Introduction

Pulmonary Ventilation-Perfusion Mismatch in COVID-19

Pulmonary ventilation is the process of air flowing into the lungs during inspiration (inhalation) and out of the lungs during expiration (exhalation). Gas exchange occurs in alveoli, and for effective gas exchange to occur, alveoli must be ventilated and perfused. Ventilation (V) refers to the flow of air into and out of the alveoli, while perfusion (Q) refers to the flow of blood to alveolar capillaries. These two variables, V and Q, determine oxygen (O2) and carbon dioxide (CO2) levels in the blood. Normal V is 4 l/min of air and normal Q is 5 l/min of blood. So normal V/Q ratio is 4/5 = 0.8 [1]. The actual values in the human lung vary depending on the position within the lung due to the gravitational effect. A VQ mismatch is the term used when the ventilation and the perfusion at the alveolar site are not matched [2, 3]. A VQ mismatch is indicative of respiratory distress or failure in COVID-19 patients [4]. Extreme VQ mismatch is the result of an area with perfusion, but no ventilation (V/Q = 0) termed a “shunt” or an area with ventilation, but no perfusion (V/Q = undefined) termed a “dead space” (Fig. 1a-c). In severe hypoxemic COVID-19 cases, hypoperfusion to the normal parenchyma due to vasoconstriction of pulmonary blood vessels results in poor gas exchange. The shifting of the blood pool from vasoconstricted small capillaries to the dilated blood vessels (redistribution) in ground-glass opacities (GGO) and consolidation results in shunting (Fig. 1d, e).

Fig. 1
figure 1

Normal ventilation and perfusion are responsible for gas exchange in alveoli (a). Alveolar sac filled with edema and exudates produces hypoxemia by decreasing the alveolar and arterial oxygen level result in shunt (b). Pulmonary embolism develops high ventilation in proportion to perfusion produce a dead space (c). Hypoperfusion to the apparently normal parenchyma due to vasoconstriction of pulmonary blood vessels results in poor gas exchange (d) and hyperperfusion through dilated capillaries to alveolar sac filled with edema and exudates result in shunt (e)

Full size image

The fluid in alveoli triggers a pulmonary shunt and although perfused, the lungs remain unventilated. Intrapulmonary shunting causes hypoxemia (inadequate blood oxygen) in which the lungs become consolidated. Even breathing 100% oxygen does not result in fully oxygenated blood. Pulmonary embolism decreases perfusion relative to ventilation results in increased dead space. Dead zones usually can be corrected by supplying 100% inspired oxygen. Even when a capillary is blocked because the pulsating nature of blood flow traveling forward and backward distributes between other capillaries that are exchanging gases efficiently [5, 6].

Imaging Lung Manifestation and VQ Mismatch in COVID-19

In COVID-19, severe hypoxemia is caused by two mechanisms: the shunt effect through diffuse alveolar damage and the dead space effect through thrombotic injuries in microvascular beds [7,8,9] (Fig. 2). Rapid multiplication and release of the virus in the host cell lead to pyroptosis of epithelial and endothelial cells. It results in vascular leakage and alveolar edema [10]. The accumulation of fluid in the damaged lung parenchyma results in a shunt. The damaged parenchyma appears as GGO, consolidation, septal thickening on conventional chest CT scans, and perfusion abnormalities on contrast-enhanced CT [11]. COVID-19 viral infection may accelerate microvascular thrombosis followed by pulmonary embolism results in VQ mismatch and hypoxemias [12].

Fig. 2
figure 2

Source—https://doi.org/10.1111/jth.14975

Normal lung (left) showing smooth blood flow and the effective gas exchange is recognized as normal ventilation and perfusion. COVID-19 infection causes an intense inflammatory reaction (right) results in shunt or dead space or both. Uncontrolled activation of lymphocytes, neutrophil, and pulmonary production of platelets cause lung tissue damages. The virulence in COVID-19 triggers pulmonary microthrombi, endothelial damage, and vascular leakage. The host intends to control the thrombi formation by vigorous fibrinolysis, and degraded fibrin (D-dimer) are released in blood stream. Image

Full size image

Although portable chest X-rays and lung ultrasound, valuable imaging tools provide limited granular details about lung parenchyma, they are unable to assess pulmonary vasculature/blood flow. Advanced imaging modalities such as high-resolution CT (HRCT), ultra-HRCT (U-HRCT), dual-energy CT (DECT), CT pulmonary angiogram (CTPA), single-photon emission tomography (SPECT)-CT, and electrical impedance tomography (EIT) are useful imaging tools to visualize ventilation-perfusion abnormalities that lead to a VQ mismatch in COVID-19 and are described in the following sections. While EIT does not involve radiation exposure, other imaging modalities involved either administration of radioactivity or exposure to X-rays and require scanning suits. A summary of these imaging modalities to assess lung manifestation in COVID-19 is presented in Table 1.

Table 1 Comparative study of imaging modalities to assess COVID-19 clinical manifestation
Full size table

Imaging Modalities

High-Resolution CT (HRCT) and Ultra-HRCT (U-HRCT)

Functional respiratory imaging (FRI) technique from Fluidda, Inc. (https://www.fluidda.com/) when combined with HRCT or U-HRCT scans could provide an assessment of pulmonary air tracks and blood volume distribution in the pulmonary blood vessels. The technique requires inspiratory and expiratory scans but does not require a contrast agent. FRI works optimally when the CT slice thickness is < 1 mm. HRCT (slice thickness: 0.625–1.25 mm) and U-HRCT (slice thickness: 0.2–0.5 mm) in which thin-slice chest images are acquired and reconstructed using a sharp algorithm are outstanding imaging tools to access blood volume and vasculature abnormalities in COVID-19 without contrast agents. In some longitudinal studies, HRCT detects early lung injury and evaluates disease severity, clinical typing, type of lesions, and post-treatment follow-up [13,14,15,16,17]. U-HRCT scanner observed superiority over HRCT in detecting not only distribution and lesions of COVID-19 but also imaging local lung volume loss [18, 19].

Some COVID-19 patients have normal lung compliance and pulmonary blood volume but significant hypoxemia. FRI combined with blood vessel (BV) segmentation algorithm on HRCT scans found that patients with COVID-19 had significantly reduced blood volume in the smaller caliber blood vessels (BV5, cross-sectional area < 5 mm2) compared with matched acute respiratory distress syndrome (ARDS) patients without COVID-19 and healthy controls [20, 21]. In the long hauler COVID-19 patients, the blood vessels remain anomalous for a long time. The mid-size vessels (BV5_10, cross-sectional area of 5–10 mm2) seem to remain dilated which could indicate sustained microvascular obstruction and endothelial damage. In acute COVID-19, they had a significantly higher proportion of blood volume within large-caliber vessels (BV10, cross-sectional area of > 10 mm2) (Fig. 3). Impaired gas exchange in the lungs may be due to redistribution of blood away from the small-caliber pulmonary vessels. This mechanism could be due to increased vascular resistance caused by either vasoconstriction of distal pulmonary arteries or the presence of microthrombi or both. The vasculopathy in the lungs reduces gas diffusion capacity causing hypoxemia. FRI technique reaffirms that high levels of positive end-expiratory pressure (PEEP) may not alone be enough to inflate alveoli and facilitate gas exchange, but anticoagulant or pulmonary vasodilators therapies should assist to reduce hypoxemia in these patients [20, 21]. A major challenge in the pulmonary rehabilitation process of COVID-19 long-haulers is persisting symptoms long after recovery from infection. The persistent symptoms often include brain fog, fatigue, headaches, dizziness, and shortness of breath [22]. Tracking pulmonary structural and functional information during treatment of COVID-19 long-haulers, FRI possibly could evaluate the development of an optimal therapeutic solution.

Fig. 3
figure 3

Source—Fluidda Inc

Functional respiratory imaging on HRCT scans found that the COVID-19 patients had significantly reduced blood volume in the smaller caliber blood vessels (BV5, cross-sectional area < 5 mm2) compared with the long hauler COVID-19 subjects and healthy controls. In the long haulers, the blood vessels remain anomalous for a long time. The mid-size vessels, indicated in yellow (BV5_10, cross-sectional area of 5–10 mm2), seem to remain dilated which could indicate sustained microvascular obstruction and endothelial damage. In acute COVID-19, they had a significantly higher proportion of blood volume within large-caliber vessels (BV10, cross-sectional area of > 10 mm2) and mainly projected towards the posterior part of the lungs. Impaired gas exchange in the lungs seen in COVID-19 may be partially a result of redistribution of blood away from the small-caliber pulmonary vessels. Image

Full size image

Dual-Energy CT (DECT) and CT Pulmonary Angiogram (CTPA)

Both DECT and CTPA are used to diagnose pulmonary perfusion and embolism in COVID-19 patients. Intravenous administration of iodine-containing contrast agent image pulmonary arteries. Compared with CTPA, DECT allows a significant 50% reduction in iodine dose while improving signal intensity, maintaining signal-to-noise ratio with a comparable radiation dose [23]. DECT utilizes high- and low-kilovoltage series X-ray energy spectra during image acquisitions and detects perfusion abnormalities. The technique reduces radiation exposure, increases the attenuation of blood vessels at lower energies, and simulates true unenhanced images [11, 24,25,26,27].

In the early phase, pulmonary DECT angiography revealed a hyperperfused dilated capillary system in GGO in COVID-19 patients. Enlarged vessels entering GGO are suggested to occur due to hyperemia and vasodilation through activation of bradykinin receptors. Increased blood flow in poorly ventilated areas results in shunt (Fig. 4A,C) [11, 25]. In a prolonged disease course, GGO becomes denser (consolidation). The hypoperfusion in consolidated areas suggests obstruction or blockage of capillaries due to thrombus formation. The consolidation was found encircled by hyperperfused blood vessels (Fig. 4B) [24, 27]. During the long hospitalization of the patients, DECT evaluates the possible onset of pulmonary infarcts secondary to pulmonary thromboembolism. In this clinical scenario, dual-energy pulmonary angiography allows differentiation between GGO and ischemic or infarcted areas (Fig. 4B) [19]. Pulmonary perfusion evaluated by iodine concentration maps shows extreme heterogeneity in COVID-19 patients and lower iodine levels in normal parenchyma. DECT revealed a significant number of pulmonary ischemic areas although thrombosis remained undetected on images (Fig. 4C) [25].

Fig. 4
figure 4

source—https://doi.org/10.1016/S1473-3099(20)30367-4. In panel 4B, (a, c) axial CT image shows wedge-shaped bilateral opacities with surrounding GGO; (b) iodine map image shows a triangular peripheral area of decreased perfusion (yellow arrow) in the right lower, distal to pulmonary embolism (red arrow) lobe compatible with pulmonary infarction; (d) iodine map images showing a peripheral, triangular, and hypoperfused area in the left lower lobe (yellow arrow) suggestive of pulmonary infarction. Image source—https://doi.org/10.1016/j.rec.2020.04.013. In panel 4C, (a) axial CT image shows central GGO and peripheral consolidation a right inferior lobe. (b) Axial iodine color map shows high iodine concentrations in consolidations, and hypoperfusion in left medio-basal segment (dotted line), secondary to the thrombosis of corresponding segmental pulmonary artery (white arrow). (c) Axial CT image shows consolidation in a right inferior lobe and GGO in both inferior lobes. (d) Axial iodine color map shows hypoperfused area in the middle lobe (dotted line). Right inferior lobe consolidation shows high and heterogeneous iodine levels. Image source—https://doi.org/10.21037/qims-20-708. In panel 4D, (a) Axial chest CT of the right upper lobe with subpleural pneumonia (red arrows), surrounded by small GGOs. (b) CTPA shows multiple small subpleural perfusion defects (red arrows) and a larger perfusion defect dorsal in the normal ventilated right upper lobe (Δa), due to microvascular obstruction (Δb). Pulmonary emboli in the right pulmonary upper lobe were not observed. Image source—https://doi.org/10.1259/bjr.20200718

Panels display perfusion CT scans with dual energy CT (4A, 4B, and 4C) and conventional CT (4D) in COVID-19 patients. In panel 4A, (a) Peripheral GGO and consolidation within the right upper lobe and smaller GGO in the posterior left upper lobe (green arrowheads) are accompanied by dilated subsegmental vessels proximal to, and within, the opacities (green arrows). (b) The accompanying pulmonary blood volume image shows higher perfusion (green arrows) in GGO and consolidation. Image

Full size image

CTPA detected impaired pulmonary ventilation, pulmonary embolism, and hypoperfusion areas in a COVID-19 patient with elevated D-dimer level [28], and revealed multiple pulmonary perfusion defects without demonstrable pulmonary embolism even before D-dimer elevation (Fig. 4D) [29].

SPECT-CT

VQ scintigraphy was recommended to detect pulmonary embolism and ventilation-perfusion mismatch in COVID-19 [30,31,32,33], where CT confronts limitations. Perfusion scans are performed using hybrid SPECT-CT cameras with intravenous injection of Technetium 99mTc macro-aggregated albumin (99mTc-MAA) and ventilation scans with inhalation of 81mKr-Krypton gas or Technetium 99mTc-diethylen-tetraamino-pentaacetate (99mTc-DTPA) or ultra-fine dispersion of 99mTc-Carbon (Technegas®) [34]. The biodistribution of radiotracers in the lungs is then measured to observe ventilation and perfusion defects. Decreased radiotracer uptake and no mismatch of V and Q SPECT scans are indicative of a low possibility of pulmonary embolism. The regions with decreased radiotracer uptake are examined for peripheral parenchymal GGO on CT scans of COVID-19 patients (Fig. 5A) [31]. Ventilation SPECT-CT scan revealed increased Technegas® uptake in the tracheobronchial tract in COVID-19 patients suggesting blockage of airway paths (bronchus, bronchi, and bronchioles) in the lungs and cautioning signs of a tracheobronchitis. In the absence of previous pulmonary disease, ventilation SPECT-CT might have diagnostic and therapeutic applications (Fig. 5B) [30]. In the early phase, COVID-19 patients have shown decreased ventilation and normal perfusion in the lung regions where GGO or consolidation was observed on CT scans (Fig. 5C) [31].

Fig. 5
figure 5

source—https://doi.org/10.1007/s00259-020-04920-w, https://doi.org/10.1007/s00259-020-04834-7

In panel 5A, the axial images showing decreased radiotracer uptake in both perfusion and ventilation scan (white arrow) in fused SPECT-CT (a, b) indicating low probability of pulmonary embolism. SPECT (c, d) observed defect matches with the posterior segment of the right upper lobe alveolar filling (black arrow) on the CT scan (e). In panel 5B, on coronal images, decreased dual radiotracer uptake (green arrows) in ventilation (a) and perfusion (b) SPECT scans indicates low possibility of pulmonary embolism. Increased tracheobronchial tract uptake of (blue arrows), with marked intensity on the proximal bronchi suggesting tracheobronchitis or chronic obstructive pulmonary disease. In panel 5C, the axial SPECT images showing normal perfusion scan (a) and defects in ventilation scan (b) in the posterior segment of the right upper lobe (white arrow) that matches with the GGO on CT scan (c, black arrow). Image

Full size image

Ventilation SPECT scans are recommended if CTPA is inconclusive, contraindicated, and in exceptional circumstances in COVID-19 confirmed or suspected cases [35, 36]. VQ SPECT-CT not only performs diagnosis of pulmonary embolism but also assesses the degree of pulmonary involvement avoiding the risk of renal failure linked to the injection of iodinated contrast agents to the patients at risk [37]. Perfusion SPECT-CT avoids unnecessary risk related to aerosolization of respiratory droplets and possible viral spread—a risk factor associated with ventilation SPECT scan. It is therefore recommended as a choice for detecting pulmonary embolism in COVID-19 patients [38, 39]. In pregnancy, perfusion SPECT without CT reduces exposure to radiation dose. However, in certain cases, if the pregnant woman falls in the clinical “vulnerable group” for COVID-19 infection, a perfusion SPECT with low-dose CT is advocated [34, 35].

Electrical Impedance Tomography (EIT)

Imaging tools like EIT can monitor lung ventilation and perfusion bedside while the patient is receiving treatment. It is a non-invasive radiation-free functional imaging modality that allows continuous bedside monitoring of lung ventilation and perfusion in real time [40]. EIT uses electric current and evaluates the distribution of alternating current conductivity within the thoracic cavity. It is based on the repeated measurement of the surface voltages (40–50 cycles/s) resulting from a rotating high frequency (50–80 kHz) and low intensity (5–10 mA) alternating current that circulates between the electrodes around the thorax. The spatial resolution can be improved with the increasing number of electrodes. With a computer and electrodes applied to the patient’s chest, the bioimpedance (resistance) signal can be processed allowing the description of tissue characteristics (conductivity) located in the selected body circumference. The electrodes collect information on impedance by forming a relative image based on a reference. While the tissues allow the passage of current with little resistance, the air in the lungs offers high resistance [41]. The impedance of the lung during inspiration is double that of expiration. EIT allows the estimation of the tidal volume, the percentages of the tidal volume distribution in each quadrant or layer, the minimum impedance (end-expiratory lung impedance), or maximum impedance (total lung capacity) [40]. The quantification of pulmonary perfusion by EIT can be performed by a brief breath-hold, followed by rapid infusion of hypertonic saline (10 ml, 5–7.5% w/v NaCl), which dramatically reduces chest impedance, thus acting as an intravascular contrast [40]. In the pigs, pulmonary perfusion maps by EIT were found to well agree with perfusion maps produced using SPECT [42] and positron emission tomography [43].

EIT was found helpful to characterize the etiology of hypoxemia of COVID-19 patients with respiratory failure (Fig. 6) [44]. COVID-19 pneumonia requires different ventilator strategies. For patients who have VQ mismatch, EIT could assist in identifying different pneumonia patterns and setting optimal parameters on a mechanical ventilator [45]. COVID-19 patients often have airway secretions that severely compromise ventilation. Therapeutic bronchoalveolar lavage removes mucus in the small airways and improves regional ventilation. EIT was found useful at the bedside to monitor the ventilation treatment and anticoagulant treatment in COVID-19 [14, 46].

Fig. 6
figure 6

source—https://doi.org/10.1164/rccm.202005-1801IM

In this study, EIT showed real-time noninvasive bedside ventilation and perfusion in hypoxemic COVID-19 patients with respiratory failure. All the three COVID-19 patients intubated for acute hypoxic respiratory failure (PaO2/FiO2 < 300) had different respiratory system compliance. EIT was used to determine regional ventilation and perfusion distribution. On CT scans, case 1 shows peripheral and basilar GGO, while case 2 and 3 describe diffuse bilateral GGO. In case 1, EIT showed severe right-lung perfusion anomalies, homogenous ventilation, and a moderate decrease in respiratory compliance (40 ml/cm H2O). Case 2 and case 3 showed a progressive decrease of respiratory compliance, without major perfusion disturbances. PaO2/FiO2 means ratio of arterial oxygen partial pressure in mmHg to fractional inspired oxygen expressed as fraction. Image

Full size image

The procedure to titrate PEEP in COVID-19 or associated ARDS patients is unknown. The survival of patients is improved if higher PEEP successfully recruited partly or completely collapsed lung tissue [47]. However, excessive PEEP caused alveolar overdistention, resulting in reduced patient survival [48]. Therefore, PEEP should be personalized to maximize alveolar recruitment and minimize the amount of alveolar overdistention. In a case series of COVID-19 patients with moderate to severe ARDS, EIT successfully provided a reliable bedside approach to detect both alveolar overdistention and alveolar collapse [49, 50].

Concluding Remarks

COVID-19 hypoxemic patients exhibit abnormal hyperperfusion in areas of lung GGO and show abnormally decreased perfusion in areas of the apparently normal lung parenchyma. Hypoperfusion of apparently healthy areas could be a consequence of vasoconstriction due to the accumulation of angiotensin II that shifts vascular flow towards areas of non-aerated hyperperfused lung in moderate to severe COVID-19 cases. In most COVID cases, the posterior part of the lungs is severely affected. Prone sleeping position shifts blood flow through gravity effect towards hypoperfused anterior part of the apparently normal lungs and improves oxygenation. The severity of the disease and degree of VQ mismatch in COVID patients makes them different from one another clinically.

FRI on HRCT revealed that the total pulmonary blood volumes as measured do not vary significantly between healthy and COVID-19 patients [21]. Throughout the pulmonary vasculature, the blood flow was redistributed from small vessels to medium and larger blood vessels entering in GGO and consolidation. The measurements of blood volume distribution, anatomical changes of pulmonary vasculature may offer a powerful means to evaluate effective treatments in acute COVID-19 patients or recovering long haulers [20, 21]. FRI findings correlate with the DECT and CTPA iodine perfusion maps where high iodine concentration was observed in GGO and lower iodine levels in normal parenchyma with or without evidence of thrombosis or pulmonary embolism [11, 25].

DECT allows a significant reduction in iodine dose while improving the intravascular signal intensity and maintaining a signal-to-noise ratio with comparable radiation dose. Although the pulmonary vasculature and embolism can be seen in exquisite detail with CTPA and DECT, they are reserved for symptomatic patients with specific clinical indications [51]. VQ SPECT scan with low-dose CT identifies the dual pathology of pulmonary embolism and COVID-19 related lung parenchymal abnormalities. In pregnant patients with suspected massive pulmonary embolism, only perfusion scintigraphy is recommended. A low-dose CT with SPECT perfusion scan increases sensitivity to ~ 97%, compared with CTPA, and a significant increase in specificity (97.2–100%) to detect pulmonary embolism in COVID patients [35]. The absorbed fetal dose from the low-dose CT is negligible (< 0.1 mSv), while radiation exposure reduces 5 times compared with 9 mSv dose received from CTPA [52, 53]. Allergy to iodinated contrast agents and renal failure in COVID patients restricts the use of iodinated contrast agents on the CT scan [54]. In this case, pulmonary scintigraphy is recommended for diagnosing pulmonary embolism in COVID-19. The radiation exposure and the need for decontaminated scanning room limit the serviceability of SPECT or CT for routine use. In this situation, EIT is emerging as a promising clinical imaging tool for continuous and real-time monitoring of pulmonary ventilation and perfusion in COVID-19 patients. The complete discussion of the specific pros and cons associated with the different imaging modalities is outside the scope of this review. We emphasize the clinical usefulness of imaging in the settings of COVID-19.