Abstract
This article reviews the common pulmonary complications seen in the pediatric oncology population and our approach to diagnosis, management, and therapy considerations in this specialized population, including patients receiving chemotherapy, radiation, and hematopoietic stem cell transplantation. Although infections cause the most significant complications in this population, non-infectious complications, including acute lung injury from chemotherapy or radiation, idiopathic interstitial pneumonia, diffuse alveolar hemorrhage, bronchiolitis obliterans, and cryptogenic organizing pneumonia, also occur commonly. With improvements in survival of childhood cancer, there are now a growing number of adults who are childhood cancer survivors who may be encountered by therapists in adult hospitals. We also review the growing literature on the emerging late pulmonary findings in these adult childhood cancer survivors.
- childhood cancer
- chemotherapy
- radiation
- hematopoietic stem cell transplantation
- interstitial pneumonia
- pulmonary hemorrhage
- bronchiolitis oblterans
- pnemonia
Introduction
Childhood cancer therapy is associated with a wide variety of pulmonary complications during therapy. Most care of children with cancer is done in specialized cancer centers, so most respiratory therapists may not be involved in the care of these complex patients. This article reviews the common pulmonary complications seen in our large children's cancer center and our approach to diagnosis, management, and therapy considerations in this specialized population. With improvements in survival of childhood cancer, there is now a growing number of adults who are childhood cancer survivors who may be encountered by therapists in adult hospitals. Therefore, we also review the growing literature on the late pulmonary findings in these adult childhood cancer survivors.
Historical Perspective
In the 1970s, large numbers of children began to survive their primary malignancy, only to die with infectious complications, including Pneumocystis jirovecii (formally carinii) pneumonia. Initially, there was great reluctance to intubate cancer patients due to fear of worse pulmonary complications. Drs Sam Sanyal and Walter Hughes at St Jude Children's Research Hospital pioneered the use of negative pressure to avoid intubation and saved many P. jirovecii pneumonia patients during that period1 (Fig. 1). With the discovery of effective P. jirovecii pneumonia prophylaxis in 1975 using trimethoprim-sulfamethoxazole, this infection ceased to be a major cause of mortality.2 However, with more intensive chemotherapy and radiation regimens, fungal and viral infections, such as Aspergillus, cytomegalovirus, and varicella zoster, emerged as the primary pulmonary infectious pathogens in the 1980s.3,4 Although the prognosis for pediatric cancer and hematopoietic stem cell (bone marrow) transplant patients was poor during this early period, later advances in antifungal and antiviral therapy and supportive pediatric critical care for cancer patients has significantly improved survival since the 1990s.5–7 With improved survival, we have seen the development of more chronic lung injury syndromes in oncology patients who require long-term respiratory support, including tracheostomy and prolonged mechanical ventilation.
Lung, Airway, and Chest-Wall Involvement in Primary Malignancy
Although primary lung tumors are rare in childhood, primary and secondary tumor involvement can cause significant problems during initial therapy for childhood cancer. These problems include airway obstruction from large mediastinal tumors or lymph nodes. The management of large mediastinal tumors or malignant lymph nodes may involve the respiratory therapist, since simple sedation can lead to life-threatening airway obstruction without careful planning.8 In addition, the chest wall may be involved in primary malignancies, and resection of ribs and chest wall can contribute to respiratory complications. The most common primary pediatric lung tumors include pleuropulmonary blastoma (chest masses) and carcinoid tumors (endobronchial obstruction).9 Large malignant pleural effusions can also complicate initial management of childhood cancers and may require chest tube insertion or tapping for diagnostic purposes or relief of respiratory distress. The lung is more likely to be involved with metastatic disease with pediatric tumors, particularly for malignancies such as Wilms' tumor; osteosarcomas; other sarcomas, such as Ewing's sarcoma and rhabdomyosarcoma; and hepatoblastoma. Surgical resection of these metastatic lesions can contribute to reduced lung function, particularly after resection of multiple lesions.10
Opportunistic and Usual Respiratory Infections in the Childhood Cancer Patient
Some common pulmonary pathogens are listed in Table 1. The term opportunistic infection usually refers to organisms that typically are seen primarily in patients with a compromised immune status (eg, neutropenia or depressed T- or B-cell immunity). Infections such as cytomegalovirus, Pneumocystis, and fungi, such as Candida and Aspergillus, are usually seen in patients with both primary and acquired immunodeficiency states, such as during cancer treatment. In addition, cancer patients can also be infected with common bacterial and viral respiratory pathogens, such as Staphylococcus aureus, influenza, respiratory syncytial virus (RSV), human metapneumovirus, adenovirus, and varicella zoster, but these common infections can lead to catastrophic pulmonary infection in the immunocompromised population.3,11–14 The degree of residual host defense also plays a role in the severity of the lung process, and the phenomenon of worsening lung disease with recovery of immune function following cancer therapy is well described for a number of respiratory pathogens. Many neutropenic children with acute respiratory failure experience worsening of their oxygenation during neutrophil recovery.15
Common pulmonary infections in pediatric cancer patients include viral infections like cytomegalovirus. Cytomegalovirus pneumonia can occur as a primary infection in individuals with no evidence of prior infection by serum antibody levels or as reactivation in individuals previously infected. The impact of cytomegalovirus has been mitigated somewhat by aggressive monitoring of early cytomegalovirus disease using molecular detection techniques (polymerase chain reaction titers) and by use of prophylaxis (acyclovir) or treatment with ganciclovir, foscarnet, and galganciclovir. RSV and human metapneumovirus are common causes of severe and prolonged lower respiratory disease in infants and children during cancer therapy and may occur as co-pathogens in association with other respiratory infections, particularly in the bone marrow transplantation patients. Palivizumab administration for patients at risk of RSV infection helps to reduce the risk of severe illness from RSV. Adenovirus is another common respiratory viral pathogen that can cause disseminated disease with lung, liver, and gastrointestinal involvement. Cidofovir or CMX001, a new orally bioavailable lipid conjugate of cidofovir, can be used for treatment.12,16 P. jirovecii (formerly carinii, now reserved for a similar infection in rats) is one of the fungal pathogens that can still be seen in this population if there is poor adherence with prophylaxis and can lead to a severe infection with hypoxemia and respiratory failure. Other invasive fungal organisms commonly seen in pediatric cancer patients include Aspergillus species (primarily A. fumigatus), Mucor, and Rhizopus. These organisms can cause severe tracheobronchitis, pneumonia, and, in the case of Mucor, devastating direct invasion of pulmonary blood vessels. Diagnosis is usually made by direct examination of tissue obtained by open or needle biopsy. Bronchoscopy/bronchoalveolar lavage (BAL) has historically had a low yield in fungal infections, particularly early in the infection, but the use of galactomannan detection in serum and BAL may be helpful in improving the early detection of Aspergillus.17 Because of the difficulty in making a tissue diagnosis, patients often receive prolonged empiric antifungal therapy with oral antifungal agents, such as itraconazole, voriconazole, micafungin, or posaconazole, with amphotericin B reserved for refractory cases. Surgical intervention with removal of pulmonary nodules is sometimes sought when the pulmonary fungal disease is not disseminated.18 Candida species are common colonizers of the respiratory tract in cancer patients (particularly with prolonged antibiotic therapy), but primary Candida pneumonia is relatively uncommon. Invasive sepsis and secondary pulmonary involvement is more likely, typically with C. albicans and C. tropicalis. In endemic areas, fungal organisms, such as histoplasmosis, can also be an important cause of pulmonary involvement in immunocompromised hosts or cause disseminated disease in normal infants. Many patients with histoplasmosis are also referred to St Jude because of concern about hilar adenopathy to rule out malignancy.19
Acute Lung Injury From Chemotherapy and Radiation
Patients receiving radiation involving the lung are likely candidates for acute radiation-induced changes in lung function. Frank radiation pneumonitis occurs in up to 15% of older patients. With modern radiation dosing and targeting, these acute pulmonary effects of radiation are usually minimized and are able to be managed with supportive care, although radiation remains a significant risk factor for late pulmonary effects.20 Several chemotherapy agents (eg, bleomycin and dactinomycin) increase the risk of radiation pneumonia. Adriamycin and actinomycin D are also associated with a recall pneumonia when administered after radiation. Radiation injury may also affect subsequent lung and chest-wall growth.21
Later Cardiopulmonary Complications of Cancer Therapy
Both radiation and chemotherapy are well described as causing both cardiac and pulmonary late effects (Table 2).22–24 For cardiac effects, the classic agents associated with late cardiac toxicity are the anthracyclines in a dose-dependent manner. A number of chemotherapy agents are associated with late pulmonary injury, including bleomycin, carmustine, lormustine, busulfan, and cyclophosphamide.
Lung and Cardiac Function in Adult Childhood Cancer Survivors
With improved therapy for childhood malignancy, there is a growing population of childhood cancer survivors, and a significant literature now exists on late effects of childhood cancer therapy. Evidence-based guidelines for risk-based assessment of cardiac and pulmonary outcomes in childhood cancer survivors have been developed.25 Studies in this population of survivors from St Jude Children's Research Hospital (St Jude Lifetime Cohort) have documented a wide range of systemic complications, but pulmonary complications are among the most common.26,27 For the most common childhood cancer, acute lymphocytic leukemia, the pulmonary outcomes are excellent, since therapy does not generally include drugs with significant pulmonary toxicity.28 In the past, therapy for acute lymphocytic leukemia included craniospinal irradiation, and since a portion of the lung is included, some effect on the lung function can be seen.29 Other therapies that include chemotherapy agents that affect the lung or radiation are associated with significant late lung function abnormalities. A recent report of 606 childhood cancer survivors from St Jude (median elapsed time from diagnosis of 21.9 y) described pulmonary function abnormalities (49% with FEV1 and 45% with FVC lower than the lower limit of normal) with 32% having restrictive lung defects (total lung capacity < 75% predicted).27 Risk factors for reduced lung function include the volume of lung receiving radiation and elapsed time from diagnosis, as well as age at diagnosis (for FVC and total lung capacity). Abnormal pulmonary function tests were also associated with decreased 6-min walk distance.27 As noted, cardiac function is affected by a variety of agents, primarily anthracyclines, that affect cardiac function. Impaired cardiac function can be a significant problem during cancer therapy, particularly when patients have secondary problems with fluid overload, acute infections, ARDS, or respiratory or multisystem failure. Although the major problem with anthracyclines is cardiomyopathy, findings on cardiac echocardiogram suggestive of pulmonary hypertension have also been identified in adult childhood cancer survivors, although the exact mechanism for these changes is still being elucidated.30
Exercise
The effects of cancer therapy on reducing exercise performance are well described in the St Jude Lifetime Cohort study and other follow-up studies.25,27 The effects of therapy on exercise performance are multifactorial and include reduced cardiopulmonary function, as well as decreased muscle strength. Generalized muscle weakness is common during cancer therapy, but how this relates to acute pulmonary complications, such as pneumonia or atelectasis, has not been studied. Some evidence suggests that exercise interventions can improve cardiopulmonary fitness, strength and flexibility, and physical function in this population.31 In cancer survivors, 18.1% reported deficits in physical performance, whereas 10.5% reported deficits in emotional health, both significantly impacting health-related quality of life.32
Pulmonary Complications of Hematopoietic Stem Cell Transplantation
Hematopoietic stem cell transplantation is a well-established treatment now for both refractory malignant and non-malignant conditions. There is extensive literature on the diverse pulmonary complications in hematopoietic stem cell transplantation that occur during acute therapy or appear later in hematopoietic stem cell transplantation survivors. Patients undergoing hematopoietic stem cell transplantation have generally been pretreated with chemotherapy or radiation for their primary disease as well as receiving immunosuppressive regimens to prepare for the new bone marrow cells. These treatments lead to injury to the lung and altered lung function before transplant. The increased risk associated with low lung function before hematopoietic stem cell transplantation was demonstrated by Srinivasan et al,33 who examined lung function in a large cohort of 410 hematopoietic stem cell transplantation subjects. In this study, 42% developed pulmonary complications, and lower lung function at baseline before transplant significantly increased the risk of transplant. T cell depletion of the hematopoietic stem cell transplantation graft, acute grade 3–4 graft-versus-host disease and chronic graft-versus-host disease increased the risk of pulmonary complications, and pulmonary complications resulted in a 2.8-fold increased risk of mortality.33
Infectious Complications
Pulmonary infections remain a major cause of morbidity and mortality in hematopoietic stem cell transplantation patients, and the types of infectious complications are usefully classified according to the time period after transplant (Table 1).4
Non-Infectious Complications
Non-infectious complications in hematopoietic stem cell transplantation include severe mucositis and pulmonary edema (due to fluid overload, cardiac compromise, or renal toxicity) (Fig. 2). Engraftment initially can be associated with pulmonary edema and may lead to hypoxia secondary to inflammatory response. Idiopathic pneumonia syndrome is characterized by dyspnea; cough; hypoxemia, with no detection of infectious agent on BAL; and restrictive pulmonary function changes.34 Immunosuppressive therapy with corticosteroids and anti-tumor necrosis factor agents like etanercept are usually mainstays of treatment. Diffuse alveolar hemorrhage (DAH) is a less common complication but historically has had a very poor prognosis.35 If DAH is associated with multi-organ failure, mortality can be > 80%. The diagnosis of DAH is generally made by BAL with increasingly hemorrhagic lavage with each BAL aliquot and exclusion of underlying pulmonary infection as a cause of hemorrhage. Treatment includes using corticosteroids, correction of coagulopathy and thrombocytopenia, administration of aminocaproic acid to stabilize clots, and judicious use of higher PEEP on mechanical ventilation. Pulmonary and hepatic veno-occlusive disease presents with pulmonary hypertension, signs of right-heart failure, and pulmonary infiltrates. It is characterized by obstruction of the small veins and venules secondary to fibrosis.36 Obliterative bronchiolitis is the most common late pulmonary complication of hematopoietic stem cell transplantation and is usually associated with other manifestations of graft-versus-host disease, such as skin or hepatic involvement. Post-transplant infections, including viral infections (like adenoviral infection) may increase the risk of obliterative bronchiolitis. The diagnosis of obliterative bronchiolitis is now usually made by a combination of fixed airway obstruction (for patients old enough to do pulmonary function tests) and high-resolution CT with inspiratory/expiratory views showing areas of patchy hyperinflation, alternating with areas of increased density (so-called mosaic pattern).37,38 Cryptogenic organizing pneumonia, formerly known as BOOP (bronchiolitis obliterans-organizing pneumonia) is another late complication of hematopoietic stem cell transplantation and differs from obliterative bronchiolitis in biopsy appearance (with areas of organizing pneumonia, as the name suggests) with features of obliterative bronchiolitis, restrictive lung defects rather than obstructive, and improvement with systemic corticosteroids.39 Pulmonary alveolar proteinosis has been described as a secondary complication in some childhood cancers and in hematopoietic stem cell transplantation due to accumulation of surfactant protein in the alveolar space secondary to defective macrophages. The diagnosis is usually made by BAL showing the typical milky white fluid of alveolar proteinosis.40 Treatment usually involves the use of granulocyte-macrophage colony-stimulating factor.
Respiratory Therapy Treatments and Interventions in the Childhood Cancer Patient
Aerosol Therapies
The obstructive lung disease of obliterative bronchiolitis is usually not associated with reversible airway obstruction, but some patients do appear to have a reversible component that may benefit from bronchodilator therapy. The use of so-called FAM therapy (high dose inhaled fluticasone-azithromycin-montelukast) for obliterative bronchiolitis has been reported to improve the need for corticosteroids in some small series, but more research is needed, particularly in the pediatric hematopoietic stem cell transplantation population.41 In patients with DAH, administration of activated factor VII has been used in pediatric patients by direct instillation or aerosolization, but more studies are needed because of the potential for adverse events like endotracheal tube obstruction by clot.42 The use of inhaled ribavirin in immunocompromised hosts with RSV is controversial, although the greatest benefit appears to be when started early in the course of lower respiratory infection; whether it has a role in the intubated patient with severe RSV pneumonia is unclear.13,43 Human metapneumovirus also demonstrates in vitro susceptibility to ribavirin.
Noninvasive Ventilation
Noninvasive ventilation with CPAP and bi-level positive airway pressure has generally replaced negative-pressure ventilation for supportive care of pediatric cancer patients, and many patients have successfully avoided intubation with early use of noninvasive ventilation.44 High-frequency chest oscillation using an external chest shell (Hayek oscillator) is the modern equivalent of the 1970s iron lung and has been used with success in other acute respiratory failure settings, although there is insufficient literature to recommend it at this time.45 For cancer patients with increased risk of infection, less invasive approaches may decrease the risk of ventilator-associated pneumonia.
Positive-Pressure Ventilation
Intubation and positive-pressure ventilation is the procedure of choice for ventilatory support in this population. Current approaches to mechanical ventilation include a protective lung strategy with the use of low tidal volumes and permissive hypercapnia, prone positioning, conservative fluid management, and new modes of conventional ventilation that have improved outcomes in most ICUs providing care for cancer patients. A recent report from our institution emphasized the importance of a team approach, particularly for hematopoietic stem cell transplantation patients, for patients who require mechanical ventilation (Fig. 3).46 Other techniques, such as high-frequency oscillation, are usually used as a salvage therapy, and whether the earlier initiation of high-frequency ventilation will improve outcome is still unknown.
Tracheostomy
As children with cancer survive their complicated acute pulmonary complications, the question of whether to perform a tracheostomy may become an important issue. Tracheostomy does not appear to carry a significantly higher risk than in other populations, although children have higher risk of infection with the common bacterial organisms, such as Pseudomonas, that colonize most tracheostomy patients secondary to their immunosuppressed status.
Prolonged Mechanical Ventilation
Pediatric cancer patients who need support due to severe lung injury or central hypoventilation (brain tumor patients) are candidates for long-term mechanical ventilation, and the same considerations for preparing these patients and families apply as for other populations.47 For some patients with poor prognosis from their underlying malignancy, home mechanical ventilation may have different goals for the patient and family, such as allowing a return to home for a period of time in palliative mode.
Lung Transplantation
For patients with end-stage lung disease due to pulmonary fibrosis or obliterative bronchiolitis or after severe infections, the question of possible lung transplantation may be raised. Because of the shortage of lung donors, patients must generally be in remission. Varying time period requirements that a patient must be in remission before lung transplantation would be considered differ among different lung transplant centers, with the minimum period being 2 years. One review of lung transplantation in pediatric hematopoietic stem cell transplantation patients with end-stage obliterative bronchiolitis reported that the outcome of patients post-lung transplant was comparable with other transplant recipient populations.48
Discussion
Sweet:
Thanks, that was a great overview. I don't know this field very well, but with obliterative bronchiolitis after a stem cell transplant, have people looked for the same risk factors that we're starting to see with lung transplant, such as autoantibodies and human leukocyte antigen antibodies?
Stokes:
Some similar work has been identified in the hematopoietic stem cell transplantation (HSCT) population, but I'm not very familiar with that literature.1 We tend to see them later when they present. FAM therapy—fluticasone, azithromycin, and montelukast—is the newest therapy for which some data are available but with limited pediatric experience.2 There is so much work that needs to be done in this area.
Panitch:
I took care of a young man from childhood through early adulthood who had had a Wilms' tumor. He had a tiny chest with normal lungs by CT imaging and an FVC of about 40% predicted. I noticed in your talk that chest-wall growth abnormalities have been identified as a long-term issue, so I was wondering if anyone is thinking about expansion surgeries, especially for children who receive their damaging therapies early in life, and you can follow their chest-wall growth? Is there a role for thoracic expansion surgery to improve lung function in those patients?
Stokes:
I'm not aware of anybody doing that, but it's a really interesting question. Identifying this group early enough to make an intervention worthwhile would be essential to that. Again, I think some of the worst abnormalities of the chest wall were related to older regimens of radiation therapies. I think they've become better at targeting where they need to target, so we don't see the gross chest-wall deformities that we used to see back in the ‘80s. What to do with that group, I'm not sure anybody has thought about using chest expansion.
Panitch:
Just as a follow-up, have you looked at or measured chest-wall compliance in the children who have required repeated thoracic interventions?
Stokes:
Do you know how to do that, Howard? Because nobody remembers anymore how to drop esophageal balloons and measure chest-wall compliance except for us old-timers. We actually have some data from the St Jude LIFE study3 that Dr Kirsten Ness has collected looking at restriction in chest-wall expansion and relating it to functional changes. There just isn't that much about the mechanics of the chest wall. When oncologists see a restrictive defect, they see the lung and assume it's pulmonary fibrosis. But I think one of the things we bring to this area is thinking more broadly about diaphragm, chest-wall, and respiratory muscles that could influence restrictive change, not just fibrosis. I'll also mention that we have a new special interest group on pulmonary complications of childhood cancer therapy in the American Thoracic Society, and if you or anybody in your group is interested in pulmonary complications and late effects, we would love to have more members who have an interest in this area. I think it's a great area for research for the next 20 years.
Smallwood:
This might be perhaps stretching too far into the future, but I recall a group of researchers at Northeastern University who published a study4 in Science Translational Medicine. What was interesting to me was the intervention in an animal model of an oncologic process was supplying 40-60% O2. They demonstrated in their cohort of mice that they could awaken anti-tumor T lymphocyte cells just by using inhaled O2. Essentially, breathing in high concentrations of oxygen helped to fight cancer. Although it's translational at this point, do you have any thoughts?
Stokes:
Unfortunately, we do a lot of O2 therapy, but it's usually for the wrong reasons in the ICU and not for its potential antitumor effects. Ira [Cheifetz], I know the PALISI (Pediatric Acute Lung Injury and Sepsis Investigator) group has looked at bone marrow transplant and done a lot of nice work trying to tease out the more acute complications and how to manage that better.
Cheifetz:
You provided an excellent summary, thank you. Most of the work by the stem cell transplant subgroup of the PALISI Network5 has investigated acute complications; we have not done as much work on long-term effects.6 The one thing I would like to mention, and I am sorry if I missed it, is diffuse alveolar hemorrhage. This is one of the important clinical entities that we still wrestle with in the ICU. We saw one child recently who, out of nowhere, had a catastrophic pulmonary bleed. As much as we have tried over the years to figure out the right recipe of therapies, both ventilator and non-ventilatory, we still struggle with this population. Kyle [Rehder] has quite a bit of experience with this population, and I would be curious to hear his thoughts as well.
Stokes:
I did include DAH in the paper, but one of the therapies that respiratory therapists might be involved with is use of nebulized factor VII. I don't know if you have tried that in any cases? There are limited case series in pediatrics, and we've been asked a number of times to directly instill factor VII in patients with diffuse alveolar hemorrhage at bronchoscopy. It's definitely an area where the outcome is terrible, and that's about all you can say. It tends to come in runs, and we get these runs of DAH, and when we ask the transplant people what they're doing differently, they don't know. But it has a bad outcome, particularly if it's associated with an underlying infection.
Rehder:
That was my sense as well from our population. We have tried nebulized factor VII, but with concerns that we would form a large obstructive clot in the airways and then have a new issue. The patient tolerated the nebulized factor VII, but as you stated, it was a rescue therapy in a patient with a very poor prognosis, and the patient did not ultimately survive. And we've tried systemic factor VII as well in the setting of alveolar hemorrhage, but it seems like it comes down to what is the underlying reason for the bleed, and if it's infection, it often feels like we can't get on top of it.
Stokes:
The reference in the paper is one of the most recent case reports. You probably saw it, and it was exactly like you said, a pediatric patient who they gave factor VII to and developed a clot in the endotracheal tube and acutely decompensated, and the tube had to be taken out emergently. That's something people need to be aware of, that this is not a therapy without potential downsides.
Footnotes
- Correspondence: Dennis C Stokes MD MPH, University of Tennessee Health Science Center College of Medicine, 49 North Dunlap, Memphis, TN 38103. E-mail: dstokes4{at}uthsc.edu.
The authors have disclosed no conflicts of interest.
Dr Stokes presented a version of this paper at the 55th Respiratory Care Journal Conference, “Pediatric Respiratory Care,” held June 10-11, 2016, in St Petersburg, Florida.
- Copyright © 2017 by Daedalus Enterprises