A retrospective comparison between digital to conventional drainage systems for secondary spontaneous pneumothorax related to diffuse interstitial lung disease

Abstract Introduction Secondary spontaneous pneumothorax (SSP) occurs as one of the complications associated with interstitial pneumonia (IP). Chest drainage is performed when there is a large volume of air in the pleural space. Notably, SSP with IP (SSP‐IP) is frequently not curable by chest drainage only. A digital drainage system (DDS) provides an objective evaluation of air leakage and maintains a pre‐determined negative pressure, compared to an analog drainage system (ADS). Few studies have reported the effectiveness of DDS in the treatment of SSP‐IP. This study aimed to assess the usefulness of DDS for SSP‐IP. Methods This retrospective study included patients with SSP‐IP who had undergone chest drainage. We reviewed the included patients' medical records, laboratory data, computed tomography findings, and pulmonary function data. Results DDS was used in 24 patients and ADS in 49 patients. The mean duration of chest drainage was 11.4 ± 1.9 days in the DDS group and 14.2 ± 1.3 days in the ADS group, which was not significantly different (p = 0.218). Surgery, pleurodesis, and/or factor XIII administration were performed in 40 patients. Additionally, five (20.8%) patients in the DDS group and nine (18.4%) in the ADS group had a recurrence of pneumothorax within 4 weeks (p = 1.000). One patient (14%) in the DDS group and six (12.2%) in the ADS group (p = 0.414) were cured of pneumothorax but later died. Conclusion DDS did not demonstrate a significant difference in the shortening of chest drainage duration. Further study is needed to validate the results of this study.


| INTRODUCTION
Pneumothorax is defined as the presence of air in the pleural space between the lung and chest wall. 1 Secondary spontaneous pneumothorax (SSP) occurs as a complication of various lung diseases and is often experienced in daily clinical practice. Interstitial pneumonia (IP) is one of the lung diseases causing SSP, with risk factors including the presence of computed tomography (CT) abnormalities, lower vital capacity, lower body mass index, receiving long-term oxygen therapy, and methylprednisolone pulse therapy. [2][3][4] Additionally, SSP with IP (SSP-IP) has been reported to occur in 20.2% of patients with idiopathic pulmonary fibrosis (IPF) and is considered a poor prognostic factor. 5 Chest drainage should be performed for patients with pneumothorax affecting a large area of the pleural space. 6,7 Notably, SSP-IP is frequently not curable by chest drainage only. Surgery is one of the most common additional treatments and is the first choice for pneumothorax not cured by chest drainage alone. 6 However, SSP-IP often recurs even after surgery, and the postoperative mortality rate is high as 2.9-4.1%. [8][9][10] Furthermore, some patients with SSP-IP do not tolerate surgery under general anesthesia. Therefore, patients with SSP-IP tend to avoid surgery and receive alternative treatments. A previous study reported that 38.1% of cases of SSP-IP required alternative treatments such as pleurodesis in addition to chest drainage. 11 A digital drainage system (DDS), Thopaz ® (Medela), has been introduced into clinical practice for use in chest drainage. Compared to a conventional analog drainage system (ADS), this system enables objective evaluation of air leaks and maintains a pre-determined negative pressure without the influence of position changes or obstruction of tubes. [12][13][14] In addition, the ability to continuously record air leaks reduces variability by physicians in the timing of drain removal. 15,16 This DDS was expected to shorten the drainage and hospitalization periods and hospitalization costs. Several studies reported the effectiveness of DDS in spontaneous pneumothorax or after thoracic surgery. 13,17,18 On the other hand, only one case report has reported its effectiveness in the treatment of SSP-IP. 19 Therefore, this study aimed to assess the usefulness of DDS for SSP-IP.

| Study design and patients
We conducted a retrospective chart review of patients with SSP-IP who had undergone chest drainage at the Chiba University Hospital or the Seirei Hamamatsu General Hospital between April 2016 and March 2022. The exclusion criteria were as follows: (1) patients aged <20 years, (2) patients who died with uncured pneumothorax, and (3) patients whose drainage system was changed within 48 h after chest drainage.
All analyses were performed in accordance with the amended Declaration of Helsinki. Written informed consent for chest drainage was obtained from each patient. Data anonymization and privacy issues were strictly addressed. The study protocol was approved by the Human Ethics Committee of our institution (approval number M10306), and individual consent for this retrospective analysis was waived.

| Clinical data collection
We reviewed the included patients' medical records, laboratory data, CT findings, and pulmonary function data. The CT images were reviewed by two pulmonologists, and the findings were classified into fibrosis, cysts, fibrosis and cysts, and others. Pulmonary function data were obtained within 1 year of pneumothorax onset.

| Treatment of pneumothorax
The management of chest drain depends on a decision made by each clinician. After chest drainage, the chest tube was connected to the DDS or ADS. The intrathoracic pressure was between À8 cm H 2 O and À40 cm H 2 O in the DDS group and 0 cm H 2 O and À20 cm H 2 O in the ADS group.
The chest tube was removed when an air leak was not detected and lung expansion on the chest X-ray was obtained sufficiently. No air leak was set at lower than 30 mL/min of airflow in DDS and as a lack of an air bubble in ADS. In many cases, the clump test was performed to check that the chest tube was clumped, and the lung was not collapsed.
When pneumothorax was not cured by chest drainage alone, other additional treatments, including surgery, were performed based on the patient's condition and the clinical course.

| Statistical analysis
The median and range of continuous data were calculated. Counts and percentages were determined for categorical data. Continuous data were analyzed using the Wilcoxon rank-sum test and categorical data were analyzed using the Fisher's exact test. Two-sided p values of <0.05 were considered statistically significant. All analyses were performed using SAS software v.9.4 for Windows (SAS Institute Inc., Cary, NC, USA) and JMP pro 16.0.0 software (SAS Institute Inc. Cary, NC, USA).

| Study population and characteristics
In total, 99 patients with SSP-IP underwent chest drainage. Of these, 26 patients were excluded according to the criteria, and a total of 73 patients were included in the study (Figure 1). Patient characteristics are summarized in Table 1. Digital drainage was performed in 24 (32.9%) patients and analog drainage in 49 (67.1%) patients. There were 49 (67.1%) males and 24 (32.9%) females, with a mean age of 65.3 years. The CT image showed fibrosis in 52 (71.2%) cases and cysts in 42 (57.5%) cases. Prednisolone was administrated for 21 (28.8%) patients and anti-fibrotic medication for 16 (21.9%) patients for the treatment of IP. Regarding pneumothorax conditions, 52 (71.2%) cases were first episode of pneumothorax, and the right and left sides were affected in almost half cases each. Significant differences were observed in age and CT findings between the DDS and ADS groups.  Table 2 shows the treatments and outcomes of pneumothorax. The mean duration of drainage tended to be shorter in the DDS group than in the ADS group; however, the difference was not significant (11.4 ± 1.9 days in the DDS group vs. 14.2 ± 1.3 days in the ADS group, p = 0.218). Surgery, pleurodesis, and/or factor XIII administration were performed for treating pneumothorax in 40 (54.8%) patients. Five (20.8%) patients in the DDS group and nine (18.4%) in the ADS group had a recurrence of pneumothorax within 4 weeks (p = 1.000). One (4.2%) patient in the DDS group and six (12.2%) in the ADS group ( p = 0.414) were cured of pneumothorax but died.

| Treatments and outcomes of pneumothorax in subgroups
The subgroup analysis was performed on the group of cases excluding surgery and those with IPF. Table 3 shows the results of the analysis of 62 patients, excluding those who underwent surgery. The mean duration of chest drainage was 11.4 ± 1.8 days in the DDS group and 13.4 ± 1.3 days in the ADS group, which was not significantly different (p = 0.369). There was no significant difference between the two groups in treatments, recurrence, and death. Eighteen cases with SSP with IPF were analyzed, and the results are summarized in Table 4. The mean duration of chest drainage was 11.3 ± 4.7 days in the DDS group and 13.8 ± 3.3 days in the ADS group. This result was also not significantly different (p = 0.668). Treatments, recurrence, and death showed no significant differences between the two groups.

| DISCUSSION
This is the first study to report a comparison of DDS and ADS for treating SSP-IP in multiple patients, and the duration of chest drainage was approximately 3 days shorter with DDS, although the difference was not statistically significant. Lung compliance in patients with interstitial pneumonia decreases due to the destruction of T A B L E 2 Treatments and outcomes of pneumothorax in the digital and the analog drainage group.  13 This helps avoid excessive suctioning and thus, lung injury may be less likely to occur. These may explain the difference in the chest drainage duration. However, this study did not show any significant differences between DDS and ADS for the treatment of SSP-IP. One reason for this result could be due to the different characteristics of the patients. In this study, the details of CT findings differed between the DDS and ADS groups. The ADS group may have had a more complicated disease than the digital drainage system group, because of more use of prednisolone and/or antifibrotic drugs. Both fibrosis and cysts, the latter of which is represented by chronic obstructive pulmonary disease, cause lung fragility. 22 However, lung compliance differs between fibrosis and cysts; therefore, the different characteristics of the patients might lead to no significant differences between DDS and ADS for the treatment of SSP-IP. Another reason is the lack of criteria for clamping the chest tube before chest drain removal. Although clamping is unnecessary in cases of DDS for providing quantitative air leak, 16 clamping was performed in some cases with DDS in this study. Therefore, a study in which these factors are aligned should be performed.
In cases with SSP and IPF, the chest drainage duration tended to be shorter in the DDS group. Pneumothorax is a poor prognostic factor of IPF. Nishimoto et al reported that 20.2% of patients with IPF developed pneumothorax, and the median survival time from the onset of SSP was 13.3 months. 5 Although it is not clear whether shortening the drainage duration contributes to improved survival, it may prevent deterioration of the patients' general condition. Since the number of cases included in this study is small, analysis of a larger population is needed.
In the current study, additional treatments, surgery, and/or pleurodesis, were performed as much as previously reported. 11 Although surgery should be sought in cases of persistent air leakage, 6 it tends to be avoided for pneumothorax with IP because of the possibility of causing acute exacerbation of IP. 23,24 However, previous reports indicated that surgery could be performed safely in such cases. 25 Pleurodesis, which is recommended as an alternative treatment in cases where surgery cannot be performed, 1,6 also has the risk of acute exacerbation of IP and has a higher rate of the recurrence of pneumothorax than surgery. 25 Therefore, these treatment choices should be made with careful consideration. Porcel recommends that studies comparing treatments for SSP-IP are needed. 26 In the present study, patients received various treatments for SSP-IP, and the drainage duration was shorter in the DDS group than in the ADS group, with or without surgery. In addition, there were no deaths among patients who underwent surgery. Indicating what treatment should be performed is beyond the scope of the present study. However, in any case, it is important to perform appropriate multidisciplinary treatment, including surgery, in combination with DDS.
Prolonged chest drainage may cause complications such as pain, intrapleural infection, and drain-related visceral injury, 27 contributing to the deterioration of the patient's condition. These events lead to a decline in activities of daily living, prolonged hospitalization, and an increase in medical costs. 28 There are some limitations to our study. First, this was a retrospective study. Second, the severity of IP varied in this study. The CT findings of fibrosis are different between the ADS and DDS groups, which is a major limitation of this study. Surgery, which is an invasive procedure may have been avoided for the severe cases of IP. A T A B L E 4 Treatments and outcomes of pneumothorax with idiopathic pulmonary fibrosis in the digital and the analog drainage group. prospective, randomized, multicenter study with a large sample size should be performed. Finally, this study did not analyze the length of hospital stay and costeffectiveness. These items should be analyzed together in future studies. Needless to say, the patients' backgrounds must be aligned so as not to affect the length of hospital stay.
In conclusion, the current study aimed to assess the usefulness of DDS for SSP-IP by retrospectively analyzing data of patients who received chest drainage. The DDS group did not demonstrate a significant difference in the shortening of chest drainage duration. Further studies are needed that multicenter prospective studies with aligned patient backgrounds.

AUTHOR CONTRIBUTIONS
Kohei Shikano, Mitsuhiro Abe, and Takuji Suzuki contributed to the study concept and design. Kohei Shikano, Ryutaro Hirama, and Shinsuke Kitahara examined the enrolled patients in the hospital. In addition, Kohei Shikano performed the statistical analyses. All authors have read and approved the final manuscript.