Management of Severe Acute Pulmonary Hemorrhage in Children

Abstract

Pulmonary hemorrhage is a potentially life-threatening condition with a variety of causes that pediatricians commonly encounter. Quality clinical trials are lacking in children, limiting the evidence base to observational data and adult studies. The overall management strategy should address control of symptomatic bleeding, identification of the bleeding source, and treatment of the underlying cause. Flexible bronchoscopy is an important tool to identify the cause and site of bleeding, perform interventional procedures, and directly instill medications to affected areas. Medications to control bleeding include vasoconstrictors, antifibrinolytics, and recombinant factor VIIa. Definitive treatment often requires the use of immunomodulatory medications, bronchial artery embolization, or surgery. In this review, we summarize the most recent evidence pertaining to medical, interventional, and surgical treatments of pulmonary hemorrhage in children.

Introduction

Pulmonary hemorrhage is a rare but potentially life-threatening condition that occurs in children of all ages. The lungs receive a blood supply from two circulatory systems. The bronchial circulation arises from the aorta and intercostal arteries and supplies the conducting airways.^1,2 Although only 1% of total pulmonary blood flow comes from the bronchial circulation, hemorrhage from this circuit may be catastrophic because it is under high pressure.¹ The rest of the lungs are perfused by the pulmonary circulation, which is a high volume, low pressure system.^1,2 “Massive hemoptysis” has been variably defined in the adult literature, ranging from 100 mL to 1,000 mL expectorated within a 24-hour period. However, the routine use of this term is misleading because the clinical consequences are far more important than the volume of blood. Furthermore, true quantification of hemoptysis over a 24-hour period is challenging. Some have argued that the term “life-threatening hemoptysis” is a more clinically meaningful term to use.³

Bleeding may be localized or diffuse, and the severity can range from relatively mild with no clinical consequences to life-threatening. The potential causes of pulmonary hemorrhage are numerous, and several are listed in the Panel. Symptoms may vary in children. They may present with hemoptysis and cardiorespiratory compromise or with unexplained anemia and no respiratory symptoms.^1,4 The diagnostic evaluation will differ depending on the history, physical exam, and severity of presentation. In addition to differentiating between localized and diffuse pulmonary hemorrhage, classifying disorders with or without pulmonary capillaritis (PC) is helpful in directing therapy decisions, although diagnosing PC requires a lung biopsy and may not be feasible during an episode of acute hemorrhage.⁵ Although several interventions may symptomatically control bleeding regardless of the etiology, identifying the diagnosis is paramount in addressing the underlying cause of pulmonary hemorrhage and preventing recurrence.

Panel. Causes of pulmonary hemorrhage in children. HSCT, hematopoietic stem cell transplant. DAH, diffuse alveolar hemorrhage.

Rheumatologic/Vasculitis

Anti-glomerular basement membrane disease

Microscopic polyangiitis

Granulomatosis with polyangiitis

Henoch-Schonlein Purpura

Pulmonary capillaritis

Sarcoidosis

Systemic lupus erythematosus

Other

Trauma (including non-accidental)

Coagulopathy

Factitious hemoptysis

Idiopathic pulmonary hemosiderosis

Celiac disease/Lane Hamilton syndrome

Post-HSCT DAH

Pulmonary edema

Asphyxiation

Local

Bronchiectasis

Bronchial tumor

Pulmonary arteriovenous malformation

Bronchial Dieulafoy lesion

Foreign body aspiration

Infectious

Bronchitis

Pneumonia

Mycobacterial infection

Lung abscess

Iatrogenic

Suctioning trauma

Endobronchial or transbronchial biopsy

Tracheostomy associated bleeding

Cardiac

Pulmonary hypertension

Multiple aortopulmonary collateral vessels

Pulmonary embolism

Mitral valve stenosis

In any patient in which there is concern for substantial pulmonary hemorrhage, imaging with a chest radiograph, CT angiogram (CTA) of the chest, and flexible bronchoscopy are often required to understand the diagnosis, assess severity, and localize the source of bleeding. It is vital that the CT protocol enables visualization of the bronchial arteries.¹ While several adult studies have compared the utility of CT scan and bronchoscopy in the evaluation of hemoptysis,^6,7 fewer data are available to guide the pediatric clinician as to when both CT scan and bronchoscopy are warranted. The approach to evaluation of pulmonary hemorrhage in children should be patient-specific and involve collaboration with several subspecialists.

Initial stabilization in severe pulmonary hemorrhage

The most acute danger to a child with hemoptysis is asphyxiation rather than exsanguination.^1,7 Severe impairments in gas exchange occur when enough blood accumulates to fill the conducting airways. Patients with hemoptysis and hypoxemia or respiratory distress should be promptly intubated with the largest possible endotracheal tube to control the airway and allow the passage of a bronchoscope or large-bore suction catheter.⁷ High positive end-expiratory pressure (PEEP) should be applied to improve oxygenation and tamponade bleeding vessels.⁸ This is based on the finding that high PEEP reduces both bronchial⁹ and alveolar capillary¹⁰ perfusion, which would therefore be expected to mitigate active bleeding from either of these circulations. Volume resuscitation with crystalloid or blood products and correction of coagulation and platelet derangements are paramount.¹¹ At some centers with expertise in management of hemorrhage with rigid bronchoscopy, this tool can be used to secure the airway, maintain ventilation, and assist with suctioning of blood.^7,11 Dramatic increases in intrathoracic pressure (up to 400 cm H₂O in adults¹²) that occur during cough may lead to rupture of damaged bronchopulmonary capillaries¹³ and worsening pulmonary hemorrhage. Therefore, neuromuscular blockade to suppress coughing should be considered until active hemorrhage is controlled.

The source and laterality of bleeding should be determined as rapidly as possible.⁷ Although chest radiograph is relatively insensitive in localization of pulmonary hemorrhage¹⁴, it is an easy test that can be performed quickly at the bedside. If the site of bleeding is unknown, CTA and bronchoscopy should be performed to attempt to localize the bleeding.^7,11 If the bleeding is unilateral, and the laterality can be determined, the bleeding lung should be placed in the dependent position to prevent blood from filling the unaffected lung. Additionally, immediate efforts should be made to obstruct the main bronchus of the bleeding lung.^7,11 The main bronchus of the non-bleeding side may be selectively intubated, under direct visualization via bronchoscopy, to protect the healthy lung.⁷ It is important to note that when selectively intubating the right lung, the right upper lobe bronchus may be occluded by the endotracheal tube.¹⁵ An inflatable bronchial blocker (Figure 1A) can be placed using bronchoscopic guidance and inflated when active bleeding occurs. It can then be deflated when bleeding has stopped or when exploration of the airways distal to the balloon is indicated.¹⁵ If a bronchial blocker is not available or the child’s airway is too small to accommodate it, a Fogarty embolectomy catheter (Figure 1B) can be passed through the endotracheal tube, guided into the appropriate bronchus under direct visualization using a bronchoscope, and inflated to occlude the bronchus.¹⁶ The balloons in bronchial blockers and Fogarty catheters deflate within hours and therefore need to be re-inflated regularly. In older children, a dual lumen endotracheal tube (Figure 1C) can be placed, where one lumen is positioned in a main bronchus and the other lumen in the trachea to ventilate the other lung. There are several disadvantages of the double lumen endotracheal tubes, including technical difficulty in placement (especially when pulmonary hemorrhage is present), dislodgement of the bronchial tube, occlusion of the small lumens by blood clots, and inability to perform effective bronchoscopy in some cases due to the small lumen diameter.^7,15,16

Figure 1.

Devices used to isolate the healthy lung. (A) Bronchial blocker. An adapter is used to allow for insertion of the bronchial blocker and the flexible bronchoscope into the endotracheal tube. The snare on the end of the catheter enables the blocker to be advanced under direct visualization with the bronchoscope. (B) Fogarty balloon catheter. The catheter is advanced the flexible bronchoscope through the endotracheal tube. (C) Double-lumen endotracheal tube. After intubation, a bronchoscope is inserted through the tracheal lumen to visualize appropriate positioning of the bronchial lumen in the mainstem bronchus

Medications

Cold saline

Although cold saline is commonly used to treat pulmonary hemorrhage, it has not been well-studied in adults or children. In one of the only studies of cold saline in pulmonary hemorrhage, Conlan et al. reported on the use of 50 mL aliquots of cold saline (4°C) through a rigid bronchoscope to treat massive hemoptysis in 12 adults. Complete cessation of bleeding was observed in all patients after instilling an average of 500 mL of cold saline in total.¹⁷ The mechanism by which cold saline induces hemostasis is likely local vasoconstriction.^7,18 Its wide availability and presumed safety make it an attractive option, although the optimal volume is not known. However, in a clinical trial comparing the use of instilled tranexamic acid (TXA) with epinephrine to treat endobronchial bleeding following treatment failure with cold saline, three applications of 5 mL of cold saline were able to control bleeding in only 3 of 20 episodes of severe bleeding.¹⁹ This indicates that small volumes may be less effective. The British Thoracic Society recommends using 5–10 mL of 4°C saline to treat bleeding that arises during bronchoscopy for adults.²⁰ A European Respiratory Society (ERS) position statement on interventional bronchoscopy states that cold saline may be used to control hemoptysis, but more studies are needed.²¹

Epinephrine

Epinephrine, instilled through a bronchoscope, is commonly used to treat endobronchial bleeding.^7,20 There are no placebo-controlled studies examining its efficacy as a treatment in adults or children with pulmonary hemorrhage. However, one adult trial found that the use of prophylactic epinephrine was more effective than normal saline in preventing severe bleeding associated with transbronchial biopsy (8·7% vs. 24%, p=0·04).²²

Several reports have described cardiac arrhythmias that have occurred in adults as a result of topical epinephrine applied to the airways,^23–25 and one adult suffered coronary vasospasm.²⁶ This has led to a wide variety of doses being recommended in different textbooks and society guidelines.^21,23 The ERS statement on pediatric bronchoscopy mentions the use of epinephrine 1:10,000 or 1:20,000 solution without recommending a specific dose.²¹ The volume of saline in which epinephrine is diluted, and the site of administration in the airway may also impact plasma levels of epinephrine and therefore the risk of cardiac arrhythmias.^27,28 Because of the uncertainty surrounding the optimal dose and diluent volume, endobronchial epinephrine should probably be avoided in patients with an increased risk of cardiac arrhythmia.²³ Because of this, our practice is to be cautious in using this medication to treat pulmonary hemorrhage in children with cyanotic congenital heart disease when other agents are available, especially during the peri-operative period.

Tranexamic acid

Tranexamic acid (TXA), a synthetic analog of lysine, is an antifibrinolytic medication that inhibits plasminogen activation, which in turn prevents the conversion of plasminogen to plasmin. As a result, the fibrin clot remains intact.²⁹ TXA can be administered systemically or in a nebulized form. Systemic TXA has shown promise in reducing perioperative blood loss in children undergoing surgery for craniosynostosis, cardiac disease, and scoliosis.³⁰ The use of nebulized TXA is attractive because it can be administered rapidly before IV access is obtained, achieve higher concentrations at the site of bleeding, work quickly, and reduce the risk of side effects.^31,32

In general, TXA is extremely well-tolerated.²⁹ Due to its mechanism of inhibiting fibrinolysis, venous thromboembolism (VTE) is a theoretical concern, although data regarding this risk are conflicting.^29,32 Animals studies have shown that topical application of TXA to the central nervous system induces seizures in a dose-dependent fashion, and high-dose systemic TXA use during cardiac bypass surgery has been associated with increased risk of post-operative seizures in retrospective studies.²⁹ Overall, these serious adverse effects are extremely rare and have not been reported with use of inhaled TXA.

In adults, several studies have examined the efficacy of systemic TXA to treat hemoptysis, with conflicting results. A Cochrane review of two randomized control trials (RCTs) found a statistically significant reduction in bleeding time in those who received TXA (−19·47 hours) compared to placebo,³³ although a subsequent RCT found that the only significant improvement in adults who received systemic TXA was in the visual analogue scale (VAS), in which patients rated their hemoptysis between a range of “no hemoptysis” and “worst imaginable hemoptysis.”³⁴ A large retrospective study using propensity score matching (n = 9933 matched pairs) of adults in Japan treated with systemic TXA for hemoptysis found that the use of TXA was associated with lower in-hospital mortality (11·5 vs. 9·0%) and decreased length of hospitalization (16 vs. 18 days). This study is limited by the inability to control for several unreported confounders, including a large proportion of crossover between the groups.³⁵

Two clinical trials in adults demonstrated that inhaled TXA is effective in treating pulmonary hemorrhage. Wand et al. randomized 47 adults with hemoptysis to receive nebulized TXA or placebo for up to 5 days. The TXA group experienced reduced volume of hemoptysis, (p < 0·01), increased resolution of hemoptysis within 5 days (96% vs. 50%, p < 0·0005), shorter length of hospitalization (5·7 vs. 7·8 days, p = 0·046), and reduced recurrence rate of hemoptysis. There were no clinically important adverse effects. Notably, patients with massive (> 200 mL within 24 hours) hemoptysis or hemodynamic instability were excluded from the study. The rate of resolution of hemoptysis (96%) in the inhaled TXA group was remarkable, and no patients required an interventional procedure to control bleeding (compared with 18·2% in the placebo group).³⁶

Gopinath et al. performed an open-label randomized control trial comparing nebulized TXA to IV TXA in 110 adults presenting to the emergency department (ED) with non-life-threatening hemoptysis. Hemoptysis cessation at 30 minutes following TXA administration was significantly higher in the nebulized group compared to the IV group (72·72% vs. 50·91%, p = 0·0019). The nebulized group also had less hemoptysis (volume) at 6, 12, and 24 hours following TXA administration. More patients in the IV TXA group required emergent bronchial artery embolization (BAE) compared to the nebulized group (47·72% vs. 25·49%, p=0·024). Mild bronchospasm occurred in two patients in the nebulized group, which was treated with bronchodilator medication.³¹

In a meta-analysis of studies examining the use of TXA (nebulized or systemic) to treat adults with hemoptysis, TXA was associated with a reduction in bleeding time (mean −24·61 hours), short-term mortality (RR = 0·78), length of hospitalization (−1·94 days), and need for interventional procedure (RR = 0·38). No adverse effects were associated with TXA in this very large analysis that included close to 20,000 patients.³⁷

TXA can also be instilled through a bronchoscope directly onto a site of bleeding. Two clinical trials in adults comparing endobronchial TXA with epinephrine to treat bleeding encountered during bronchoscopy suggest that these medications are equally effective in this setting.^19,38

Studies regarding the use of TXA in children to treat pulmonary hemorrhage are limited. O’Neil et al. retrospectively studied of the use of nebulized TXA in 19 critically ill children (median age 72 months) with pulmonary hemorrhage. Eighteen (95%) patients experienced complete cessation of bleeding within 48 hours. No adverse events occurred.³⁹ In a retrospective study of children with DAH following hematopoietic stem cell transplantation (HSCT), the use of inhaled TXA in 27 patients was associated with reduced mortality.⁴⁰

In a prospective pilot trial, Bafaqih et al. utilized a two-step protocol to administer nebulized TXA and nebulized recombinant factor VIIa (rFVIIa) to 18 children with intractable diffuse alveolar hemorrhage (DAH). Initially, subjects received nebulized TXA, and if they did not improve after three to four doses or if bleeding worsened at any time, nebulized rFVIIa was added. Ten (55·6%) children experienced complete or near-complete resolution of bleeding with nebulized TXA alone within 24 hours, while six of the remaining eight children experienced resolution of bleeding when rFVIIa was added. There were no treatment-related adverse effects.⁴¹

Overall, there is convincing evidence that inhaled TXA is effective in treating pulmonary hemorrhage in adults when compared both to placebo and to systemic TXA. While uncontrolled studies in children suggest that inhaled TXA may be effective, clinical trials in the pediatric population are needed to determine both efficacy and safety.

Recombinant Factor VIIa

Recombinant factor VIIa (rFVIIa) promotes hemostasis by two mechanisms (Figure 2). In the tissue factor (TF) dependent pathway, FVIIa binds to TF at the side of injured endothelium, and this complex activates factors IX and X. This leads to the conversion of prothrombin to thrombin (“thrombin burst”), which in turn leads to increased fibrin production and clot formation. At higher doses (80–100 mcg/kg), rFVIIa works via a TF-independent pathway by directly activating factors IX and X on the surface of activated platelets.^42–44 In DAH, inflammation leads to the release of tissue factor, making rFVIIa a theoretically attractive treatment option.⁴⁴

Figure 2.

Medications used to treat pulmonary hemorrhage. TXA and EACA inhibit plasminogen activation, preventing fibrinolysis. rFVIIa binds tissue factor and binds factors IX and X on the surface of activated platelets, leading to increased thrombin production. Epinephrine, cold saline, oxymetazoline, phenylephrine, and vasopressin cause vasoconstriction. FVIIa: recombinant factor VIIa. TXA: tranexamic acid. EACA: Ɛ-Aminocaproic acid.

Although originally used to treat bleeding in patients with hemophilia with inhibitors or factor VII deficiency⁴⁵, rFVIIa has been used extensively off label. While authors of pediatric registry studies suggest that off label use of rFVIIa may be effective in treating bleeding from a variety of causes, these studies should be interpreted with extreme caution due to their observational nature.^46,47 Additionally, there is a small but real risk of arterial and venous thromboembolism associated with the use of systemic rVIIa.⁴⁷ The literature that describes the use of rFVIIa to treat pulmonary hemorrhage does so in the context of DAH. Our ability to determine efficacy in this context is limited because most of these studies are case reports, case series, and retrospective cohort studies.

Several case reports and series report the use of systemic rFVIIa in adults to treat DAH due to a variety of etiologies. In many instances, bleeding was life-threatening and refractory to other therapies, and patients experienced a dramatic and rapid improvement shortly after receiving a single dose of rFVIIa.^48–51 However, publication bias in case reports limits the routine application of these results to clinical practice. In a retrospective cohort study of critically ill adults with DAH following HSCT, the addition of rFVIIa to corticosteroids did not improve outcomes.⁵²

Park reviewed 37 reported cases in the literature of pulmonary hemorrhage in children without hemophilia that were treated with either intravenous (IV) or intrapulmonary rFVIIa. While clinical improvement was reported in many of these cases, the small sample size, pooling together of multiple case reports, and lack of control groups in general limit this report.⁴³

Direct endobronchial instillation of rFVIIa via bronchoscopy for refractory DAH was first reported by Heslet et al. in 2006 in a case series of 6 critically ill children, all of whom had already been treated with endotracheal or systemic TXA. They received 50 mcg/kg rFVIIa diluted in 50 mL of normal saline, and 25 mL aliquots were delivered to each of the main bronchi. The dose was repeated in 24 hours if adequate response did not occur with the first dose. Three patients experienced complete resolution of bleeding after the first dose, and the other three after repeated doses.⁴⁴ Subsequently, other case reports and series describe successful use of endobronchial rFVIIa using different dosing preparations to treat pulmonary hemorrhage in children and adults.^43,53–55

In addition to endobronchial instillation, rFVIIa may also be nebulized.⁵⁶ This is attractive because it obviates the need for a bronchoscopy to instill the medication. In the pilot study by Bafaqih et al., six of eight children with DAH who initially did not improve with nebulized TXA experienced resolution of hemorrhage when nebulized rFVIIa was added.⁴¹ Similarly, in a retrospective study of children with DAH following HSCT, nebulized rFVIIa in 10 patients was associated with decreased mortality.⁴⁰

Heslet et al. argue that intrapulmonary administration of rFVIIa is preferred to the systemic route because of the need to reach drug receptors on the “air side” of the alveolar capillary membrane. rFVIIa is too large to easily pass through this membrane from capillaries when administered systemically, impairing its ability to bind to tissue factor in the alveoli and achieve hemostasis. Furthermore, because inhaled rFVIIa is too large to pass through the membrane to reach the blood, there will be minimal, if any, systemic adverse effects. They surmise that daily inhaled rFVIIa may be an effective therapy for chronic causes of DAH that are traditionally managed with long-term systemic corticosteroids, although this has not been studied.⁵⁷

An important risk to consider unique to the setting of pulmonary hemorrhage is the development of an intrabronchial clot and airway obstruction; fatal cases have been reported in a child and adult.^58,59

One important consideration regarding the use of rFVIIa is its cost. The wholesale price of rFVIIa is $3·22 per microgram.⁶⁰ Therefore, using a dose of 50 mcg/kg in a 20 kg child would cost $3,220 per dose. Conversely, TXA costs $2·40 per milligram.⁶¹ A dose of 250 mg would cost roughly $600.

While uncontrolled studies, case reports, and case series report promising results regarding the use of rFVIIa to treat pulmonary hemorrhage in children, better evidence is needed before its use is widely adopted both because of its high cost and potential adverse effects. In cases where bleeding is refractory to other therapies, however, it may be considered.

Immunomodulatory medications

When pulmonary hemorrhage is associated with capillaritis or in idiopathic pulmonary hemosiderosis (IPH), immunomodulatory therapies may be required to achieve clinical stability.⁶² Pulse dose glucocorticoids are commonly used in immune-mediated pulmonary hemorrhage, although they are not well-studied, particularly in children. When capillaritis is present, more aggressive immunosuppressive therapy may be required.⁶² Definitive diagnosis of capillaritis has traditionally required open lung biopsy. However, we have had recent success diagnosing capillaritis using transbronchial cryobiopsy, which is a tool that is being increasingly used in pediatric centers for histopathologic evaluation of lung disease.⁶³ Specific therapies used to treat a particular systemic disease are often used to treat pulmonary hemorrhage associated with that disease.⁶⁴ In DAH due to systemic lupus erythematosus (SLE), the most commonly used therapies are high-dose corticosteroids and cyclophosphamide.⁶⁵ IPH is by definition not associated with capillaritis, but nonetheless patients characteristically improve when administered immunomodulatory therapy. Based on expert opinion and observational data, acute severe episodes of DAH due to IPH are treated with high dose (10–30 mg/kg/day) intravenous methylprednisolone. Additional immunomodulatory medications are added in cases refractory to high dose corticosteroids.⁶⁶

DAH is a devastating complication of HSCT that usually occurs within two months of transplant and is associated with a high mortality rate.⁶⁷ While the pathogenesis is not clearly defined, it is likely that infection, pulmonary toxicity related to conditioning regimens, graft versus host disease, transplant-associated thrombotic microangiopathy, and aberrant inflammatory responses play roles in causing endothelial injury and subsequent alveolar hemorrhage.^40,67 Identification and treatment of these concomitant conditions is vital in addressing the pulmonary hemorrhage itself.⁶⁷ Evidence regarding the impact of high dose (15–30 mg/kg/day) methylprednisolone on mortality is mixed, although prospective, controlled studies have not been performed.^40,67,68 As mentioned above, inhaled TXA and rFVIIa were both associated with improved survival in a retrospective study of children with DAH following HSCT.⁴⁰

Other

Several other medications have been used to manage pulmonary hemorrhage. As these are mainly documented in small case series or case reports, there is limited literature available regarding efficacy and safety of these agents in adults or children. Oxymetazoline is an alpha-1 adrenergic agonist that induces vasoconstriction. It is commonly used as an intranasal spray to treat nasal congestion and epistaxis.⁶⁹ Occasionally, it can cause hypertension.⁷⁰ To our knowledge, there is no literature describing its use in treating endobronchial bleeding in adults or children. At our institution, we commonly instill 1–2 mL of 0·05% oxymetazoline through the bronchoscope as our first-line agent to treat endobronchial bleeding. In our experience, this medication is very effective and well-tolerated. However, prospective studies are needed to assess its safety and efficacy. Phenylephrine is also an alpha-1 adrenergic agonist that may be instilled through a bronchoscope to treat pulmonary hemorrhage. One study showed that endobronchial phenylephrine increased mean arterial pressure in some patients, and it therefore should be used with caution in patients with hypertension.⁷¹

Ɛ-Aminocaproic acid (EACA), like TXA, is a synthetic lysine analog that prevents fibrinolysis by blocking the conversion of plasminogen to plasmin. It is effective in treating hemorrhage due to various causes in children,⁷² although its use in pediatric pulmonary hemorrhage has not been evaluated. One retrospective study of adults with DAH following HSCT found that systemic EACA was not associated with decreased mortality.⁷³ Simon et al. report successful use of endobronchial EACA to treat pulmonary hemorrhage in 46 adults. However, this retrospective study is weakened by multiple sources of confounding (medications, hemostatic maneuvers).⁷⁴

Vasopressin is a hormone produced by the pituitary gland that causes vasoconstriction when used at high doses. It may also increase factor VII and von Willebrand factor levels.^75,76 Nebulized vasopressin was successfully used in case reports to treat hemoptysis in a child⁷⁶ and adults.⁷⁵

The use of topical thrombin and fibrinogen injected through a bronchoscope to treat hemoptysis have been reported to be effective in small observational studies in adults.^77,78 Corcoran et al. describe the use of fibrin sealant just distal to an endobronchial valve that successfully controlled life-threatening pulmonary hemorrhage in a child with polychondritis and pulmonary aspergillosis.⁷⁹

Bronchoscopic instillation of absorbable materials such as gelatin and thrombin slurry⁸⁰ and cellulose mesh⁸¹ to treat severe hemoptysis has also been used with reported success in small retrospective adult studies.

Intrapulmonary medications used to treat pulmonary hemorrhage are summarized in Table 1, with mechanisms of action depicted in Figure 2.

Table 1.

Instilled/nebulized medications used to treat pulmonary hemorrhage in children. TXA, tranexamic acid. rFVIIa, recombinant factor VIIa. EACA, Ɛ-Aminocaproic acid.

Medication	Formulation	Mechanism(s) of action	Evidence for efficacy	Possible adverse effects	Dose(s) used	Frequency
Cold saline	Instilled	Vasoconstriction	Case reports and series’ in adults1, 2	- Transient bradycardia - Hypoxemia if large volume used	Unknown in children	During bronchoscopy
Epinephrine	Instilled	Vasoconstriction	Anecdotal	- Cardiac arrhythmia3 - Coronary vasospasm4	- Unknown in children - Increased diluent volume may increase toxicity5, 6	During bronchoscopy
TXA	Nebulized Instilled	Inhibits plasminogen activation, preventing fibrinolysis	RCTs in adults demonstrate efficacy.7–10 Low quality observational data suggest efficacy in children11, 12	- Bronchospasm9 - Seizures, VTE reported rarely with systemic TXA	Nebulized: - 250 mg11, 12 - 500 mg11, 12	Q6h11, 12 Q8h11 Q12h11 Q24h11
rFVIIa	Nebulized Instilled	- Binds to tissue factor, leading to thrombin burst which increases fibrin production - Directly binds factors IX and X on surface of activated platelets at higher doses	Very low-quality evidence based on observational studies12–14	- VTE with systemic rFVIIa - Intrabronchial thombus15	Instilled intrapulmonary: - 50 mcg/kg13, 14, 16 often diluted in 50 mL saline, with 25 mL instilled into each main bronchus Nebulized: - < 25 kg: 35 mcg/kg in 2 mL saline12 - 50 mcg/kg in 3 mL saline12, 13	Usually a single dose, may be repeated once 24 hours later17
Oxymetazoline	Instilled	Vasoconstriction	Anecdotal	Hypertension	0.5–2 mL of 0.05% (anecdotal)	During bronchoscopy
Phenylephrine	Instilled	Vasoconstriction	Anecdotal	Hypertension	Unknown	During bronchoscopy
EACA	Instilled	Inhibits plasminogen activation, preventing fibrinolysis	Low quality observational data in adults18	Unknown	Unknown in children	During bronchoscopy
Vasopressin	Nebulized	- Vasoconstriction - Increase factor VII and vWF levels	Case reports19, 20	Unknown	Nebulized: - 10 units diluted in 4 mL saline20 (one case report)	Q8h20 (one case report)

Plasmapheresis

Plasmapheresis may be used in life-threatening immune mediated disorders causing pulmonary hemorrhage based on the theory that removal of pathogenic antibodies will lead to improvement in inflammation that is causing the bleed. Plasmapheresis is standard of care in anti-glomerular basement membrane (anti-GBM) antibody disease⁸² and therefore should also be considered in hemorrhage associated with anti-GBM disease. Evidence regarding the efficacy of plasmapheresis in anti-neutrophilc cytoplasmic antibodies (ANCA)-associated vasculitis in adults is mixed.^83–85

Bronchoscopy

In addition to localization of bleeding and guiding placement of lung isolation devices, bronchoscopy is helpful in removing blood from the airways.¹¹ Fresh blood can be suctioned through the working channel of the flexible bronchoscope. When blood clots are present and are causing massive atelectasis or gas exchange impairment, a cautious effort should be undertaken to remove them.¹¹ A large bore suction catheter can be used to remove clots from the artificial airway. However, for clots located in the more distal airways, the cryoprobe is the first line tool used at our institution. Although most of the literature reports on the use of cryoextraction of clots in adults⁸⁶, the advent of the 1·1 mm cryoprobe has made it possible for the pediatric pulmonologist to easily extract blood clots from the airways of children large enough to accommodate a 3·5 mm inner diameter endotracheal or tracheostomy tube.^87,88 Bronchoscopy also allows for sampling of the lower airways to detect infection as a contributor to bleeding, and assessment for the presence of hemosiderin laden macrophages, which can differentiate between chronic and acute pulmonary hemorrhage.⁸⁹

Endobronchial valves (EBVs) are devices that allow air and mucus to exit the lung segment distal to the valve, but do not allow air in. They are FDA-approved for endoscopic lung volume reduction in adults with emphysema and are used off-label to treat bronchopleural fistula.^90,91 More recently, these devices have been used in children to treat bronchopleural fistulas.⁸⁷ EBVs have been used in several case reports in adults to treat refractory pulmonary hemorrhage, and in one case report in a child.^79,92–94 Using an EBV to prevent blood from spilling out of a lung segment seems counterintuitive, because the valve opens during exhalation and therefore would allow blood to exit. Several theories to explain the observed efficacy in these cases include creation of a nidus for coagulation and atelectasis of the affected segment leading to tamponade of bleeding vessels in addition to hypoxic vasoconstriction, both of which would lead to improvement of bleeding.⁹⁴ Because the EBV can be left in place for several months and removed when necessary, this could be a useful tool in centers with experience placing EBVs.

Bronchial artery embolization

Angiography with BAE is used to both identify the cause of and definitively treat pulmonary hemorrhage that occurs due to abnormally dilated bronchial arteries or non-bronchial systemic collateral vessels.⁹⁵ Because this procedure cannot be performed at the bedside, stabilization of the patient and at least some control of the bleeding must be achieved prior to BAE.⁷ However, in cases of severe pulmonary hemorrhage amenable to BAE, it should be performed as soon as is safely possible because it is a definitive therapy.⁷ Controversy exists regarding the decision to perform bronchoscopy or CTA prior to proceeding with BAE.⁹⁵ Both CTA and bronchoscopy may be helpful in localizing the site of bleeding which will inform the interventionalist performing BAE. These studies should be considered on a case-by-case basis in consultation with the interventionalist.⁹⁵

Embolization can be performed with embolic particles such as microspheres, polyvinyl alcohol (PVA) particles, or glue. Alternatively, coils may be used.^95,96 The choice of embolic agent depends on the size of the vessels being targeted and the preference of the operator. At our institution, PVA particles are often used because they effectively occlude small vessels while minimizing the risk of tissue ischemia.

While BAE has been extensively reported in adults with pulmonary hemorrhage,⁹⁵ the literature describing the use of BAE in children is limited. A retrospective study of 41 children in China who underwent BAE to treat pulmonary hemorrhage describes bleeding-free survival rates of 92·5%, 83·9%, 83·9%, and 70·8% at 6, 12, 24, and 36 months following the procedure.⁹⁶ A rare but serious complication of BAE is spinal cord infarction. In about 5% of the population, the anterior spinal artery arises from a bronchial artery, and proximal embolization of that bronchial artery will lead to spinal cord ischemia and infarction.⁹⁷ The rate of spinal cord infarction in children undergoing BAE is unknown. In a large study of over 1500 adults who underwent BAE for pulmonary hemorrhage, it occurred in 0·19%.⁹⁷

Surgery

There are several instances in which surgical resection of the bleeding lung portion is necessary. In adults, surgery is most often performed when bleeding recurs after BAE or BAE was technically unsuccessful.⁹⁸ Surgery performed for treatment of pulmonary hemorrhage in children has not been well-studied. In adults, emergent surgery is associated with higher morbidity and mortality compared to elective surgery performed after control of bleeding is achieved.^98,99 Commonly reported complications in adults include bronchopulmonary fistula and acute intraoperative hemorrhage.^98,99 The risk of complications depends on initial stability of the patient, surgical approach, and amount of lung resected.⁹⁹

In cases of pulmonary hemorrhage resulting in severe respiratory failure, extracorporeal life support (ECLS) has been used successfully to support patients while the underlying cause of hemorrhage is treated, despite the increased risk of bleeding associated with ECLS.¹⁰⁰

The specific management approach for pediatric pulmonary hemorrhage will vary depending on the patient history, bleeding severity and clinical stability, and institutional experience. We have presented one possible approach in Figure 3.

Figure 3.

Suggested algorithm for management of severe acute pulmonary hemorrhage. ETT, endotracheal tube. PEEP, positive end-expiratory pressure. TXA, tranexamic acid. rFVIIa, recombinant factor VIIa. CTA, computed tomography angiography. EACA, Ɛ-Aminocaproic acid. IR, interventional radiology

Conclusion

Pulmonary hemorrhage is a life-threatening condition that arises from a variety of causes. In addition to securing the airway and correcting anemia and coagulopathy, localizing the bleed with imaging and bronchoscopy is vital. Several techniques can then be used to protect the healthy lung in the setting of unilateral pulmonary hemorrhage. While many systemic and topical medications can be used to treat the bleeding, very few controlled studies have been performed in children. Bronchial artery embolization is the definitive treatment of choice in hemorrhage caused by abnormal bronchial arteries, although bleeding recurrence is common. The evidence base for most treatments of pediatric pulmonary hemorrhage is weak, and there is an urgent need to conduct prospective, controlled studies in this population. Because pulmonary hemorrhage is rare in children, multicenter collaboration will likely be necessary to adequately power clinical trials.

Key messages:

• Pulmonary hemorrhage may be life-threatening in children due to large airway obstruction and subsequent asphyxiation

• Securing the airway, identifying the bleeding source, and isolating the bleeding lung are the most important steps in acute management

• Bronchoscopy is essential to localize the bleed, clear blood from the airways, guide placement of bronchial blockers, and instill hemostatic medications

• Vasoconstrictors, antifibrinolytics, or recombinant factor VIIa may be used to treat acute pulmonary hemorrhage

• Immunomodulatory medications are used when the underlying cause of hemorrhage is pulmonary capillaritis or idiopathic pulmonary hemosiderosis

• Bronchial artery embolization is a definitive therapy for hemorrhage caused by enlarged bronchial arteries or bronchial-pulmonary collateral vessels