Features of radiological and physiological findings in pulmonary capillary hemangiomatosis: an updated pooled analysis of confirmed diagnostic cases

Abstract

Pulmonary capillary hemangiomatosis (PCH) is a very rare and refractory disease characterized by capillary angioproliferation. The updated classification of pulmonary hypertension categorizes PCH into a subgroup of pulmonary arterial hypertension (PAH) alongside pulmonary veno-occlusive disease (PVOD). However, the definitive diagnosis of PCH only with noninvasive tools remains difficult. The aim of this study was to elucidate the radiological and physiological characteristics of PCH. We searched for cases of pathologically confirmed PCH in the English literature published between 2000 and 2018. We identified 26 cases among 39 studies. Then, we extracted and evaluated the relevant clinical information in all cases with available data. On chest computed tomography (CT), ground-glass opacities (GGOs) were observed in 92% of the cases, in which poorly defined nodular pattern was the most common (88%). GGOs in a bat-wing distribution were observed in one case. Septal lines and lymph node enlargement were observed less frequently (each 19%, 12%). Seven cases (27%) had overlapping abnormalities. Diffusing capacity of the lung for carbon monoxide (DL_CO) was remarkably decreased. Alveolar hemorrhage by histological findings or bronchoalveolar lavage (BAL) was observed in seven cases. The present study showed that the most characteristic findings of CT in PCH was centrilobular GGOs with a poorly defined nodular pattern, and septal lines and lymph node enlargement were seen less frequently. Alveolar hemorrhage detected by BAL and decreased DL_CO may also be helpful to recognize the possibility of PCH like PVOD.

Keywords

pulmonary capillary hemangiomatosis (PCH)pulmonary veno-occlusive disease (PVOD)computed tomography

Introduction

Pulmonary capillary hemangiomatosis (PCH) is a rare form of pulmonary hypertension (PH) (<100 reported cases) that is characterized by capillary angioproliferation in the alveolar septa.¹ Although histological findings are required for the definitive diagnosis of PCH, obtaining adequate lung samples is usually difficult due to severe PH. Additionally, since its clinical characteristics are poorly defined, patients with PCH are frequently misdiagnosed,² and diagnosis is made after death or lung transplantation.³ Nevertheless, the prognosis is poor, and median survival from the onset of the initial symptom is reported as approximately three years.² Furthermore, as the common PCH symptoms, such as progressive dyspnea, cough, chest pain, and hemoptysis,⁴ are nonspecific, diagnostic clues for PCH by noninvasive modalities are much needed.

In the updated classification of PH from Sixth World Symposium on Pulmonary Hypertension (2018), PCH is categorized in the World Health Organization group 1.6, namely “pulmonary arterial hypertension (PAH) with overt features of venous/capillaries involvement” alongside pulmonary veno-occlusive disease (PVOD), because PVOD/PCH and PAH have been considered as a continuum of pulmonary vascular disease.⁵ Recently, some reports of PVOD/PCH have demonstrated that high-resolution chest tomography (HRCT) images, pulmonary function tests (PFTs), and bronchoalveolar lavage (BAL) can be useful modalities for noninvasive diagnostic examinations.^4,6 The characteristic findings of PVOD/PCH in HRCT are centrilobular ground-glass opacities (GGOs), septal lines, and mediastinal lymph node enlargement.⁷ PFTs, including forced expiratory volume in 1 second (FEV₁) and FEV₁/forced vital capacity (FVC) ratio, are normal, but the diffusing capacity of the lung for carbon monoxide (DL_CO) is remarkably decreased.⁸ Bronchoscopy is not usually performed, but BAL can reveal hemosiderin-laden macrophages in PVOD.⁹ However, although the clinical findings of “PVOD” or “combined PVOD and PCH” are occasionally reported, previous reports of clinical characteristics which focus on PCH were very limited. Therefore, the features of these noninvasive examinations in the PCH remain uncertain.

The aim of this study was to elucidate the radiological and physiological characteristics of PCH. Developing noninvasive diagnostic tools may help to raise suspicion of the presence of PCH and lead to beneficial effects for the patient with PCH, such as early treatment and improved prognosis and outcomes.

Methods

Literature search

To identify PCH cases, we searched English literature published between 2000 and 2018 via PubMed/Medline, the Cochrane Central Register of Controlled Trials, and the Cochrane Database of Systematic Reviews, using the keywords “pulmonary capillary hemangiomatosis” and/or “PCH.” Other keywords, including “pulmonary veno-occlusive disease” and/or “PVOD,” related to PCH were also searched in these databases.

The study was conducted in accordance with the ethical principles of 1964 Helsinki Declaration and subsequent amendments. The requirement for the approval by the Ethical Review Board of the Graduate School of Medicine of Chiba University was waived by the ethics committee because the analysis of the present study was limited to preexisting literature data.

Selection criteria

We selected studies or case reports on pathologically diagnosed PCH which described the radiological and physiological findings, such as chest radiograph, CT scan, PFTs, and/or bronchoscopic findings. Reports with (1) cases of infant, (2) no CT imaging information, (3) no pathological evidence of PCH, and (4) no evidence of PH were excluded. As a result, we identified 26 cases in 20 reports among 39 studies or case reports (Table 1). Then, clinical information, hemodynamic data, radiological findings (CT), physiological findings (PFTs and BAL), histological findings, treatment, and outcomes were extracted and evaluated in all cases with available data.

Table 1.

Review of 26 selected cases with pulmonary capillary hemangiomatosis.

Case	Reference	Age	Sex	CT findings				Alveolar hemorrhage	DL_CO (%)	Hemodynamics (mmHg)	Treatment	Diagnosis	Outcome
Case	Reference	Age	Sex	GGOs	Septal lines	LN enlargement	Others	Alveolar hemorrhage	DL_CO (%)	Hemodynamics (mmHg)	Treatment	Diagnosis	Outcome
1	Gugnani et al.¹⁷	66	F	(−)	(−)	(−)		(+)	50	92/32 (52)	PGI₂	Autopsy	Dead
2	Ito et al.¹⁸	25	M	Nodular pattern	NR	NR		NR	NR	61/33	PGI₂, NO	Autopsy	Dead
3	Lawler and Askin¹⁴	29	M	Nodular pattern	NR	NR		NR	50	NA	NR	OLB	NR
4	El-Gabaly et al.¹⁹	26	F	Nodular pattern	NR	(+)	Pericardial effusion	NR	40	NA	NA	OLB	Dead
5	Frazier et al.¹²	12	M	Nodular pattern	Few	(−)		NR	NR	NR	NR	Pathology^a	NR
6	Frazier et al.¹²	52	F	Nodular pattern	Few	(−)	Pericardial effusion	NR	NR	NR	NR	Pathology^a	NR
7	Frazier et al.¹²	22	F	Nodular pattern	(−)	(−)		NR	NR	NR	NR	Pathology^a	NR
8	Frazier et al.¹²	27	F	Nodular pattern	(−)	(−)	Pericardial effusion	NR	NR	NR	NR	Pathology^a	NR
9	Kothari et al.²⁰	13	M	Nodular pattern	(−)	(−)		NR	NR	RVSP 100	NR	LB	Dead
10	McGuire et al.²¹	52	F	Nodular pattern	NR	(+)		NR	40	72/39	PGI₂, PSL	Autopsy	Dead
11	Lee et al.²²	48	F	Nodular pattern	NR	NR		NR	NR	RVSP 80	PGI₂, ERA, diuretics	LT	Dead
12	Sakashita et al.²³	47	M	(+)	(+)	(−)	Emphysema	(+)	NR	RVSP 90	PSL, OT	Autopsy	Dead
13	Xie et al.¹⁶	28	M	Nodular pattern	NR	NR	Emphysema	(+)	21	RVSP 86	Diuretics, OT	OLB	Dead
14	Xie et al.¹⁶	19	F	Nodular pattern	NR	NR		(+)	NR	49/25 (35)	OT	VATS	Alive
15	Miura et al.¹⁵	11	M	Nodular pattern	(−)	(+)		NR	64	mPAP 52	PGI₂	Autopsy	Dead
16	Miura et al.¹⁵	17	F	Nodular pattern	(+)	(−)		NR	NA	NR	NA	Autopsy	Dead
17	Miura et al.¹⁵	25	F	Nodular pattern	(+)	(+)		NR	36	mPAP 55	PGI₂	LT	NR
18	Faria et al.¹³	25	M	Nodular pattern	(−)	NR		NR	20	mPAP 65	Diuretics	Autopsy	Dead
19	Adachi et al.²⁴	43	M	Nodular pattern	(+)	(+)		NR	22.5	64/22 (35)	PGI₂, PDE5-I, ERA, imatinib	Autopsy	Dead
20	Babu et al.²⁵	19	F	Nodular pattern	NR	(−)		(+)	NR	RVSP 71	Diuretics, OT	VATS	Alive
21	Nayyar et al.²⁶	63	F	(+)	NR	NR	Pleural effusion	NR	52	70/38	PGI₂, PDE5-I, ERA, imatinib	LB	Dead
22	Rea et al.¹¹	53	M	(−)	(+)	NR	Pleural effusion	BAL(+)	NR	RVSP 70	Carvedilol, diuretics	LB	Alive
23	Odronic et al.²⁷	55	M	Nodular pattern	NR	NR		NR	NR	NR	PGI₂, PDE5-I, ERA	LT	Alive
24	Diao et al.²⁸	65	F	Nodular pattern	NR	NR		NR	NR	RVSP 78	PDE5-I, PSL	OLB	Alive
25	Sharma et al.²⁹	14	M	Nodular pattern	(−)	NR		NR	NR	NR	NR	LB	NR
26	Wada et al.¹⁰	27	F	Bat-wing like	Few	(−)		BAL(+)	24	67/25 (45)	PGI₂, PDE5-I, ERA	LT	Alive

BAL: bronchoalveolar lavage; CT: computed tomography; DL_CO: diffusing capacity of the lung for carbon monoxide; ERA: endothelin receptor antagonists; F: female; GGOs: ground-glass opacities; LN: lymph node; M: male; mPAP: mean pulmonary arterial pressure; LB: lung biopsy; LT: lung transplantation; NA: not applicable; NO: nitric oxide; NR: not reported; OLB: open lung biopsy; OT: oxygen therapy; PDE5-I: phosphodiesterase type 5 inhibitor; PGI₂: prostacyclin; PSL: prednisolone; RVSP: right ventricular peak pressure; VATS: video-assisted lung surgery.

Not reported for methods.

Results

Clinical characteristics of PCH

Clinical characteristics are described in Tables 1 and 2. The percentage of males and females included was almost equal (ratio, 1:1.2), and the mean age at diagnosis was 34.0 ± 17.8 (range, 11–66 years). Hemodynamic parameters including mean pulmonary arterial pressure and right ventricular systolic pressure indicated severe PH. Treatments were reported in 17 patients. Prostanoids were the most popular, and some patients were treated with imatinib, nitric oxide, or prednisolone. Oxygen therapy and diuretics were other treatments prescribed. Various treatments were not effective for 10 out of 17 patients. In several cases, intravenous prostanoid or imatinib were temporarily effective. In two cases, acute pulmonary edema was observed after initiation of epoprostenol. Only three cases showed good control by symptomatic treatment with oxygen therapy, diuretics, and/or heparinization without vasodilators. Two cases underwent lung transplantation successfully while receiving vasodilators. In many cases, the diagnosis was established by autopsy (31%), and four patients underwent lung transplantation. The disease outcome for many patients was death, and the time from onset to death was approximately two years.

Table 2.

Clinical characteristics of 26 cases with pulmonary capillary hemangiomatosis.

Variables	Results
Male/female (n)	12/14
Age at diagnosis (years)^a	34.0 ± 17.8
Mean PAP (mmHg)^a	48.4 ± 10.9
Mean RVSP (mmHg)^a	82.1 ± 10.7
%DL_CO (%)^a	38.1 ± 14.9
Alveolar hemorrhage, n (%)	7 (27)
Treatment (n)
Prostanoids	10 (38)
PDE5 inhibitors	5 (19)
ERAs	5 (19)
Imatinib	2 (8)
Nitric oxide	1 (4)
Prednisolone	3 (12)
Definitive diagnostic criteria, n (%)
Autopsy	8 (31)
Lung biopsy	4 (15)
Open lung biopsy	4 (15)
Video-assisted thoracoscopic lung biopsy	2 (8)
Lung transplantation	4 (15)
Outcome
Dead/alive (n)	13/6
Prognosis (month)^a	22.8 ± 18.8

DL_CO: diffusing capacity of the lung for carbon monoxide; ERAs: endothelin receptor antagonists; PAP: pulmonary arterial pressure; PDE5: phosphodiesterase type 5; RVSP: right ventricular peak pressure.

Results are expressed as mean ± SD.

Radiological findings

Radiological characteristics are described in Table 3. GGOs were observed in 24/26 cases (92%), in which a poorly defined nodular pattern was the most common (21/24, 88%). GGOs in a diffuse bat-wing distribution were observed in the lung field of one patient (case 26). Septal lines (19%) and lymph node enlargement (12%) were observed less frequently than GGOs. Seven patients (27%) presented two or three common abnormalities (centrilobular GGOs, septal lines, and lymph node enlargement). Perfusion lung scans were performed in seven cases. Most scans had been done as screening for chronic thromboembolic PH. In case 26, perfusion lung scans showed slight subsegmental defects.

Table 3.

Radiological characteristics of 26 cases with pulmonary capillary hemangiomatosis.

Computed tomography findings	Results
Ground-glass opacities, n (%)	24/26 (92)
Nodular pattern	21/24 (88)
Bat-wing pattern	1/24 (4)
Septal lines	5/26 (19)
Lymph node enlargement	3/26 (12)
Others, n (%)
Pleural effusion	2 (8)
Pericardial effusion	3 (12)
Emphysema	2 (8)
Pulmonary fibrosis	1 (4)

Regarding the bat-wing distribution of GGOs, this was observed in a patient with PCH reported from our hospital. HRCT of the chest showed a rare radiological form of GGOs distributed in a diffuse bat-wing pattern¹⁰ in the lung field, but sparing the subpleural parenchyma (Fig. 1a). There were a few thickened interlobular septa in the lower lobes. Lymph node enlargement was not seen. BAL revealed hemosiderin-laden macrophages. Based on these clinical findings, we diagnosed as PVOD with secondary pulmonary hemosiderosis. She was treated with sildenafil, bosentan, and epoprostenol. After three years from the onset of symptoms, she underwent double-lung transplantation. Histological findings demonstrated capillary proliferation with multiple layers in patchy areas, arterial intimal fibrosis, medial hypertrophy, and hemosiderin-laden macrophages in the alveolar spaces. There were a few pulmonary venous occlusions. Because these pathological findings were consistent with PCH, she was finally diagnosed with PCH.

Fig. 1.

Typical pattern of CT in PCH and PVOD. (a) Unique pattern of PCH: GGOs distributed diffusely in a bat-wing pattern in the lung field, sparing subpleural parenchyma,¹⁰ (b) typical pattern of PCH: diffuse centrilobular GGOs with poorly defined nodular pattern,¹³ and (c) typical pattern of PVOD: centrilobular GGOs with poorly defined nodular pattern, septal lines, and mediastinal lymph node enlargement (white dotted lines).⁶

Physiological findings (PFTs and BAL)

In the present analysis, the findings of remarkably decreased %DL_CO were identified in 11 cases (38.1 ± 14.9%, range 20%–64%). Many cases with available data showed normal or mild decreased values of FEV₁ and FEV₁/FVC ratio. Additionally, hemosiderosis detected by histological findings or BAL was observed in 7/26 cases (Table 2), which might be related to diffuse or nodular GGOs in CT. Among these cases, only two cases reported the findings of BAL.^10,11 Conversely, no cases reported transbronchial lung biopsy (TBLB). However, “lung biopsy” (not including open lung biopsy) was performed in four cases (Table 2).

Pathological findings

In the present study, histological confirmation of PCH was obtained in all cases, and detailed description of pathological findings was provided for 22 cases. All the 22 cases showed capillary proliferation. Pulmonary arterial changes with intimal hyperplasia and/or fibrosis without plexiform lesions were described in 9 cases. In addition, alveolar hemorrhage was shown in five cases, and no venous occlusion was confirmed in six cases.

Discussion

The present study explored the features of radiological and physiological findings in pathologically diagnosed PCH. The major findings of this study were that (1) the most characteristic findings of CT in PCH is centrilobular GGOs with poorly defined nodular pattern (21/26 cases, 81%), (2) septal lines and lymph node enlargement are less frequently observed (each 19%, 11%), and (3) evidence of alveolar hemorrhage using BAL (2/2, 100%) and decreased DL_CO on PFTs (10/11 cases, 91%) also appear to be helpful noninvasive assessments for the recognition of PCH.

In 1989, Faber et al.³⁰ first reviewed the clinical and pathologic features of PCH. They described that a diffuse bilateral reticulonodular pattern was typical in chest radiographs of patients with PCH.³⁰ Then, several cases of PCH^31,32 and literature about HRCT findings of PCH were recently reported.^1,12 In these reports, the most common characteristics of PCH compared with PAH are centrilobular GGOs, septal lines, and lymph node enlargement in HRCT images. GGO patterns in HRCT of PCH patients might also have some specific characteristics, because GGOs with a poorly defined nodular pattern, diffusely, bilateral in distribution, and sparing of the periphery are reported in PCH cases^13,31,32 (Fig. 1b). Frazier et al.¹² also mentioned that CT findings of PCH were characterized by widespread ill-defined centrilobular nodules of GGOs in addition to main pulmonary arterial enlargement. In the present study, centrilobular GGOs with nodular patterns was the most common radiological feature, which is consistent with these previous reports.^13,31,32 Centrilobular GGOs may represent focal tumor-like capillary angioproliferation in the alveolar septa,^14,32 locally increased perfusion to the lesion, and secondary alveolar hemorrhage (i.e., hemosiderosis) in the airspace.³⁰ Regarding bat-wing pattern of GGOs, this type of GGO is sometimes observed in pulmonary edema, interstitial lung disease, PH, lymphoproliferative disorder, and hemosiderosis. We speculate that this bat-wing pattern in which GGOs were distributed is in correlation with both the primary lesion of PCH and occult alveolar hemorrhages because histological findings of the transplanted lung showed capillary angioproliferation in patchy areas and abundant hemosiderin-laden macrophages (case 26; see Fig. 1a). In the present study, other CT findings, such as pleural effusion (8%), pericardial effusion (12%), and emphysema (8%) were less frequent. Therefore, these CT findings may be less helpful to recognize the possibility of PCH.

As the present study analyzed radiological and physiological characteristics of PCH only, and not PVOD, the difference in such findings between PCH and PVOD was not clarified. Regarding histological findings, PCH is characterized by abnormal capillary angioproliferation at least two layers thick and extends along septal lines.^1,33 The major feature distinguishing PCH from other diseases is the infiltration of capillary vessels into other pulmonary structures, including the walls of arteries and veins, intralobular fibrous septa, and bronchi.³³ On the other hand, main lesion of PVOD is venous occlusion, and capillary angioproliferation is also found as secondary change of venous occlusion.^3,6 In PVOD, the infiltration into other structures is not observed, and capillary angioproliferation is seen at single layer.³³ In the present study, three characteristics of HRCT images in PCH were shown: centrilobular GGOs, septal lines, and lymph node enlargement. Nonetheless, these are also known as characteristics of PVOD³⁴ (Fig. 1c). Patterns of GGOs in PVOD may vary from a minimal patchy appearance to an extensively diffuse one,³⁵ while in the present study, the typical pattern of GGOs in PCH comprised a poorly defined nodular pattern. Additionally, septal lines are observed in 93% of patients with PVOD and adenopathy, in 80% of those with PVOD.³⁴ When there are two or three common findings in HRCT, PVOD can be distinguished from PAH with a 75% sensitivity and an 84.6% specificity.⁷ Conversely, our study of PCH cases demonstrated that septal lines and lymph node enlargement were observed less frequently (19% and 12%, respectively) than in PVOD. Therefore, when a patient with PH is suspected to have PVOD/PCH, we speculate that centrilobular GGOs with poorly defined nodular pattern and absence of septal lines or lymph node enlargement may be more suggestive of PCH than PVOD. However, as confirmed diagnostic cases of PCH are very limited, the difference in these CT findings between PCH and PVOD was uncertain, and further investigation is needed. Indeed, previous report indicated that patients with PCH show larger size of GGOs in HRCT compared to those with PVOD.¹⁵ Histologically, PCH presents tumor-like capillary angioproliferation, and PVOD presents dilated capillary vessels due to changes secondary to pulmonary venous occlusion. Thus, the different condition of the capillaries between PCH and PVOD seems to lead to differences in the sizes of centrilobular GGOs that are observed via HRCT.¹⁵

Among the physiological examinations based on noninvasive modalities, PFTs are known as a useful tool to recognize the possibility of PVOD/PCH. In PVOD, some previous reports have shown that the features of PFTs are normal values of FEV₁ and FEV₁/FVC and remarkably decreased DL_CO (frequently DL_CO < 55%).^7,8 DL_CO/VA in patients with PVOD is significantly decreased compared with PAH.⁷ In PVOD, interstitial edema and/or pulmonary capillary angioproliferation due to chronic pulmonary venous obstruction may be associated with severe reduction of DL_CO.⁹ On the other hand, severe reduction of DL_CO in PCH may be explained by the underlying pathophysiology; the proliferative capillaries and alveolar hemorrhage fill the alveolar septa in PCH. The gas exchange mechanism may be impaired in such pathophysiology. Previously, it has been reported that the gas exchange becomes more severe in PCH when there is pulmonary artery narrowing, venous occlusion; subsequently, alveolar hemorrhage appears in addition to capillary angioproliferation.¹⁴ However, there are only few review articles specifically describing PFT findings in PCH. Almagro et al.² demonstrated that the findings of PFTs, FVC, FEV₁, and FEV₁/FVC in PCH were almost normal and that there was a marked reduction of DL_CO, similar to those in PVOD. In the present study, the %DL_CO decreased markedly in almost all cases. Therefore, we speculate that the severe reduction of DL_CO may be one of a meaningful finding in PCH. Another noninvasive tool is bronchoscopy; however, this is not usually examined in patients with PAH, especially those with severe PH. Although the findings of CT are helpful for the diagnosis of PVOD/PCH, other lung diseases, such as interstitial lung disease, infections, or lymphoproliferative disorders, also present GGOs. Hemosiderin-laden macrophages in BAL correlate with occult alveolar hemorrhage.⁶ Because the main lesion of PVOD is postcapillary pulmonary vessels, alveolar hemorrhage is a common characteristic in patients with PVOD. Reportedly, patients with PVOD present a higher percentage of hemosiderin-laden macrophages and Golde score compared with those with idiopathic PAH.⁹ Therefore, BAL is one of the essential noninvasive tools for the detection of PVOD. Kano et al. reported that a child with PVOD showed GGOs distributed diffusely in a bat-wing pattern in the lung field but were sparing the lung periphery. That patient’s BAL also revealed hemosiderin-laden macrophages.³⁶ Bat-wing GGO is also observed in cases of idiopathic pulmonary hemosiderosis, which is characterized by hemoptysis and alveolar capillary hemorrhage.³⁷ It has been speculated that GGOs distributed diffusely in a bat-wing pattern were caused not only by PVOD itself but also by massive alveolar hemorrhage. Recently, it has been proposed that hemosiderin-laden macrophages in the sputum may also support the diagnosis of PVOD and distinguish PVOD from other forms of PH. The reported sensitivity was 100% and the specificity was 97% for the clinical diagnosis of PVOD.³⁸ Conversely, there is little evidence on whether BAL supports the clinical diagnosis of PCH. Xie et al.¹⁶ reported that among 64 cases, some cases showed hemosiderin-laden macrophages in BAL, and BAL may be a useful tool for the diagnosis of PCH. In the present study, we only identified two cases in which pulmonary alveolar hemorrhage by BAL was detected. The main lesion in PCH, capillary angioproliferation, also seems to cause an alveolar hemorrhage. Therefore, alveolar hemorrhage detected by BAL may be considered as one of the useful noninvasive diagnostic tools for PCH. As the number of cases is very limited in the present study, further analysis with more number of cases is required to evaluate BAL in PCH cases. Conversely, TBLB (reported as “lung biopsy”) was performed in four cases (Table 2). As the histological evaluation to identify characteristic changes of PCH requires an extensive lung tissue evaluation (a minimum of five blocks from surgical lung biopsies), TBLB is not appropriate for adequate sampling.³

This study has several limitations. First, most of the selected articles were single case reports, and the number of cases was very limited because confirmed PCH cases are rare. Second, we could not verify images of CT or other information (PFTs, BAL, hemodynamics, diagnostic tools, treatments, and outcomes) in all cases. Third, we did not analyze the findings of PCH compared to PVOD directly. Fourth, CT images were not scanned using the same equipment or in similar settings.

In conclusion, our study showed that the most characteristic findings of CT in PCH were centrilobular GGOs with a poorly defined nodular pattern, while septal lines and lymph node enlargement were seen less frequently. Severe reduction of DL_CO on PFTs and alveolar hemorrhage detected by BAL may also help to establish the diagnosis of PCH like PVOD. However, if these findings are obtained, we always need to rule out other diseases to make a definitive diagnosis of PCH. Though the clinical diagnosis of PCH is difficult, developing noninvasive diagnostic tools may be useful in suggesting the diagnosis of PCH and lead to beneficial effects, such early detection and treatment, which may help improve the prognosis and outcomes of patients with PCH.

Footnotes

Acknowledgments

We sincerely thank Dr Takashi Ogura for his advice regarding the interpretation of HRCT in a case with PCH and Mrs Chieko Handa, Mrs Tamie Hirano, and Miki Sakurai for their technical assistance and general support.

Author contributions

All authors contributed to the conceptual design of the analysis. RA, JT, and AS collected data and wrote the paper. RA and JT performed the analysis. All authors took part in drafting and revising the article. All authors have read and approved this final manuscript.

Conflict of interest

The author(s) declare that there is no conflict of interest.

Funding

This study was partly supported by the Japanese Ministry of Health, Labour and Welfare research grants specifically designated to the Respiratory Failure Research Group and Cardiovascular Diseases and the Pulmonary Hypertension Research Group from the Japan Agency for Medical Research and Development (No. 16ek0109127h0002). The funder had no role in the study’s design, collection of data or their analysis, decision to publish, or preparation of the manuscript.

ORCID iD

Ayako Shigeta

References

O’Keefe

Post

. Pulmonary capillary hemangiomatosis: a rare cause of pulmonary hypertension. Arch Pathol Lab Med 2015; 139: 274–277.

Almagro

Julia

Sanjaume

, et al. Pulmonary capillary hemangiomatosis associated with primary pulmonary hypertension: report of 2 new cases and review of 35 cases from the literature. Medicine (Baltimore) 2002; 81: 417–424.

Lantuéjoul

Sheppard

Corrin

, et al. Pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis: a clinicopathologic study of 35 cases. Am J Surg Pathol 2006; 30: 850–857.

Chaisson

Dodson

Elliott

. Pulmonary capillary hemangiomatosis and pulmonary veno-occlusive disease. Clin Chest Med 2016; 37: 523–534.

Simonneau

Montani

Celermajer

, et al. Haemodynamic definitions and updated clinical classification of pulmonary hypertension. Eur Respir J 2019; 53: pii: 1801913.

Montani

Price

Dorfmuller

, et al. Pulmonary veno-occlusive disease. Eur Respir J 2009; 33: 189–200.

Montani

Achouh

Dorfmuller

, et al. Pulmonary veno-occlusive disease: clinical, functional, radiologic, and hemodynamic characteristics and outcome of 24 cases confirmed by histology. Medicine (Baltimore) 2008; 87: 220–233.

Holcomb

Jr Loyd

Ely

, et al. Pulmonary veno-occlusive disease: a case series and new observations. Chest 2000; 118: 1671–1679.

Rabiller

Jais

Hamid

, et al. Occult alveolar haemorrhage in pulmonary veno-occlusive disease. Eur Respir J 2006; 27: 108–113.

10.

Wada

Nakajima

Suzuki

, et al. Pulmonary capillary hemangiomatosis diagnosed by pathology of explanted lungs: a unique etiology serves as a key of clinical diagnosis. Gen Thorac Cardiovasc Surg 2019; 67: 332–335.

11.

Rea

Valente

de Rosa

, et al. Pulmonary capillary hemangiomatosis: a diagnostic challenge. Arch Bronconeumol 2015; 51: 98–99.

12.

Frazier

Franks

Mohammed

, et al. From the archives of the AFIP: pulmonary veno-occlusive disease and pulmonary capillary hemangiomatosis. Radiographics 2007; 27: 867–882.

13.

Faria

Carneiro

Tiradentes

, et al. Pulmonary capillary hemangiomatosis: an uncommon cause of pulmonary hypertension. J Bras Pneumol 2013; 39: 390–392.

14.

Lawler

Askin

. Pulmonary capillary hemangiomatosis: multidetector row CT findings and clinico-pathologic correlation. J Thorac Imaging 2005; 20: 61–63.

15.

Miura

Akagi

Nakamura

, et al. Different sizes of centrilobular ground-glass opacities in chest high-resolution computed tomography of patients with pulmonary veno-occlusive disease and patients with pulmonary capillary hemangiomatosis. Cardiovasc Pathol 2013; 22: 287–293.

16.

Xie

Dai

Jin

, et al. Clinical features and imaging findings in pulmonary capillary hemangiomatosis: report of two cases and a pooled analysis. Chin Med J (Engl) 2012; 125: 3069–3073.

17.

Gugnani

Pierson

Vanderheide

, et al. Pulmonary edema complicating prostacyclin therapy in pulmonary hypertension associated with scleroderma: a case of pulmonary capillary hemangiomatosis. Arthritis Rheum 2000; 43: 699–703.

18.

Ito

Ichiki

Ohi

, et al. Pulmonary capillary hemangiomatosis with severe pulmonary hypertension. Circ J 2003; 67: 793–795.

19.

El-Gabaly

Farver

Budev

, et al. Pulmonary capillary hemangiomatosis imaging findings and literature update. J Comput Assist Tomogr 2007; 31: 608–610.

20.

Kothari

Jagia

Gupta

, et al. Images in cardiovascular medicine. Pulmonary capillary hemangiomatosis. Circulation 2009; 120: 352–354.

21.

McGuire

Kennelly

Tillack

, et al. Pulmonary capillary hemangiomatosis associated with CREST syndrome: a case report and review of the literature. Respiration 2010; 80: 435–438.

22.

Lee

Suh

Krishnam

, et al. Recurrent pulmonary capillary hemangiomatosis after bilateral lung transplantation. J Thorac Imaging 2010; 25: W89–W92.

23.

Sakashita

Motooka

Suganuma

, et al. A case of pulmonary capillary hemangiomatosis with pulmonary fibrosis associated with MMP-9 related pulmonary remodeling. Pathol Int 2011; 61: 306–312.

24.

Adachi

Hirashiki

Kondo

, et al. Imatinib is partially effective for the treatment of pulmonary capillary hemangiomatosis. Intern Med 2014; 53: 603–607.

25.

Babu

Supraja

Singh

. Pulmonary capillary haemangiomatosis: a rare cause of pulmonary hypertension. Indian J Chest Dis Allied Sci 2014; 56: 259–262.

26.

Nayyar

Muthiah

Kumarasinghe

, et al. Imatinib for the treatment of pulmonary arterial hypertension and pulmonary capillary hemangiomatosis. Pulm Circ 2014; 4: 342–345.

27.

Odronic

Narula

Budev

, et al. Pulmonary capillary hemangiomatosis associated with connective tissue disease: a report of 4 cases and review of the literature. Ann Diagn Pathol 2015; 19: 149–153.

28.

Diao

Jin

. Pulmonary capillary hemangiomatosis associated with CREST syndrome: a challenge of diagnosis and treatment. Chin Med J (Engl) 2017; 130: 2645–2646.

29.

Sharma

Pandey

Kumar

. Pulmonary capillary haemangiomatosis causing pulmonary arterial hypertension: a clinician’s conundrum. BMJ Case Rep 2018. doi: 10.1136/bcr-2018-227393.

30.

Faber

Yousem

Dauber

, et al. Pulmonary capillary hemangiomatosis. A report of three cases and a review of the literature. Am Rev Respir Dis 1989; 140: 808–813.

31.

Eltorky

Headley

Winer-Muram

, et al. Pulmonary capillary hemangiomatosis: a clinicopathologic review. Ann Thorac Surg 1994; 57: 772–776.

32.

Lippert

White

Cameron

, et al. Pulmonary capillary hemangiomatosis: radiographic appearance. J Thorac Imaging 1998; 13: 49–51.

33.

Peacock

Rubin

. Pulmonary circulation: diseases and their treatment, 2nd ed. London: Arnold, 2004.

34.

Resten

Maitre

Humbert

, et al. Pulmonary hypertension: CT of the chest in pulmonary venoocclusive disease. AJR Am J Roentgenol 2004; 183: 65–70.

35.

Swensen

Tashjian

Myers

, et al. Pulmonary venoocclusive disease: CT findings in eight patients. AJR Am J Roentgenol 1996; 167: 937–940.

36.

Kano

Nakamura

Sakamoto

. Pulmonary veno-occlusive disease in an 11-year-old girl: diagnostic pitfalls. Pediatr Int 2014; 56: 119–122.

37.

Khorashadi

Betancourt

, et al. Idiopathic pulmonary haemosiderosis: spectrum of thoracic imaging findings in the adult patient. Clin Radiol 2015; 70: 459–465.

38.

Lederer

Muggli

Speich

, et al. Haemosiderin-laden sputum macrophages for diagnosis in pulmonary veno-occlusive disease. PLoS One 2014; 9: e115219.