Keywords

Introduction

In 1983, the team in Toronto led by Joel Cooper performed the first successful lung transplant (Cooper et al. 1986). Although the introduction of cyclosporine was a significant factor in this success, most of the failures predating this were related to complications at the bronchial anastomosis. The first attempt at lung transplantation in a child was actually a heart-lung transplant performed by Denton Cooley, in 1968, in a 2-month-old infant with pulmonary hypertension and a complete AV canal. This patient survived for less than 12 h following the procedure (Cooley et al. 1969). Lung transplantation in children was firmly established in the early 1990s, although very few centers offered this. Over the past 20 years, pediatric lung transplantation has grown to the point that over 100 transplants are performed per year in more than 40 centers throughout the world (Benden et al. 2012). The 22nd Pediatric Lung and Heart-Lung Transplantation Report 2019 reported that a total of 2,514 pediatric lung and 733 pediatric heart-lung transplants were performed up to June 30, 2018 (Hayes Jr et al. 2019). Long-term success is limited by the availability of donor organs and wait list mortality pretransplant, and an increased infection risk because of immunosuppression, and most importantly late complications such as chronic lung allograft dysfunction and medication nonadherence (Benoit and Benden 2019; Lancaster and Eghtesady 2018). It remains the only option for end-stage lung disease failing medical therapy.

Indications

Lung transplantation is reserved for those with progressive pulmonary parenchymal or pulmonary vascular disease, where the life expectancy is less than 2 years and no alternative therapies are available. Predicting the life expectancy of someone with a chronic lung disease is by its nature difficult, particularly in the setting of progress in the medical research providing new and innovative therapies for these diseases. Submitting a patient to a lung transplant “early” could actually lessen his/her life expectancy. On the other hand, waiting until a patient is in critical condition lends itself to a higher risk patient less likely to survive the transplant procedure itself. The indications are clearly age-related and differ significantly based upon that factor (Benden et al. 2012). Pulmonary hypertension (idiopathic and congenital heart disease-related) and surfactant protein B deficiency are the most common indications in small children. Cystic fibrosis is the overwhelmingly most common indication in children between the ages of 6 and 18, accounting for approximately 65% of all transplants in this age group (Fig. 1). These are the general considerations to make when evaluating potential candidates.

Fig. 1
figure 1

The indications for lung transplantation in children vary significantly based upon the age at presentation. Cystic fibrosis predominates in the teenage category, while pulmonary hypertension and pulmonary fibrosis are more common in the young children and infants

Full size image

Cystic Fibrosis

Cystic fibrosis is by far the most common disease for which lung transplantation is performed in children (Benden et al. 2019). It accounts for two-thirds of all teenagers undergoing lung transplantation (Bryant 3rd et al. 2017; Dellon et al. 2017). The median survival of patients with cystic fibrosis is currently approximately 40 years of age. However, some deteriorate more rapidly than others, resulting in the need for transplantation during their childhood years. The timing of listing is important and is based largely on a study by Karem et al. and updated by Mayer-Hamblett et al. (Karem et al. 1992; Mayer-Hamblett et al. 2002). Patients are listed when the forced expiratory volume in one second (FEV1) on standard pulmonary function tests drops to below 30% of predicted, the pCO2 is rising on arterial blood gases, the patient becomes oxygen-dependent, and hospitalizations for pulmonary exacerbations increase in frequency.

Pulmonary Hypertension

Pulmonary vascular disease is the second most common diagnosis for which children require lung transplantation. This diagnosis is split between idiopathic pulmonary hypertension (IPH) and that related to congenital heart disease (PH/CHD). PH/CHD includes patients with Eisenmenger’s syndrome, but is not limited to that. The causes of death in both IPH and PH/CHD are progressive right heart failure, ventricular arrhythmias, and severe hemoptysis. Although the causes of death and the pulmonary histology are the same, IPH and Eisenmenger’s syndrome have much different long-term prognoses. A retrospective analysis by Hopkins of 100 adults with severe pulmonary hypertension due to either Eisenmenger’s syndrome or IPH revealed marked differences in the survival from the time of diagnosis. The patients with IPH had an actuarial survival of 77%, 69%, and 35% at 1, 2, and 3 years, respectively. In contrast, survival was 97%, 89%, and 77% in those patients with Eisenmenger’s syndrome over the same intervals (Hopkins et al. 1996). Predicting a life-threatening arrhythmia or episode of hemoptysis is impossible. However, progressive right heart failure is heralded by clinical signs along with a rising central venous pressure and falling cardiac output. These patients should be evaluated with regularly scheduled cardiac catheterizations while undergoing medical treatment. The frequency of lung transplantation for IPH and PH/CHD has decreased over the past decade as improvements in drug therapy have been developed. The models developed to predict life expectancy in IPH are no longer valid as they were developed prior to the availability of a whole host of drugs that provide some degree of selective pulmonary vasodilatation. These drugs include prostacyclin and its analogues, endothelin-1 antagonists, and phosphodiesterase inhibitors. Nonetheless, when the patient develops rising central venous pressure and dropping cardiac output while on maximal medical therapy, it is appropriate to place this patient on the prospective recipient list (Ivy et al. 2013).

Patients with PH/CHD include those with Eisenmenger’s syndrome, those with an “inadequate pulmonary vascular bed” (tetralogy of Fallot with pulmonary atresia and multiple aortopulmonary collaterals), those with repaired congenital heart disease and persistent pulmonary hypertension, and patients with pulmonary vein stenosis. The timing for transplantation in this rather broad group is the same as with IPH. Patients with pulmonary hypertension and repair congenital heart disease follow a clinical course more like IPH than those with Eisenmenger’s (Hopkins et al. 1996). In the largest review of transplantation for Eisenmenger’s syndrome, the underlying congenital cardiac diagnosis was a relatively straightforward lesion (atrial septal defect, ventricular septal defect, patent ductus arteriosus) in 85% and multiple congenital anomalies in the remaining 15% (Waddell et al. 2002). The transplant options for this group are isolated lung transplant with repair of the congenital cardiac lesion or heart-lung transplant. The advantage of heart-lung transplant is that the operation is not complicated by the need for isolated heart surgery with the inherent need for a period of myocardial ischemia. The disadvantages are the longer wait for a suitable donor, the possibility of early cardiac graft failure, the potential for isolated cardiac rejection, and the long-term risks of transplant coronary artery disease. The advantages of isolated lung transplant with repair of the congenital cardiac lesion are the patient retains his/her own heart with no chance of cardiac rejection, easier access to donor organs resulting in shorter time on the wait list, and a more economical allocation of a scarce resource – the heart can be used for another recipient. The disadvantages of isolated lung transplant with cardiac repair are that the cardiopulmonary bypass time is increased and an experienced congenital heart surgeon must be available at the time of the transplant. When a ventricular septal defect (VSD) is the underlying cause of Eisenmenger’s, and the patient is undergoing lung transplant with repair of the VSD, one must be mindful to divide the muscle bundles in the right ventricular outflow tract. It is very common for these muscle bundles to become so hypercontractile with relief of the pulmonary hypertension that a dynamic obstruction occurs post-repair. Waddell and colleagues reviewed the combined registries of the United Network of Organ Sharing and the registry of the International Society for Heart and Lung Transplantation. They found that heart-lung transplantation had a more favorable early survival posttransplant than did lung transplant with repair of the congenital cardiac lesion. The differences were consistently seen in the first month posttransplant; the survival curves were parallel beyond that (Waddell et al. 2002). Each center will have to decide how to approach these patients. For the very straightforward lesions, it would seem quite reasonable to repair the cardiac lesion and do an isolated lung transplant. Any added complexity to the cardiac operation would have to be considered a major risk factor.

The one specific type of PH/CHD for which there is no medical therapy is pulmonary venous stenosis. This presents as an isolated congenital lesion or in associated with another congenital abnormality, usually total anomalous pulmonary venous return (TAPVR). The incidence of pulmonary vein stenosis following repair of TAPVR is reported to be 15–20% (Karamlou et al. 2007). Some of these patients may be palliated with interventions such as balloon dilatation and stent placement (Tomita et al. 2003), but the recurrence rate is very high. Surgical treatment with the so-called sutureless technique has had reasonably good results for patients with pulmonary vein stenosis following repair of TAPVR (Yun et al. 2005). Those that fail to respond or have recurrent disease will deteriorate rapidly and need urgent listing. Pulmonary vasodilators are of no benefit.

Pulmonary Fibrosis

Pulmonary fibrosis is quite uncommon in children and is often a mix of fibrosis and bronchiolitis obliterans. It may be idiopathic or, more often, related to some sort of pulmonary injury such as a viral infection, radiation, chemotherapy, or following bone marrow transplant with graft versus host disease. Because of the mixed histology and unusual etiologies, it is difficult to develop a predictive model that reliably indicates need for lung transplantation. The survival at 5 years for a diffuse group of children with interstitial lung disease was 64% following the onset of symptoms. The development of pulmonary hypertension is a poor prognostic sign. That and the progression of symptoms on medical therapy are indications for listing for transplantation (Fan et al. 1997).

Surfactant Genetic Anomalies

Genetic abnormalities affecting surfactant protein production, metabolism, and function have recently been recognized as a cause of respiratory distress in otherwise normal term infants. The interstitial lung disease resulting from this is a consequence of mutations in genes that control surfactant protein B, surfactant protein C, ABCA3 transporter protein, and thyroid transcription factor. In many cases the respiratory failure is progressive, and there is no medical treatment available. This is particularly true for surfactant protein B deficiency. The clinical course of patients with surfactant protein C deficiency and abnormalities in ABCA3 transporter proteins is less predictable. The diagnosis can be made with genetic testing of a peripheral blood sample and should be suspected in an otherwise normal newborn with interstitial lung disease. Transplantation is the only treatment option for these infants with progressive respiratory failure but is complicated by multiple issues including small size of the recipients (infants), ventilator dependence (usually very high settings), unclear neurologic status (often sedated and paralyzed at the time of evaluation), and difficulty with posttransplant surveillance (Turca et al. 2013).

Other pulmonary diseases seen in the newborn for which transplantation has been applied include alveolar capillary dysplasia and bronchopulmonary dysplasia. Alveolar capillary dysplasia is a disease characterized by malalignment of the pulmonary capillaries with the alveolar sacs resulting in inability to provide adequate gas exchange. Most of these infants are critically ill from birth and die within a few days due to hypoxia and pulmonary hypertension. Treatment with nitric oxide and extracorporeal membrane oxygenation (ECMO) provide some degree of stability but do not result in long-term survival. The diagnosis is made with lung biopsy only (Steinhorn et al. 1997; Lazar et al. 2012). The only effective treatment for these patients is lung transplantation. The problem is maintaining a stable patient while awaiting a donor. Nearly all of these neonates will require ECMO during their course. A mechanical alternative to ECMO has been used to successfully bridge an infant with this diagnosis to lung transplantation (Hoganson et al. 2014). By far, the most common pulmonary disease in infants is bronchopulmonary dysplasia related to premature birth. Those most affected by this are those born very early with very low birth weight. In spite of extreme prematurity, few will develop end-stage lung disease although most will have some degree of pulmonary impairment (Hacking et al. 2013). Unfortunately, those that do have progressive pulmonary insufficiency also suffer from significant neurologic injury and are not suitable candidates for lung transplantation on that basis (Kurland 1996).

Retransplantation

Retransplant for either early or late graft failure is another indication. Early graft failure is usually the consequence of severe reperfusion injury to the transplanted lungs. With the lung allocation score currently in place, it is possible to obtain donor offers in these gravely ill patients in a timely fashion. However, the results of retransplantation in this situation are generally poor (Kawult et al. 2008). One should be very selective when considering a patient for retransplant in this situation. Late graft failure with respiratory failure is usually due to bronchiolitis obliterans. Retransplantation for these patients is only slightly worse than first-time transplant recipients. Again, careful patient selection is crucial in these situations.

Contraindications

There are absolute and relative contraindications to lung transplantation in children. These are listed in Table 1. Centers may vary somewhat in the consideration of these factors. All would agree that the presence of an active malignancy is an absolute contraindication. However, there may be some differences in considering how long the patient with a history of malignancy should be in remission prior to listing. The presence of multisystem organ failure can be altered with adequate medical support including the use of mechanical respiratory assistance, such as ECMO. Isolated dysfunction of the liver or kidneys is another issue to consider. The liver disease associated with cystic fibrosis may result in significant derangements of synthetic liver function concomitant with respiratory failure. The combination of liver-lung transplantation may be considered, and the results with this procedure have generally been good (Faro et al. 2007). The acceptable degree of renal dysfunction is open to some discussion. Given that nephrotoxic drugs will likely be necessary posttransplant, a serum creatinine of greater than 2.0 mg/dl or a glomerular filtration rate of less than 50 ml/min should give one pause. Active autoimmune diseases pose a risk to the transplanted lungs by causing vasculitis and other inflammatory changes.

Table 1 Contraindications to lung transplantation
Full size table

Colonization of the host with highly resistant organisms is a somewhat controversial issue. This applies primarily to patients with cystic fibrosis where the colonization is not just in the lower respiratory tract but also in the sinuses. The organisms of concern include Pseudomonas aeruginosa, Burkholderia cepacia, and atypical mycobacterial species. It is commonplace for cystic fibrosis patients to be colonized with multidrug-resistant pseudomonas organisms by the time transplantation is being considered. In general, the outcomes for these patients are no different than other diagnostic groups (Rosenblatt 2009). However, infection with Burkholderia cenocepacia (genomovar III) is associated with a very poor outcome; in one series there was 75% mortality in the first posttransplant year (De Soyza et al. 2010). Atypical mycobacterium species colonize the airways in approximately 10–15% of patients with cystic fibrosis evaluated for transplantation. Mycobacterium avium complex is the most common of these. Its presence does not appear to worsen outcomes and therefore is not a contraindication to transplantation (De Vrankrijker et al. 2010). Mycobacterium abscessus, however, is a much more difficult organism to treat and if not eradicated with aggressive therapy should be considered a contraindication (Zaidi et al. 2009).

Several other factors listed in Table 1 must be taken into consideration when evaluating any individual patient. At one time, the need for mechanical ventilation was considered a contraindication. It is clearly a risk factor for early and late failure (Benden et al. 2012). However, it is no longer considered a contraindication in most centers. Mechanical support with ECMO or some other lung assist device also places the patient at some added risk, but most do not consider this a contraindication presently. Lung transplantation requires significant commitment on the part of the patient and his/her caregivers. Psychosocial factors, particularly when accompanied by a history of poor compliance, should raise red flags and prompt thorough evaluation. Failure in this regard will universally result in a poor outcome.

Listing and Donor Evaluation

Beginning in 2005, a lung allocation score was assigned to all patients over the age of 12 years. This was developed as an alternative to the “first-come, first-served” method previously in place that was based only on time accrued on the waiting list. Critically ill patients are now given priority resulting in a reduction in wait list mortality in adults; a similar trend has not yet been noted in children (Sweet 2009). The risk inherent in this allocation system is that the outcome posttransplant will be worse due to sicker candidates being transplanted. Thus far that has not been noted.

Donors are matched on the basis of ABO blood typing and size. Using height as a surrogate of chest size is typically more accurate than weight for size matching. 20% above and below the recipient will provide a reasonable size match. Larger lungs can be trimmed to fit if necessary using a stapling device to go around the edges. Donor lobectomy may be required in some instances. Donors are excluded if there is a history of positive HIV serology, malignancy, active hepatitis, asthma, tuberculosis, or other significant pulmonary disease. A history of limited cigarette smoking is probably acceptable if other parameters of the evaluation fall within the guidelines. The upper age limit for donors is generally 55 years, but successful lung transplants have been performed using donors older than 60.

Donor lung function is evaluated with chest radiograph, arterial blood gas on 100% oxygen, and flexible bronchoscopy. The chest film should be free of infiltrates. Any atelectasis should be aggressively treated with chest physiotherapy. The arterial blood gas should have a pO2 of greater than 300 mm Hg with an inspiratory FiO2 of 1.0. Bronchoscopy is performed looking for purulent secretions or evidence of aspiration.

The donor harvest procedure is generally part of a multi-organ retrieval. A midline sternotomy is performed. Both pleural spaces are opened for visual inspection and palpation. The trachea is dissected out between the superior vena cava and the ascending aorta. It may be helpful to develop the interatrial groove to allow for better division of the left atrial tissue anterior to the right pulmonary veins, an area that must be shared with the cardiac retrieval team. The steps to the procurement beyond this are (Barr et al. 2001) anticoagulation with high-dose intravenous heparin (300 units/kg), (Battafarano et al. 2000) bolus injection of prostaglandin E1 (50–70 μg/kg), (Benden et al. 2019) decompression of the right and left sides of the heart by dividing the inferior vena cava and left atrial appendage, (Benden et al. 2012) cross-clamping the aorta, (Benden et al. 2005) high volume (50 ml/kg) low-pressure flush of cold pulmonary preservation solution via a cannula in the main pulmonary artery, (Benoit and Benden 2019) topical application of cold saline and slush to the lungs, and (Bhorade and Stern 2009) continued ventilation of the lung with low volumes and low pressures using an FiO2 of 0.4. After the cardiac retrieval team has removed the heart, retrograde pulmoplegia should be given via all four pulmonary veins; this serves to flush out any pulmonary emboli that might have occurred in the donor while in the ICU. It is not at all uncommon to see this. The lungs are removed en bloc with the mediastinal tissue including the esophagus and descending thoracic aorta. The esophagus is isolated with a stapling device to avoid contamination. The trachea is stapled, while the lungs are gently inflated at a low inflation pressure (15–20 cm H2O) with an FiO2 of 0.4. The lungs are then explanted and placed into a container of the preservative solution which is then placed into sterile bags of cold saline and ice. This is then placed into a cooler for transport. A number of preservative solutions are available including a modified Euro-Collins solution, University of Wisconsin solution, Perfadex, and Celsior. None of these solutions has clear superiority over another.

Transplant Procedure

The operation is performed using a trans-sternal bilateral anterior thoracotomy incision, the so-called clamshell incision. This provides excellent access to both pleural spaces as well as access to the heart for cardiopulmonary bypass and intracardiac repair of any associated cardiac anomalies if present. Most lung transplants in children are bilateral, rather than single lung transplants. Single lung transplants in cystic fibrosis would likely result in contamination of the transplanted lung with infected material from the contralateral lung. For non-septic lung diagnoses, bilateral lung transplantation is preferred because of better long-term results overall and concerns over the growth potential for transplanted lungs. Cardiopulmonary bypass is used in most patients because of the technical difficulties of single lung ventilation in small children that would be necessary if transplantation were performed without cardiopulmonary bypass. Bilateral recipient pneumonectomies are performed. The transplants are bilateral sequential lung transplants with bronchial, pulmonary artery, and pulmonary vein anastomoses. The bronchial anastomoses should be wrapped with viable tissue in the general vicinity, such as donor and recipient peribronchial tissue or pericardium. Following completion of the anastomotic connections, the patient is weaned from cardiopulmonary bypass, appropriate drains are placed, and the chest is closed.

When a particularly large donor is accepted with the plan to use lobes only as whole lungs in a smaller recipient, the lower lobes are generally preferred because the anatomy is technically more suitable. A technique has been devised by which a single left lung may be partitioned to use the upper lobe on the right and the lower lobe on the left (Couetil et al. 1997). The circumstances under which one might employ this technique would be quite unusual – a desperately ill small child with a donor offer in which the right lung is unavailable. Nonetheless, this is another attempt at maintaining flexibility in dealing with transplant organ shortage.

Posttransplant Immunosuppression

So-called induction immunosuppression is generally begun soon after completion of the transplant procedure using a cytolytic agent, such as antithymocyte globulin or a lymphocyte interleukin-2 receptor-binding monoclonal antibody such basiliximab or daclizumab. The theory is that with intense early immunosuppression, there will be less acute graft rejection early following lung transplantation. The use of cytolytic agents fell into disfavor because of the observed increase in infections, particularly cytomegalovirus. The interleukin-2R agents work by blocking a critical pathway in the activation of lymphocytes involved in cellular rejection, rather than indiscriminately killing lymphocytes. The low infection rate associated with these drugs has increased the popularity of this strategy. Maintenance immunosuppression is usually a triple-drug regimen consisting of either cyclosporine or tacrolimus, azathioprine or mycophenolate mofetil, and steroids (Bhorade and Stern 2009).

Nearly all patients will remain on this triple drug regimen for the remainder of their lives. A few may be weaned from the steroids. The side effects of these agents are significant. For obvious reasons there will be risk of infections. Cyclosporine and tacrolimus are both associated with risks of renal insufficiency, hypertension, and seizures. Tacrolimus and corticosteroids may increase the risk of diabetes mellitus, particularly in patients with cystic fibrosis whose pancreatic function is already abnormal. Cyclosporine may cause hirsutism and gingival hyperplasia. Azathioprine is associated with leukopenia. Mycophenolate mofetil may cause a variety of gastrointestinal symptoms. The advantage of the triple drug approach is that lower doses of these drugs in combination provide adequate immunosuppression and limit side effect exposure.

Posttransplant Management and Complications

These patients require close hemodynamic monitoring with every effort made to maintain adequate cardiac output while keeping the intravascular volume as low as possible. Because there is no lymphatic drainage from the transplanted lungs, pulmonary edema can occur at a lower left atrial pressure than in otherwise normal patients. Patients transplanted for pulmonary hypertension require particularly close monitoring and support as they enter the procedure with poor right ventricular function and are prone to hemodynamic instability.

Flexible bronchoscopy is performed within 24 h of the transplant to assess the gross appearance of the bronchial anastomosis and sample the secretions for infection surveillance. A perfusion scan of the lungs is also performed within that time period to assess the relative blood flow distribution to both lungs. These two tests are generally adequate to assure that the bronchial and vascular anastomotic connections are satisfactory. If the perfusion scan shows a significant mismatch in the relative distribution of flow to each lung, this should be investigated further as soon as possible. This could be due to stenosis in either the pulmonary arterial or venous anastomoses. Echocardiography is a reasonable screening tool, and it is often helpful to perform this as a transesophageal approach as the images from the transthoracic views may not be adequate. If this is still inconclusive, a cardiac catheterization should be performed. In some circumstances it may be possible to stent a stenotic anastomosis in the cardiac catheterization lab. Otherwise, the finding of a significant stenosis in either the arterial or venous connection should prompt immediate return to the operating room for revision.

Posttransplant antibiotic prophylaxis is based upon pretransplant sputum cultures for patients with septic lung diseases. For others, a broad-spectrum antibiotic is generally satisfactory. Cytomegalovirus infection following lung transplantation is a significant source of morbidity (Danziger-Isakov et al. 2003). Prophylaxis against cytomegalovirus infection is recommended for all patients with any combination of seropositive donor or recipient. Prophylaxis against fungal infections with either oral nystatin or fluconazole is recommended. Trimethoprim/sulfamethoxazole or inhaled pentamidine is provided for prevention of Pneumocystis jirovecii (formerly carinii) pneumonia.

Primary graft dysfunction (PGD) is a common complication following transplantation. This is also referred to as reperfusion injury. The precise cause is unknown and is probably multifactorial including inflammatory and immunological injury-repair responses. It is characterized by pulmonary edema with diffuse alveolar damage that manifests clinically as progressive hypoxemia with radiographic pulmonary infiltrates. The incidence is on the order of 20–25% depending upon the precise definition used (Christie et al. 2005a). The degree of graft failure varies from mild to life threatening. However, the 30-day mortality for those with significant PGD is around 40% versus 6% for those without this (Christie et al. 2005b). The treatment is supportive therapy as no specific medication has demonstrated efficacy (Shargall et al. 2005). These supportive measures include inhaled nitric oxide, diuresis, higher levels of positive end-expiratory pressures while maintaining lung-protective ventilation strategies, and ECMO if necessary. Earlier intervention with ECMO is recommended to avoid injury to the newly transplanted lungs (Shargall et al. 2005).

Rejection is relatively common following lung transplantation, perhaps more so than any other solid organ transplant. The lung has a much larger endothelial surface on which major histocompatibility antigen expression occurs compared with other solid organs. The clinical manifestations are generally non-specific; some patients may be completely asymptomatic. Therefore, the diagnosis must be substantiated by histologic evidence on tissues samples acquired from transbronchial biopsies (Fig. 2) performed on a regularly scheduled basis. The incidence of rejection is highest in the first 6 months following transplantation and diminishes beyond that. A typical protocol for surveillance bronchoscopy is every 2 weeks for the first month, monthly for the next 2 months, and every 3 months for the next 2 years. Any clinical change should prompt bronchoscopic evaluation as well. Rejection is treated with bolus steroids, generally in a dose of 10 mg/kg of methylprednisolone daily for 3 days. Follow-up bronchoscopy within 2 weeks of treatment is recommended to confirm resolution. Unresolved or recurrent rejection should be treated more aggressively with cytolytic agents and a change in the baseline immunosuppression regimen.

Fig. 2
figure 2

This depicts the histologic findings in acute pulmonary rejection. There is an accumulation of mononuclear inflammatory cells around vascular structures

Full size image

Infections are generally more common following lung transplant than other solid organs with the respiratory tract being the usual source. With the exception of intestinal transplants, the lung is the only transplanted organ constantly exposed to the outside world. The donor organ comes from an individual on the ventilator with an endotracheal tube in an airway colonized by whatever organisms are prevalent in that particular unit. Posttransplant the recipient is also on the ventilator with an endotracheal tube bypassing some of the natural defenses of the respiratory tract. The transplanted lung is denervated resulting in diminished cough reflex. These factors and others, along with immunosuppression, predispose the transplanted lung to infections. The infectious agent may be viral, bacterial, or fungal but the most common are bacterial. Bronchoscopy done as a regularly scheduled procedure or with clinical indications should be accompanied by bronchoalveolar lavage to screen for infections of all sorts. Patients with cystic fibrosis will likely have infections caused by the same organisms colonizing the airways and sinus pretransplant. The viruses causing the most difficulty in children are adenovirus and parainfluenza. Cytomegalovirus as a primary infection is more likely to be severe than when it occurs as reactivation. Primary infection with Epstein-Barr virus is significant as it is an important risk factor for posttransplant lymphoproliferative disease. Fungal infections are uncommon but can be devastating, particularly in the case of invasive Aspergillus.

Airway complications occur in approximately 10–15% of all airway anastomoses at risk (Choong et al. 2006). Airway dehiscence, once the bane of lung transplantation, is now quite uncommon. Partial dehiscence may occur and generally heals without significant sequelae. Airway stenosis, however, is relatively common. This is treated with dilatation of the stenotic airway using the same balloon-tipped catheters used in the cardiac catheterization for vascular dilatation. Usually multiple dilatations on a regularly scheduled basis are necessary as the airway heals. Airway stents may occasionally be necessary when more conservative therapy fails; however, stents can be problematic due to risk of granulation tissue and problems clearing secretions. The preferred stent is a self-expanding covered metal stent placed under fluoroscopic control via bronchoscopy. These do not need to stay in long term but can be removed within 6 months to a year. During that time, it is expected that the airway will have healed in the size and shape of the stent. Reoperation directly on the stenotic bronchus may be necessary in rare instances.

Hoarseness due to injury to the left recurrent laryngeal nerve has an incidence of approximately 10%. This likely occurs during dissection of the left pulmonary artery in the region of the ligamentum arteriosum or along the proximal length of the left vagus nerve. The diagnosis is made clinically and confirmed at the time of flexible bronchoscopy by direct examination of the vocal cords. In most cases this will improve within 6 months without specific therapy. This does place the patient at risk of aspiration, however.

Phrenic nerve injury occurs in 20% of lung transplants and is particularly more common during retransplants because of the blurring of dissection planes consequent to adhesions from the prior operation (Ferdinande et al. 2004). This is perhaps more common on the right than the left in part because of the proximity of the nerve to the pulmonary artery and the need for greater retraction on that side necessary for exposure of the hilar structures for the bronchial and vascular connections. Recovery of phrenic nerve function later on is typical although there may the occasional patient that might benefit from diaphragmatic plication.

Gastrointestinal complications are relatively common with gastroesophageal reflux being the most common of these (Benden et al. 2005). This occurs in approximately 50% of infants undergoing lung transplantation. The diagnosis is made by 24-h esophageal pH probe, upper gastrointestinal barium studies, or evidence of aspiration by the presence of lipid-laden macrophages on bronchoalveolar lavage taken during routine bronchoscopy. This may be related to injuries to both vagus nerves during the transplant procedure. It should be treated aggressively as there may be some relation to the development of bronchiolitis obliterans. Patients with cystic fibrosis are at risk for distal intestinal obstruction syndrome (Minkes et al. 1999). This can be avoided by the aggressive use of osmotic cathartics following transplantation. Prophylaxis against ulcers using appropriate agents is recommended given the stress associated with the procedure and the use of high-dose steroids.

Other early complications include bleeding, atrial arrhythmias, and seizures. These patients are at particular risk of bleeding because of the frequent association of at least mild liver dysfunction, the need for relatively prolonged cardiopulmonary bypass for the transplant procedure, and the presence of adhesions due to prior pulmonary infections or prior operations. These adhesions are particularly vascular in patients with cyanotic congenital heart disease and prior thoracotomies. The most common atrial arrhythmia is atrial flutter and may occur in up to 10% of recipients (Gandhi et al. 1996). Some may require long-term treatment for this. Neurologic complications following lung transplant occur in nearly half and include headaches, seizures, tremors, and focal neurologic deficits (Wong et al. 1999). These are generally transient, and most are related to posterior reversible encephalopathy syndrome (previously referred to as leukoencephalopathy) caused by calcineurin inhibitors. Assurance that the calcineurin levels are at the lowest level possible for therapeutic effect is the mainstay of therapy. Some patients may require antieleptic medications.

Late Complications

Bronchiolitis obliterans is by far the most important and the most common late complication of lung transplantation. This disease is characterized histologically by fibrous obliteration of the distal bronchioles (Fig. 3). The pulmonary function tests typically show a falling forced expiratory volume in 1 s. The diagnosis of bronchiolitis obliterans syndrome is made when the FEV1 is less than 80% of the peak measures value posttransplant. More than 50% of all lung transplant recipients surviving to 5 years will have acquired this. The cause is unknown although suspected to be related to chronic rejection (Verleden et al. 2005). A number of risk factors have been suggested including cytomegalovirus infection (Scott et al. 2005), prior episodes of acute rejection (Husain et al. 1999), history of primary graft dysfunction (Daud et al. 2007), gastroesophageal reflux (Cantu III et al. 2004), and others. A variety of treatments have been proposed. One difficulty with analyzing any therapy is that the bronchiolitis obliterans typically has a rapid early decline in pulmonary function followed by a less steep decline over the ensuing months. The impact of therapy is often assessed by the change in the slope of the decline in FEV1. The usual treatment is enhanced immunosuppression. Initially this is in the form of cytolytic therapy with either antithymocyte globulin or anti-lymphocytic globulin. Following this, there is usually a change in the maintenance immunosuppression as well. Other treatments include azithromycin, which has shown promise in limited studies without placebo controls (Gottlieb et al. 2008). Photopheresis has gained some popularity recently with some promising results (Jaksch et al. 2012). Finally, retransplantation may be appropriate for those failing these therapies.

Fig. 3
figure 3

Bronchiolitis obliterans is characterized by fibrous and inflammatory cell plugging of bronchioles

Full size image

Although the incidence of acute cellular rejection declines over the first year posttransplant, antibody-mediated rejection may be an ongoing concern. The precise incidence of this form of rejection is unknown because of the difficulty in making a diagnosis. The combination of inflammatory changes, positive staining for complement 4d, and the presence of circulating donor specific antibody is generally used to confirm the diagnosis. This diagnosis is complicated further by issues of appropriate treatment. Removal of antibody is typically accomplished by plasmapheresis. However, the cells responsible for production remain. Therefore, some have advocated rituximab in addition to plasmapheresis in an effort to impact the function of type B lymphocytes. The impact of any of these therapies is unknown (Glanville 2013; Chih et al. 2013). It seems likely to play a significant role in the development of bronchiolitis obliterans, thus underscoring its importance in long-term survival.

Malignancy occurs in approximately 10% of all long-term survivors. This is usually of lymphatic origin and is referred to as posttransplant lymphoproliferative disease. Epstein-Barr viral infection as a primary infection is a particularly strong risk factor for this (Reams et al. 2003). Treatment of this includes a reduction of immunosuppression, which may be effective if early in the disease. Rituximab, a monoclonal antibody to the CD20 protein found on type B lymphocytes, is particularly effective in treatment. Chemotherapy has also been advocated (Gross et al. 2012).

Renal insufficiency posttransplant is a time-related complication. There are often some difficulties with renal function early posttransplant related to cardiopulmonary bypass, the use of nephrotoxic antibiotics, and hemodynamic issues. This often resolves. However, over time the degree of renal insufficiency worsens related to the long-term use of calcineurin drugs (cyclosporine or tacrolimus). At 5 years posttransplant, approximately 10% of all patients will have some degree of renal insufficiency; that figure jumps to 29% at 10 years (Benden et al. 2012).

A number of other long-term morbidities are a consequence of lung transplantation in children. These include hypertension, hyperlipidemia, and diabetes mellitus. These are associated with increasing rates of frequency over time following transplantation. By 5 years, over 60% of children require treatment for hypertension, 17% have hyperlipidemia, and 36% have diabetes mellitus (Benden et al. 2012). Many of those with diabetes mellitus had pretransplant glucose intolerance. Nonetheless, these statistics illustrate the ongoing healthcare needs of these patients beyond their lung and immunosuppression management.

Survival

The 1-, 5-, and 10-year survival following lung transplantation in children is approximately 75%, 50%, and 36% respectively. This is comparable to the results in adults (Fig. 4). Table 2 lists the causes of death based upon the timing following transplantation. Graft failure is the most common cause of death early following lung transplantation. Bronchiolitis obliterans is the most common cause of death late following transplantation. There has been some improvement in survival comparing an earlier era (1988–1999, 42% 5-year survival) to a more recent era (2000–2011, 53% 5-year survival), although the differences appear related to the first month following transplantation where the survival curves diverge (Fig. 5). This is likely the result of better candidate selection, improved treatment/prevention of reperfusion injury, and perhaps better donor selection. The impact of bronchiolitis obliterans on survival appears unchanged.

Fig. 4
figure 4

The survival following lung transplantation in children is similar to that seen in adults with approximately 50% survival at 5-year posttransplant. There may be some divergence favoring the pediatric age group at 10 years following lung transplantation. (Taken from the Registry of the International Society of Heart and Lung Transplantation 2013 Power Point slides)

Full size image
Table 2 Causes of death following lung transplantation in children
Full size table
Fig. 5
figure 5

Survival appears improved in the most recent decade of lung transplantation compared to the earlier era. This all appears related to the early results; beyond that the curves are nearly parallel. (Taken from the Registry of the International Society of Heart and Lung Transplantation 2013 Power Point slides)

Full size image

Special Considerations

Living donor lung transplantation. This procedure involves using the lobes (usually lower) of two acceptable donors who are generally significantly bigger than the recipient (Barr et al. 2001). These two lobes are then transplanted as whole lungs into the recipient. This procedure was adopted as a bail out for recipients felt to be too ill to wait for a suitable cadaveric donor. The implementation of the lung allocation scoring system for organ allocation has nearly eliminated the need for this procedure. In fact, fewer than five such procedures have been performed since 2005, the year the lung allocation score was implemented. The survival of these patients is similar to that seen with cadaveric donors. However, Battafarano et al. report that 60% of donors had some sort of complication, 15% of which were considered major (Battafarano et al. 2000). For these reasons this procedure is rarely practiced today.

Infant lung transplantation. Infants have particularly difficult issues for a variety of reasons (Huddleston et al. 1999). Nearly all recipients have been ventilated from birth and are often sedated and paralyzed for much of their lives due to the severity of the lung disease. This makes a careful pretransplant evaluation very difficult, particularly the neurologic status. Although imaging studies are difficult to obtain because of the tenuous clinical status of the patient, an MRI or CT scan of the head should be done in the absence of a reliable examination. Postoperative surveillance is complicated by small airways preventing consistent acquisition of adequate tissue via transbronchial biopsies for rejection evaluation. Small flexible bronchoscopes with corresponding small channels for biopsy forceps are now available. However, the tissue samples obtained may not be suitable for analysis. Standard pulmonary function tests cannot be performed. Infant pulmonary function tests are available, but consistent values are difficult to obtain, and there has not been validation of findings with the presence of bronchiolitis obliterans. In spite of these obstacles, the survival of infants is nearly equivalent to older children. The explanation may lie in the observation that the incidence of acute rejection is less than that seen in older recipients. Likewise, the incidence of bronchiolitis obliterans is less, and the onset is later posttransplant (Ibrahim et al. 2002). Infection is the most common cause of death in this age bracket. This begs the question as to whether infants can be managed with less immunosuppression and thus reduce their risks of the sequelae of a lifetime of these medications.

Mechanical support. Although mechanical ventilation pretransplant is recognized as a risk factor in outcomes (Elizur et al. 2007), the impact on survival in infants is less clear, particularly since most infants listed have such significant lung disease that mechanical ventilation is necessary for survival. Beyond that, other more invasive support systems such as ECMO have been associated with very poor outcomes in the past (Puri et al. 2010). However, more recent experience has shown that the use of ECMO support with a goal toward rehabilitation of patients has had reasonably good results (Hoopes et al. 2013). Patient selection is very important. An alternative to ECMO is a pumpless paracorporeal lung device. The experience with this as a bridge to transplantation is limited, but it has been effective in a small number of patients (De Perrot et al. 2011).

Growth. This outcome measure is unique to pediatrics. Most children come to transplantation with growth failure. There is generally some “catch-up” growth that occurs after transplant, but most recipients have some delay, presumably due to the suppressive effect of corticosteroids (Sweet et al. 1997). Growth of the transplanted lungs has been much more difficult to assess, and the results are somewhat mixed (Cohen et al. 1999; Sritippayawan et al. 2002). Children receiving their transplanted lungs during infancy have shown reasonable lung capacity as they grow to school age suggesting that growth has occurred. Infant pulmonary function tests measure functional residual capacity, a reasonable surrogate for lung volume. In a single study looking at lung growth in infants following lung transplantation, the average functional residual capacity per centimeter in height at 3-month posttransplant was 2.3 ml/cm. Through 15-month posttransplant, this remained between 2.1 and 2.8 ml/cm. During this time substantial somatic growth occurred in these infants suggesting that lung growth appropriate for size occurred (Cohen et al. 1999). This was an observational data only, and further study is necessary to confirm these findings.

Conclusions and Future Directions

What lies ahead for lung transplantation in children? There is little question that the status quo is unacceptable. 50% survival at 5-year posttransplant does not constitute a good result, particularly given the investment of time and resources into this therapy for end-stage lung disease. To be sure, lung transplantation is the only therapeutic option for children with end-stage pulmonary parenchymal and vascular disease. Much research is ongoing in the area of increasing the donor pool, immunosuppression, primary graft dysfunction, and bronchiolitis obliterans; these are the major hurdles in the quest for improved long-term results. Although there are encouraging reports, no clinically applicable progress has been made in any of these areas of concern.

Perhaps the most intriguing development in the past decade is in the area of ex vivo lung perfusion (EVLP). In this procedure, the donor lungs are retrieved and placed inside a sterile dome with an endotracheal tube inserted into the trachea and a cannula inserted into the main pulmonary artery. At normothermia, the lungs are perfused with a hyperosmotic, balanced electrolyte solution while being ventilated at around 5–10 ml/kg (donor weight) at a relatively low rate. Using this system, some marginal lungs have been “revitalized” and became acceptable for transplantation. Bronchoscopy with sampling and clearing of airway secretions is possible during the perfusion. Although the original intent was to see if unacceptable lungs could become acceptable, it is now evident that there may be some therapeutic advantage of EVLP over “standard” lung procurement. There is evidence that genes involved in inflammation are downregulated during EVLP. In fact, the incidence of PGD in the limited human experience to date is quite low. Another early finding has been that the incidence of rejection is lower with EVLP than with standard lung procurement, perhaps related to the removal of donor-derived leukocytes. Finally, this procedure may allow for the introduction of a variety of treatments to the airways and vascular tree. These include stem cells, gene vector therapy, and a variety of drugs which otherwise could not be given in either a directed fashion or in the safe doses required if given systemically. How all this evolves over the next decade is obviously unknown but is perhaps the most exciting area of lung transplant research to come along since the first successful operation in 1983 (Roman et al. 2013).

Cross-References