mL×mmHg^-1 in CARDS patients vs. 2.8 mL × mmHg^-1 in ARDS patients. There was a negative correlation between PVR and C_PA. The product of PVR × C_PA was thus similar between groups (P > 0.05); however, the higher C_PA in CARDS patients indicated sufficient RV-PA coupling. Indeed, no CARDS patient had right heart failure.

In both groups, cardiac output was in a normal range. Patients with CARDS had significantly higher cardiac output (P < 0.05) and slightly higher stroke volume (P > 0.05) than ARDS patients. Systemic vascular resistance (SVR) was significantly higher in patients with ARDS than in patients with CARDS (P < 0.05). No CARDS patient received inotropes. In the CARDS group, 4 patients received norepinephrine during measurement, and in the ARDS group, 6 patients received norepinephrine during measurement.

Physiological Parameters Derived From Pulmonary Artery Catheter Measurements and Blood Gas Analysis in Patients With CARDS and ARDS

As patients with CARDS presented with elevated p_aCO₂, we calculated the estimated CO₂ production for both groups (VCO_2est.). Indeed, patients with CARDS had a statistically significantly higher estimated CO₂ production. Hence, the statistically significant higher minute ventilation noted in CARDS patients could not compensate for the higher production. The ventilatory ratio (VR) was higher in patients with CARDS, but the difference did not reach statistical significance. The same was true for physiological dead space (V_D/V_T). V_D/V_T was strongly correlated with VR in patients with CARDS (P < 0.05; Figure 1A ) but not in patients with ARDS. In CARDS patients, there was a negative correlation between P_aO₂/F_iO₂ and ventilatory ratio (P < 0.05; Figure 1 B ). The calculated shunt fraction was not different between groups.

Figure 1.

(A) Correlation between ventilatory ratio (VR) and dead space fraction (VD/VT). There was a statistically significant association between VR and VD/VT in CARDS patients (P < 0.05). (B) Correlation between PaO2/FiO2 ratio and VR.

The ratio of the venous–arterial CO₂ gap to arterial-venous oxygen content difference (P_(v-a)CO₂/C_(a-v)O₂ ratio) was not different between groups; however, patients with CARDS had a higher ratio than patients with ARDS (2.0 vs. 1.6). Additionally, the difference between central-venous and arterial CO₂ (▵pCO₂) was different between groups, with a higher difference in CARDS patients (6.2 vs. 4.7 mmHg), but this difference did not reach statistical significance. Lactate levels, mixed venous oxygenation and alveolar-arterial oxygen difference (AaDO₂) were similar in both groups. The parameters are summarized in Table 4 .

Table 4.

Physiological Parameters Derived From Measurements in CARDS Patients (N = 10) and ARDS of Other Causes (N = 10) Acquired Swan-Ganz Catheterization and Blood Gas Analysis. Barometric Pressure Was Recorded on the Day of the Actual Swan-Ganz Catheterization.^a

Variable	CARDS; N = 10 median (range; mean ± SD)	ARDS; N = 10 median (range; mean ± SD)	P-value
P_(v-a)CO₂/C_(a-v)O₂	2.0 (0.5-5.6; 2.2 ± 1.5)	1.6 (0.3-3.3; 1.7 ± 0.8)	>0.05
▵pCO₂ [mmHg]	6.2 (1.2-9.8; 5.5 ± 2.8)	4.7 (2.5-5.3; 4.7 ± 1.1)	>0.05
s_vO₂ [%]	73 (69-79; 73 ± 3)	68 (46-79; 68 ± 9.6)	>0.05
Lactate (art.)	1.4 (0.7-1.9; 1.4 ± 0.3)	2.1 (0.6-3.4; 2 ± 1)	>0.05
AaDO₂ [mmHg]	278 (30-551; 298 ± 146)	298 (79-546; 282 ± 159)	>0.05
VCO_2est [mL/min]	347 (220-382; 336 ± 47)	277 (217-374; 287 ± 51)	<0.05
VE [L/min]	13.0 (6.7-20.3; 13.4 ± 4.2)	10.5 (7.5-13.2; 10.1 ± 2)	<0.05
Ventilatory ratio	1.9 (1.0-3.2; 2.0 ± 0.8)	1.5 (0.8-2.6; 1.5 ± 0.6)	>0.05
Shunt Fraction [%]	23 (8-50; 26 ± 13)	19 (7-55; 23 ± 15)	>0.05
V_D/V_T [%]	50 (11-69; 46 ± 20)	41 (10-68; 43 ± 18)	>0.05

For calculation of parameters, please see methods section.

^a P values <0.05 were considered significant.

Discussion

The present study compared patients with CARDS to patients with ARDS. The main findings of our study are as follows: (i) pulmonary hemodynamics are not profoundly altered in CARDS compared to usual ARDS, (ii) autoregulation of lung vessels seems to be largely intact in patients with SARS-CoV2-related lung failure, and (iii) mechanical ventilation intensity was significantly higher in CARDS compared to usual ARDS.

Usual ARDS is defined by decreased diffusion capacity and ventilation deficit as a consequence of inflammation and low compliance,⁷ resulting in a “baby lung.” In addition, there is often pulmonary hypertension due to increased vascular resistance with the risk of consecutive right heart failure.⁴ In COVID-19 patients admitted to our ICU, we could not completely confirm these pathophysiological findings.

Further p_aCO₂ in COVID-19 patients was rather similar compared to patients with ARDS at 52 mmHg and 41 mmHg, respectively. Our data show that all COVID-19 patients had tidal volumes and compliance resulting in minute ventilation of 6 L/min to 20 L/min. These findings might not be typical for patients with ARDS. In these patients, tidal volume is low because lung compliance is decreased.¹³ As CO₂ has a higher diffusion coefficient than O₂, hypercapnia is normally a problem of ventilation rather than diffusion.¹⁴ In usual ARDS, hypercapnia can be explained by low tidal volume and the subsequent lack of alveolar ventilation. However, in CARDS patients, minute ventilation was not decreased. Our group of CARDS patients had significantly higher tidal volume and higher dynamic compliance than ARDS patients. Matching these 2 groups by P_aO₂ and oxygenation index, these parameters were quite similar in both groups. The question that arises is the pathophysiology behind hypercapnia despite high minute ventilation and preserved compliance. V/Q mismatch could be an explanation for the ineffective gas exchange in CARDS patients. Additionally, non-effective regulation of hypoxic vasoconstriction¹⁰ could be another reason why blood volume passes non-ventilated areas in the lungs and is not able to participate in gas exchange. However, we could not find differences between CARDS and ARDS with regard to calculated physiological dead space and shunt fraction. The calculated dead space in CARDS and ARDS patients was in the range expected according to the P_aO₂/F_iO₂ ratio.¹⁵ However, we found a significant difference in the estimated CO₂ production, which might explain at least some of the CARDS features. Indeed, the intensity of ventilation was significantly higher in CARDS patients than ARDS patients. Nevertheless, the intensity in CARDS patients was not surprisingly high and comparable to previously reported data in patients with ARDS.¹⁶

In ARDS patients, there is often pulmonary hypertension as a consequence of hypoxia or hypercapnia. Furthermore, there is a higher rate of right heart failure and lower cardiac output in ARDS patients because of a higher right ventricular afterload.^17,18 In CARDS patients, we did not observe signs of right heart failure or decreased cardiac output. The cardiac output in CARDS patients was 7.8 L/min in median compared to 6.1 L/min in ARDS patients (cardiac index 3.9 L/min/m² vs. 3.1 L/min/m²). Pulmonary vascular resistance was significantly lower in CARDS patients than in ARDS patients (133 dyn × sec × cm^-5 in COVID-19 vs. 248 dyn × sec × cm^-5 in ARDS patients). One reason for increased PVR is higher PEEP. It has been shown previously that increasing PEEP increases PVR in mechanically ventilated patients.¹⁹ In our study, PEEP was similar in both groups, PAOP did not differ, and there was no correlation between PEEP and PVR. The difference in PVR may thus simply result from increased cardiac output of CARDS patients. CARDS patients probably had an increased cardiac output in comparison to ARDS patients due to inflammation and low systemic vascular resistance.

Mean pulmonary arterial pressure was similar in COVID-19 patients and in patients with ARDS; nevertheless, both groups showed pulmonary arterial hypertension to some degree. Both groups had postcapillary forms of pulmonary hypertension with PAOP > 15 mmHg, but the median diastolic pressure gradient was 7 mmHg in CARDS patients vs. 8 mmHg in ARDS patients. With a DPG less than 7 mmHg, there is no precapillary component in pulmonary hypertension according to the actual guidelines.¹² In our study, the median DPG was 7 mmHg, and PVR was 1.7 WU × m² (vs. DPG 8 mmHg, PVR 3.1 WU × m² in ARDS), meaning that pulmonary arterial hypertension in CARDS might be more a result of a postcapillary than a precapillary component. The transpulmonary gradient (TPG) that may be useful to differentiate between pulmonary hypertension as result of passive or reactive mechanisms did not differ between the 2 groups. In both groups the TPG was 15 mmHg, indicating a reactive component of pulmonary hypertension. Hence, there was only a slight difference, and a defect in pulmonary vascular regulation in the presence of hypoxemia and hypercapnia unable to avoid a pathologically high right-left shunt²⁰ can be ruled out. We did not calculate an unexpectedly high shunt in CARDS patients. Rather, elevated CO₂ production might explain differences between CARDS and ARDS. Patients presented with prolonged fever and unexpected long periods of elevated inflammatory parameters, both of which might contribute to increased VCO₂.

Early recommendations for COVID-19 patients with respiratory failure consisted of early invasive ventilation strategies to avoid contagious aerosols and the use of a high PEEP analogue to usual ARDS.²¹ These therapies may be most effective in patients with lower diffusion capacity and compliance, but the benefit in ventilation/perfusion mismatch needs to be questioned. These recommendations may be effective in ARDS patients; however, in CARDS patients, an important factor according to our data might be to limit excessive CO₂ production, e.g. by rigorous temperature control.

Moreover, veno-venous extracorporeal life support (vvECMO) should be considered to treat these patients. Six patients in the CARDS group received vvECMO therapy later because of increasing hypercapnia and pH < 7.1 at exhausted mechanical ventilation.

Limitations

Our study has several limitations that need to be considered. First, the sample size of this analysis is relatively small. However, PAC is only used in special situations in ARDS patients in clinical practice, rendering sample sizes small. Differences between groups in important values might be more pronounced if the sample size was larger. A critical parameter such as P_(v-a)CO₂/C_(a-v)O₂ might in fact reach statistical significance in larger samples. The fact that SARS-CoV-2 is still causing a global pandemic could be a favor in recruiting more patients. Further, some of the physiological parameters are derived from calculations and e.g. calculation of CO₂ using Harris-Benedict formula is critical, since predicted VCO₂ has a large variation.

Second, the comparator group for pulmonary hemodynamics could be more consistent and homogeneous. We compared the hemodynamics of COVID-19 patients with ARDS that was not exclusively due to viral infection. However, hypoxic pulmonary vasoconstriction has been expected to work independently of a causative organism in the past.

Third, the time of PAC insertion in CARDS patients was distinctly later than in ARDS patients, and the measurement was only at one point in time. This was mainly because as a tertiary care center, we received patients with COVID-19 at the beginning of the pandemia from other hospitals where patients had already been treated for several days in ICUs. It seems that COVID-19 has different phenotypes according to the time of evaluation, being less usual at the beginning and adopting a more usual phenotype after days of mechanical ventilation. Interestingly, despite the differences in time, we could not detect relevant differences between groups. It is possible that the pathophysiology of COVID-19 is simply prolonged compared to that of other respiratory diseases.

Conclusion

Our data do not support the assumption that CARDS (at a later stage) leads to changes in pulmonary vasculature distinct from ARDS due to pulmonary infections. The mechanical power applied to the lungs of CARDS patients by ventilation was significantly higher than that in ARDS patients. The data presented here could be useful to guide CARDS therapy, as we assume that the results also apply for other cohorts of patients with severe SARS-CoV2 infection. Further investigations are needed to elucidate the mechanisms behind SARS-CoV2-induced pulmonary pathology.

Footnotes

Authors’ Note

Members of the PACovid Study Group: Albert Omlor, MD (Department of Internal Medicine V—Pneumology, Allergology and Critical Care Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany; Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany), Sebastian Mang, MD (Department of Internal Medicine V—Pneumology, Allergology and Critical Care Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany; Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany), Christophe Jentgen, MD (Department of Internal Medicine V—Pneumology, Allergology and Critical Care Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany; Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany), Holger Wehrfritz, MD (Department of Internal Medicine V—Pneumology, Allergology and Critical Care Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany; Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany), Thilo Mertke, MD (Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany; Department of Anaesthesiology, Critical Care Medicine and Pain Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany), Heinrike Wilkens, MD (Department of Internal Medicine V—Pneumology, Allergology and Critical Care Medicine, University Hospital of Saarland and Saarland University, Homburg/Saar, Germany; Interdisciplinary COVID-19-Center, University Hospital of Saarland, Saarland University, Homburg/Saar, Germany).

Author Contributions

André Becker, Frederik Seiler, Guy Danziger, Sebastian Mang, Maren Kamphorst, Thilo Mertke, and Philipp M. Lepper contributed to the collection, review, and/or analysis of the data; André Becker and Philipp M. Lepper drafted the manuscript, Ralf M. Muellenbach, Christopher Lotz, the members of the PACovid Study Group and Robert Bals revised the manuscript for important intellectual content. All authors have seen and approved the final version of the manuscript. Philipp M. Lepper is the guarantor of the manuscript.

Availability of Data and Materials

Data can be provided on request addressed to the corresponding author. All data sharing statements are subject to conformity with German data protection legislation and rules (Datenschutzgrundverordnung—DGSVO).

Ethics Approval and Consent to Participate

Informed consent for pulmonary artery catheterization as part of standard intensive care was obtained by the patient or legal representative. The institutional review board (Ärztekammer des Saarlandes) registered the study to use a PAC in patients with ARDS, irrespective of the infectious agent, in 2019 under the number 278/19. Informed consent for the analysis of data is waived by the institutional review board due to the anonymous and retrospective analysis of data.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article: All other authors have no conflicts of interest to declare.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: COVID-19 research at the University Hospital of Saarland is funded by unrestricted grants of the Federal State of Saarland, Saarland University and Dr Rolf M. Schwiete Foundation. The funders had no role with regard to this study in the design of the study and collection, analysis, and interpretation of data or in writing the manuscript. Robert Bals received funding from AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Grifols, Novartis, CSL Behring, German Federal Ministry of Education and Research (BMBF) Competence Network, Sander Stiftung, Dr Rolf M. Schwiete Foundation, German Cancer help (Krebshilfe) and Mukoviszidose e.V.

ORCID iD

Philipp M. Lepper

References

Grasselli

Zangrillo

Zanella

, et al. Baseline characteristics and outcomes of 1591 patients infected with SARS-CoV-2 admitted to ICUs of the Lombardy Region, Italy. JAMA. 2020;323(16):1574–1581.

Marini

Gattinoni

. Management of COVID-19 respiratory distress. JAMA. 2020;323(22):2329–2330.

Varga

Flammer

Steiger

, et al. Endothelial cell infection and endotheliitis in COVID-19. Lancet. 2020;395(10234):1417–1418.

Zapol

Snider

. Pulmonary hypertension in severe acute respiratory failure. N Engl J Med. 1977;296(9):476–480.

Ashbaugh

Bigelow

Petty

Levine

. Acute respiratory distress in adults. Lancet. 1967;2(7511):319–323.

Beiderlinden

Kuehl

Boes

Peters

. Prevalence of pulmonary hypertension associated with severe acute respiratory distress syndrome: predictive value of computed tomography. Intensive Care Med. 2006;32(6):852–857.

Ferguson

Fan

Camporota

, et al. The Berlin definition of ARDS: an expanded rationale, justification, and supplementary material. Intensive Care Med. 2012;38(10):1573–1582.

Roza

Shizgal

. The Harris Benedict equation reevaluated: resting energy requirements and the body cell mass. Am J Clin Nutr. 1984;40(1):168–182.

Sinha

Fauvel

Singh

Soni

. Ventilatory ratio: a simple bedside measure of ventilation. Br J Anaesth. 2009;102(5):692–697.

10.

Acute Respiratory Distress Syndrome Network, Brower

Matthay

, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301–1308.

11.

Force

ADT

Ranieri

Rubenfeld

, et al. Acute respiratory distress syndrome: the Berlin definition. JAMA. 2012;307(23):2526–2533.

12.

Galie

Humbert

Vachiery

, et al. 2015 ESC/ERS Guidelines for the diagnosis and treatment of pulmonary hypertension: The Joint Task Force for the Diagnosis and Treatment of Pulmonary Hypertension of the European Society of Cardiology (ESC) and the European Respiratory Society (ERS): Endorsed by: Association for European Paediatric and Congenital Cardiology (AEPC), International Society for Heart and Lung Transplantation (ISHLT). Eur Heart J. 2016;37(1):67–119. Epub 2015 Aug 29.

13.

Hager

Krishnan

Hayden

Brower

Network

ACT

. Tidal volume reduction in patients with acute lung injury when plateau pressures are not high. Am J Respir Crit Care Med. 2005;172(5):1241–1245.

14.

Petersson

Glenny

. Gas exchange and ventilation-perfusion relationships in the lung. Eur Respir J. 2014;44(4):1023–1041.

15.

Radermacher

Maggiore

Mercat

Fifty years of research in ards. gas exchange in acute respiratory distress syndrome. Am J Respir Crit Care Med. 2017;196(8):964–984.

16.

Wang

Sun

, et al. Mechanical ventilation strategy guided by transpulmonary pressure in severe acute respiratory distress syndrome treated with venovenous extracorporeal membrane oxygenation. Crit Care Med. 2020;48(9):1280–1288.

17.

Mekontso Dessap

Boissier

Charron

, et al. Acute cor pulmonale during protective ventilation for acute respiratory distress syndrome: prevalence, predictors, and clinical impact. Intensive Care Med. 2016;42(5):862–870.

18.

Zochios

Parhar

Tunnicliffe

Roscoe

Gao

. The right ventricle in ARDS. Chest. 2017;152(1):181–193.

19.

Tyberg

Grant

Kingma

, et al. Effects of positive intrathoracic pressure on pulmonary and systemic hemodynamics. Respir Physiol. 2000;119(2-3):171–179.

20.

Dunham-Snary

Sykes

, et al. Hypoxic pulmonary vasoconstriction: from molecular mechanisms to medicine. Chest. 2017;151(1):181–192.

21.

Alhazzani

Moller

Arabi

, 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):e440–e469.