Pulmonary Rehabilitation for Bronchiolitis Obliterans Syndrome after Hematopoietic Stem Cell Transplantation

Abstract

Bronchiolitis obliterans syndrome (BOS) is a progressive, insidious lung disease affecting allogeneic hematopoietic stem cell transplant (HSCT) recipients. Unfortunately, there is no standardized approach for treatment of BOS in post HSCT patients. Pulmonary rehabilitation is a standard treatment in emphysema, an irreversible obstructive lung disease secondary to tobacco abuse. The National Emphysema Treatment Trial (NETT) demonstrated improved exercise tolerance, decrease dyspnea, and increase of quality of life in patients with severe emphysema after pulmonary rehabilitation. We hypothesized that pulmonary rehabilitation may benefit patients with BOS. Patients with BOS were identified retrospectively from January 2005 to the present. Patients who enrolled in pulmonary rehabilitation were included in the study. We obtained summaries via chart review of each patient’s progress after pulmonary rehabilitation enrollment from their respective rehabilitation centers. Six minute walk distances, spirometry, and pulmonary symptoms were compared before and after the completion of pulmonary rehabilitation. We identified 11 patients with BOS documented from their pulmonologist’s clinical notes that were enrolled into pulmonary rehabilitation. Ten of the 11 patients completed pulmonary rehabilitation. All patients had improvement in their 6 minute walk distances after the completion of pulmonary rehabilitation with an average improvement in distance of 307 feet (p value = 0.005). Six of the 10 patients completed a Short Form-36 questionnaires prior to and after rehab. There was a significant improvement in the physical functioning score (p value =0.029). Pulmonary rehabilitation appears to improve 6 minute walk distance, subjective symptoms of dyspnea and exercise tolerance in patients with BOS. This may be an important adjunctive therapy for a debilitating disease with limited treatment options.

Introduction

Bronchiolitis obliterans syndrome (BOS) is the most common and serious form of chronic pulmonary graft versus host disease (GVHD). The prevalence varies depending on the definition used but was recently estimated at 5.5% using a modified NIH definition [1]. The disease negatively and substantially contributes to morbidity and mortality [2]. Patients typically develop symptoms within 18 months of transplantation and report symptoms of dyspnea, cough and wheezing [2, 3]. Diagnosis can be established clinically based on symptoms, new fixed airflow obstruction and air trapping on imaging in a patient usually with other evidence of chronic graft versus host disease [3, 4]. Unfortunately, the disease remains difficult to manage and mortality remains high even with attempts of increasing immunosuppression. Other pharmacologic and non-pharmacologic therapies need to be investigated in this patient population including pulmonary rehabilitation.

Pulmonary rehabilitation is a standard treatment in emphysema, an irreversible obstructive lung disease secondary to tobacco abuse. The National Emphysema Treatment Trial (NETT) and prior randomized controlled trials demonstrated improvement in exercise tolerance and quality of life along with decreased dyspnea in COPD patients [5–8]. Recent COPD guidelines recommend pulmonary rehab in all symptomatic patients with COPD and FEV1 < 50% [9]. The NIH consensus statement on chronic GVHD recommends pulmonary rehabilitation in patients with BOS based on expert opinion but there are no studies looking at this treatment [10]. Evaluating the effect of a standardized exercise program on patients identified to have presence of BOS could potentially offer a treatment that improves exercise tolerance and quality of life in a disease with few options.

Materials and Methods

Patient Identification

Patients who were being seen at the outpatient pulmonary clinic were identified retrospectively from January 2005 to the present if they had bronchiolitis obliterans after allogeneic SCT and had been enrolled in pulmonary rehabilitation. All patients diagnosed with bronchiolitis obliterans syndrome fit the NIH criteria which included an FEV1/FVC ratio <70% and an FEV1<75% and either evidence of air trapping by CT or elevated RV (>120%)[11].Data were collected from an institutional database and from retrospective chart review. The study was approved by The Ohio State University Institutional Review Board.

Pulmonary Rehabilitation

The majority of patients completed rehabilitation at our institution but four patients completed the program at an outside institution due to the proximity to their homes. All patients participating in our program completed 24 sessions of rehabilitation over an 8 week duration. Exercise classes covered nutrition, medication and oxygen safety, pursed lip breathing and other breathing techniques, use and care of metered dose inhaler. Exercise included upper and lower body strength training using a combination of free weights and weight machines. Cardiovascular exercise included the recumbent bike, treadmill or step machine and the upper body bike. Individuals usually started the program with using the treadmill for 10 minutes, bike for 8 minutes and upper body bike for 6 minutes. The initial exercise prescription was guided by cardiopulmonary exercise testing or the 6MW testing and then speed, incline, or resistance were gradually increased as tolerated individually with a goal of 45 minutes of cardiovascular exercise per session. Strength training was usually about 20 min per session but varied from person to person. Of the 11 patients enrolled in the study, 4 performed their rehabilitation at 3 outside institutions. Outside programs were reviewed and were not found to be significantly different in content than our institution’s program. The duration of outside programs ranged from 8–12 weeks, and the 4 patients completed from 21 – 36 sessions.

Data Collection

We reviewed spirometry/pulmonary function tests, 6 minute walk tests, and medical records from the Information Warehouse at the Ohio State University (our computerized record storehouse). The walk tests and SF-36 scores were assessed during the patient’s rehabilitation entrance and exit visits. The timing of spirometry varied, however the most recent spirometry prior to rehabilitation and after treatment portion of rehabilitation was used for comparison. Symptoms of dyspnea, cough, wheezing and fatigue were retrospectively evaluated by documentation of subjective symptoms at their most recent pulmonary clinic visit prior to pulmonary rehabilitation and the next visit after the completion of the treatment portion of pulmonary rehabilitation.

Statistical Analysis

Pre- and post-rehab six minute walk distances and SF-36 subscores were compared using a paired t-test. A p-value of <0.05 was considered significant. All analyses are run using Stata 10.2 or 11.1, Stata Corporation, College Station, Texas.

Results

Eleven patients were included in the study, however only 10 patients completed the rehabilitation program. One patient did not complete the program due to chronic pain issues. Patient demographics are noted in Table 1. The average age was 48 years old and a majority of the patients were male. Three of the patients had a significant smoking history and 2 of the three had a documented history of chronic obstructive pulmonary disease. Only one patient had a severely depressed ejection fraction on echocardiogram.

Table 1.

Demographics and Transplantation Information

Subject number	Age	Gender	malignancy	conditioning regimen	ablation	transplant cell source	donor	HLA matching	time to diagnosis of BOS (days)	acute GVHD grade	Chronic GVHD	Pre- transplant FEV1 Liters (% predicted)	Post- transplant FEV1 Liters (% predicted)	Drop in FEV1 (%)
1	38	F	AML	Cytoxan, TBI	yes	blood	sibling	NA	~10000	NA	no	NA	0.55 (24)
2	51	M	CLL	fludarabine busulfan and ATG	no	blood	unrelated	10	392	0	severe	2.52 (67)	1.86 (51)	26
3	63	F	transformed follicular lymphoma	Fludarabine, Busulfan, ATG	no	blood	unrelated	9	427	0	severe	2.82 (111)	1.3 (52)	54
4	53	M	mantle cell lymphoma	Busulfan and Cytoxan	yes	blood	sibling	10	315	2	severe	2.74 (74)	1.1 (30)	60
5	47	M	NK/T cell NHL	fludarabine and busulfan	no	blood	sibling	10	303	2	severe	3.67 (87)	2.38 (58)	35
6	58	M	HD	busulfan, fludarabine and ATG	no	blood	unrelated	9	1163	4	severe	2.96 (94)	1.8 (53)	39
7	27	M	ALL	Cytoxan, TBI	yes	blood	unrelated	10	1335	4	severe	4.75 (100)	2.39 (49)	50
8	40	M	AML	fludarabine, busulfan and ATG	no	blood	unrelated	9	448	1	severe	2.71 (81)	1.05 (32)	61
9	52	F	diffuse large B cell lymphoma	busulfan and cyclophosphamide	yes	blood	sibling	10	1242	3	severe	3.62 (124)	0.91 (33)	75
10	60	M	AML	fludarabine and busulfan	no	blood	sibling	10	388	2	severe	2.9 (77)	2.08 (56)	28
11	38	M	AML	Cytoxan and TBI	yes	blood	sibling	10	218	2	severe	3.07 (76)	2.77 (69)	10

Abbreviations: FEV1 = forced expiratory volume in 1 second, AML = Acute Myelocytic Leukemia, ALL = Acute Lymphoblastic Leukemia, NHL= Non Hodgkin’s Lymphoma, HD = Hodgkin’s disease, TBI = total body irradiation, ATG = anti-thymocyte globulin.

The most common reason for transplantation in the population studied was acute myelocytic leukemia (36%). Transplant characteristics and GVHD data are shown in Table 1. The most common conditioning regimen was Fludarabine/Busulfan with or without ATG (54.5%). Ten of the 11 transplants were peripheral blood stem cell transplants and 6 were sibling matched donors whereas 5 were unrelated. HLA typing was done by high resolution DNA typing for HLA-A, -B, -C, -DRB1, -DQB1. Three of the eleven patients had a 9/10 HLA match, and the rest were 10/10.

Eight of the 11 patients had evidence of acute GVHD, the information on patient number 1 was unavailable. Ten of the 11 patients had severe chronic GVHD as assessed by the NIH Consensus criteria [11]. The other patient did not have any organ involvement besides pulmonary disease. There was a wide range of days to diagnosis of BOS after transplant. All patients with pre-transplant pulmonary function tests demonstrated a drop of at least 10% in FEV1 after transplant, and most had a drop of >25%. Patient number 1 did not have pre-transplant PFTs because she was transplanted at age 7.

Pulmonary function test values and 6 minute walk distances for each patient before and after rehab are shown in Figure 1. Summary data can be seen in Table 2. There was no significant change in spirometry when comparing pre and post rehab (p=0.446 for FEV1, and p=.822 for FVC, Table 2). All patients who completed pulmonary rehabilitation improved their 6 minute walk distance (p=0.005). Six minute walk distance increased an average of 307 feet post rehab. Short Form (SF)-36 surveys were performed prior to and after rehabilitation in 5 of the 10 patients. There was a significant improvement in the physical functioning score by a mean of 14.4 points (p value =0.029).

Figure 1.

A. FEV1 (liters) for each subject pre-transplant, pre-rehabilitation and post-rehabilitation. B. FVC (liters) for each patient pre-transplant, pre-rehab and post-rehab. C. FEV1/FVC ratio (percentage) for each patient pre-transplant, pre-rehabilitation and post-rehabilitation. D. Six-minute walk distance (feet) pre-rehabilitation and post-rehabilitation.

Table 2.

Summary outcomes for SF-36 domains, six-minute walk distance, FEV1 and FVC.

Clinical outcome	Mean (SD)		p-value1
	Pre rehab	Post rehab
FEV1	1.48 (0.65)	1.37 (0.79)	0.446
FVC	2.51 (1.02)	2.54 (0.91)	0.822
Six – minute walk, ft	1291 (373)	1598 (333)	0.005
SF-36 Domains
Physical functioning	43.9 (15.3)	56.7 (16.8)	0.029
Role functioning – physical	18.2 (21.0)	20.9 (24.7)	0.180
Role functioning – emotional	77.4 (25.7)	77.9 (20.7)	0.961
Bodily pain	63.2 (18.5)	58.4 (22.3)	0.713
General health	42.4 (9.0)	41.2 (12.8)	0.742
Vitality	37.3 (20.3)	53.5 (7.4)	0.071
Social functioning	57.3 (16.0)	59.8 (12.1)	0.688
Mental health	74.2 (17.1)	75.1 (13.1)	0.784
Reported health transition	2.3 (1.4)	2.5 (1.05)	0.771

¹p-value based on a paired t-test

Discussion

This study demonstrates a marked improvement in objective exercise tolerance, subjective symptoms and quality of life. There was improvement in 6 minute walk distance after pulmonary rehabilitation. Patients improved their 6MW distance by an average of 307 ft which is well above the 76 ft improvement seen in the NETT [5]. The majority of patients experienced less dyspnea based on clinic notes and half of the patients had subjective improvement in their exercise tolerance post-rehabilitation. The SF-36 physical functioning score improved in the patients who completed the form prior to and after rehab, although there were not significant changes in the other SF-36 domains. There was little change on spirometry following rehab which is consistent with the findings in the previous rehabilitation trials in COPD [5, 6].

The rationale behind the positive effects seen in our patients can be extrapolated from other patients with chronic lung diseases. Pulmonary rehabilitation has been evaluated predominantly in patients with COPD. Pulmonary rehabilitation typically involves weeks of supervised exercise, strength training, education, nutritional assessment and psychosocial evaluation [12]. Studies have demonstrated that pulmonary rehabilitation increases exercise capacity, improves quality of life and decreases dyspnea in patients with COPD [5–8]. It is thought that the exercise limitation seen in COPD is due to multiple factors including: hypoxemia, hyperinflation, cardiac dysfunction and potentially respiratory muscle weakness [8, 12]. Skeletal muscle weakness also plays a role in exercise intolerance and has been attributed to underlying inflammation, frequent use of corticosteroids and deconditioning among other factors [8, 12]. By addressing these factors along with the other systemic manifestations of COPD, pulmonary rehabilitation has improved important endpoints in COPD patients. A recent guideline update from the American College of Physicians (ACP), American College of Chest Physicians (ACCP), American Thoracic Society (ATS) and European Respiratory Society (ERS) recommends pulmonary rehabilitation in patients with an FEV1<50% but can also be considered for patients above that threshold [9].

Like COPD, BOS is a chronic pulmonary condition characterized by irreversible air flow obstruction and is associated with hyperinflation [2, 3]. Also similar to COPD patients, it has been shown that patients undergoing stem cell transplantation in general suffer from skeletal and respiratory muscle weakness. White et al, noted that maximal inspiratory muscle strength, maximal expiratory muscle strength and grip strength were decreased in a significant number of patients even before undergoing transplantation [13]. These patients also were noted to have decreased exercise tolerance prior to transplantation as evidenced by a decreased 6MW distance [13]. While the etiology for this is not specifically known, it has been attributed to the patient’s underlying disease and/or treatment [13, 14]. At the same institution, it was demonstrated that respiratory and skeletal muscle weakness continues to be a significant issue following transplantation [14].

Given these findings, exercise therapy has been evaluated in SCT patients around the time of transplant with evidence of benefit. Studies have demonstrated improved quality of life [15, 16], muscle strength [15], physical fitness [15–18] and fatigue measures [16, 18, 19] with exercise around the time around transplant. There is also one small retrospective study of patients that showed an association between the number of steps a patient could take with the duration of hospitalization [20]. However, there are no data to our knowledge evaluating pulmonary rehabilitation in patients that develop BOS post-transplantation until now.

There are limitations of our study to note. First, it is a small retrospective study and comparison to a control population was not feasible. There are also limitations as the clinic visits were not at standardized times and the time from rehabilitation to the pre- and post-visits varied. Ideally, patients would be seen at a specific interval before and after rehabilitation for measurement of physiologic measures and symptoms. Another issue was that we did not have one center providing the rehabilitation which limited some of the analysis. All of the centers were accredited, reputable and utilized the basic multidisciplinary approach; however, homogeneity would have provided easier comparison of the testing pre and post rehabilitation. This also limited our ability to analyze quality of life surveys, since different centers had differing preferences for which survey they utilized. The data used to compare quality of life was obtained using the SF-36 survey which has not been validated in this specific patient population so may not be an accurate reflection of quality of life domains. Larger controlled studies would ideally be performed to confirm our findings but this may be difficult given the low prevalence of this disease.

Despite these limitations, the improvement in aspects of quality of life and walk distance in all 10 patients are important outcomes in this patient population. With few treatment options available for patients with bronchiolitis obliterans, our study demonstrates improvement in symptoms and exercise capacity from a historically low risk treatment. Pulmonary rehabilitation should be strongly considered as an adjunct treatment for patients who develop bronchiolitis obliterans which is compatible NIH Consensus recommendations for supportive care of chronic GVHD [10].