Severe pulmonary hypertension in histiocytosis X: long-term improvement with bosentan

To the Editors:

Pulmonary Langerhans’ cell histiocytosis (LCH)/histiocytosis X is a rare smoking-related interstitial lung disease that predominantly affects adults aged 20–40 yrs. It is commonly associated with irreversible airflow obstruction and highly variable clinical outcomes. The disease may regress spontaneously or following smoking cessation; however, in 20–30% of cases it progresses to chronic respiratory insufficiency. Despite encouraging results in case series and reports 1, no established efficacy on disease outcomes has been demonstrated with corticosteroids, vinblastine or cladribine.

Although rare in patients with early disease, severe precapillary pulmonary hypertension (PH; group 5 in the clinical classification of PH) 2 is frequent in patients with advanced pulmonary LCH. Precapillary PH was present in all 21 consecutive pulmonary LCH patients 3, and in 92% of 39 patients referred for lung transplantation 4 in two series. Lung transplantation is considered the therapy of choice for end-stage pulmonary LCH and chronic respiratory insufficiency, especially among patients who exhibit severe PH 4. Whether therapy for PH may affect long-term outcomes of disease and impact timing of transplantation is currently unknown. Here, we report the long-term improvement of LCH-associated PH in a patient treated using the dual endothelin receptor antagonist bosentan, obviating the need for lung transplantation.

A male aged 25 yrs was referred to our centre in 1987 with pulmonary LCH. He had smoked for <3 yrs and discontinued smoking when aged 24 yrs. Diagnosis of LCH was made at 22 yrs, with central nervous system (diabetes insipidus, hypogonadism), pulmonary (recurrent bilateral pneumothoraces) and biopsy-proven cutaneous involvement. Pleurodesis with lung biopsy was performed by open surgical parietal pleurectomy. Lung biopsy confirmed LCH, with end-stage destruction of distal bronchiolar walls. Neither Langerhans’ cell infiltration nor granulomas of pulmonary artery walls were observed. Computed tomography of the chest showed typical nodules and numerous small thick- and thin-walled cysts predominating in the upper lobes of the lungs. Forced vital capacity and forced expiratory volume in 1 s were 65% of predicted values (table 1). The carbon monoxide diffusion coefficient was normal (92% pred), with moderate hypoxaemia on room air (arterial oxygen pressure (P_a,O2) 8.7 kPa).

In 1994, the patient’s residual volume had increased and hypoxaemia worsened; the patient was given glucocorticoid therapy during 1 yr, without improvement of pulmonary LCH. The patient successively received vinblastine, chlormethine, carmustine, thalidomide, vepeside and cladribine for LCH skin lesions; long-term improvement was observed on cladribrine. The course of disease was characterised by progressive worsening of lung function parameters and chronic respiratory failure (table 1). Long-term nasal oxygen was started in 1999.

In 2003, the patient presented with increased dyspnoea (New York Heart Association (NYHA) functional class III), with no evidence of right heart failure. Pulmonary function testing showed a marked obstructive defect with increased residual volume (table 1), severe hypoxaemia (6.1 kPa on room air), and severe impairment of pulmonary gas and carbon monoxide transfer. His 6-min walk distance was 335 m, with a decrease in oxygen saturation during assessment on room air to 73%.

At this time, the patient was referred for lung transplantation. Transthoracic echocardiography showed increased estimated systolic right ventricular pressure (62 mmHg) with dilated right heart cavities. Right heart catheterisation demonstrated precapillary PH, with a mean pulmonary arterial pressure of 41 mmHg, a cardiac index of 1.9 L·min⁻¹·m⁻², and a pulmonary vascular resistance of 649 dyn·s·cm⁻⁵. Isotopic measurement of right systolic ejection fraction was 15%. Pulmonary embolism was excluded by lung perfusion scan and spiral high-resolution computed tomography. Results of haematological and biochemical tests were within normal ranges. Serology for human immunodeficiency virus was negative. Given the severity of disease, heart and lung transplantation were considered; however, therapy for PH was proposed and inscription on the transplantation list delayed until the next planned assessment. The patient was treated using diuretics and warfarin, and bosentan was started in June 2003.

Bosentan therapy was followed by improved dyspnoea (to NYHA functional class II), 6-min walking distance (by +109 m) and haemodynamics (table 1). Adjunction of sildenafil in 2007 did not provide further improvement and was discontinued after 3 months for claimed intolerance.

To the present day, clinical improvements with bosentan therapy have been maintained for a total of 7 yrs. Due to the observed improvements in clinical status, inscription on the transplantation list was postponed.

Although the prevalence of PH in patients with pulmonary LCH is not precisely known, PH is reported to be present in the vast majority of patients with end-stage disease undergoing evaluation for lung transplantation 3, 4, with an interval from the first respiratory symptoms to referral for transplantation of 11±2.6 yrs (range 1.3–25 yrs). PH in the context of pulmonary LCH may be severe, with pulmonary arterial pressures ≥35 mmHg observed in 72% of patients, and ≥45 mmHg observed in 45% of patients 4. The severity of PH may be markedly greater in patients with pulmonary LCH than with idiopathic pulmonary fibrosis (despite similar P_a,O2) and chronic obstructive pulmonary disease 3. When present, PH may contribute to exercise limitation and dyspnoea and possibly reduced survival 5.

The pathophysiology of PH in pulmonary LCH is likely to be multifactorial, with contributions from chronic hypoxaemia, abnormal pulmonary mechanics and vascular remodelling. The pulmonary granuloma in LCH produces a number of cytokines and growth factors implicated in PH, including platelet-derived growth factor 6. Pulmonary haemodynamics in LCH-related PH are similar to those in idiopathic pulmonary arterial hypertension, are unrelated to lung function variables 3, 7 and are associated with severe histopathological vascular changes including medial hypertrophy, intimal fibrosis, proliferation leading to lumen obstruction and frequent venous involvement 3. These observations support the hypothesis of vascular remodelling and intrinsic pulmonary vascular disease independent of involvement of small airways and lung parenchyma. However, PH has not been reported in the absence of advanced parenchymal lung disease. Of note, severe PH developed while our patient was treated with supplemental oxygen during 4 yrs with adequate correction of hypoxaemia, ruling out further haemodynamic improvement as a result of oxygen. Taken together, these observations provide a rationale for the use of PH-specific therapy in the management of patients with pulmonary LCH and severe PH, although this has not yet been adequately evaluated.

Fartoukh et al. 3 reported that two patients with pulmonary LCH and severe PH who received intravenous epoprostenol presented with acute pulmonary oedema, which probably resulted from increased pulmonary blood flow in the setting of veno-occlusive disease with prominent pulmonary venous involvement by LCH. Epoprostenol may therefore be hazardous in this patient population due to the frequency of veno-occlusive disease 8.

As substantial evidence links endothelin-1 to the pathogenesis of both PH and pulmonary fibrosis 9 and no alternative therapy other than transplantation was available, our patient was treated with the oral dual endothelin receptor antagonist bosentan on a compassionate use basis. Dramatic and long-term improvements in clinical manifestations and haemodynamic parameters were observed. Although this patient also received the phosphodiesterase type-5 inhibitor sildenafil, it was poorly tolerated and discontinued by the patient within 3 months, and could not therefore account for the observed clinical improvement. It is important to note that improvement of PH, despite unchanged pulmonary function, has been reported in a patient with pulmonary LCH and PH who quit smoking and received corticosteroids 10. Whether antiproliferative therapy such as cladribine might affect PH in patients with pulmonary arterial wall involvement by Langerhan's cells is unknown.

In conclusion, these observations demonstrate that long-term improvement of PH may be obtained using bosentan in patients with pulmonary LCH, further supporting the theory of pulmonary vasculopathy. Bosentan therapy may be considered in individual cases as a bridge to lung transplantation, which remains the therapy of choice in patients with pulmonary LCH and severe PH.