The three peaks in age distribution of females with pneumothorax: a nationwide database study in Japan

The three peaks in age distribution of females with pneumothorax: a nationwide database study in... Abstract OBJECTIVES Women are the minority among patients with spontaneous pneumothorax, but catamenial pneumothorax (CP) is unique to them. We aimed to clarify the clinical characteristics of female patients with spontaneous pneumothorax using a nationwide database. METHODS Medical records from the Japanese Diagnosis Procedure Combination database for inpatients with pneumothorax between July 2010 and March 2016 were retrospectively reviewed. Age, underlying diseases, body mass index, smoking status, laterality, number of hospitalizations and treatments were studied. RESULTS We identified 157 087 patients with pneumothorax, including 27 716 (17.6%) women and 129 371 (82.4%) men. The age distribution of female patients with pneumothorax had 3 peaks: 18 years, around 40 years and 80 years; male patients had 2 peaks: 18 years and 79 years. We identified 873 patients with CP; this number was not sufficient to account for the female-specific peak around 40 years. The characteristics of female patients of reproductive age were significantly different between those with and without CP. The patients with CP were older (average age: 37.9 ± 7.7 years vs 31.3 ± 11.5 years, P < 0.001), were right side dominant (right: 64.9%, left: 6.5%), had more hospitalizations (average number of hospitalizations: 1.6 ± 0.9 vs 1.3 ± 0.6, P < 0.001) and had more frequently undergone surgery (57.1% vs 37.3%, P < 0.001). CONCLUSIONS The age distribution of women with pneumothorax had 3 distinct peaks while that of men had 2. CP has different characteristics from other types of pneumothorax, thus requiring different treatment strategies for women of reproductive age. Female pneumothorax, Catamenial pneumothorax, Epidemiology of pneumothorax INTRODUCTION Spontaneous pneumothorax is a common disease that mostly affects men [1–6]. Being thin and tall in stature is a risk factor for spontaneous pneumothorax among young men [7–9]; the clinical features of pneumothorax among women remain unknown. Catamenial pneumothorax (CP), a type of female-specific pneumothorax, is defined as recurrent pneumothorax that occurs within 72 h of menstruation [10]. Although there are several reports of female pneumothorax, including CP, the difference in the characteristics of pneumothorax between men and women remains unclear. This is because most reports are based on relatively small amounts of data collected from a single institution, or only from surgically treated patients [10–14]. In the present study, we aimed to use large-scale data from a national inpatient database in Japan to examine the differences in patient backgrounds (i) between men and women with pneumothorax and (ii) between CP and other types of pneumothorax in female patients of reproductive age. MATERIALS AND METHODS Data source For this study, we used the Diagnosis Procedure Combination database, a national administrative claim and discharge abstract database in Japan. The details of this database have been described elsewhere [15], and numerous clinical studies based on this database have been published [16, 17]. Data collection for this registry has only been consecutive since July 2010, and approximately 8 million inpatients are entered into the database each year, a number that accounts for approximately 55% of all acute-care hospitalizations in Japan. The database includes the following data: patient age and gender; main diagnosis, comorbidities at admission and complications after admission, recorded according to the International Classification of Diseases 10th Revision (ICD-10) and text data in Japanese; procedures coded with the Japanese original coding system; body height and weight and smoking status. The present study was approved by the Institutional Review Board at the University of Tokyo (No. 3501). The requirement for informed consent was waived because of the anonymous nature of the data. Patient selection and data We identified patients with pneumothorax or CP as the main diagnosis or comorbidity occurring between 1 July 2010 and 31 March 2016. The case selection and definition for each group are presented in Fig. 1. Patients with pneumothorax were identified with the ICD-10 code J93 (‘pneumothorax’). Patients with CP were identified with code N948 (‘other specified conditions associated with female genital organs and menstrual cycle’), along with the text data ‘catamenial pneumothorax’ written in Japanese. We excluded ‘suspected pneumothorax’, iatrogenic pneumothorax (T812) and traumatic pneumothorax (S27). Cases in which a patient was rehospitalized in the same facility were recognized by the patient’s unique hospital identification number as a returning patient, and data could be aggregated. However, if a patient was rehospitalized in a different institution, then data would be seen as 2 separate cases because the patient would have been registered under different hospital identification numbers at different hospitals. Figure 1: View largeDownload slide Case selection and definition for each group. Asterisk indicates overlapping of some of the cases. CP: catamenial pneumothorax. Figure 1: View largeDownload slide Case selection and definition for each group. Asterisk indicates overlapping of some of the cases. CP: catamenial pneumothorax. Patient background data included age, underlying diseases, body mass index (kg/m2) classification [18], the affected side and smoking status (current/ex-smokers or non-smokers). Underlying diseases included chronic obstructive pulmonary disease, interstitial pneumonia, primary lung cancer, lung metastasis, asthma, tuberculosis, sarcoidosis, Marfan syndrome, Ehlers–Danlos syndrome, Birt–Hogg–Dubé syndrome and lymphangioleiomyomatosis. We compared the backgrounds of male and female patients. We classified female patients into the following 3 age categories: young (13–30 years), middle aged (31–53 years) and aged (>53 years). We compared the background of patients with the number of hospitalizations among these 3 categories. We also compared patient background, cases of endometriosis (any endometriosis and thoracic endometriosis) and the number of hospitalizations and treatments (chemical pleurodesis and surgery) between CP and other types of pneumothorax in female patients of reproductive age. Surgical procedures included lung resection, thoracoscopic lung resection, diaphragm resection and suture and resection of the parietal pleura. Reproductive age was defined as 13–53 years, according to the age distribution of CP in the present study. Statistical analysis The Mann–Whitney U-test or Kruskal–Wallis test was used to compare continuous variables (age and number of hospitalizations) between the groups. The χ2 test was used to compare the categorical variables (underlying diseases, body mass index classification, affected side, smoking status, pleurodesis and surgery). After conducting a multivariate analysis of the risk factors for CP in female pneumothorax of reproductive age, we performed a logistic regression analysis and calculated the odds ratios and 95% confidence intervals. The threshold for significance was P-value <0.05. All statistical analyses were performed using SPSS Statistics version 22.0 (IBM SPSS Inc., Armonk, NY, USA). RESULTS From all inpatient data (42 million hospitalizations), we identified 192 858 hospitalizations with ICD-10 code J93. After iatrogenic and traumatic cases (84 and 964, respectively, with 1 overlap) and suspected pneumothorax cases were excluded, identical patient cases were aggregated. Ultimately, we identified 157 087 eligible patients with spontaneous pneumothorax, including 129 371 (82.4%) male patients and 27 716 (17.6%) female patients (Fig. 1). In terms of the total number of cases, the most common underlying disease was chronic obstructive pulmonary disease, especially among male patients. Male patients were significantly more likely to have primary lung cancer than were female patients, while female patients were significantly more likely to have interstitial pneumonia, asthma, lung metastasis, tuberculosis, sarcoidosis, Marfan syndrome, Birt–Hogg–Dubé syndrome, CP and lymphangioleiomyomatosis (Table 1). Lymphangioleiomyomatosis and CP accounted for 0.9% and 3.1% of female pneumothorax, respectively. The female patients with pneumothorax were thinner than were the male patients with pneumothorax. There was no apparent laterality in male patients (right, 27.1%; left, 26.7%), whereas female patients with pneumothorax appeared to exhibit right-sided dominance (right, 22.6%; left, 18.2%). Among the male patients, approximately half were current or ex-smokers. Table 1: Characteristics of patients with pneumothorax, by gender (n = 157 087)   Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001    Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001  Continuous data are presented as the mean ± standard deviation; categorical data are presented as the n (%). Table 1: Characteristics of patients with pneumothorax, by gender (n = 157 087)   Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001    Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001  Continuous data are presented as the mean ± standard deviation; categorical data are presented as the n (%). The age distribution of all patients was biphasic (Fig. 2A); the age distribution of male patients alone was also biphasic (Fig. 2B). For both groups, the younger peak was observed at 18 years, and the older peak was observed at 79 years. In contrast, the age distribution of female patients had a peak around 40 years, in addition to peaks at 18 and 80 years (Fig. 2C). The ages of patients with CP ranged from 13 to 53 years, with a peak around 40 years (Fig. 2D). Figure 2E shows the age distribution of female patients without CP. The peak around 40 years of age was not diminished by the exclusion of patients with CP. Figure 2: View largeDownload slide Age distribution of patients with pneumothorax in Japan between July 2010 and March 2016. (A) All patients (n = 157 087); (B) male patients (n = 129 371); (C) female patients (n = 27 716); (D) catamenial pneumothorax (n = 873) and (E) female patients without catamenial pneumothorax (n = 26 843). Figure 2: View largeDownload slide Age distribution of patients with pneumothorax in Japan between July 2010 and March 2016. (A) All patients (n = 157 087); (B) male patients (n = 129 371); (C) female patients (n = 27 716); (D) catamenial pneumothorax (n = 873) and (E) female patients without catamenial pneumothorax (n = 26 843). The characteristics of female patients with pneumothorax according to age group are shown in Table 2. Focusing on the fact that there were 3 age peaks for female patients, we subdivided the adult female pneumothorax patients into 3 groups—young (premenopausal, younger): 13–30 years; middle-aged (premenopausal, older): 31–53 years and aged (postmenopausal): >53 years. The aged group was more likely to have most of the underlying diseases, while the proportion of patients with Marfan syndrome was highest in the young group and that of CP was highest in the middle-aged group. The proportion of underweight patients was highest in the young group. The middle-aged group had the highest smoking rate. Table 2: Differences in characteristics of female patients with pneumothorax, by age group (n = 27 125) Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 2: Differences in characteristics of female patients with pneumothorax, by age group (n = 27 125) Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 3 presents the comparison between CP and other types of pneumothorax in female patients of reproductive age. The average age of patients with CP was significantly higher than that of patients with other types of pneumothorax. The proportions of underweight patients and smokers were significantly lower in the CP group. All types of endometriosis and thoracic endometriosis were found in 28.8% and 9.4% of CP patients, respectively. In contrast, the 2 types were found in 1.4% and 0.0% of patients with other types of pneumothorax, respectively. In both groups, the right side was more commonly affected; however, the tendency was stronger in the CP group (CP: right, 64.9%; left, 6.5%; Other: right, 32.7%; left, 30.4%). Female patients with CP were more frequently hospitalized than other pneumothorax patients (mean number of hospitalizations: 1.6 ± 0.9 vs 1.3 ± 0.6, P < 0.001). Patients with CP were also more likely to undergo pleurodesis (8.5% vs 4.5%, P < 0.001) and surgical treatment for pneumothorax (57.1% vs 37.3%, P < 0.001) compared with patients with other types of pneumothorax. Patients with CP required multiple surgical treatments more frequently than did those with other types of pneumothorax (7.6% vs 5.4%, P < 0.001). Table 3: Characteristics and treatments of female patients of reproductive age, by type of pneumothorax (n = 11 599) Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 3: Characteristics and treatments of female patients of reproductive age, by type of pneumothorax (n = 11 599) Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). We performed a multivariate analysis to determine the predictors of CP in female pneumothorax of reproductive age (Table 4). Being aged >25 years (especially 36–45 years), having coexisting endometriosis, right-sided pneumothorax or a history of hospitalization for pneumothorax was associated with a higher probability of having a CP diagnosis; however, being under- or overweight or a current or ex-smoker was associated with a lower probability of having CP. Table 4: Multivariate analysis for predictors of catamenial pneumothorax in female pneumothorax of reproductive age Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  CI: confidence interval. Table 4: Multivariate analysis for predictors of catamenial pneumothorax in female pneumothorax of reproductive age Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  CI: confidence interval. DISCUSSION The present study investigated the differences in patients with spontaneous pneumothorax according to sex. The male-to-female ratio in the present study was 4.7:1, which was consistent with similar ratios reported previously [1–6]. The present study showed a biphasic age distribution for all patients and for male patients alone and a triphasic distribution for female patients. The reason the age distributions for both total and male data sets were biphasic was that men comprised more than 80% of the cohort. A previous epidemiological study reported that female idiopathic pneumothorax was most common in the 3rd decade of life and that secondary pneumothorax was more common in the 4th decade of life [3]. In a study of the French national inpatient database [6], age distributions for both men and women had 2 peaks; among women, the 1st peak of incidence was older than that of men and incidence remained stable up to 40 years. The present study confirmed that the age distribution of female pneumothorax had a distinct peak around 40 years, along with those at 18 and 79 years; the latter 2 peaks exactly matched those seen in men. The present study also showed that the peak age at which CP occurred was around 40 years, a finding that is consistent with those in previous reports [11, 13]. CP and pneumothorax related to lymphangioleiomyomatosis are unique to women of reproductive age. However, the present study showed that the number of these diseases diagnosed was insufficient to account for the peak that was observed around 40 years of age. The fact that the 3rd peak was maintained even without CP patients suggests that the number of CP cases was underdiagnosed in the data set. CP is clinically defined as recurrent pneumothorax that occurs within 72 h of menstruation [10]. Some patients with CP have histologically demonstrated ectopic endometrial tissues on their diaphragms or pleura, although others do not [10, 11, 13]. When histological findings of thoracic endometriosis are observed in a patient with pneumothorax, she is diagnosed as having thoracic endometriosis-related pneumothorax (TERP), regardless of whether the onset of pneumothorax met the criteria for CP [19]. Because these diagnoses are not stated definitely in the ICD-10, CP may have included both original CP and TERP in the present study. The present study showed that the proportion of CP was 3.1% among all cases of female pneumothorax. Previously reported proportions of either CP or TERP among all cases of female pneumothorax have varied widely. According to a Japanese nationwide study on female pneumothorax in 1986 [3], only 0.9% of all patients with pneumothorax and 2.8% of patients with secondary pneumothorax had CP. The most common cause of secondary pneumothorax at that time was pulmonary tuberculosis. In several previous studies of surgically treated patients [10–14, 19, 20], 21–23% of the female patients with pneumothorax had CP or TERP [12, 19]. Among the patients of reproductive age [10, 11, 13, 14], 24–41% had CP or TERP. In our latest nationwide study, CP accounted for only 7.5% of the cases of pneumothorax in female patients of reproductive age. Even considering that these previous studies only included surgically treated patients in expert centres and, therefore, selection bias was inevitable, CP might still be underdiagnosed by general clinicians. The present study showed a lower proportion of underweight patients with CP than with other types of pneumothorax. However, approximately one-third of the patients with CP were classified as underweight, a proportion that was much larger than that of healthy Japanese women (14.5%) of the same age [21]. This implies that patients with CP are generally thin, similar to patients with endometriosis [22]. In our study, 28.8% of CP patients had endometriosis. Previous studies have shown that the proportions of patients with pelvic endometriosis among those with CP or TERP vary broadly, with values ranging from 18.0% to 58.7% [19, 20]. We speculate that the diagnosis of endometriosis may have been under-reported, possibly because most of the clinicians who treated CP were not gynaecologists. CP is known to be a right side-dominant pneumothorax, and left-sided CP was very rare in previous small studies [11–13, 19]. Our results showed that left-sided CP was rare, but certainly existed. It should be noted that male pneumothorax had no apparent laterality, whereas female patients of reproductive age without CP still appeared to exhibit right-sided dominance. Thus, in the non-CP group, there may have been some patients with underdiagnosed CP. We found that patients with CP were more frequently hospitalized than were other female pneumothorax patients. In other words, they were more likely to experience recurrence than those with other types of pneumothorax. Surgical interventions were more frequently performed on CP patients to stop air leakage and/or to confirm histological diagnosis; however, repeat surgery was required more frequently in these patients than in patients with other types of pneumothorax. On the other hand, less than 10% of CP patients underwent chemical pleurodesis. Because our data included only chemical pleurodesis (by talc powder, minocycline or picibanil) but did not include autologous blood pleurodesis or pleural ablation, more procedures would have been performed than present results indicate. In Japan, pleurodesis treatment for pneumothorax is not actively performed, except in patients with high surgical risk; this is because many surgeons have concerns regarding difficulties associated with reoperation after pleurodesis [23]. For recurrent CP, some surgeons perform surgical pleurodesis using an absorbable mesh [24]. In Western countries, the standard practice is to perform pleurodesis after the second episode of pneumothorax, and talc pleurodesis has been reported to provide better results in terms of preventing recurrence of CP than pleural ablation [11]. The effectiveness of pleurodesis for CP should be reconsidered. Several previous studies have determined the predictors of CP and TERP. They identified several factors as predictors: age (>28 years [25] or ≥31 years [19]), never smoked [19] or not a current smoker [25], a history of pelvic endometriosis [19], right-sided pneumothorax [19], infertility and a history of pelvic surgery or uterine scraping [25]. Indeed, these factors were strong predictors, but because of the small sample size and because CP had likely been underdiagnosed, they may have missed other suspicious features of CP. We found that the independent predictors of CP in female pneumothorax of reproductive age were as follows: aged >25 years (especially 36–45 years), normal weight, never smoked, coexisting endometriosis, right-sided pneumothorax and a history of hospitalization for pneumothorax. To prevent underdiagnosis of CP, we should take detailed anamneses and pay particular attention to female pneumothorax patients presenting with these factors. Limitations Several limitations of the present study should be acknowledged. First, as the large database used in this retrospective study is primarily intended for the request of medical fees, some data may not have been as well-validated as data in planned prospective studies, particularly for categories such as comorbidity, the affected side and smoking status, which are not directly related to any medical fee. For example, the amount of missing data pertaining to the affected side was large (25.3–55.9%), and whether or not data were truly missing at random was unverifiable. Second, the database did not include outpatient data, and we were, therefore, unable to follow-up patients after discharge unless they were rehospitalized at the same institution for a second episode of pneumothorax. Assuming that this limitation was the same among the groups, it is acceptable to compare the mean numbers of hospitalizations instead of the tendency of recurrence. Because the proportion of CP in female pneumothorax may be underestimated in our study, we would like to leave the precise comparison between idiopathic spontaneous pneumothorax and CP to future research based on detailed medical history taking and accurate pathological diagnosis. CONCLUSION In conclusion, the present large-scale study using a national inpatient database showed a distinct difference in age distribution of pneumothorax according to sex. We confirmed that the age distribution of all female patients with pneumothorax had a distinct peak around 40 years of age, along with those at 18 and 80 years. CP had different characteristics and thus requires different treatment strategies compared with other types of pneumothorax in women of reproductive age. Funding This work was supported by the Ministry of Health, Labour and Welfare, Japan [H29-Policy-Designated-009 and H29-ICT-Genral-004]; Ministry of Education, Culture, Sports, Science and Technology, Japan [17H04141]; and the Japan Agency for Medical Research and Development. Conflict of interest: none declared. REFERENCES 1 Gupta D, Hansell A, Nichols T, Duong T, Ayres JG, Strachan D. Epidemiology of pneumothorax in England. Thorax  2000; 55: 666– 71. Google Scholar CrossRef Search ADS PubMed  2 Primrose WR. Spontaneous pneumothorax: a retrospective review of aetiology, pathogenesis and management. Scott Med J  1984; 29: 15– 20. Google Scholar CrossRef Search ADS PubMed  3 Nakamura H, Konishiike J, Sugamura A, Takeno Y. Epidemiology of spontaneous pneumothorax in women. Chest  1986; 89: 378– 82. Google Scholar CrossRef Search ADS PubMed  4 Sadikot RT, Greene T, Meadows K, Arnold AG. Recurrence of primary spontaneous pneumothorax. Thorax  1997; 52: 805– 9. Google Scholar CrossRef Search ADS PubMed  5 Guo Y, Xie C, Rodriguez RM, Light RW. Factors related to recurrence of spontaneous pneumothorax. Respirology  2005; 10: 378– 84. Google Scholar CrossRef Search ADS PubMed  6 Bobbio A, Dechartres A, Bouam S, Damotte D, Rabbat A, Regnard JF et al.   Epidemiology of spontaneous pneumothorax: gender-related differences. Thorax  2015; 70: 653– 8. Google Scholar CrossRef Search ADS PubMed  7 Tan J, Yang Y, Zhong J, Zuo C, Tang H, Zhao H et al.   Association between BMI and recurrence of primary spontaneous pneumothorax. World J Surg  2017; 41: 1274– 80. Google Scholar CrossRef Search ADS PubMed  8 Kawakami Y, Irie T, Kamishima K. Stature, lung height, and spontaneous pneumothorax. Respiration  1982; 43: 35– 40. Google Scholar CrossRef Search ADS PubMed  9 Melton LJ3rd, Hepper NG, Offord KP. Influence of height on the risk of spontaneous pneumothorax. Mayo Clin Proc  1981; 56: 678– 82. Google Scholar PubMed  10 Alifano M, Roth T, Camilleri-Broet S, Schussler O, Magdeleinat P, Regnard JF. Catamenial pneumothorax: a prospective study. Chest  2003; 124: 1004– 8. Google Scholar CrossRef Search ADS PubMed  11 Alifano M, Jablonski C, Kadiri H, Falcoz P, Gompel A, Camilleri-Broet S et al.   Catamenial and noncatamenial, endometriosis-related or nonendometriosis-related pneumothorax referred for surgery. Am J Respir Crit Care Med  2007; 176: 1048– 53. Google Scholar CrossRef Search ADS PubMed  12 Fukuoka M, Kurihara M, Haga T, Ebana H, Kataoka H, Mizobuchi T et al.   Clinical characteristics of catamenial and non-catamenial thoracic endometriosis-related pneumothorax. Respirology  2015; 20: 1272– 6. Google Scholar CrossRef Search ADS PubMed  13 Legras A, Mansuet-Lupo A, Rousset-Jablonski C, Bobbio A, Magdeleinat P, Roche N et al.   Pneumothorax in women of child-bearing age: an update classification based on clinical and pathologic findings. Chest  2014; 145: 354– 60. Google Scholar CrossRef Search ADS PubMed  14 Mehta CK, Stanifer BP, Fore-Kosterski S, Gillespie C, Yeldandi A, Meyerson S et al.   Primary spontaneous pneumothorax in menstruating women has high recurrence. Ann Thorac Surg  2016; 102: 1125– 30. Google Scholar CrossRef Search ADS PubMed  15 Yasunaga H, Matsui H, Horiguchi H, Fushimi K, Matsuda S. Clinical epidemiology and health services research using the diagnosis procedure combination database in Japan. Asian Pac J Dis Manage  2013; 7: 19– 24. 16 Hasegawa W, Yamauchi Y, Yasunaga H, Sunohara M, Jo T, Matsui H et al.   Clinical features of 280 hospitalized patients with lymphangioleiomyomatosis in Japan. Respirology  2015; 20: 160– 5. Google Scholar CrossRef Search ADS PubMed  17 Nagano M, Ichinose J, Sasabuchi Y, Nakajima J, Yasunaga H. Surgery versus percutaneous transcatheter embolization for pulmonary arteriovenous malformation: analysis of a national inpatient database in Japan. J Thorac Cardiovasc Surg  2017; 154: 1137– 43. Google Scholar CrossRef Search ADS PubMed  18 WHO Expert Consultation. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Consultation. WHO Technical Report Series Number 854. World Health Organization, Geneva, 1995, 312– 74. 19 Haga T, Kataoka H, Ebana H, Otsuji M, Seyama K, Tatsumi K et al.   Thoracic endometriosis-related pneumothorax distinguished from primary spontaneous pneumothorax in females. Lung  2014; 192: 583– 7. Google Scholar CrossRef Search ADS PubMed  20 Joseph J, Sahn SA. Thoracic endometriosis syndrome: new observations from an analysis of 110 cases. Am J Med  1996; 100: 164– 70. Google Scholar CrossRef Search ADS PubMed  21 2014 National Health and Nutrition Survey Japan. 2016. http://www.mhlw.go.jp/bunya/kenkou/kenkou_eiyou_chousa.html (25 January 2018, date last accessed). 22 Vigano P, Somigliana E, Panina P, Rabellotti E, Vercellini P, Candiani M. Principles of phenomics in endometriosis. Hum Reprod Update  2012; 18: 248– 59. Google Scholar CrossRef Search ADS PubMed  23 Kurihara M, Kataoka H, Ishikawa A, Endo R. Latest treatments for spontaneous pneumothorax. Gen Thorac Cardiovasc Surg  2010; 58: 113– 9. Google Scholar CrossRef Search ADS PubMed  24 Ikeda T, Sasaki M, Sakon K, Koshiji T. An effective method of pleurodesis involving absorbable mesh for repetitive catamenial pneumothorax. Eur J Cardiothorac Surg  2012; 42: 370– 2. Google Scholar CrossRef Search ADS PubMed  25 Rousset-Jablonski C, Alifano M, Plu-Bureau G, Camilleri-Broet S, Rousset P, Regnard JF et al.   Catamenial pneumothorax and endometriosis-related pneumothorax: clinical features and risk factors. Hum Reprod  2011; 26: 2322– 9. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Cardio-Thoracic Surgery Oxford University Press

The three peaks in age distribution of females with pneumothorax: a nationwide database study in Japan

Loading next page...
 
/lp/ou_press/the-three-peaks-in-age-distribution-of-females-with-pneumothorax-a-aIY41TlSIw
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.
ISSN
1010-7940
eISSN
1873-734X
D.O.I.
10.1093/ejcts/ezy081
Publisher site
See Article on Publisher Site

Abstract

Abstract OBJECTIVES Women are the minority among patients with spontaneous pneumothorax, but catamenial pneumothorax (CP) is unique to them. We aimed to clarify the clinical characteristics of female patients with spontaneous pneumothorax using a nationwide database. METHODS Medical records from the Japanese Diagnosis Procedure Combination database for inpatients with pneumothorax between July 2010 and March 2016 were retrospectively reviewed. Age, underlying diseases, body mass index, smoking status, laterality, number of hospitalizations and treatments were studied. RESULTS We identified 157 087 patients with pneumothorax, including 27 716 (17.6%) women and 129 371 (82.4%) men. The age distribution of female patients with pneumothorax had 3 peaks: 18 years, around 40 years and 80 years; male patients had 2 peaks: 18 years and 79 years. We identified 873 patients with CP; this number was not sufficient to account for the female-specific peak around 40 years. The characteristics of female patients of reproductive age were significantly different between those with and without CP. The patients with CP were older (average age: 37.9 ± 7.7 years vs 31.3 ± 11.5 years, P < 0.001), were right side dominant (right: 64.9%, left: 6.5%), had more hospitalizations (average number of hospitalizations: 1.6 ± 0.9 vs 1.3 ± 0.6, P < 0.001) and had more frequently undergone surgery (57.1% vs 37.3%, P < 0.001). CONCLUSIONS The age distribution of women with pneumothorax had 3 distinct peaks while that of men had 2. CP has different characteristics from other types of pneumothorax, thus requiring different treatment strategies for women of reproductive age. Female pneumothorax, Catamenial pneumothorax, Epidemiology of pneumothorax INTRODUCTION Spontaneous pneumothorax is a common disease that mostly affects men [1–6]. Being thin and tall in stature is a risk factor for spontaneous pneumothorax among young men [7–9]; the clinical features of pneumothorax among women remain unknown. Catamenial pneumothorax (CP), a type of female-specific pneumothorax, is defined as recurrent pneumothorax that occurs within 72 h of menstruation [10]. Although there are several reports of female pneumothorax, including CP, the difference in the characteristics of pneumothorax between men and women remains unclear. This is because most reports are based on relatively small amounts of data collected from a single institution, or only from surgically treated patients [10–14]. In the present study, we aimed to use large-scale data from a national inpatient database in Japan to examine the differences in patient backgrounds (i) between men and women with pneumothorax and (ii) between CP and other types of pneumothorax in female patients of reproductive age. MATERIALS AND METHODS Data source For this study, we used the Diagnosis Procedure Combination database, a national administrative claim and discharge abstract database in Japan. The details of this database have been described elsewhere [15], and numerous clinical studies based on this database have been published [16, 17]. Data collection for this registry has only been consecutive since July 2010, and approximately 8 million inpatients are entered into the database each year, a number that accounts for approximately 55% of all acute-care hospitalizations in Japan. The database includes the following data: patient age and gender; main diagnosis, comorbidities at admission and complications after admission, recorded according to the International Classification of Diseases 10th Revision (ICD-10) and text data in Japanese; procedures coded with the Japanese original coding system; body height and weight and smoking status. The present study was approved by the Institutional Review Board at the University of Tokyo (No. 3501). The requirement for informed consent was waived because of the anonymous nature of the data. Patient selection and data We identified patients with pneumothorax or CP as the main diagnosis or comorbidity occurring between 1 July 2010 and 31 March 2016. The case selection and definition for each group are presented in Fig. 1. Patients with pneumothorax were identified with the ICD-10 code J93 (‘pneumothorax’). Patients with CP were identified with code N948 (‘other specified conditions associated with female genital organs and menstrual cycle’), along with the text data ‘catamenial pneumothorax’ written in Japanese. We excluded ‘suspected pneumothorax’, iatrogenic pneumothorax (T812) and traumatic pneumothorax (S27). Cases in which a patient was rehospitalized in the same facility were recognized by the patient’s unique hospital identification number as a returning patient, and data could be aggregated. However, if a patient was rehospitalized in a different institution, then data would be seen as 2 separate cases because the patient would have been registered under different hospital identification numbers at different hospitals. Figure 1: View largeDownload slide Case selection and definition for each group. Asterisk indicates overlapping of some of the cases. CP: catamenial pneumothorax. Figure 1: View largeDownload slide Case selection and definition for each group. Asterisk indicates overlapping of some of the cases. CP: catamenial pneumothorax. Patient background data included age, underlying diseases, body mass index (kg/m2) classification [18], the affected side and smoking status (current/ex-smokers or non-smokers). Underlying diseases included chronic obstructive pulmonary disease, interstitial pneumonia, primary lung cancer, lung metastasis, asthma, tuberculosis, sarcoidosis, Marfan syndrome, Ehlers–Danlos syndrome, Birt–Hogg–Dubé syndrome and lymphangioleiomyomatosis. We compared the backgrounds of male and female patients. We classified female patients into the following 3 age categories: young (13–30 years), middle aged (31–53 years) and aged (>53 years). We compared the background of patients with the number of hospitalizations among these 3 categories. We also compared patient background, cases of endometriosis (any endometriosis and thoracic endometriosis) and the number of hospitalizations and treatments (chemical pleurodesis and surgery) between CP and other types of pneumothorax in female patients of reproductive age. Surgical procedures included lung resection, thoracoscopic lung resection, diaphragm resection and suture and resection of the parietal pleura. Reproductive age was defined as 13–53 years, according to the age distribution of CP in the present study. Statistical analysis The Mann–Whitney U-test or Kruskal–Wallis test was used to compare continuous variables (age and number of hospitalizations) between the groups. The χ2 test was used to compare the categorical variables (underlying diseases, body mass index classification, affected side, smoking status, pleurodesis and surgery). After conducting a multivariate analysis of the risk factors for CP in female pneumothorax of reproductive age, we performed a logistic regression analysis and calculated the odds ratios and 95% confidence intervals. The threshold for significance was P-value <0.05. All statistical analyses were performed using SPSS Statistics version 22.0 (IBM SPSS Inc., Armonk, NY, USA). RESULTS From all inpatient data (42 million hospitalizations), we identified 192 858 hospitalizations with ICD-10 code J93. After iatrogenic and traumatic cases (84 and 964, respectively, with 1 overlap) and suspected pneumothorax cases were excluded, identical patient cases were aggregated. Ultimately, we identified 157 087 eligible patients with spontaneous pneumothorax, including 129 371 (82.4%) male patients and 27 716 (17.6%) female patients (Fig. 1). In terms of the total number of cases, the most common underlying disease was chronic obstructive pulmonary disease, especially among male patients. Male patients were significantly more likely to have primary lung cancer than were female patients, while female patients were significantly more likely to have interstitial pneumonia, asthma, lung metastasis, tuberculosis, sarcoidosis, Marfan syndrome, Birt–Hogg–Dubé syndrome, CP and lymphangioleiomyomatosis (Table 1). Lymphangioleiomyomatosis and CP accounted for 0.9% and 3.1% of female pneumothorax, respectively. The female patients with pneumothorax were thinner than were the male patients with pneumothorax. There was no apparent laterality in male patients (right, 27.1%; left, 26.7%), whereas female patients with pneumothorax appeared to exhibit right-sided dominance (right, 22.6%; left, 18.2%). Among the male patients, approximately half were current or ex-smokers. Table 1: Characteristics of patients with pneumothorax, by gender (n = 157 087)   Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001    Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001  Continuous data are presented as the mean ± standard deviation; categorical data are presented as the n (%). Table 1: Characteristics of patients with pneumothorax, by gender (n = 157 087)   Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001    Female patients  Male patients  P-value  Number of patients (%)  27 716 (17.6)  129 371 (82.4)    Age (years)  55.9 ± 25.6  51.1 ± 25.6  <0.001  Underlying diseases   Chronic obstructive pulmonary disease  869 (3.1)  11 620 (9.0)  <0.001   Interstitial pneumonia  2706 (9.8)  8256 (6.4)  <0.001   Primary lung cancer  1420 (5.1)  8163 (6.3)  <0.001   Asthma  1378 (5.0)  5592 (4.3)  <0.001   Lung metastasis  974 (3.5)  2340 (1.8)  <0.001   Tuberculosis  143 (0.5)  510 (0.4)  0.002   Sarcoidosis  70 (0.3)  122 (0.1)  <0.001   Marfan syndrome  58 (0.2)  110 (0.1)  <0.001   Ehlers–Danlos syndrome  6 (0.0)  10 (0.0)  0.048   Birt–Hogg–Dubé syndrome  13 (0.0)  14 (0.0)  <0.001   Catamenial pneumothorax  873 (3.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  231 (0.9)  2 (0.0)  <0.001  Body mass index classification (kg/m2)      <0.001   Underweight (<18.5)  11 344 (40.9)  42 897 (33.2)     Normal (18.5–24.9)  11 638 (42.0)  68 437 (52.9)     Overweight (≥25.0)  1337 (4.8)  6247 (4.8)     Missing data  3395 (12.2)  11 774 (9.1)    Affected side      <0.001   Right  6275 (22.6)  35 124 (27.1)     Left  5041 (18.2)  34 533 (26.7)     Bilateral  917 (3.3)  6473 (5.0)     Missing data  15 483 (55.9)  53 241 (41.2)    Smoking status      <0.001   Current or ex-smoker  4591 (16.6)  65 302 (50.5)     Never smoked  20 560 (74.2)  51 085 (39.5)     Missing data  2565 (9.3)  12 984 (10.0)    Treatments   Pleurodesis  2060 (7.4)  12 108 (9.4)  <0.001   Surgery  5635 (20.3)  37 598 (29.1)  <0.001  Continuous data are presented as the mean ± standard deviation; categorical data are presented as the n (%). The age distribution of all patients was biphasic (Fig. 2A); the age distribution of male patients alone was also biphasic (Fig. 2B). For both groups, the younger peak was observed at 18 years, and the older peak was observed at 79 years. In contrast, the age distribution of female patients had a peak around 40 years, in addition to peaks at 18 and 80 years (Fig. 2C). The ages of patients with CP ranged from 13 to 53 years, with a peak around 40 years (Fig. 2D). Figure 2E shows the age distribution of female patients without CP. The peak around 40 years of age was not diminished by the exclusion of patients with CP. Figure 2: View largeDownload slide Age distribution of patients with pneumothorax in Japan between July 2010 and March 2016. (A) All patients (n = 157 087); (B) male patients (n = 129 371); (C) female patients (n = 27 716); (D) catamenial pneumothorax (n = 873) and (E) female patients without catamenial pneumothorax (n = 26 843). Figure 2: View largeDownload slide Age distribution of patients with pneumothorax in Japan between July 2010 and March 2016. (A) All patients (n = 157 087); (B) male patients (n = 129 371); (C) female patients (n = 27 716); (D) catamenial pneumothorax (n = 873) and (E) female patients without catamenial pneumothorax (n = 26 843). The characteristics of female patients with pneumothorax according to age group are shown in Table 2. Focusing on the fact that there were 3 age peaks for female patients, we subdivided the adult female pneumothorax patients into 3 groups—young (premenopausal, younger): 13–30 years; middle-aged (premenopausal, older): 31–53 years and aged (postmenopausal): >53 years. The aged group was more likely to have most of the underlying diseases, while the proportion of patients with Marfan syndrome was highest in the young group and that of CP was highest in the middle-aged group. The proportion of underweight patients was highest in the young group. The middle-aged group had the highest smoking rate. Table 2: Differences in characteristics of female patients with pneumothorax, by age group (n = 27 125) Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 2: Differences in characteristics of female patients with pneumothorax, by age group (n = 27 125) Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Variables  Young (13–30 years)  Middle aged (31–53 years)  Aged (>53 years)  P-value  Number of patients  5562  6037  15 526    Underlying diseases   Chronic obstructive pulmonary disease  39 (0.7)  80 (1.3)  748 (4.8)  <0.001   Interstitial pneumonia  26 (0.5)  145 (2.4)  2531 (16.3)  <0.001   Primary lung cancer  4 (0.1)  114 (1.9)  1301 (8.4)  <0.001   Asthma  149 (2.7)  272 (4.5)  911 (5.9)  <0.001   Lung metastasis  16 (0.3)  165 (2.7)  789 (5.1)  <0.001   Tuberculosis  7 (0.1)  13 (0.2)  123 (0.8)  <0.001   Sarcoidosis  1 (0.0)  13 (0.2)  56 (0.4)  <0.001   Marfan syndrome  36 (0.6)  13 (0.2)  2 (0.0)  <0.001   Ehlers–Danlos syndrome  2 (0.0)  4 (0.1)  0 (0.0)  0.010   Birt–Hogg–Dubé syndrome  1 (0.0)  8 (0.1)  4 (0.0)  0.003   Catamenial pneumothorax  141 (2.5)  732 (12.1)  0 (0.0)  <0.001   Lymphangioleiomyomatosis  48 (0.9)  168 (2.8)  15 (0.1)  <0.001  Body mass index (kg/m2)        <0.001   Underweight (<18.5)  2832 (50.9)  1979 (32.8)  6459 (41.6)     Normal (18.5–24.9)  2314 (41.6)  3386 (56.1)  5927 (38.2)     Overweight (≥25.0)  43 (0.8)  275 (4.6)  1019 (6.6)     Missing data  373 (6.7)  397 (6.6)  2118 (13.6)    Smoking status        <0.001   Current or ex-smoker  590 (10.6)  1741 (28.8)  2259 (14.5)     Never smoker  4638 (83.4)  3801 (63.0)  11 563 (74.5)     Missing data  334 (6.0)  495 (8.2)  1704 (11.0)    Number of hospitalization  1.3 ± 0.7  1.3 ± 0.7  1.1 ± 0.5  <0.001  Treatments   Pleurodesis  129 (2.3)  428 (7.1)  1498 (9.6)  <0.001   Surgery  2466 (44.3)  2035 (33.7)  1099 (7.1)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 3 presents the comparison between CP and other types of pneumothorax in female patients of reproductive age. The average age of patients with CP was significantly higher than that of patients with other types of pneumothorax. The proportions of underweight patients and smokers were significantly lower in the CP group. All types of endometriosis and thoracic endometriosis were found in 28.8% and 9.4% of CP patients, respectively. In contrast, the 2 types were found in 1.4% and 0.0% of patients with other types of pneumothorax, respectively. In both groups, the right side was more commonly affected; however, the tendency was stronger in the CP group (CP: right, 64.9%; left, 6.5%; Other: right, 32.7%; left, 30.4%). Female patients with CP were more frequently hospitalized than other pneumothorax patients (mean number of hospitalizations: 1.6 ± 0.9 vs 1.3 ± 0.6, P < 0.001). Patients with CP were also more likely to undergo pleurodesis (8.5% vs 4.5%, P < 0.001) and surgical treatment for pneumothorax (57.1% vs 37.3%, P < 0.001) compared with patients with other types of pneumothorax. Patients with CP required multiple surgical treatments more frequently than did those with other types of pneumothorax (7.6% vs 5.4%, P < 0.001). Table 3: Characteristics and treatments of female patients of reproductive age, by type of pneumothorax (n = 11 599) Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). Table 3: Characteristics and treatments of female patients of reproductive age, by type of pneumothorax (n = 11 599) Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Variables  Catamenial pneumothorax  Other pneumothorax  P-value  Number of patients (%)  873 (7.5)  10 726 (92.5)    Age (years)  37.9 ± 7.7  31.3 ± 11.5  <0.001  Body mass index (kg/m2)  <0.001   Underweight (<18.5)  271 (31.0)  4540 (42.3)     Normal (18.5–24.9)  552 (63.2)  5148 (48.0)     Overweight (≥25.0)  20 (2.3)  298 (2.8)     Missing data  30 (3.4)  740 (6.9)    Smoking status      <0.001   Current or ex-smoker  168 (19.2)  2163 (20.2)     Never smoked  670 (76.7)  7769 (72.4)     Missing data  35 (4.0)  794 (7.4)    Endometriosis   Any endometriosis  251 (28.8)  154 (1.4)  <0.001   Thoracic endometriosis  82 (9.4)  4 (0.0)  <0.001  Affected side      <0.001   Right  567 (64.9)  3508 (32.7)     Left  57 (6.5)  3258 (30.4)     Bilateral  28 (3.2)  461 (4.3)     Missing data  221 (25.3)  3499 (32.6)    Number of hospitalizations  1.6 ± 0.9  1.3 ± 0.6  <0.001  Treatments   Pleurodesis  74 (8.5)  483 (4.5)  <0.001   Surgery  499 (57.1)  4002 (37.3)  <0.001    Diaphragm  55 (11.0)  23 (0.6)  <0.001    Pleura  10 (2.0)  1 (0.0)  <0.001    Exploratory  10 (2.0)  37 (0.9)  <0.001    Lung  447 (89.6)  3957 (98.9)  <0.001    Multiple surgeries  38 (7.6)  218 (5.4)  <0.001  Continuous data are presented as mean ± standard deviation; categorical data are presented as n (%). We performed a multivariate analysis to determine the predictors of CP in female pneumothorax of reproductive age (Table 4). Being aged >25 years (especially 36–45 years), having coexisting endometriosis, right-sided pneumothorax or a history of hospitalization for pneumothorax was associated with a higher probability of having a CP diagnosis; however, being under- or overweight or a current or ex-smoker was associated with a lower probability of having CP. Table 4: Multivariate analysis for predictors of catamenial pneumothorax in female pneumothorax of reproductive age Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  CI: confidence interval. Table 4: Multivariate analysis for predictors of catamenial pneumothorax in female pneumothorax of reproductive age Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  Variables  Odds ratio (95% CI)  P-value  Age (years) (vs under 20 years)   21–25  0.7 (0.4–1.1)  0.13   26–30  3.1 (1.9–5.1)  <0.001   31–35  4.4 (2.5–7.9)  <0.001   36–40  6.8 (3.8–12.0)  <0.001   41–45  6.8 (4.1–11.5)  <0.001   >45  3.8 (2.2–6.6)  <0.001  Body mass index (kg/m2; vs normal)   Underweight (<18.5)  0.8 (0.7–0.9)  0.003   Overweight (≥25.0)  0.5 (0.3–0.8)  0.003   Missing data  0.5 (0.3–0.8)  0.005  Smoking status (vs never smoked)   Current or ex-smoker  0.6 (0.5–0.8)  <0.001   Missing data  0.6 (0.3–1.0)  0.033  Coexistence of any endometriosis (vs none)   Yes  13.7 (10.4–18.1)  <0.001  Affected side (vs left side only)   Right  5.2 (2.8–9.7)  <0.001   Bilateral  1.4 (0.9–2.4)  0.17   Missing data  2.5 (1.8–3.6)  <0.001  Number of hospitalizations (vs single)   Multiple  2.5 (1.9–3.3)  <0.001  CI: confidence interval. DISCUSSION The present study investigated the differences in patients with spontaneous pneumothorax according to sex. The male-to-female ratio in the present study was 4.7:1, which was consistent with similar ratios reported previously [1–6]. The present study showed a biphasic age distribution for all patients and for male patients alone and a triphasic distribution for female patients. The reason the age distributions for both total and male data sets were biphasic was that men comprised more than 80% of the cohort. A previous epidemiological study reported that female idiopathic pneumothorax was most common in the 3rd decade of life and that secondary pneumothorax was more common in the 4th decade of life [3]. In a study of the French national inpatient database [6], age distributions for both men and women had 2 peaks; among women, the 1st peak of incidence was older than that of men and incidence remained stable up to 40 years. The present study confirmed that the age distribution of female pneumothorax had a distinct peak around 40 years, along with those at 18 and 79 years; the latter 2 peaks exactly matched those seen in men. The present study also showed that the peak age at which CP occurred was around 40 years, a finding that is consistent with those in previous reports [11, 13]. CP and pneumothorax related to lymphangioleiomyomatosis are unique to women of reproductive age. However, the present study showed that the number of these diseases diagnosed was insufficient to account for the peak that was observed around 40 years of age. The fact that the 3rd peak was maintained even without CP patients suggests that the number of CP cases was underdiagnosed in the data set. CP is clinically defined as recurrent pneumothorax that occurs within 72 h of menstruation [10]. Some patients with CP have histologically demonstrated ectopic endometrial tissues on their diaphragms or pleura, although others do not [10, 11, 13]. When histological findings of thoracic endometriosis are observed in a patient with pneumothorax, she is diagnosed as having thoracic endometriosis-related pneumothorax (TERP), regardless of whether the onset of pneumothorax met the criteria for CP [19]. Because these diagnoses are not stated definitely in the ICD-10, CP may have included both original CP and TERP in the present study. The present study showed that the proportion of CP was 3.1% among all cases of female pneumothorax. Previously reported proportions of either CP or TERP among all cases of female pneumothorax have varied widely. According to a Japanese nationwide study on female pneumothorax in 1986 [3], only 0.9% of all patients with pneumothorax and 2.8% of patients with secondary pneumothorax had CP. The most common cause of secondary pneumothorax at that time was pulmonary tuberculosis. In several previous studies of surgically treated patients [10–14, 19, 20], 21–23% of the female patients with pneumothorax had CP or TERP [12, 19]. Among the patients of reproductive age [10, 11, 13, 14], 24–41% had CP or TERP. In our latest nationwide study, CP accounted for only 7.5% of the cases of pneumothorax in female patients of reproductive age. Even considering that these previous studies only included surgically treated patients in expert centres and, therefore, selection bias was inevitable, CP might still be underdiagnosed by general clinicians. The present study showed a lower proportion of underweight patients with CP than with other types of pneumothorax. However, approximately one-third of the patients with CP were classified as underweight, a proportion that was much larger than that of healthy Japanese women (14.5%) of the same age [21]. This implies that patients with CP are generally thin, similar to patients with endometriosis [22]. In our study, 28.8% of CP patients had endometriosis. Previous studies have shown that the proportions of patients with pelvic endometriosis among those with CP or TERP vary broadly, with values ranging from 18.0% to 58.7% [19, 20]. We speculate that the diagnosis of endometriosis may have been under-reported, possibly because most of the clinicians who treated CP were not gynaecologists. CP is known to be a right side-dominant pneumothorax, and left-sided CP was very rare in previous small studies [11–13, 19]. Our results showed that left-sided CP was rare, but certainly existed. It should be noted that male pneumothorax had no apparent laterality, whereas female patients of reproductive age without CP still appeared to exhibit right-sided dominance. Thus, in the non-CP group, there may have been some patients with underdiagnosed CP. We found that patients with CP were more frequently hospitalized than were other female pneumothorax patients. In other words, they were more likely to experience recurrence than those with other types of pneumothorax. Surgical interventions were more frequently performed on CP patients to stop air leakage and/or to confirm histological diagnosis; however, repeat surgery was required more frequently in these patients than in patients with other types of pneumothorax. On the other hand, less than 10% of CP patients underwent chemical pleurodesis. Because our data included only chemical pleurodesis (by talc powder, minocycline or picibanil) but did not include autologous blood pleurodesis or pleural ablation, more procedures would have been performed than present results indicate. In Japan, pleurodesis treatment for pneumothorax is not actively performed, except in patients with high surgical risk; this is because many surgeons have concerns regarding difficulties associated with reoperation after pleurodesis [23]. For recurrent CP, some surgeons perform surgical pleurodesis using an absorbable mesh [24]. In Western countries, the standard practice is to perform pleurodesis after the second episode of pneumothorax, and talc pleurodesis has been reported to provide better results in terms of preventing recurrence of CP than pleural ablation [11]. The effectiveness of pleurodesis for CP should be reconsidered. Several previous studies have determined the predictors of CP and TERP. They identified several factors as predictors: age (>28 years [25] or ≥31 years [19]), never smoked [19] or not a current smoker [25], a history of pelvic endometriosis [19], right-sided pneumothorax [19], infertility and a history of pelvic surgery or uterine scraping [25]. Indeed, these factors were strong predictors, but because of the small sample size and because CP had likely been underdiagnosed, they may have missed other suspicious features of CP. We found that the independent predictors of CP in female pneumothorax of reproductive age were as follows: aged >25 years (especially 36–45 years), normal weight, never smoked, coexisting endometriosis, right-sided pneumothorax and a history of hospitalization for pneumothorax. To prevent underdiagnosis of CP, we should take detailed anamneses and pay particular attention to female pneumothorax patients presenting with these factors. Limitations Several limitations of the present study should be acknowledged. First, as the large database used in this retrospective study is primarily intended for the request of medical fees, some data may not have been as well-validated as data in planned prospective studies, particularly for categories such as comorbidity, the affected side and smoking status, which are not directly related to any medical fee. For example, the amount of missing data pertaining to the affected side was large (25.3–55.9%), and whether or not data were truly missing at random was unverifiable. Second, the database did not include outpatient data, and we were, therefore, unable to follow-up patients after discharge unless they were rehospitalized at the same institution for a second episode of pneumothorax. Assuming that this limitation was the same among the groups, it is acceptable to compare the mean numbers of hospitalizations instead of the tendency of recurrence. Because the proportion of CP in female pneumothorax may be underestimated in our study, we would like to leave the precise comparison between idiopathic spontaneous pneumothorax and CP to future research based on detailed medical history taking and accurate pathological diagnosis. CONCLUSION In conclusion, the present large-scale study using a national inpatient database showed a distinct difference in age distribution of pneumothorax according to sex. We confirmed that the age distribution of all female patients with pneumothorax had a distinct peak around 40 years of age, along with those at 18 and 80 years. CP had different characteristics and thus requires different treatment strategies compared with other types of pneumothorax in women of reproductive age. Funding This work was supported by the Ministry of Health, Labour and Welfare, Japan [H29-Policy-Designated-009 and H29-ICT-Genral-004]; Ministry of Education, Culture, Sports, Science and Technology, Japan [17H04141]; and the Japan Agency for Medical Research and Development. Conflict of interest: none declared. REFERENCES 1 Gupta D, Hansell A, Nichols T, Duong T, Ayres JG, Strachan D. Epidemiology of pneumothorax in England. Thorax  2000; 55: 666– 71. Google Scholar CrossRef Search ADS PubMed  2 Primrose WR. Spontaneous pneumothorax: a retrospective review of aetiology, pathogenesis and management. Scott Med J  1984; 29: 15– 20. Google Scholar CrossRef Search ADS PubMed  3 Nakamura H, Konishiike J, Sugamura A, Takeno Y. Epidemiology of spontaneous pneumothorax in women. Chest  1986; 89: 378– 82. Google Scholar CrossRef Search ADS PubMed  4 Sadikot RT, Greene T, Meadows K, Arnold AG. Recurrence of primary spontaneous pneumothorax. Thorax  1997; 52: 805– 9. Google Scholar CrossRef Search ADS PubMed  5 Guo Y, Xie C, Rodriguez RM, Light RW. Factors related to recurrence of spontaneous pneumothorax. Respirology  2005; 10: 378– 84. Google Scholar CrossRef Search ADS PubMed  6 Bobbio A, Dechartres A, Bouam S, Damotte D, Rabbat A, Regnard JF et al.   Epidemiology of spontaneous pneumothorax: gender-related differences. Thorax  2015; 70: 653– 8. Google Scholar CrossRef Search ADS PubMed  7 Tan J, Yang Y, Zhong J, Zuo C, Tang H, Zhao H et al.   Association between BMI and recurrence of primary spontaneous pneumothorax. World J Surg  2017; 41: 1274– 80. Google Scholar CrossRef Search ADS PubMed  8 Kawakami Y, Irie T, Kamishima K. Stature, lung height, and spontaneous pneumothorax. Respiration  1982; 43: 35– 40. Google Scholar CrossRef Search ADS PubMed  9 Melton LJ3rd, Hepper NG, Offord KP. Influence of height on the risk of spontaneous pneumothorax. Mayo Clin Proc  1981; 56: 678– 82. Google Scholar PubMed  10 Alifano M, Roth T, Camilleri-Broet S, Schussler O, Magdeleinat P, Regnard JF. Catamenial pneumothorax: a prospective study. Chest  2003; 124: 1004– 8. Google Scholar CrossRef Search ADS PubMed  11 Alifano M, Jablonski C, Kadiri H, Falcoz P, Gompel A, Camilleri-Broet S et al.   Catamenial and noncatamenial, endometriosis-related or nonendometriosis-related pneumothorax referred for surgery. Am J Respir Crit Care Med  2007; 176: 1048– 53. Google Scholar CrossRef Search ADS PubMed  12 Fukuoka M, Kurihara M, Haga T, Ebana H, Kataoka H, Mizobuchi T et al.   Clinical characteristics of catamenial and non-catamenial thoracic endometriosis-related pneumothorax. Respirology  2015; 20: 1272– 6. Google Scholar CrossRef Search ADS PubMed  13 Legras A, Mansuet-Lupo A, Rousset-Jablonski C, Bobbio A, Magdeleinat P, Roche N et al.   Pneumothorax in women of child-bearing age: an update classification based on clinical and pathologic findings. Chest  2014; 145: 354– 60. Google Scholar CrossRef Search ADS PubMed  14 Mehta CK, Stanifer BP, Fore-Kosterski S, Gillespie C, Yeldandi A, Meyerson S et al.   Primary spontaneous pneumothorax in menstruating women has high recurrence. Ann Thorac Surg  2016; 102: 1125– 30. Google Scholar CrossRef Search ADS PubMed  15 Yasunaga H, Matsui H, Horiguchi H, Fushimi K, Matsuda S. Clinical epidemiology and health services research using the diagnosis procedure combination database in Japan. Asian Pac J Dis Manage  2013; 7: 19– 24. 16 Hasegawa W, Yamauchi Y, Yasunaga H, Sunohara M, Jo T, Matsui H et al.   Clinical features of 280 hospitalized patients with lymphangioleiomyomatosis in Japan. Respirology  2015; 20: 160– 5. Google Scholar CrossRef Search ADS PubMed  17 Nagano M, Ichinose J, Sasabuchi Y, Nakajima J, Yasunaga H. Surgery versus percutaneous transcatheter embolization for pulmonary arteriovenous malformation: analysis of a national inpatient database in Japan. J Thorac Cardiovasc Surg  2017; 154: 1137– 43. Google Scholar CrossRef Search ADS PubMed  18 WHO Expert Consultation. Physical status: the use and interpretation of anthropometry. Report of a WHO Expert Consultation. WHO Technical Report Series Number 854. World Health Organization, Geneva, 1995, 312– 74. 19 Haga T, Kataoka H, Ebana H, Otsuji M, Seyama K, Tatsumi K et al.   Thoracic endometriosis-related pneumothorax distinguished from primary spontaneous pneumothorax in females. Lung  2014; 192: 583– 7. Google Scholar CrossRef Search ADS PubMed  20 Joseph J, Sahn SA. Thoracic endometriosis syndrome: new observations from an analysis of 110 cases. Am J Med  1996; 100: 164– 70. Google Scholar CrossRef Search ADS PubMed  21 2014 National Health and Nutrition Survey Japan. 2016. http://www.mhlw.go.jp/bunya/kenkou/kenkou_eiyou_chousa.html (25 January 2018, date last accessed). 22 Vigano P, Somigliana E, Panina P, Rabellotti E, Vercellini P, Candiani M. Principles of phenomics in endometriosis. Hum Reprod Update  2012; 18: 248– 59. Google Scholar CrossRef Search ADS PubMed  23 Kurihara M, Kataoka H, Ishikawa A, Endo R. Latest treatments for spontaneous pneumothorax. Gen Thorac Cardiovasc Surg  2010; 58: 113– 9. Google Scholar CrossRef Search ADS PubMed  24 Ikeda T, Sasaki M, Sakon K, Koshiji T. An effective method of pleurodesis involving absorbable mesh for repetitive catamenial pneumothorax. Eur J Cardiothorac Surg  2012; 42: 370– 2. Google Scholar CrossRef Search ADS PubMed  25 Rousset-Jablonski C, Alifano M, Plu-Bureau G, Camilleri-Broet S, Rousset P, Regnard JF et al.   Catamenial pneumothorax and endometriosis-related pneumothorax: clinical features and risk factors. Hum Reprod  2011; 26: 2322– 9. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2018. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

European Journal of Cardio-Thoracic SurgeryOxford University Press

Published: Mar 27, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off