To the Editor:
A 54-year-old man presented with massive hemoptysis. Physical examination, chest radiograph and contrast-enhanced computed tomography of the thorax were unremarkable. Computed tomography angiography did not reveal any bronchial or pulmonary arterial source of bleed. On flexible bronchoscopy, 2 small nodules (<5 mm) were identified near the opening of the right upper lobe (Fig. 1A). Dieulafoy disease was suspected, and a biopsy was avoided. Subsequently, rigid bronchoscopy was performed. Convex probe endobronchial ultrasound (EBUS) showed a round anechoic structure immediately beneath the bronchial mucosa, and Doppler confirmed its vascular nature (Figs. 1C–E). We then cauterized the lesion with argon plasma coagulation (APC). There was no recurrence of hemoptysis. Bronchoscopy after 3 months showed complete resolution (Fig. 1B).
Dieulafoy disease is a vascular anomaly characterized by the presence of a dysplastic submucosal artery eroding the surface epithelium in the absence of a primary ulcer. Usually encountered in the gastrointestinal tract, similar lesions can also occur in the bronchial tree.1 Endobronchial Dieulafoy lesion is uncommon and can be overlooked by an inexperienced observer. Conventionally, the identification of a suspicious lesion (whitish nodule) on bronchoscopy, and confirmation by angiography (or histopathology) establishes the diagnosis. With the advent of EBUS, the vascular nature of the lesion can be shown without the need for angiography or histopathology.2,3
A systematic review of the PubMed and Embase databases [till September 30, 2018, using: “dieulafoy” AND (“bronchus” OR “bronchoscopy” OR bronch* OR “lung” OR “pulmonary”)] yielded 47 unique cases from 34 citations (Table S1, Supplemental Digital Content 1, http://links.lww.com/LBR/A174). Of the 48 cases (including the index case), 70.8% were males (n=34) with a mean (SD) age of 48.3 (17.9) years (Table S1, Supplemental Digital Content 1, http://links.lww.com/LBR/A174). Majority of the subjects had hemoptysis; one was asymptomatic, detected incidentally during bronchoscopy and confirmed by EBUS.4 About 47.9% (n=23) had recurrent hemoptysis. A lesion was identified or the bleeding was localized by bronchoscopy in 40 subjects (83.3%), no abnormality was detected in 1, whereas the findings were not available in 7 subjects. The abnormalities were right-sided in 28 (70%), left-sided in 10 (25%), and tracheal in 1 subject. A nonpulsatile elevated sessile lesion (2 to 10 mm in diameter, with or without a white cap covering the surface of the nodule) was the most common finding (n=25, 52%) (Table S2, Supplemental Digital Content 2, http://links.lww.com/LBR/A175), followed by a blood-filled bronchus (n=8, 16.7%) or a bleeding nodule (n=6, 12.5%). Nine patients (18.8%) had profuse bleeding (2 of them fatal) following bronchoscopy (Table S2, Supplemental Digital Content 2, http://links.lww.com/LBR/A175), usually performed suspecting an endobronchial tumor. Six of these developed after biopsy, 2 after contact with the bronchoscope, and 1 after saline instillation.5 The appearance of Dieulafoy lesion on convex probe EBUS (index case),2–4 and radial probe EBUS have been recently described (Table S2, Supplemental Digital Content 2, http://links.lww.com/LBR/A175).
Bronchial Dieulafoy disease was successfully managed with surgery (n=25, 52.1%), bronchial artery embolization (BAE; n=13, 27.1%) or bronchoscopic modalities (n=7, 14.6%) (Table S3, Supplemental Digital Content 3, http://links.lww.com/LBR/A176). The duration of follow-up varied from none to 84 months. Surgery was usually a lobectomy; however, occasionally pneumonectomy has been performed.6 BAE was successful only in 13/31 (41.9%) subjects. Technical difficulty was the most common reason for failure (n=12). Even in those patients (n=13) where the procedure was successful, the majority had no follow-up (n=8). Bronchoscopy can offer a minimally invasive treatment option for endobronchial Dieulafoy disease. Of the bronchoscopic therapies, 2 subjects were treated with APC, and 3 with Nd:YAP laser. No recurrence was noted (4 to 24 mo follow-up). Cryotherapy was unsuccessful in 1 patient, and a silicone stent (rigid bronchoscopy) had to be deployed to control the bleed. Apart from this case, rigid bronchoscopy was also used in 2 subjects; one for performing EBUS,2 and the other for Nd:YAP laser.
Historically, most patients with Dieulafoy disease have been successfully managed with surgery. In a large case series of cryptogenic hemoptysis (n=81), 9 patients required surgery and 5 of these 9 had Dieulafoy disease identified on histopathology.7 However, surgery may not be readily available. Also, a small lesion might require removal of >1 lobe, and the extent of surgery is often dictated by the location of the lesion (right middle and lower lobes may have to be removed for a small lesion in the right intermediate bronchus). On the contrary, BAE may not be technically feasible in several patients, and recurrence is common. Among the currently available bronchoscopic therapies, APC seems to be effective in the management of Dieulafoy disease. The noncontact nature of APC, and the reduced depth of penetration offer unique advantages over electrocautery (contact coagulation) and LASER, respectively. In the index case, the lesion was located at the junction of distal trachea and the opening of the right upper lobe. By using APC, we were able to avoid a complex surgical procedure. However, with only 3 cases (including the index case) reported so far, the evidence is insufficient to draw a conclusion. Importantly, the lesion may bleed on contact with the bronchoscope or during the therapeutic procedure. This potential complication was anticipated in the index case, and therefore EBUS as well as APC was performed in the operating room using rigid bronchoscopy. We did not encounter any bleeding, but facilities to handle them should be readily available.
In conclusion, bronchial Dieulafoy is a rare cause of massive hemoptysis. A high index of suspicion is required for diagnosis. The disease is probably underrecognized and underreported. Hence, the ideal treatment modality remains unclear. Nevertheless, bronchoscopy seems to be a useful modality in the diagnosis and management of Dieulafoy disease.
Valliappan Muthu, MD, DM, MRCP (UK)
Kuruswamy T. Prasad, MD, DM
Inderpaul S. Sehgal, MD, DM
Sahajal Dhooria, MD, DM
Ashutosh N. Aggarwal, MD, DM
Ritesh Agarwal, MD,
DM Department of Pulmonary Medicine Postgraduate Institute of Medical Education and Research (PGIMER), Chandigarh India
1. Sweerts M, Nicholson AG, Goldstraw P, et al. Dieulafoy’s disease of the bronchus. Thorax. 1995;50:697–698.
2. Gurioli C, Casoni GL, Gurioli C, et al. Endobronchial ultrasound in Dieulafoy’s disease of the bronchus: an additional application of EBUS. Monaldi Arch Chest Dis. 2010;73:166–168.
3. Lin C, Ginsburg M, Kuo W, et al. Dieulafoy’s lesion of the bronchus: more than meets the eye. Am J Respir Crit Care Med. 2015;191:A3100.
4. Minchole E, Penin RM, Rosell A. The utility of linear endobronchial ultrasound for the incidental finding of dieulafoy disease of the bronchus. J Bronchology Interv Pulmonol. 2018;25:e48–e50.
5. Hope-Gill B, Prathibha BV. Bronchoscopic and angiographic findings in Dieulafoy’s disease of the bronchus. Hosp Med. 2002;63:178–179.
6. Xia XD, Ye LP, Zhang WX, et al. Massive cryptogenic hemoptysis undergoing pulmonary resection: clinical and pathological characteristics and management. Int J Clin Exp Med. 2015;8:18130–18136.
7. Savale L, Parrot A, Khalil A, et al. Cryptogenic hemoptysis: from a benign to a life-threatening pathologic vascular condition. Am J Respir Crit Care Med. 2007;175:1181–1185.