Regional aeration and perfusion distribution in a sheep model of endotoxemic acute lung injury characterized by functional computed tomography imaging

New Frontiers in Functional and Molecular Imaging of the Acutely Injured Lung: Pathophysiological Insights and Research Applications

This review focuses on the advances in the understanding of the pathophysiology of ventilator-induced and acute lung injury that have been afforded by technological development of imaging methods over the last decades. Examples of such advances include the establishment of regional lung mechanical strain as a determinant of ventilator-induced lung injury, the relationship between alveolar recruitment and overdistension, the regional vs. diffuse nature of pulmonary involvement in acute respiratory distress syndrome (ARDS), the identification of the physiological determinants of the response to recruitment interventions, and the pathophysiological significance of metabolic alterations in the acutely injured lung. Taken together, these advances portray multimodality imaging as the next frontier to both advance knowledge of the pathophysiology of these conditions and to tailor treatment to the individual patient's condition.

Cardiovascular Responses During Sepsis

Sepsis is the life-threatening organ dysfunction arising from a dysregulated host response to infection. Although the specific mechanisms leading to organ dysfunction are still debated, impaired tissue oxygenation appears to play a major role, and concomitant hemodynamic alterations are invariably present. The hemodynamic phenotype of affected individuals is highly variable for reasons that have been partially elucidated. Indeed, each patient's circulatory condition is shaped by the complex interplay between the medical history, the volemic status, the interval from disease onset, the pathogen, the site of infection, and the attempted resuscitation. Moreover, the same hemodynamic pattern can be generated by different combinations of various pathophysiological processes, so the presence of a given hemodynamic pattern cannot be directly related to a unique cluster of alterations. Research based on endotoxin administration to healthy volunteers and animal models compensate, to an extent, for the scarcity of clinical studies on the evolution of sepsis hemodynamics. Their results, however, cannot be directly extrapolated to the clinical setting, due to fundamental differences between the septic patient, the healthy volunteer, and the experimental model. Numerous microcirculatory derangements might exist in the septic host, even in the presence of a preserved macrocirculation. This dissociation between the macro- and the microcirculation might account for the limited success of therapeutic interventions targeting typical hemodynamic parameters, such as arterial and cardiac filling pressures, and cardiac output. Finally, physiological studies point to an early contribution of cardiac dysfunction to the septic phenotype, however, our defective diagnostic tools preclude its clinical recognition. © 2021 American Physiological Society. Compr Physiol 11:1605-1652, 2021.

Multi-resolution convolutional neural networks for fully automated segmentation of acutely injured lungs in multiple species

Segmentation of lungs with acute respiratory distress syndrome (ARDS) is a challenging task due to diffuse opacification in dependent regions which results in little to no contrast at the lung boundary. For segmentation of severely injured lungs, local intensity and texture information, as well as global contextual information, are important factors for consistent inclusion of intrapulmonary structures. In this study, we propose a deep learning framework which uses a novel multi-resolution convolutional neural network (ConvNet) for automated segmentation of lungs in multiple mammalian species with injury models similar to ARDS. The multi-resolution model eliminates the need to tradeoff between high-resolution and global context by using a cascade of low-resolution to high-resolution networks. Transfer learning is used to accommodate the limited number of training datasets. The model was initially pre-trained on human CT images, and subsequently fine-tuned on canine, porcine, and ovine CT images with lung injuries similar to ARDS. The multi-resolution model was compared to both high-resolution and low-resolution networks alone. The multi-resolution model outperformed both the low- and high-resolution models, achieving an overall mean Jacaard index of 0.963 ± 0.025 compared to 0.919 ± 0.027 and 0.950 ± 0.036, respectively, for the animal dataset (N=287). The multi-resolution model achieves an overall average symmetric surface distance of 0.438 ± 0.315 mm, compared to 0.971 ± 0.368 mm and 0.657 ± 0.519 mm for the low-resolution and high-resolution models, respectively. We conclude that the multi-resolution model produces accurate segmentations in severely injured lungs, which is attributed to the inclusion of both local and global features.

Imaging the Injured Lung: Mechanisms of Action and Clinical Use

Acute respiratory distress syndrome (ARDS) consists of acute hypoxemic respiratory failure characterized by massive and heterogeneously distributed loss of lung aeration caused by diffuse inflammation and edema present in interstitial and alveolar spaces. It is defined by consensus criteria, which include diffuse infiltrates on chest imaging-either plain radiography or computed tomography. This review will summarize how imaging sciences can inform modern respiratory management of ARDS and continue to increase the understanding of the acutely injured lung. This review also describes newer imaging methodologies that are likely to inform future clinical decision-making and potentially improve outcome. For each imaging modality, this review systematically describes the underlying principles, technology involved, measurements obtained, insights gained by the technique, emerging approaches, limitations, and future developments. Finally, integrated approaches are considered whereby multimodal imaging may impact management of ARDS.

Use of manual alveolar recruitment maneuvers to eliminate atelectasis artifacts identified during thoracic computed tomography of healthy neonatal foals

OBJECTIVE To evaluate use of single manual alveolar recruitment maneuvers (ARMs) to eliminate atelectasis during CT of anesthetized foals. ANIMALS 6 neonatal Standardbred foals. PROCEDURES Thoracic CT was performed on spontaneously breathing anesthetized foals positioned in sternal (n = 3) or dorsal (3) recumbency when foals were 24 to 36 hours old (time 1), 4 days old (time 2), 7 days old (time 3), and 10 days old (time 4). The CT images were collected without ARMs (all times) and during ARMs with an internal airway pressure of 10, 20, and 30 cm H₂O (times 2 and 3). Quantitative analysis of CT images measured whole lung and regional changes in attenuation or volume with ARMs. RESULTS Increased attenuation and an alveolar pattern were most prominent in the dependent portion of the lungs. Subjectively, ARMs did not eliminate atelectasis; however, they did incrementally reduce attenuation, particularly in the nondependent portion of the lungs. Quantitative differences in lung attenuation attributable to position of foal were not identified. Lung attenuation decreased significantly (times 2 and 3) and lung volume increased significantly (times 2 and 3) after ARMs. Changes in attenuation and volume were most pronounced in the nondependent portion of the lungs and at ARMs of 20 and 30 cm H₂O. CONCLUSIONS AND CLINICAL RELEVANCE Manual ARMs did not eliminate atelectasis but reduced attenuation in nondependent portions of the lungs. Positioning of foals in dorsal recumbency for CT may be appropriate when pathological changes in the ventral portion of the lungs are suspected.

Inhalation therapy with the synthetic TIP-like peptide AP318 attenuates pulmonary inflammation in a porcine sepsis model

BACKGROUND

The lectin-like domain of TNF-α can be mimicked by synthetic TIP peptides and represents an innovative pharmacologic option to treat edematous respiratory failure. TIP inhalation was shown to reduce pulmonary edema and improve gas exchange. In addition to its edema resolution effect, TIP peptides may exert some anti-inflammatory properties. The present study therefore investigates the influence of the inhaled TIP peptide AP318 on intrapulmonary inflammatory response in a porcine model of systemic sepsis.

METHODS

In a randomized-blinded setting lung injury was induced in 18 pigs by lipopolysaccharide-infusion and a second hit with a short period of ventilator-induced lung stress, followed by a six-hour observation period. The animals received either two inhalations with the peptide (AP318, 2×1 mg kg(-1)) or vehicle. Post-mortem pulmonary expression of inflammatory and mechanotransduction markers were determined by real-time polymerase chain reaction (IL-1β, IL-6, TNF-α, COX-2, iNOS, amphiregulin, and tenascin-c). Furthermore, regional histopathological lung injury, edema formation and systemic inflammation were quantified.

RESULTS

Despite similar systemic response to lipopolysaccharide infusion in both groups, pulmonary inflammation (IL-6, TNF-α, COX-2, tenascin-c) was significantly mitigated by AP318. Furthermore, a Western blot analysis shows a significantly lower of COX-2 protein level. The present sepsis model caused minor lung edema formation and moderate gas exchange impairment. Six hours after onset pathologic scoring showed no improvement, while gas exchange parameters and pulmonary edema formation were similar in the two groups.

CONCLUSION

In summary, AP318 significantly attenuated intrapulmonary inflammatory response even without the presence or resolution of severe pulmonary edema in a porcine model of systemic sepsis-associated lung injury. These findings suggest an anti-inflammatory mechanism of the lectin-like domain beyond mere edema reabsorption in endotoxemic lung injury in vivo.

Low tidal volume pressure support versus controlled ventilation in early experimental sepsis in pigs

Background

In moderate acute respiratory distress syndrome (ARDS) several studies support the usage of assisted spontaneous breathing modes. Only limited data, however, focus on the application in systemic sepsis and developing lung injury. The present study examines the effects of immediate initiation of pressure support ventilation (PSV) in a model of sepsis-induced ARDS.

Methods

18 anesthetized pigs received a two-staged continuous lipopolysaccharide infusion to induce lung injury. The animals were randomly assigned to PSV or volume controlled (VCV) lung protective ventilation (tidal volume each 6 ml kg^-1, n = 2x9) over six hours. Gas exchange parameters, hemodynamics, systemic inflammation, and ventilation distribution by multiple inert gas elimination and electrical impedance tomography were assessed. The post mortem analysis included histopathological scoring, wet to dry ratio, and alveolar protein content.

Results

Within six hours both groups developed a mild to moderate ARDS with comparable systemic inflammatory response and without signs of improving gas exchange parameters during PSV. The PSV group showed signs of more homogenous ventilation distribution by electrical impedance tomography, but only slightly less hyperinflated lung compartments by multiple inert gas elimination. Post mortem and histopathological assessment yielded no significant intergroup differences.

Conclusions

In a porcine model of sepsis-induced mild ARDS immediate PSV was not superior to VCV. This contrasts with several experimental studies from non-septic mild to moderate ARDS. The present study therefore assumes that not only severity, but also etiology of lung injury considerably influences the response to early initiation of PSV.

Development of quantitative computed tomography lung protocols

The purpose of this review article is to review the process of developing optimal computed tomography (CT) protocols for quantitative lung CT (QCT). In this review, we discuss the following important topics: QCT-derived metrics of lung disease; QCT scanning protocols; quality control; and QCT image processing software. We will briefly discuss several QCT-derived metrics of lung disease that have been developed for the assessment of emphysema, small airway disease, and large airway disease. The CT scanning protocol is one of the most important elements in a successful QCT. We will provide a detailed description of the current move toward optimizing the QCT protocol for the assessment of chronic obstructive pulmonary disorder and asthma. Quality control of CT images is also a very important part of the QCT process. We will discuss why it is necessary to use CT scanner test objects (phantoms) to provide frequent periodic checks on the CT scanner calibration to ensure precise and accurate CT numbers. We will discuss the use of QCT image processing software to segment the lung and extract the desired QCT metrics of lung disease. We will discuss the practical issues of using this software. The data obtained from the image processing software are then combined with those from other clinical examinations, health status questionnaires, pulmonary physiology, and genomics to increase our understanding of obstructive lung disease and improve our ability to design new therapies for these diseases.

Qualitative and quantitative interpretation of computed tomography of the lungs in healthy neonatal foals

OBJECTIVE

To qualitatively describe lung CT images obtained from sedated healthy equine neonates (≤ 14 days of age), use quantitative analysis of CT images to characterize attenuation and distribution of gas and tissue volumes within the lungs, and identify differences between lung characteristics of foals ≤ 7 days of age and foals > 7 days of age.

ANIMALS

10 Standardbred foals between 2.5 and 13 days of age.

PROCEDURES

Foals were sedated with butorphanol, midazolam, and propofol and positioned in sternal recumbency for thoracic CT. Image analysis software was used to exclude lung from nonlung structures. Lung attenuation was measured in Hounsfield units (HU) for analysis of whole lung and regional changes in attenuation and lung gas and tissue components. Degree of lung attenuation was classified as follows: hyperinflated or emphysema, -1,000 to -901 HU; well aerated, -900 to -501 HU; poorly aerated, -500 to -101 HU; and nonaerated, > -100 HU.

RESULTS

Qualitative evidence of an increase in lung attenuation and patchy alveolar patterns in the ventral lung region were more pronounced in foals ≤ 7 days of age than in older foals. Quantitative analysis revealed that mean ± SD lung attenuation was greater in foals ≤ 7 days of age (-442 ± 28 HU) than in foals > 7 days of age (-521 ± 24 HU). Lung aeration and gas volumes were lower than in other regions ventrally and in the mid lung region caudal to the heart. CONCLUSIONS AND CLINICAL RELEVANCE-Identified radiographic patterns and changes in attenuation were most consistent with atelectasis and appeared more severe in foals ≤ 7 days of age than in older neonatal foals. Recognition of these changes may have implications for accurate CT interpretation in sedated neonatal foals with pulmonary disease.

Pulmonary blood flow increases in damaged regions directly after acid aspiration in rats

BACKGROUND

After gastric aspiration events, patients are at risk of pulmonary dysfunction and the development of severe acute lung injury and acute respiratory distress syndrome, which may contribute to the development of an inflammatory reaction. The authors' aim in the current study was to investigate the role of the spatial distribution of pulmonary blood flow in the pathogenesis of pulmonary dysfunction during the early stages after acid aspiration.

METHODS

The authors analyzed the pulmonary distribution of radiolabeled microspheres in normal (n = 6) and injured (n = 12) anesthetized rat lungs using positron emission tomography, computed tomography, and histological examination.

RESULTS

Injured regions demonstrate increased pulmonary blood flow in association with reduced arterial pressure and the deterioration of arterial oxygenation. After acid aspiration, computed tomography scans revealed that lung density had increased in the injured regions and that these regions colocalized with areas of increased blood flow. The acid was instilled into the middle and basal regions of the lungs. The blood flow was significantly increased to these regions compared with the blood flow to uninjured lungs in the control animals (middle region: 1.23 [1.1; 1.4] (median [25%; 75%]) vs. 1.04 [1.0; 1.1] and basal region: 1.25 [1.2; 1.3] vs. 1.02 [1.0; 1.05], respectively). The increase in blood flow did not seem to be due to vascular leakage into these injured areas.

CONCLUSIONS

The data suggest that 10 min after acid aspiration, damaged areas are characterized by increased pulmonary blood flow. The results may impact further treatment strategies, such as drug targeting.

Regional lung derecruitment and inflammation during 16 hours of mechanical ventilation in supine healthy sheep

BACKGROUND

Lung derecruitment is common during general anesthesia. Mechanical ventilation with physiological tidal volumes could magnify derecruitment, and produce lung dysfunction and inflammation. The authors used positron emission tomography to study the process of derecruitment in normal lungs ventilated for 16 h and the corresponding changes in regional lung perfusion and inflammation.

METHODS

Six anesthetized supine sheep were ventilated with VT=8 ml/kg and positive end-expiratory pressure=0. Transmission scans were performed at 2-h intervals to assess regional aeration. Emission scans were acquired at baseline and after 16 h for the following tracers: (1) F-fluorodeoxyglucose to evaluate lung inflammation and (2) NN to calculate regional perfusion and shunt fraction.

RESULTS

Gas fraction decreased from baseline to 16 h in dorsal (0.31±0.13 to 0.14±0.12, P<0.01), but not in ventral regions (0.61±0.03 to 0.63±0.07, P=nonsignificant), with time constants of 1.5-44.6 h. Although the vertical distribution of relative perfusion did not change from baseline to 16 h, shunt increased in dorsal regions (0.34±0.23 to 0.63±0.35, P<0.01). The average pulmonary net F-fluorodeoxyglucose uptake rate in six regions of interest along the ventral-dorsal direction increased from 3.4±1.4 at baseline to 4.1±1.5 10(-3)/min after 16 h (P<0.01), and the corresponding average regions of interest F-fluorodeoxyglucose phosphorylation rate increased from 2.0±0.2 to 2.5±0.2 10(-2)/min (P<0.01).

CONCLUSIONS

When normal lungs are mechanically ventilated without positive end-expiratory pressure, loss of aeration occurs continuously for several hours and is preferentially localized to dorsal regions. Progressive lung derecruitment was associated with increased regional shunt, implying an insufficient hypoxic pulmonary vasoconstriction. The increased pulmonary net uptake and phosphorylation rates of F-fluorodeoxyglucose suggest an incipient inflammation in these initially normal lungs.

Spatial-temporal dynamics of pulmonary blood flow in the healthy human lung in response to altered FI(O2)

The temporal dynamics of blood flow in the human lung have been largely unexplored due to the lack of appropriate technology. Using the magnetic resonance imaging method of arterial spin labeling (ASL) with subject-gated breathing, we produced a dynamic series of flow-weighted images in a single sagittal slice of the right lung with a spatial resolution of ~1 cm(3) and a temporal resolution of ~10 s. The mean flow pattern determined from a set of reference images was removed to produce a time series of blood flow fluctuations. The fluctuation dispersion (FD), defined as the spatial standard deviation of each flow fluctuation map, was used to quantify the changes in distribution of flow in six healthy subjects in response to 100 breaths of hypoxia (FI(O(2)) = 0.125) or hyperoxia (FI(O(2)) = 1.0). Two reference frames were used in calculation, one determined from the initial set of images (FD(global)), and one determined from the mean of each corresponding baseline or challenge period (FD(local)). FD(local) thus represented changes in temporal variability as a result of intervention, whereas FD(global) encompasses both FD(local) and any generalized redistribution of flow associated with switching between two steady-state patterns. Hypoxic challenge resulted in a significant increase (96%, P < 0.001) in FD(global) from the normoxic control period and in FD(local) (46%, P = 0.0048), but there was no corresponding increase in spatial relative dispersion (spatial standard deviation of the images divided by the mean; 8%, not significant). There was a smaller increase in FD(global) in response to hyperoxia (47%, P = 0.0015) for the single slice, suggestive of a more general response of the pulmonary circulation to a change from normoxia to hyperoxia. These results clearly demonstrate a temporal change in the sampled distribution of pulmonary blood flow in response to hypoxia, which is not observed when considering only the relative dispersion of the spatial distribution.

Relationship between regional lung compliance and ventilation homogeneity in the supine and prone position

BACKGROUND

The prone position (PP) improves ventilation homogeneity in acute respiratory distress syndrome. The aim of this study was to investigate whether the alleviation of ventilation inhomogeneity in PP was due to changes in regional lung compliance.

METHODS

Ten lung-lavaged piglets were mechanically ventilated in supine position (SP) and in PP. In each position, positive end-expiratory pressure (PEEP) was reduced from 20 to 6 cmH(2)O in steps of 2 cmH(2)O every 10 min after full lung recruitment. Respiratory mechanics, blood gas, haemodynamic data and whole-lung computed tomography scans were recorded at each PEEP. The compliances of normally aerated (C(normal)) and newly recruited (C(recruited)) lung regions were calculated. Open lung PEEP (OL-PEEP) was defined as the lowest PEEP to maintain full lung recruitment.

RESULTS

At OL-PEEP, PP significantly increased normally aerated lung regions, decreased poorly aerated and hyperinflated lung regions and decreased tidal recruitment and hyperinflation. C(normal) was significantly reduced in PP compared with SP (12.8 ± 4.2 ml/cmH(2)O vs. 20.1 ± 6.2 ml/cmH(2)O, P < 0.001), whereas C(recruited) was increased in PP (13.9 ± 3.9 ml/cmH(2)O vs. 9.4 ± 2.4 ml/cmH(2)O, P < 0.001). C(normal) was correlated with hyperinflated lung regions at end-expiration (rho = 0.67) and end-inspiration (rho = 0.56) at OL-PEEP. C(recruited) was correlated with normally (r(2) = 0.36) and poorly aerated lung regions (rho = -0.58) at OL-PEEP.

CONCLUSION

This surfactant-depleted model shows that the improvement of ventilation homogeneity in PP is related to an increase in C(recruited) and a decrease in C(normal).

Regional pulmonary inflammation in an endotoxemic ovine acute lung injury model

The regional distribution of inflammation during acute lung injury (ALI) is not well known. In an ovine ALI model we studied regional alveolar inflammation, surfactant composition, and CT-derived regional specific volume change (sVol) and specific compliance (sC). 18 ventilated adult sheep received IV lipopolysaccharide (LPS) until severe ALI was achieved. Blood and bronchoalveolar lavage (BAL) samples from apical and basal lung regions were obtained at baseline and injury time points, for analysis of cytokines (IL-6, IL-1β), BAL protein and surfactant composition. Whole lung CT images were obtained in 4 additional sheep. BAL protein and IL-1β were significantly higher in injured apical vs. basal regions. No significant regional surfactant composition changes were observed. Baseline sVol and sC were lower in apex vs. base; ALI enhanced this cranio-caudal difference, reaching statistical significance only for sC. This study suggests that apical lung regions show greater inflammation than basal ones during IV LPS-induced ALI which may relate to differences in regional mechanical events.

Effects of surfactant depletion on regional pulmonary metabolic activity during mechanical ventilation

Inflammation during mechanical ventilation is thought to depend on regional mechanical stress. This can be produced by concentration of stresses and cyclic recruitment in low-aeration dependent lung. Positron emission tomography (PET) with (18)F-fluorodeoxyglucose ((18)F-FDG) allows for noninvasive assessment of regional metabolic activity, an index of neutrophilic inflammation. We tested the hypothesis that, during mechanical ventilation, surfactant-depleted low-aeration lung regions present increased regional (18)F-FDG uptake suggestive of in vivo increased regional metabolic activity and inflammation. Sheep underwent unilateral saline lung lavage and were ventilated supine for 4 h (positive end-expiratory pressure = 10 cmH(2)O, tidal volume adjusted to plateau pressure = 30 cmH(2)O). We used PET scans of injected (13)N-nitrogen to compute regional perfusion and ventilation and injected (18)F-FDG to calculate (18)F-FDG uptake rate. Regional aeration was quantified with transmission scans. Whole lung (18)F-FDG uptake was approximately two times higher in lavaged than in nonlavaged lungs (2.9 ± 0.6 vs. 1.5 ± 0.3 10(-3)/min; P < 0.05). The increased (18)F-FDG uptake was topographically heterogeneous and highest in dependent low-aeration regions (gas fraction 10-50%, P < 0.001), even after correction for lung density and wet-to-dry lung ratios. (18)F-FDG uptake in low-aeration regions of lavaged lungs was higher than that in low-aeration regions of nonlavaged lungs (P < 0.05). This occurred despite lower perfusion and ventilation to dependent regions in lavaged than nonlavaged lungs (P < 0.001). In contrast, (18)F-FDG uptake in normally aerated regions was low and similar between lungs. Surfactant depletion produces increased and heterogeneously distributed pulmonary (18)F-FDG uptake after 4 h of supine mechanical ventilation. Metabolic activity is highest in poorly aerated dependent regions, suggesting local increased inflammation.

Extrapolation from ten sections can make CT-based quantification of lung aeration more practicable

PURPOSE

Clinical applications of quantitative computed tomography (qCT) in patients with pulmonary opacifications are hindered by the radiation exposure and by the arduous manual image processing. We hypothesized that extrapolation from only ten thoracic CT sections will provide reliable information on the aeration of the entire lung.

METHODS

CTs of 72 patients with normal and 85 patients with opacified lungs were studied retrospectively. Volumes and masses of the lung and its differently aerated compartments were obtained from all CT sections. Then only the most cranial and caudal sections and a further eight evenly spaced sections between them were selected. The results from these ten sections were extrapolated to the entire lung. The agreement between both methods was assessed with Bland-Altman plots.

RESULTS

Median (range) total lung volume and mass were 3,738 (1,311-6,768) ml and 957 (545-3,019) g, the corresponding bias (limits of agreement) were 26 (-42 to 95) ml and 8 (-21 to 38) g, respectively. The median volumes (range) of differently aerated compartments (percentage of total lung volume) were 1 (0-54)% for the nonaerated, 5 (1-44)% for the poorly aerated, 85 (28-98)% for the normally aerated, and 4 (0-48)% for the hyperaerated subvolume. The agreement between the extrapolated results and those from all CT sections was excellent. All bias values were below 1% of the total lung volume or mass, the limits of agreement never exceeded ± 2%.

CONCLUSION

The extrapolation method can reduce radiation exposure and shorten the time required for qCT analysis of lung aeration.