|Home | About | Journals | Submit | Contact Us | Français|
Pulmonary capillary hemangiomatosis is a rare, vascular obstructive disorder that uniformly causes pulmonary arterial hypertension. Clinically, pulmonary capillary hemangiomatosis is indistinguishable from primary pulmonary arterial hypertension and histology is required for definitive diagnosis. The distinctive histologic feature of pulmonary capillary hemangiomatosis is non-malignant extensive proliferation of capillaries in the alveolar septae. Vasodilator treatment of humans with primary arterial hypertension due to pulmonary capillary hemangiomatosis can result in fatal acute pulmonary edema. Computed tomography is thus critical to discern pulmonary capillary hemangiomatosis from other causes of pulmonary arterial hypertension prior to vasodilator therapy. This is the first report of a vasoproliferative process resembling pulmonary capillary hemangiomatosis in the feline species.
A 15-year-old, male castrated, domestic shorthair cat presented for persistent labored breathing presumptively due to congestive heart failure despite treatment with diuretics for 7 days. Echocardiography showed evidence of hypertrophic cardiomyopathy with severe pulmonary hypertension; however, a normal sized left atrium was not consistent with congestive heart failure. Thoracic computed tomography was performed and showed evidence of diffuse ill-defined nodular ground glass opacities, enlarged pulmonary arteries, and filling defects consistent with pulmonary thromboembolism. The cat acutely decompensated after a single dose of sildenafil and was euthanized. Histopathology of the lungs showed severe multifocal alveolar capillary proliferation with respiratory bronchiolar infiltration, marked type II pneumocyte hyperplasia and multifocal pulmonary arterial thrombosis.
This is the first description in a cat of a vasoproliferative disorder resembling pulmonary capillary hemangiomatosis complicated by multifocal pulmonary arterial thrombosis. Inspiratory and expiratory ventilator-driven breath holds with angiography revealed lesions predominantly characterized by ground glass opacification and vascular filling defects with absence of air trapping. The results from this report suggest that, as in humans, the cat can develop a pulmonary capillary hemangiomatosis-like disease in which vasodilator therapy to address pulmonary hypertension may lead to fatal pulmonary edema.
Pulmonary capillary hemangiomatosis (PCH) is a rare, idiopathic vascular disease that uniformly causes pulmonary arterial hypertension (PAH) . It was first described in humans in 1978 as researchers observed atypical proliferation of capillary-like channels in lung tissue that appeared to be angiomatous growths . The distinctive histologic feature of PCH is the proliferation of capillaries in the pulmonary parenchyma [1, 2]. Additional important features include evidence of invasion by the capillaries into one or more of the pulmonary veins and arteries, alveolar walls and alveolar space, interlobular fibrous septa, and bronchioles . Cytologic atypia and mitoses are absent in PCH lesions . There have been under 100 cases reported in humans in the literature . PCH as a feature of pulmonary venoocclusive disease (PVOD), an obstructive disorder affecting the post-capillary (venous) pulmonary vasculature that has been reported in humans  and dogs . To the author’s knowledge PCH has not been described in the cat.
The pathogenesis of PCH is poorly understood and is likely multifactorial. In humans, PCH has been described in association with aortic stenosis , Kartagener syndrome , systemic lupus erythematous , scleroderma , Takayasu’s arteritis , hypertrophic cardiomyopathy  and neoplasia [12, 13]. Mutations in the eukaryotic translation initiation factor 2 alpha kinase 4 (EIF2AK4) gene are a risk factor in the development of PCH . It has also been proposed that PCH is not a separate disease, but rather represents a secondary angioproliferative response to pulmonary venous hypertension as seen in PVOD .
The current clinical classification of PH in people has been categorized in five groups of disorders: pulmonary arterial hypertension (Group 1), PVOD and/or PCH (Group 1’), persistent pulmonary hypertension of the newborn (Group 1”), pulmonary hypertension due to left heart disease (Group 2), pulmonary hypertension due to chronic lung diseases and/or hypoxia (Group 3), chronic thromboembolic pulmonary hypertension (Group 4), and pulmonary hypertension due to unclear multifactorial mechanisms (Group 5) . There is a paucity of information regarding PH in cats, mainly being limited to case reports and small case series. Our proposed adaptation of the classification system used in people with PAH to cats based on the available literature is shown in Table 1 [16–38].
The prognosis for PCH in humans is poor, and without lung transplant ultimately fatal. The median survival is approximately 3 years after onset of clinical signs [1, 39]. The only definitive treatment is lung transplantation, though some therapies (i.e. imatinib [40, 41], doxycycline , interferon [alpha]-2a ) have been shown to improve quality of life and survival time. Clinical signs associated with PCH include progressive dyspnea, exertional fatigue, syncope, and sometimes a chronic cough  and are frequently seen in other forms of PAH. The commonality in presenting clinical signs makes differentiating PCH from other forms of PAH challenging. While difficult, this distinction is essential because conventional treatment with vasodilators for PAH is contraindicated in patients with PCH as it can result in fatal pulmonary edema [44–46].
Histopathologic examination of lung tissue for definitive confirmation of PCH requires invasive sampling and can be contraindicated depending on patient stability. In cats, high resolution computed tomography (HRCT) has been used to help identify interstitial lung diseases including interstitial pulmonary fibrosis , alveolar filling disorders such as pulmonary alveolar proteinosis  and vascular disease like pulmonary thromboemboli  and partial anomalous pulmonary venous connection . In humans, HRCT has been used to differentiate PAH from PCH . Main pulmonary arterial enlargement and diffuse ill-defined centrilobular nodules of ground glass opacity are commonly seen in people with PCH but the latter are not seen with PAH [1, 51]. With increased recognition of other causes of pulmonary hypertension, and in particular PAH in the cat, future use of HRCT may be an invaluable diagnostic for pulmonary hypertension in Groups 1, 3, and 4 (see Table 1). Importantly, prior to initiating vasodilator therapy, HRCT may allow differentiation of PCH and PVOD from PAH in veterinary species as it does in humans [1, 51].
A 15-year old male castrated domestic shorthair cat was referred to the University of Missouri Veterinary Health Center for evaluation of presumed congestive heart failure. The cat was clinically healthy until the age of 14 years, when he experienced an episode of abrupt, severe fatigue while playing with a toy. The cat did not lose consciousness and recovered quickly with no medical intervention. The cat remained clinically normal for approximately a year, at which time he suffered a brief syncopal event while playing with a toy. The cat was evaluated by the referring veterinarian and treated with furosemidea (2.2 mg/kg, PO, q12 hr) for presumptive congestive heart failure. The cat subsequently developed a decreased appetite and an increase in respiratory effort and rate until presentation 7 days later.
The cat was the only animal in the house and never ventured outdoors. The owner did not have any known potentially hazardous hobbies (i.e. painting, crafts, furniture restoration). There was no known exposure to noxious/irritant inhalational agents (i.e. cigarette smoke, air-freshener, incense, etc.). The cat was up to date on vaccinations as well as heartworm and flea preventative.
At presentation the cat was bright, alert, and responsive. He showed tachypnea (60 breaths/minute) with increased inspiratory effort. He was mildly hypothermic (rectal temperature 99°F), and was approximately 5% dehydrated. His heart rate was 160 beats/minute with an occasional irregular rhythm that resulted in pulse deficits. Thoracic auscultation revealed a grade III/VI left parasternal systolic murmur and normal respiratory sounds. The hair coat appeared dull and unkempt. The cat weighed 5.6 kg (12 lbs). The rest of the physical examination was unremarkable.
Thoracic radiographs showed cardiomegaly and mild enlargement of the right pulmonary artery. The left pulmonary was larger than the left pulmonary vein proximally but became ill defined at the level of the 9th rib. In addition, there was a multi-focal interstitial pattern within the ventral aspect of caudal lung lobes (Fig. 1). Serum antigen for feline heartworm and feline leukemia virus and serum antibodies for feline immunodeficiency virus were not detected.b Serum total T4 concentration was normal (2.3ug/dL; reference interval, 0.8 to 4.0 ug/dL). Serum biochemical parameters outside of the reference interval were glucose (178 mg/dL; reference interval 52 to 153 mg/dL), blood urea nitrogen (59 mg/dL; reference interval 17 to 35 mg/dL), creatinine (3.1 mg/dl; reference interval 0.5 to 2.2 mg/dL), and potassium (2.5 mEq/L; reference interval 3.0 to 4.7 mEq/L). Hematologic abnormalities were neutrophilia (13.34×103/uL; reference interval, 2.5 to 12.5×103/uL), lymphopenia (1.48×103/uL; reference interval, 1.5 to 7.0×103/uL), and monocytosis (1.15×103/uL; reference interval, 0 to 0.85×103/uL). A urinalysis obtained via cystocentesis showed a urine specific gravity of 1.019. The hematologic abnormalities and mild hyperglycemia were believed to be secondary to stress. The hypokalemia, azotemia and minimally concentrated urine were presumed to be secondary to diuretic therapy; however chronic kidney disease could not be ruled out as there was no serum biochemical panel or urinalysis available for comparison prior to initiation of diuretic therapy. Abdominal ultrasonography was unremarkable.
The echocardiogram revealed significant left ventricular hypertrophy, moderate right ventricular hypertrophy and normal left atrial dimensions (Fig. 2a-b). Color Doppler Echocardiography showed a moderate amount of tricuspid regurgitation (Fig. 3a). Continuous-wave Doppler assessment of the peak tricuspid regurgitation was measured at 4.2m/sec resulting in a tricuspid regurgitation pressure gradient between the right atrium and right ventricle of 70.2mmHg (calculated using the modified Bernoulli equation, p=4V2)(Fig. 3b). Due to the tricuspid regurgitation jet direction the true peak velocity and with that the pressure gradient may have been underestimated. By adding the estimated right atrial pressure, 75.2mmHg was calculated as the systolic pulmonary artery pressure . Due to the lack of a pulmonary arterial pressure classification scheme specific for cats, this information was extrapolated from the dog . The cat was subsequently diagnosed with hypertrophic obstructive cardiomyopathy and moderate to severe pulmonary arterial hypertension (PAH). Given the normal left atrial size it was concluded that congestive heart failure was unlikely. Based on echocardiography a primary cause for PAH was not identified. There was no evidence of structures consistent with adult D. immitis or thromboemboli in the main pulmonary artery or in close proximity to the bifurcation. However, their presence could not be excluded further distally.
The cat was housed in an oxygen chamber and treated overnight with 40% inhalational oxygen and intravenous fluids. While being treated with supplemental oxygen, the cat’s respiratory effort normalized and rate decreased.
The following day the cat had thoracic computed tomography (CT, 64-detector row Toshiba Aquilion, Toshiba America Medical Systems, Tustin, CA) to determine an identifiable underlying cause of PH (i.e. primary lung disease or thromboemboli). The cat was premedicated with butorphanolb (0.5 mg/kg IV). Following premedication, general anesthesia was induced with alfaxalonec (1.8 mg/kg IV) and maintained with intravenous alfaxalone (0.2 mg/kg/min). The cat was intubated and mechanically ventilated (Engstrom Carestation ventilator, GE Healthcare). The ventilator settings included volume-controlled ventilation with an inspired oxygen concentration of 40%, tidal volume 10 mls/kg, respiratory rate 10 breaths/minute, and positive end-expiratory pressure (PEEP) 5 cm H20. Inspiratory and expiratory breath holds were ventilator-assisted and performed in tandem with CT scans; PEEP was set to 0 cm H20 for the expiratory breath hold.
Contiguous 1–2 mm transverse images, pre-and post-contrast administration, were obtained from the level of C6 to L2 with the cat in sternal recumbency. Both inspiratory and expiratory sequences were performed as well as arterial, venous, and delayed phases. Iodinated contrastd was injected at 3 ml/s velocity with a total volume of 12 ml (2.1 mg/kg). The sure-start feature of the CT scanner was used to automatically start image acquisition when the threshold of 180 Hounsfield units (HU) inside the aortic-arch was reached thus defining the arterial phase. The image acquisition of the venous phase started automatically following termination of the arterial phase. The images acquired in the delayed phase were obtained exactly 2 min following the start of the arterial phase. On the pre-contrast inspiratory sequence, there were multiple patchy areas of ground glass opacification in the lung lobes especially in the cranial lobes that were of similar size and equally distant from each other (Fig. 4a). The most peripheral opacities were all at the same distance from the pleural surface. Ground glass opacification was also seen in the transition between areas of consolidation and normal lung attenuation in the ventral aspects of the lungs bilaterally sparing only the most cranial and most caudal portions of the lungs (Fig. 4b-c). The left and right caudal lobar pulmonary arteries were enlarged, being larger than the corresponding vein (Fig. 4c). In the post-contrast inspiratory venous sequence, the pulmonary trunk was subjectively enlarged (1.11 cm vs. 0.929 cm ascending aorta), the ascending aorta was displaced to the right by the enlarged right ventricular outflow tract/pulmonary trunk (Fig. 5a). The post-contrast arterial phase showed the presence of iodinated contrast medium in the pulmonary trunk, left and right pulmonary arteries (Fig. 5a) and the proximal aspect of the caudal lobar pulmonary arteries (Fig. 5b). Starting at the level of the 6th ribs on the left (Fig. 5c) and of the 7th ribs on the right (Fig. 5d), both caudal lobar pulmonary arteries demonstrate a major filling defect occupying the entire lumen of the artery. The lack of contrast medium in the caudal lobar pulmonary arteries is well illustrated by the abrupt termination of these arteries in relation to the pulmonary trunk and thoracic aorta on the three-dimensional (3D) volume rendering image (Fig. 5e). 3D Maximum Intensity Projection tool available on the CT workstation was used to elongate each artery in the horizontal (Fig. 5f) and sagittal (Fig. 5g-h) planes providing excellent visualization of the filling defects in relation to the length of each artery.
Following the CT, the cat was maintained in a light plane of anesthesia with 5 to 6 cm H20 PEEP for an additional hour until he was able to maintain an oxygen hemoglobin-saturation above 95% without PEEP. The cat then recovered in an oxygen chamber providing 40% inhalational oxygen. The cat continued to show tachypnea, maintained oxygen enriched hemoglobin-saturations of 90 to 95%, and had clear lung sounds on thoracic auscultation.
Based on diagnostics performed at that time it was suspected the cat had a significant pulmonary vasculopathy that resulted in PAH. The etiology of the vasculopathy was unknown at that time but suspected secondary to thromboembolic disease as well as uncharacterized primary lung disease. The cat was treated with one dose of sildenafild (0.9mg/kg, PO). Approximately 45 min following administration of sildenafil the cat developed respiratory distress. His respiratory rate increased to 80 breaths/min, he assumed an orthopneic posture, and was breathing with an open mouth. On thoracic auscultation the cat had harsh, loud crackles in all lung fields.
Because of the progressively worsening respiratory distress as well as the grave prognosis, the owner elected euthanasia. Immediately after euthanasia the lungs were removed with owner consent and fixed for histopathology. In an effort to prevent the collapse, deflation, and disruption of lung structures and to avoid fixation artifacts, each lung lobe was inflation-fixed with 10% formalin . Each lobar bronchus was cannulated with a 3 French red rubber catheter and fixed with a ligature. The lungs were inflated via the cannula by gentle infusion of 10% formalin. After each lobe was sufficiently inflated, the lobar bronchus was tied off with a ligature and the lungs were placed in jars containing 10% formalin (ratio at least 1 part tissue to 9 parts formalin).
In the subgross fixed lung lobes multifocal sharply demarcated dark red foci were evident within the lung parenchyma (Fig. 6). Histopathology of these regions were characterized by severe multifocal alveolar capillary proliferation (Figs. 7, ,88 and and9).9). Foci of infiltration of respiratory bronchioles was frequent and marked type II pneumocyte hyperplasia was present in the remodeled lung. In addition, there was moderate focally extensive alveolar congestion with intra-alveolar hemorrhage and fibrin. Furthermore, there was evidence of multifocal pulmonary arterial thrombosis and alveolar carcinomas. The apparent capillary proliferation with infiltration of respiratory bronchioles is suggestive of PCH. Although PCH often occurs along with PVOD, there was no evidence of PVOD in this cat. The epithelial neoplasms were considered to be incidental findings in this case.
Given the timing of the sildenafil to the cat’s acute respiratory decompensation (45 min prior) it is suspected that arteriolar dilation with a fixed capillary obstructive lesion led to flooding of the alveoli from increases in hydrostatic pressure. In humans with post-capillary PAH (i.e. PCH and PVOD) treatment with vasodilators can result in a florid and fatal pulmonary edema. It is also possible the cat’s abrupt respiratory decline included additional acute pulmonary thromboemboli; anaphylactic response to the compounded sildenafil or the onset of acute respiratory distress syndrome were not supported by histopathologic findings.
The cause for thromboembolic disease was not determined but could have been secondary to cardiac disease. Interestingly, a recent report evaluating the microcirculation in people with chronic thromboembolic pulmonary hypertension found a strong association with capillary hemangiomatosis-like lesions .
Pulmonary capillary hemangiomatosis is a rare, idiopathic angioproliferative disease that results in PAH . The unique histologic feature of PCH is the proliferation of capillaries in the pulmonary parenchyma as well as evidence of invasion by the capillaries into one or more of the pulmonary veins, arteries, alveolar walls, alveolar space, interlobular fibrous septa, and bronchioles . Histologic evaluation of lung parenchyma from the cat in this report revealed severe multifocal alveolar capillary proliferation with infiltration of respiratory bronchioles suggestive of PCH. The pathogenesis of PCH is poorly understood and is likely multifactorial. In humans, PCH has been described in association with several disease processes includingaortic stenosis, Kartagener syndrome, systemic lupus erythematous, scleroderma, Takayasu’s arteritis, hypertropic cardiomyopathy and neoplasia [6–13]. Echocardiographic evaluation of the cat reported here revealed significant left ventricular hypertrophy and normal left atrial dimensions. In humans, it is postulated that long standing chronic passive congestion of the lung as is seen with hypertrophic cardiomyopathy results in development of PCH . The normal left atrial dimensions in this cat make chronic pulmonary venous congestion unlikely. It has also been proposed that PCH is not a separate disease, but rather represents a secondary angioproliferative response to pulmonary venous hypertension as seen in PVOD . However, there was no evidence of PVOD in this cat. In humans, mutations in the EIF2AK4 gene are a risk factor in the development of PCH . Future studies investigating the aforementioned EIF2AK4 gene mutation are needed to establish this association in dogs and cats with PCH.
Pulmonary hypertension is a syndrome characterized by altered blood flow resulting in elevated pulmonary arterial pressures. In veterinary medicine, PH is classified based on the origin of altered blood flow, that is, pre-capillary (pulmonary arterial hypertension) or post-capillary (pulmonary venous hypertension . The clinical picture was complicated by pulmonary thromboembolism, which also contributed to PH. Thus the PH identified in the cat described in this report would likely be classified as both Group 1’ and Group 4 (Table 1).
In humans, clinical signs associated with PCH include progressive dyspnea, exertional fatigue, syncope, and sometimes a chronic cough  and are frequently seen in other forms of PAH. The commonality in presenting clinical signs makes differentiating PCH from other forms of PAH challenging. While difficult, this distinction is essential because conventional treatment with vasodilators for PAH is contraindicated in patients with PCH as it can result in fatal pulmonary edema [44–46]. It is currently recommended that patients with presumed primary pulmonary hypertension undergo a HRCT examination before initiation of vasodilator therapy . The CT in this case corroborates common CT findings in humans with PCH; diffuse, ill-defined nodular ground glass opacities and enlarged pulmonary arteries. Given the timing of the sildenafil to the cat’s acute respiratory decompensation (45 min prior) it is suspected that arteriolar dilation with a fixed capillary obstructive lesion led to flooding of the alveoli from increases in hydrostatic pressure.
The authors would like to thank TJ Stockton for CT acquisition and image reconstruction and Kate Anderson for assistance with a literature search.
No funding declare.
All necessary data is included in the body of this manuscript.
JJ was the primary author of the manuscript and primary clinician on case. CR was the supervising faculty on the case and primary editor of the manuscript. IM interpreted the thoracic CT images, provided figures and figure legends, and reviewed the manuscript. MK performed echocardiogram, provided figures and figure legends, and reviewed the manuscript. KW interpreted lung histopathology, provided figures and figure legends, and reviewed the manuscript. All authors read and approved the final manuscript.
The authors declare that they have no competing interests that could inappropriately bias the results.
Informed consent was obtained from the client.
J.A. Jaffey, Email: ude.iruossim@jyeffaj.
K. J. Williams, Email: ude.usm.hapcd@3721lliw.
I. Masseau, Email: moc.em@iuaessam.
M. Krueger, Email: email@example.com.
C. Reinero, Email: ude.iruossim@corenier.