The lung is a highly vascularized organ with a dual arterial supply via the pulmonary and bronchial arteries.1 Recent studies have shown that the analysis of perfusion patterns may be helpful in characterizing and diagnosing the cause of lung consolidation.2-4 Pathological studies in animals demonstrated that, in lung, neoplasias were vascularized by neoangiogenesis, which develops mainly via bronchial arteries.5, 6 Imaging methods like angiography, contrast-enhanced computed tomography (CT), color Doppler sonography (CDS), and contrast-enhanced ultrasound (CEUS) show changes that are due to the patterns of vascular supply and are routinely used in clinical practice to differentiate between neoplasia and acute inflammatory processes.2, 4, 7-9 A recent CEUS study has confirmed that pleural-based bronchial carcinomas predominately reveal bronchial-arterial (BA) perfusion-related enhancement patterns as an indirect sign of BA tumor neovascularization.3 For these results, BA supply of a peripheral lesion detected by CDS or CEUS is highly suggestive of a malignant consolidation.2-4, 10, 11

However, malignant neoplasias are not the only reason for BA neoangiogenesis.1 It has already been proven in pathological and angiographic studies that various inflammatory processes may lead to BA neoangiogenesis with the proliferation of new capillary networks.1, 5 The main targets of neoangiogenesis are vascular endothelial cells, whereas bronchial artery has a strong capacity and the pulmonary artery a very limited capacity for neoangiogenesis.5

Chronic granulomatous inflammatory diseases are a heterogeneous group with a tropism for the lung tissue.12-18 Pulmonary involvement is observed in up to 90% of patients with sarcoidosis or granulomatous vasculitis with polyangiitis (former Wegener's disease) and in more than 90% of patients with extrapulmonary tuberculosis.12, 14

However, to the best of our knowledge, the perfusion pattern of peripheral pulmonary granulomas has not yet been investigated by CEUS.

The aim of the current study is to describe the perfusion patterns of histologically confirmed peripheral pulmonary granulomatous lesions (PPGLs) by B-mode lung ultrasound (LUS) and CEUS and to correlate the sonographic findings with histopathological features and vascular patterns represented by immunohistochemical (CD34) endothelial staining in corresponding lesions.

Patients and Methods

From January 2007 until September 2020, 10 patients with highly suspected or known granulomatous diseases and peripheral pulmonary lesions were investigated by B-mode LUS and CEUS. Informed consent was obtained from all patients for the CEUS examination, and the study was approved by the local ethics committee and performed in accordance with the revised Helsinki Declaration.

A histopathological examination of all pulmonary lesions was performed and a diagnosis of PPGL was made. In addition to the tissue samples of the study patients, 10 samples of normal lung tissue obtained by autopsy, and 10 samples of lung tissue with acute pneumonia obtained by autopsy, were immunohistochemically stained with CD34 antibody.

Ultrasound Examination

The B-mode LUS examinations were performed with an ACUSON SEQUOIA 512 GI Siemens (Germany) ultrasound machine and a 4C1 curved array transducer with a frequency of 4 MHz.

The CEUS investigations were conducted with the same transducer in contrast-specific mode (1.5 MHz) and in accordance with the EFSUMB guidelines.19 A bolus injection of 2.4 ml of the contrast medium SonoVue® (Bracco Imaging S.p.A., Milan) was performed via peripheral venous access. This was followed by 10 ml of 0.9% NaCl. For the first 30 seconds (arterial phase), the perfusion patterns of the lesions were continuously examined and recorded by a clip. Subsequently, several short examinations were performed at 1-minute intervals up to 3 minutes (parenchymal phase) and the changes in the perfusion pattern were saved as images. All ultrasound examinations were performed in the upright sitting position and horizontal to the ribs. The following B-mode LUS data and CEUS parameters were evaluated retrospectively.

B-Mode Lung Ultrasound The echogenicity of the lesion was classified as hypoechoic or iso-/hyperechoic, compared with echogenicity of parenchymal organs as an in vivo reference 3. The border of the lesion was classified as smooth or irregular delineated. The size of the peripheral pulmonary lesion was classified as lesions of ≥2 cm or of <2 cm. The configuration of peripheral pulmonary granulomas was classified as round or oval. Contrast-Enhanced Ultrasound The time to enhancement (TE) of the contrast agent after intravenous injection was determined and classified as early pulmonary-arterial (PA) pattern of enhancement (contrast enhancement of the lesion before the arrival of contrast agent in the thoracic wall) versus delayed BA pattern of enhancement (contrast enhancement of the lesion simultaneous with the arrival of contrast agent in the thoracic wall or parenchymal organs).3, 20 The extent of enhancement (EE) in the arterial phase was categorized as reduced EE (echo-free/hypoechoic) versus marked EE (isoechoic).4, 20 The homogeneity of enhancement (HE) was classified in the arterial phase as homogeneous enhancement versus inhomogeneous enhancement of the PPGLs.4, 20-22 A perfused lesion with coexisting non-perfused areas (NPAs) was defined as an inhomogeneous enhancement.20-22 The decrease of enhancement (DE) in the parenchymal phase (washout) was classified as a rapid washout (<120 seconds) or a late washout (≥120 seconds).11

The arterial phase was defined as the time from the earliest arrival of the contrast agent at the lesion to the peak of contrast agent enhancement of the lesion. The parenchymal phase was defined as the time from the peak of contrast agent enhancement of the lesion to washout of contrast agent from the lesion. Splenic tissue was considered as an in vivo reference for the evaluation of EE, HE, and DE (washout) of the contrast agent.3, 4, 23

All examinations were performed by a German Society for Ultrasound in Medicine (DEGUM) Level III qualified examiner (C.G., internal medicine).24 The B-mode LUS and CEUS data were evaluated retrospectively by two independent, experienced investigators (C.T., C.G.). In the event of discrepancies, the final decision was made by a third experienced investigator (E.S.).

Histopathological Examination

All formalin-fixed, paraffin-embedded tissue samples were subjected to hematoxylin–eosin staining. In addition, all samples were immunohistochemically stained with an anti-human CD34 monoclonal antibody QBEnd10 (Agilent Dako, Waldbronn, Germany). The CD34 antigen is expressed by endothelial cells.25 As a control group, 10 samples of normal lung tissue and 10 samples of lung with acute pneumonia were obtained by autopsy in tissue specimens in which a chronic as well as a malignant disease was histologically excluded. All tissue samples were identified microscopically by an experienced pathologist (C.K.) as healthy lung tissue and lung tissue with acute pneumonia.

The following histopathological data were analyzed by the local pathological institute: The presence of avascular areas in the PPGLs in all study patients. The vascular patterns of PPGLs, healthy lung tissue, and acute pneumonia were classified between two patterns, A or B, according to the CD34 antigen expression of vascular endothelium. A regular alveolar vascular pattern along the linearly arranged alveoli corresponding to the pulmonary capillary vascular pattern in healthy lung tissue was defined as pattern A (Figure 1A),26, 27 while a disorganized and chaotic vascular pattern with various diameters similar to tumor neoangiogenesis in malignant lung tumors was defined as pattern B (Figure 1B).27, 28

A, Graphical representation of pattern A, with a regular alveolar vascular pattern along the linearly arranged alveoli, which is the typical pattern of capillaries in healthy lung tissue. B, Graphical representation of pattern B, with a disorganized and chaotic vascular pattern with different diameters, which is typical of tumor neoangiogenesis.

Results Demographic and Clinical Data

Of the 10 study patients, n = 6 patients were female and n = 4 were male. The mean age of the patients was 51.4 years (range 21–68 years).

In their medical history, all patients had been strongly suspected to have a granulomatous disease. The following were documented in the medical histories: 5 (50%) histologically confirmed extrapulmonary granulomas, 1 (10%) cytologically suspected (by thoracentesis) granulomatous disease, 2 (20%) radiologically suspected granulomatous disease, 1 (10%) clinically suspected to be due to active rheumatoid arthritis, and 1 (10%) laboratory chemically suspected granulomatous disease.

B-Mode Lung Ultrasound Data

On B-mode LUS, all PPGLs were hypoechoic (Figure 2). In 9/10 cases, PPGLs showed a smooth border, and, in 1/10 cases, PPGLs revealed an irregular delineated border. The size of the PPGLs was ≥2 cm in 6 and <2 cm in 4 cases. The peripheral pulmonary granulomas were round in 6 (60%) and oval in 4 (40%) cases (Table 1).

A 46-year-old male patient (patient 8) with histologically confirmed extrapulmonary granulomas on the left thigh in his medical history. A, B-mode lung ultrasound shows a 2 × 2 cm round, smooth delineated, inhomogeneous, hypoechoic pleural-based pulmonary nodule on the right upper lobe of the lung. B, An ultrasound-guided 18G-core needle biopsy of the lung lesion was performed. The histopathological examination of the lesion presents the diagnosis of granulomatous vasculitis with polyangiitis (former Wegener's disease).

Table 1. B-Mode Lung Ultrasound Data of Peripheral Pulmonary Granulomatous Lesions of the 10 Study Patients Patient Number Echogenicity of PPGLs Border of PPGLs Size of PPGLs Configuration of PPGLs 1 Hypoechoic Smooth ≥2 cm Oval 2 Hypoechoic Smooth <2 cm Round 3 Hypoechoic Smooth ≥2 cm Oval 4 Hypoechoic Smooth ≥2 cm Round 5 Hypoechoic Smooth <2 cm Round 6 Hypoechoic Smooth <2 cm Round 7 Hypoechoic Smooth ≥2 cm Oval 8 Hypoechoic Smooth ≥2 cm Round 9 Hypoechoic Smooth <2 cm Round 10 Hypoechoic Irregular delineated ≥2 cm Oval PPGLs, peripheral pulmonary granulomatous lesions. Contrast-Enhanced Ultrasound Data

Regarding TE, all lesions revealed a delayed enhancement due to BA perfusion. Furthermore, all lesions showed a reduced EE and an inhomogeneous arterial enhancement (HE) with evidence of NPAs on CEUS. The DE (washout time) was rapid (<120 seconds) in all PPGLs (Table 2) (Figure 3).

Table 2. Contrast-Enhanced Ultrasound Data of Peripheral Pulmonary Granulomatous Lesions and Its Histopathological Correlation in CD34 Antibody Staining for the 10 Study Patients Patient Number 1. CEUS TE (PA/BA) Corresponding Antibody Staining 2. CEUS EE Corresponding Antibody Staining 3. CEUS HE Corresponding Antibody Staining 4. CEUS DE Corresponding Antibody Staining 1 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 2 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 3 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 4 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 5 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 6 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 7 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 8 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 9 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable 10 BA Vascular pattern B Reduced Extent of neoangiogenesis NPA CAV Rapid In the capillary levels not detectable CEUS, contrast-enhanced ultrasound; TE, time to enhancement; BA, bronchial-arterial perfusion; EE, extent of enhancement; HE, homogeneity of enhancement; DE, decrease of enhancement; NPA, non-perfused areas; CAV, central avascular areas in granulomas.

A 46-year-old male patient (patient 8) with histologically confirmed granulomatous vasculitis with polyangiitis (former Wegener's disease). Illustration of a pleural-based pulmonary nodule on the right upper lobe of the lung on contrast-enhanced ultrasound (CEUS). A, After 10 seconds, CEUS shows an inhomogeneous reduced bronchial-arterial enhancement of the lesion simultaneous to the thoracic wall. B, Higher magnification of the area marked by a red rectangle in panel A. Arrowheads demark the lesion in the arterial phase. C, After 69 seconds, CEUS shows a rapid washout of the lesion. D, Higher magnification of the area marked by a red rectangle in panel C. Arrowheads demark the lesion in the parenchymal phase.

Histopathological Data

All PPGLs were histologically confirmed and included 4 (40%) sarcoidosis (Figure 4A), 4 (40%) granulomatosis with polyangiitis (GPA), 1 (10%) rheumatoid nodule, and 1 (10%) tuberculosis. The histological confirmation was performed by sonographically guided biopsy in 8 (80%) cases (Figure 2), by both sonographically guided biopsy and thoracic surgical intervention in 1 (10%) case and by a thoracic surgical intervention in 1 (10%) case. All sonographically guided biopsies were performed within 1 week after CEUS examination. The thoracic surgical interventions were performed in one case 3.5 months after CEUS examination and in another case 10.5 months after CEUS examination.

Histology of the vascularization of (A and B) a peripheral pulmonary granulomatous lesion in a 65-year-old female patient (patient #5) with the diagnosis of sarcoidosis; (C and D) a postmortem-obtained acute inflammatory lung tissue in a male patient, who died at age 58 due to acute pneumonia; and (E and F) a postmortem-obtained healthy lung tissue in a female patient, who died at age 80 due to sudden cardiac death. Lung biopsy specimen with granulomas (A, hematoxylin–eosin staining) showing central sclerosis with low vessel density (*) and peripheral inflammatory regions with high vessel density (B, CD34 immunohistochemistry, labeling vessels in brown, arrows). Acute pneumonia (C, hematoxylin–eosin staining) as a different inflammatory condition shows preserved reticular alveolar lung vasculature (D, CD34 immunohistochemistry) in between numerous inflammatory cells crowding the alveoli (#). Intact lung tissue control with regular reticular alveolar vascular architecture (E, hematoxylin–eosin staining; F, CD34 immunohistochemistry). Scale bars, 200 μm, respectively.

On immunohistochemical staining with CD34 antibody, central avascular areas in granulomas were identified in all PPGLs (Figure 4B). In addition, all PPGLs showed an absence of lung tissue and a vascular pattern B similar to angiogenesis in lung tumors (Figure 4B).

In both control groups (Figure 4, C and E), no avascular areas were present in any tissue sample (Figure 4, D and 4F). In healthy lung tissue (Figure 4E) and as well as in tissue infected due to acute pneumonia (Figure 4C), vascular pattern A was presented (Figure 4, D and F).

Discussion

For the proper evaluation of the sonographic phenomenology of pathological processes, a histological correlation in the learning phase is essential. The different patterns of echogenicity of tissues in B-mode LUS can be assigned to corresponding pathological processes. However, in principle, an evaluation of the nature of the lung lesions based solely on the B-mode LUS morphological features is not possible.20 Therefore, to evaluate the malignancy of the lesions, the different vascularization patterns of benign and malignant lesions could be used, visualized as perfusion patterns in CDS and CEUS.2, 4

The correlation of sonographic perfusion parameters in CDS and perfusion patterns in CEUS with histological vascularization patterns is a challenge. Therefore, the pathologist cannot clearly assign smaller capillary vessels in lung specimens to pulmonary or bronchial vessels using conventional methods.1 However, to correlate sonographically detected perfusion patterns with pathological vascularization patterns of lesions, the detection of neoangiogenesis by immunohistochemical examination can be helpful.

In this study, we examined the perfusion pattern of peripheral pulmonary granulomas using CEUS and the histopathological correlation with the CD34 antibody for neoangiogenesis. Pathological studies have shown that a disorganized and chaotic vascular pattern in immunohistochemical staining is the typical vascular pattern for tumor neoangiogenesis in malignant lesions of the lung.27

In our study, all peripheral pulmonary granulomas presented an identical perfusion pattern on CEUS regarding TE, EE, HE, and DE.

Regarding TE, all lesions showed a delayed systemic enhancement on CEUS in the arterial phase, suggestive of a BA supply.3, 4, 11 In the corresponding tissue sample, the histopathological examination showed a disorganized and chaotic vascular pattern similar to that of tumor neoangiogenesis.27 Based on the chaotic and disorganized vascular patterns in CD34 antibody staining that were found in PPGLs as an indication of neoangiogenesis and on the corresponding CEUS pattern, this neoangiogenesis could be considered as neoangiogenesis through bronchial arteries.1, 5 Our findings are in accordance with those of an angiographic study conducted in 1978, in which Babo et al. showed a BA vascularization of pulmonary tuberculosis granulomas.29 Furthermore, Linde et al. demonstrated in a retrospective study that acute pneumonias predominantly show a pulmonary arterial supply.30 These findings are in accordance with vascularization patterns of acute pneumonia in the immunohistochemical staining in our study. In contrast to acute inflammation such as pneumonia or pleurisy, granulomas are characterized by chronic inflammation.31 Here, it must be assumed that there is a loss of PA vascular supply with an increasing neovascularization from bronchial arteries caused by hypoxia, similar to that of malignant tumors.1, 5 The “shift of vascularization” of a primarily PA vascular supply to a BA vascular supply is therefore more an expression or indication of a chronic process.

Regarding EE, all PPGLs had reduced enhancement on CEUS in comparison with splenic enhancement. Extent of enhancement depends on the type of vascular supply. Thus, pulmonary-artery-supplied pathologies such as pneumonia and compression atelectasis commonly show a marked enhancement.20 In the case of BA supply, the enhancement of lesions correlates with the extent of tumor neoangiogenesis, which depends on the underlying pathology and the size of the lesion.32

Regarding HE, all PPGLs showed inhomogeneous enhancement on CEUS with evidence of centrally located NPAs. These NPAs could be correlated and demonstrated in all histopathological specimens due to the evidence of granulomas. Several peripheral pleural based lesions like peripheral pulmonary embolism and consolidation due to COVID-19 infection are described to have an inhomogeneous enhancement on CEUS. In these consolidations, the NPA are identified by histological confirmation to be infarcted areas and located strictly pleural based.21, 22

Regarding DE, all PPGLs had a rapid decrease of enhancement (<120 seconds) in the parenchymal phase on CEUS. The reason for the rapid DE with a rapid washout (<120 seconds) in the parenchymal phase could be speculated to be due to the presence of arteriovenous shunting, similar to that found in malignant tissue with tumor neoangiogenesis.33, 34 This phenomenon of decreased enhancement of PPGLs could not be correlated with a specific histopathological vascularization pattern in CD34 antibody staining. The detection of arteriovenous shunting was not possible by light microscopy.

The major limitation of our study is the small sample size. We examined only 10 histologically confirmed PPGLs with CEUS. Therefore, generalization of our results is limited.

Conclusion

In summary, we found that PPGLs on CEUS show an identical perfusion pattern with a reduced inhomogeneous BA pattern of enhancement and a rapid washout. Furthermore, for the first time, neoangiogenesis could be demonstrated as a histopathological correlation with BA pattern of enhancement on CEUS. In granulomas, the regular lung structure is completely destroyed, so that “lung foreign” tissue is present and therefore can imitate a malignant process in imaging methods.3, 11 Therefore, in the presence of the typical CEUS pattern suspected for a malignant pulmonary nodule, a granulomatous nodule is basically an important differential diagnosis and vice versa. A histological examination of all these lesions is strongly warranted.

Acknowledgments

The work of Dr. Ehsan Safai Zadeh is funded by part of a research grant from the Anneliese Pohl Foundation (Anneliese Pohl Stiftung), and we gratefully acknowledge this support. We thank Prof. Dr. Carsten Denkert, the Director of Institute of Pathology and Cytology at the University Hospital Gießen and Marburg in Marburg for supporting this study. Furthermore, we thank Viktoria Wischmann for providing expert technical assistance. Open Access funding enabled and organized by Projekt DEAL.