Spontaneous Pneumothorax and Air Travel in Pulmonary Langerhans Cell Histiocytosis: A Patient Survey

Abstract

Background

The optimal approach for management of spontaneous pneumothoraces (SPs) and the safety of air travel in patients with pulmonary Langerhans cell histiocytosis (PLCH) are not well established.

Methods

Patients with PLCH were recruited from the Rare Lung Diseases Clinic Network and the Histiocytosis Association, and surveyed about disease manifestations and safety of air travel.

Results

A total of 94 patients completed the survey. Median age at diagnosis of PLCH was 40 years (range: 15–67 years). Average interval between symptom onset and diagnosis was 2.9 years (range: −4 to 31 years). Twenty-two patients (23%) had at least one SP, of which 14 (64%) had at least one additional SP that showed either an ipsilateral recurrence (10 patients; 45%) or a contralateral recurrence (8 patients; 36%). Mean age at the time of first SP was 29 years. SP was the presenting manifestation that led to the diagnosis of PLCH in 19% of patients, typically after the second episode. Surgical pleurodesis reduced the recurrence rate of SP by half in comparison with conservative management (29% vs. 65%, p = 0.025). Two patients experienced an episode of SP during air travel, consistent with an air travel-related pneumothorax rate of 2.4% per patient and 0.27% per flight.

Conclusions

SP is a common manifestation of PLCH, can be seen in approximately one-fourth of the patients, and has a high recurrence risk. Surgical pleurodesis leads to a substantial reduction in the SP recurrence risk. The risk of an air travel-related SP in patients with PLCH is about 2–3 per thousand flights.

Trial Registry

clinicaltrials.gov:

1. Introduction

Pulmonary Langerhans cell histiocytosis (PLCH) is a rare, diffuse cystic lung disease (DCLD) that primarily affects young to middle-aged patients, and shows a strong association with exposure to cigarette smoke [1, 2]. Patients usually present with nonspecific symptoms such as dyspnea, cough, and/or constitutional features such as asthenia, fever, night sweats, and weight loss, but many cases are discovered incidentally on imaging in asymptomatic individuals [1–3]. Spontaneous pneumothorax (SP) is another common manifestation and has been reported in 10%−20% of patients with PLCH [2–7]. Patients who develop SP are at a high risk of recurrence, with a reported recurrence rate of up to 60% in one study [4]. However, the optimal approach for management of pneumothoraces to prevent recurrent episodes in patients with PLCH is not well established. Moreover, patients with DCLDs may show an increased risk of development of SP during air travel due to cyst expansion associated with atmospheric pressure fluxes [8–10]. However, the risk of air travel-related pneumothorax in patients with PLCH is unknown. We conducted this study to evaluate the risk of SP, determine the optimal approach for management of pneumothorax, and better understand the safety of air travel in patients with PLCH.

2. Methods

2.1. Patient Recruitment and Survey Design

After obtaining approval for the study protocol from the National Institutes of Health (NIH) and the University of Cincinnati Institutional Review Board (IRB number 2015–6129), a survey consisting of 62 questions was sent out to patients with PLCH. These patients were recruited from the Rare Lung Diseases Clinic Network, a geographically distributed consortium of clinics that was initially organized by the Lymphangioleiomyomatosis (LAM) Foundation and adopted for the study of other rare lung diseases by the Rare Lung Diseases Consortium (RLDC). A link to the survey was also distributed to PLCH patients registered with the Histiocytosis Association and posted at clinicaltrials.gov (). Prior to distribution, the survey was independently tested by the authors and two medical and non-medical volunteers to ensure clarity and ease of understanding by a lay audience.

Patients were informed about the details of the study, and signed informed consent was obtained from those willing to participate. In a web-based survey, responders were asked to provide details regarding their demographics, disease manifestations, occurrence and management of pneumothoraces, and air travel experiences. Detailed history regarding any secondary cause of pneumothorax and the management of pneumothorax was obtained from those who had an episode of pneumothorax. For the purpose of our study, we considered all ipsilateral pneumothoraces reported within the same month as one episode to minimize the confusion between persistent and recurrent pneumothorax. In case more than one intervention (e.g., chest tube followed by chemical pleurodesis followed by surgical pleurodesis) was used to treat a pneumothorax, then the most invasive intervention (e.g., in this case, surgical pleurodesis) was listed as the treatment of record for that event. Because retrospective distinction between true flight-related pneumothorax and an unrelated SP occurring a few days after air travel can be problematic, we chose to limit our definition of flight-related pneumothorax to those occurring either during an air flight or within 24 hours after a flight.

2.2. Data Collection and Analysis

Study data were collected and managed using the Research Electronic Data Capture (REDCap) tools hosted at the University of Cincinnati. REDCap is a secure, web-based application designed to support data capture for research studies and provides audit trails for tracking data manipulation [11]. De-identified data were stored in a password-protected format at the Data Management and Coordinating Center at the University of South Florida as part of the NIH-supported RLDC.

Respondent characteristics and data are reported descriptively using frequencies, percentages, central tendency (mean or median), and range. Statistical significance was set at p < 0.05 and determined by the chi-square test. Data analysis was performed using SAS version 9.4 software (SAS Institute, Cary, NC).

3. Results

3.1. Baseline Characteristics

A total of 94 patients completed the survey (Table 1). The mean age of the patients at the time of completing this survey was 48 years (range: 24–74 years). Eighty-six (91%) responders had ever smoked, of which 67 (80%) were former smokers and 19 (20%) were active smokers at the time of survey. Among the patients who smoked (active and former), the median tobacco use was 20 pack-years (range: 1.5–175 pack-years). In addition, 23% of the responders reported smoking marijuana. In 72% of the patients (68 out of 92), the diagnosis of PLCH was confirmed by surgical lung biopsy, and in seven patients (8%), the diagnosis was established by biopsy of extrapulmonary tissue (bone in five and lymph node in two patients). In the remaining 20% of the patients, the diagnosis of PLCH was established on the basis of characteristic radiological features noted on chest high-resolution computed tomography (HRCT) with or without bronchoscopy. Average age of diagnosis of PLCH was 40 years (range: 15 to 67 years), and there was a delay of 2.9 years (range: −4 to 31 years) between symptom onset and PLCH diagnosis. Twenty-two (23%) patients were asymptomatic at the time of diagnosis.

Table 1:

Demographics and baseline characteristics of our cohort.

Variables	Data
Total respondents	94
Females	78 (79%)
Race
Caucasian	95%
African-American	3%
American-Indian	1%
Others	1%
Age at the time of survey - median (range)	47 (24–74) years
Age at diagnosis of PLCH - median (range)	40 (15–67) years
Mean delay between symptom onset* and diagnosis	2.9 (−4 to 31) years
Diagnosis
Surgical lung biopsy	72%
Biopsy of extrapulmonary tissue**	8%
CT scan of chest +/− bronchoscopy	20%
Disease manifestation of PLCH
Dyspnea on exertion	89%
Cough	72%
Chest pain	65%
Constitutional symptoms#	42%
Weight loss	32%
Diabetes insipidus	29%
Pulmonary hypertension	29%
Bone lesions or bone pain	26%
Lymphadenopathy	23%
Spontaneous pneumothorax	23%
Hemoptysis	8%
Smoking History
Ever smokers	86 (91%)
Active smokers at time of survey	19 (20%)
Tobacco use - median (range)	20 (1.5–175) pack-years

Abbreviations: PLCH = pulmonary Langerhans cell histiocytosis, CT = computed tomography

^*Symptoms related to PLCH mentioned under disease manifestations

^**bone in 5 and lymph node in 2

^#fever, sweats, and/or chills

3.2. Disease Manifestations

The most common disease manifestation reported by the PLCH patients in our cohort was dyspnea on exertion, which was present in 84 (89%) patients with a median severity of 1 (range: 0 to 4) on the modified Medical Research Council (mMRC) dyspnea scale. The next most common symptom experienced by the PLCH patients in our cohort was fatigue (67 out of 90, 74%), followed by cough (57 out of 79, 72%), chest pain (58 out of 90, 65%), constitutional symptoms such as fever, sweats, and/or chills (38 out of 90, 42%), and weight loss (29 out of 90, 32%). Other disease manifestations of PLCH reported by our cohort included diabetes insipidus (26 out of 90, 29%), pulmonary hypertension (26 out of 90, 29%), bone lesions or bone pain (23 out of 90, 26%), lymphadenopathy (21 out of 90, 23%), and hemoptysis (7 out of 90, 8%).

In our cohort 17% (16 out of 93) of the patients also had a diagnosis of asthma and 30% (28 out of 93) were diagnosed with chronic obstructive pulmonary disease (COPD). 60% (56 out of 94) of the responders were on inhalers, 22% (20 out of 93) reported being on supplemental oxygen, of which half were using it continuously and the other half were using it with exercise and/or sleep.

3.3. Pneumothorax

Twenty-five patients (27%) reported experiencing at least one episode of pneumothorax. Of these, three patients had a secondary cause for the pneumothorax (iatrogenic following a bronchoscopy in one patient and traumatic in two patients). These events were excluded from our analysis. Thus, 22 patients reporting at least one episode of SP (Table 2) were assessed in this study. On average, patients who had an SP experienced a total of 2.4 (range: 1 to >6) episodes of pneumothoraces. 14 (64%) patients experienced another SP after the sentinel pneumothorax, 10 (45%) had an ipsilateral recurrence and 8 (36%) experienced a contralateral SP (Figure 1).

Table 2:

Pneumothorax rates among the patients in our cohort.

Variables	Data
Total number of SPs (confirmed by chest imaging)*	53
Number of patients with at least one SP	22 (23%)
Number of patients with multiple (>1) SP	14 (64% of patients with a SP)
Number of patients with an ipsilateral recurrence of SP	10 (45% of patients with a SP)
Number of patients with a contralateral occurrence of SP	8 (36% of patients with a SP)
Location of SP
Right side only	10 (45%)
Left side only	4 (18%)
Both right and left	8 (36%)**
Patients with pneumothorax as presenting symptom of PLCH	18 (19%)
Age at the development of first SP - median (range)	29 (15 to 54) years
Number of SPs per patient - median (range)	2 (1 to>6)
Number of pneumothoraces before the diagnosis of PLCH# - median (range)	1.5 (1 to >6)
Number of ipsilateral SPs prior to undergoing pleurodesis (chemical or surgical)## - median (range)	2 (1 to 4)
Ipsilateral SP recurrence rate following conservative management	20 out of 31 (65%)
Ipsilateral SP recurrence rate following chemical pleurodesis	2 out of 4 (50%)
Ipsilateral SP recurrence rate following surgical pleurodesis	4 out of 14 (29%)

Abbreviations: PLCH = pulmonary Langerhans cell histiocytosis, CT = computed tomography, SP = Spontaneous pneumothorax

^*Upto 6 pneumothoraces per patient were recorded

^**One patient reported having a synchronous bilateral pneumothorax.

^#In patients diagnosed with PLCH after 1st pneumothorax

^##In patients who underwent pleurodesis

Figure 1:

Incidence and recurrence of ipsilateral and contralateral pneumothoraces in PLCH. SP = Spontaneous pneumothorax.

Most patients developed an SP in the second and third decade (Figure 2); the average age at the first SP was 29 years (range: 15 to 54 years). Eighteen patients (81%) had experienced an SP prior to their diagnosis of PLCH, with an average of two episodes of SP prior to getting diagnosed with PLCH. The average time to diagnosis was 6.3 years (range: 0 to 27 years) after the first SP in these patients. Information regarding the timing of chest computed tomography (CT) was available in 15 of these 18 patients, and the imaging was performed after a median of two pneumothoraces (range: 1–4).

Figure 2:

Histogram depicting the relationship between age (in years) at first pneumothorax and second pneumothorax and age at diagnosis of PLCH. Most patients developed a pneumothorax in the 2^nd and 3^rd decade, usually before the diagnosis of PLCH was established.

We collected information regarding the first six SPs from each of the patients in our cohort. A total of 53 episodes of SPs were noted in these 22 patients, and details of pneumothorax treatment and recurrence were available for 49 episodes. 31 (63%) episodes were managed conservatively by observation (n = 14) or chest tube without pleurodesis (n = 17), 4 (8%) were treated with chemical pleurodesis and 14 (29%) with surgical pleurodesis. The recurrence rate of SP was 65% after conservative management (with observation or chest tube without pleurodesis), 50% after chemical pleurodesis, and 29% after surgical pleurodesis (p = 0.025 when comparing surgical pleurodesis and conservative management; comparisons between chemical and surgical pleurodesis, and chemical pleurodesis versus conservative management were not statistically significant). Efficacies of various modalities for the management of first and second episodes of SPs in our cohort are depicted in Figure 3.

Figure 3:

Effectiveness of the treatment of pneumothorax in patients with PLCH. Each hemithorax was considered separately. The highest recurrence rates of pneumothorax occurred with conservative therapy, but surgical and chemical pleurodesis procedures also frequently failed. After the first recurrence episode, all conservatively managed patients experienced a recurrent pneumothorax.

^* Includes observation, simple aspiration, and tube thoracostomy without pleurodesis

# Intervation and/or recurrence details not available for 1 patient with first pneumothorax and 1 patient with second pneumothorax

3.4. Air Travel

Eighty-two (87%) respondents had flown at least once in their lifetime, for an estimated total of 742 flights following the diagnosis of PLCH. Constitutional symptoms were reported by 58% (44 out of 76) of the patients, including headache (36%), anxiety (36%), shortness of breath (30%), chest pressure (32%), chest pain (26%), dizziness (18%), unusual fatigue (17%), nausea (15%), and peripheral cyanosis (5%). Two patients experienced an episode of SP during air travel, consistent with a flight-related pneumothorax risk of 2.4% per patient and 0.27% per flight (2/742) (Table 3). Both patients developed symptoms during air travel and denied having any symptoms prior to boarding.

Table 3:

Details regarding patients with a flight-related pneumothorax.

	Patient 1	Patient 2
Gender	Female	Female
Age at time of flight-related SP	22	37
Side of SP	Right	Right
Prior SP	No	Yes*
Diagnosed with PLCH at time of fight-related SP	No	Yes
Prior pleurodesis	No	No
Flight phase at symptom onset	Descent	Cruise altitude
In-flight treatment	None	Supplemental Oxygen
Mode of confirmation	CxR	CxR
Management after landing	Surgical pleurodesis	Observation
Recurrence on same side	Yes	Unknown
Approximate flight duration	1 hour	6 hours
Type of aircraft	Propeller plane	Large Jet
Symptoms 24–48 hours prior to boarding	No	No

Abbreviations: PLCH = pulmonary Langerhans cell histiocytosis, CT = computed tomography, SP = Spontaneous pneumothorax, CxR = Chest x-ray

^*Five episodes, three on right and two on left

18% of the patients (15 out of 82) reported to have significantly reduced their frequency of air travel, and 15% (12 out of 82) avoided air travel altogether after being diagnosed with PLCH. The reasons for avoiding air travel were variable and included personal assessment of risk of air travel (42%, 8 out of 19), physician’s advice (32%, 6 out of 19), and reasons unrelated to PLCH (37%, 7 out of 19). Patients were given variable recommendations by their clinicians regarding the safety of future air travel following an episode of SP: greater than half of the patients (69%) were given no specific recommendations regarding the safety of future air travel, 19% were told it is safe to fly one month following SP, and 12% of the patients were advised against any future air travel.

4. Discussion

The major findings from our study are as follows: (1) SP is common in patients with PLCH and can be the presenting manifestation that leads to the diagnosis of PLCH; (2) patients with PLCH are at increased risk of recurrence of SP, with recurrence rates in excess of 60%; and (3) the recurrence risk is reduced by over 50% after surgical pleurodesis.

Similar findings of increased recurrence risk following conservative management of SP have been reported in PLCH previously [4]. However, in contrast to the previous literature suggesting almost complete mitigation of the future risk of SPs following surgical pleurodesis [4], we found that while surgical pleurodesis is effective in reducing future recurrent SPs, recurrent SPs can happen even after surgical pleurodesis. Similar findings with regard to the pneumothorax recurrence risk in PLCH have been reported by a recent study [7]. The pneumothorax recurrence and pleurodesis efficacy results for PLCH in our study are similar to those seen in patients with other DCLDs such as LAM and Birt–Hogg–Dubé syndrome (BHD) [12, 13], and we submit that patients with PLCH should be considered for pleurodesis following the first episode of SP rather than waiting for a recurrent event, similar to the recent recommendations for LAM patients [14].

The majority of patients in our study did not undergo pleurodesis until after they had experienced at least one recurrent SP. Potential explanations for not undergoing early pleurodesis in our study may include patient preferences [15], or physician perception of the impact of pleurodesis on the future need for lung transplantation. Although disease-specific data for PLCH are lacking, recent guidelines from the International Society for Heart and Lung Transplantation state that previous pleurodesis is not considered a contraindication to future lung transplant [16].

There was a remarkable delay in establishing the diagnosis of PLCH, even in those who had recurrent pneumothoraces. Pneumothorax can be the presenting manifestation in approximately one-fifth of patients with PLCH, and can provide a gateway to achieving timely diagnosis of PLCH. In a recent analysis, the strategy of performing screening HRCT in patients presenting with an apparent primary SP was found to be cost-effective in facilitating a timely diagnosis of PLCH and other DCLDs such as LAM and BHD [17].

In accordance with the Boyle’s law of inverse relationship between gas volume and pressure, the volume of gas in a non-communicating lesion is predicted to increase by up to 38% during commercial flights, which are usually pressurized to 8,000 feet (2,438 m) above sea level [9, 18]. This gas expansion can pose an increased risk of SP in patients with DCLDs. Air travel-associated SP has been reported in 1.1%–2.6% per flight and 2.8%−4% per person in LAM patients [19–21], and 0%–0.63% per flight and 0%−9% per person in BHD patients [22–24]. We found a similar risk of 0.27% per flight and 2.4% per patient in PLCH patients. To the best of our knowledge, this is the first study to evaluate the risk of air travel-related SP in patients with PLCH. The risk, although small, is almost certainly higher than the negligible risk of air travel-associated SP in the general population [19, 25–27].

About one-third of the patients in our cohort significantly reduced or abandoned air travel after the diagnosis of PLCH. The reasons for this change in behavior could be multifactorial. Patients frequently experienced symptoms such as headache, anxiety, and respiratory symptoms such as shortness of breath and chest pain during air travel that could have untowardly influenced their decision toward air travel. It is important to note, however, that these symptoms are frequently reported by healthy individuals undertaking air travel [28]. Lack of studies evaluating the safety of air travel in patients with PLCH may have influenced physicians’ recommendations regarding air travel. For instance, more than half of the patients in our cohort were not given any clear recommendations regarding safety of air travel following an SP. No clear guidelines exist with regard to the optimal timing to safely undertake air travel following a SP; most medical societies tend to follow the British Thoracic Society guidelines to wait at least 1 week after radiographic resolution of SP before air travel [18, 29].

Several limitations of the study must be noted. The survey data were collected from patient responses and are subject to recall bias. The extent of recall bias is context-dependent, is likely higher with respect to information pertaining to air travel details, and may not be a major factor when recalling medical events such as SP and pleurodesis. Although the study had a decent sample size for an ultra-rare disease such as PLCH, the estimates of the efficacy of pleurodesis following SP are derived from very small patient numbers and should be interpreted with caution. In the absence of primary patient records and chest imaging data, we were unable to correlate the risk of development of SP with the physiological and anatomical characteristics of individual patients. Identification of risk factors that predispose patients to a higher risk of development of SP is key to make sound management decisions, and should be a high-priority research area for patients with PLCH. The study premise of ascertaining risk of SPs and safety of air travel may have led to selection bias by over-representation of patients with worse pulmonary symptoms. Though no gender predilection has been shown in adults with PLCH [4, 30, 31], more than 75% of our study respondents were females, which may represent a gender bias toward response to surveys [32]. The definition of air travel-related pneumothorax has not been well established. We arbitrarily chose to limit our definition of air travel-related pneumothorax to an SP occurring within 24 hours of air travel in order to minimize the confusion between a true flight-related pneumothorax and an unrelated SP that might have occurred a few days after air travel. It is important to note, however, that delayed presentation of SP following air travel has been reported previously [33]. Our inclusion of all ipsilateral pneumothoraces reported within the same month into one episode might have led to underestimation of pneumothorax recurrence rates.

5. Conclusions

SP is a common phenotypic manifestation of PLCH and has an extremely high rate of recurrence. Surgical pleurodesis can lead to a substantial reduction in the risk of recurrent SPs. The risk of a flight-related pneumothorax in patients with PLCH is about 2–3 per thousand flights.

Abbreviations:

BHD

Birt–Hogg–Dubé syndrome

COPD

Chronic obstructive pulmonary disease

CT

Computed tomography

DCLD

Diffuse cystic lung disease

HRCT

High-resolution computed tomography

IRB

Institutional Review Board

LAM

Lymphangioleiomyomatosis

NIH

National Institutes of Health

PLCH

Pulmonary Langerhans cell histiocytosis

RLDC

Rare Lung Diseases Consortium

SP

Spontaneous pneumothorax