Investigating the efficacy and safety of bronchoscopic removal of aspergilloma in PTB-CCPA Authors: Hua Li, Haihao Xiao, Weiming Fang, Duohua Su, Tianyi Li, Yu Li, Ling Zhang, Wenjie Su, Ling Wu, Jiahua Chen, Yi Xie, Yikai Xie, Dehu Peng, Ning Su, Jinxing Hu, Yanbin Zhang, Chunmei Tang Published: 2026 ABSTRACT AbstractBackgroundGiven high symptom recurrence rates after discontinuation of antifungal therapy (AF), a comparison was made between AF alone and in combination with bronchoscopic aspergilloma removal (BRA) for treating post-tuberculosis chronic cavitary pulmonary aspergillosis (PTB-CCPA) patients, focusing on fungal ball clearance rate, relapse rate, and relapse risk factors.MethodsThis retrospective cohort study was conducted at Guangzhou Chest Hospital, China, from January 2014 to December 2021, with follow-up through October 2023. Patients with PTB-CCPA who met diagnostic criteria (cavitary tuberculosis history, CT-confirmed fungal balls, and microbiological/serological Aspergillus evidence) were included.ResultsThe enrolled patients were divided into two groups: the control group (32 patients, 36.4%) received only AF, and the treatment group (56 patients, 63.6%) received a combination of AF and BRA (AF+BRA). The aspergilloma clearance rate in the cavity was 14.8% in the AF group and 100% in the AF+BRA group (p < 0.001). For patients who received AF less than 8 months, Kaplan-Meier survival curve analysis showed that the AF+BRA group had a significantly higher cumulative recurrence-free survival rate (p < 0.05). Notably, AF+BRA shortened the duration of oral AF compared with the AF treatment. Further analysis identified age ≥60 years (HR = 0.12, 95% CI 0.03–0.57, p = 0.007) and the disappearance or reduction in size of the aspergillomas (HR = 0.07, 95% CI 0.01–0.58, p = 0.014) as protective factors against recurrence, while more than one pulmonary cavity increased recurrence risk (HR = 7.4, 95% CI 1.67–32.69, p = 0.008).ConclusionsAF+BRA can eliminate Aspergillus fungal balls, may cure PTB-CCPA, shorten antifungal treatment duration, and reduce recurrence and mortality, suggesting that it is a superior treatment approach. Introduction Chronic Pulmonary Aspergillosis (CPA), caused by Aspergillus, leads to cough, hemoptysis, and has a 5-year mortality of 32% [1,2]. Half of cases link to post-TB cavities, with 10% progressing to post-tuberculosis chronic cavitary pulmonary aspergillosis (PTB-CCPA) [3–8]. This stems from inhalation of Aspergillus spores, that colonize the pulmonary cavities and form fungal balls by the aggregation of fungal hyphae, inflammatory cells, fibrin, and sloughed epithelial cells, which adheres to the cavity wall, exhibiting movements within the cavity during repeated coughing and contraction episodes [9]. Aspergilloma leads to chronic infection accompanied by cavity wall fibrosis, pleural thickening, and loss of lung function. In severe cases, this can result in life-threatening massive hemoptysis [2]. While antifungal therapy (AF) temporarily relieves symptoms, it fails to eradicate Aspergillus (relapse rate: 17-41%), often requiring lifelong medication [10], which imposes a significant physical and economic burden [4,11,12]. Surgical resection removes infected tissue, it carries 1–23% mortality risk and is often contraindicated in patients with poor lung function [13–16]. Based on this, we hypothesized that the fungal ball can be effectively cleared via bronchoscopy. What’s more, multiple reports have demonstrated that bronchoscopic removal of the aspergilloma (BRA) can effectively clear the fungal ball from the cavity [17–21], and the ESCMID/ERS 2016 clinical guideline for chronic pulmonary Aspergillosis recommends that patients not suitable for surgery should consider bronchoscopy for fungal ball removal [22]. However, to date, no large-scale study has evaluated the efficacy and safety of this method. Over the past 20 years, the physicians at our hospital have observed that BRA, when combined with AF, yields better outcomes than AF alone and established a safe and practical method. Therefore, we conducted a retrospective analysis to compare the two treatment methods: AF alone and AF combined with BRA, by evaluating the fungal ball clearance rate, patient relapse rates, and investigating relapse risk factors, to guide the application of this method in the treatment of PTB-CCPA patients. Methods Study population 328 PTB-CCPA patients (2014–2021) at the Guangzhou Chest Hospital met diagnostic criteria (cavitary tuberculosis history; respiratory symptoms for ≥3 months; CT-confirmed fungal balls; and microbiological/serological evidence of Aspergillus) [23,24]. Exclusion criteria: recent corticosteroid use (≤3 months), AIDS, prolonged antifungal therapy (>3 weeks during the 6 months preceding study enrollment), concurrent pulmonary infections, other aspergillosis forms, such as invasive aspergillosis, allergic bronchopulmonary aspergillosis (ABPA), or chronic fibrosing pulmonary aspergillosis (CFPA), and pregnancy. Sample size reflected eligible cases during study period. The study was conducted in accordance with the Declaration of Helsinki. The study protocol was reviewed and approved by the Ethics Committee of Guangzhou Chest Hospital (Ethics No. Chest Med Ethics (2023)35). Given the retrospective nature of the study and the use of anonymized patient data, the requirement for individual informed consent was waived by the ethics committee. The committee determined that the study posed minimal risk to patient privacy and confidentiality, and that obtaining informed consent would not be feasible due to the retrospective design and the use of historical data. All patient information was anonymized and de-identified before analysis. Patient management and outcome measurements The patients’ demographic data, pulmonary diseases, clinical symptoms and laboratory examination results were recorded. Oral voriconazole constituted the primary antifungal agent for the most patients, calculated according to the patient’s weight as per the instructions. The exceptions were one patient in the AF+BRA group who sequentially received oral itraconazole and voriconazole on alternate days, and another patient who started with itraconazole for 6 months, with the entire treatment course lasting 12 months. The chest CT scans were reviewed before and after treatment, and during follow-up. Seven patients were excluded from the CT data analysis because their CT scans were performed in other hospitals and no electronic records were available. The size of the cavities and aspergillomas was calculated depending on the maximum diameter on the axial CT images. For patients who received AF+BRA treatment, bronchoscopy (OLYMPUS BF-Q260/OLYMPUS BF-Q290/BF-P290/BF-XP290) was conducted under sedation to obtain biopsy specimens and bronchoalveolar lavage fluid to determine the diagnosis. Comorbidities were assessed for surgical suitability. The surgery was generally performed 2–4 weeks after the patient began oral drug therapy, and if bronchoscopic examination showed mild local inflammation and low bleeding risk. All procedures were performed under routine sedation achieved with intravenous midazolam or propofol administered by an anesthesiologist. Patient vital signs (blood pressure, heart rate, oxygen saturation) were continuously monitored. Local anesthesia should be administered with lidocaine throughout the procedure. The procedures were generally well-tolerated under this sedation regimen. In most patients, the bronchoscope was inserted via the nose, with oral insertion used for special cases. The aspergilloma was fragmented using biopsy forceps. The fragments were then dislodged and mobilized through a combination of forceps grasping, irrigation with saline lavage, and application of negative pressure suction through the bronchoscope. If the drainage bronchus or lumen of the cavity was narrowed, hindering the bronchoscope’s entry, a balloon catheter of appropriate size was employed to dilate the bronchus. The surgical procedure lasts approximately 30 min and does not exceed 120 min. No procedure was terminated prematurely due to patient intolerance. Staged bronchoscopic removal approach was performed every 4–6 weeks until complete clearance. In the last bronchoscopic examination, the cavities were smooth and clean. Consequently, considering the results of CT examination scans and symptoms, the treatment was terminated. The process and outcomes of the bronchoscopic treatment were retrospectively recorded by the bronchoscopists. The fungal clearance rate at the end of treatment was defined as the disappearance of the original aspergilloma within the target cavity, as assessed by imaging within one month after treatment. Specifically, for the AF group, evaluation was based on CT imaging, while for the AF+BRA group, it was based on direct visualization during the final bronchoscopy. In this retrospective study, recurrence was identified through two pathways: (1) patients returning for medical evaluation due to symptoms such as cough or hemoptysis; (2) aspergilloma regrowth within the cavity was detected on CT scans, which were performed either during scheduled follow-up or for other clinical reasons. All recurrence diagnoses were confirmed by pulmonologists after excluding other respiratory diseases. To ascertain the patients’ final outcomes, research nurses conducted a final telephone follow-up from May 1 to October 31, 2023, which included 72 patients. The follow-up aimed to: (1) confirm the patient’s survival status; (2) screen for respiratory symptoms suggestive of possible recurrence (e.g. new or worsening cough or hemoptysis); (3) inquire whether the patient’s recent CT results were normal; and (4) invite and encourage all surviving patients to return to our hospital for a repeat CT scan. Recurrence duration was calculated from the start of treatment refer to other similar studies, for the patients with different duration of treatment. The primary outcome was fungal clearance rate, and the secondary outcome was time to recurrence. Statistical analysis Analyses used SPSS v26.0 and R (survminer, ggplot2, forestploter). Data presented as mean (SD), median (IQR), or % (95% CI). Group comparisons: Mann-Whitney U/t-tests (continuous), chi-square (categorical). Kaplan-Meier curves evaluated mortality and recurrence time, with between-group comparisons performed using the log-rank test. Binary logistic regression identified recurrence risk factors. P-value < 0.05 was considered statistically significant. Results Comparison of the baseline characteristics of patients between the two groups of PTB-CCPA patients A total of 328 patients with PTB-CCPA were enrolled, of whom 88 patients completed the treatment and had main primary data were included in the final analysis (Figure 1). The baseline characteristics at the time of diagnosis are presented in Table 1. The AF+BRA group had fewer elderly patients (23.2% vs. 46.9%, p = 0.022) but more severe hemoptysis (p = 0.002) and less respiratory symptoms (p < 0.05). Among the 60 patients with positive pathological results, 53 patients (94.6%) in the AF+BRA group. The AF group had a larger median cavity size than the AF+BRA group (p = 0.016). The proportion of PTB-CCPA patients with more than one pulmonary cavities, bilateral aspergilloma and fibrosis were higher in the AF+BRA group than in the AF group (p = 0.040, p = 0.011, p = 0.001). Except for one patient in the AF+BRA group who received oral itraconazole and voriconazole in succession, and another who was initiated with itraconazole, all other patients received oral voriconazole throughout. The treatment duration was slightly shorter in the AF+BRA group compared with the AF group (Supplementary Figure 1). The proportion of patients who received AF treatment for less than 10 months was higher in the AF+BRA group (p = 0.007). There were no other significant differences between the two groups. Study Flowchart. Table 1.Comparison of characteristics of patients with Post-Tuberculosis chronic Cavitary pulmonary aspergillosis according to the treatment methods. All patients (N = 88)AF (N = 32)AF+BRA (N = 56)P-valueCharacteristics Age, years53.0 (38.0– 62.0)58.5 (34.0–64.5)51.5 (38.5– 59.0)0.243Age (≥60)28 (31.8%)15 (46.9%)13 (23.2%)0.022Sex 0.427Male75 (85.2)26 (81.3)49 (87.5) Female13 (14.8)6 (18.7)7 (12.5) Current or ex-smoker45 (51.1)15 (46.9)30 (53.6)0.545Smoking time (years)30 (20–30)30 (20–30)30 (10–30)0.420History of pulmonary tuberculosis Times of pulmonary tuberculosisn = 86* (97.7)n = 31 (96.9)n = 55 (98.2) 151 (59.3)22 (71.0)29 (52.7)0.098≥235 (40.7)9 (29.0)26 (47.3) Pulmonary tuberculosis to diagnosis PTB-CCPA (years)6 (3–10.2); n = 855.5 (3–10); n = 306.5 (3–11); n = 550.8661-540 (47.1)15 (50.0)25 (45.4)0.7595–1024 (28.2)9 (30.0)15 (27.3) >1021 (24.7)6 (20.0)15 (27.3) Other comorbidities Bronchiectasis34 (38.6)11 (34.4)23 (41.1)0.535Chronic obstructive pulmonary disease7 (8.0)3 (9.4)4 (7.1)0.710Pulmonary bulla4 (4.5)1 (3.1)3 (5.4)0.629Respiratory symptom Degree of cough 0.048None4 (4.5)4 (12.5)0 (0.0) Moderate52 (59.1)16 (50)36 (64.3) Severe32 (36.4)12 (37.5)20 (35.7) Degree of sputum <0.001None3 (3.4)3 (9.4)0 (0.0) Moderate68 (77.3)18 (56.3)50 (89.3) Severe17 (19.3)11 (34.4)6 (10.7) Hemoptysis <0.001No16 (18.2)13 (43.8)3 (5.4) Hemosputum30 (34.1)6 (18.8)24 (42.8) Hemoptysis42 (47.7)13 (43.8)29 (51.8) No. of hemoptysis episodes per weekn = 79 (89.8)n = 31 (96.9)n = 48 (85.7)0.045021 (26.6)13 (41.9)8 (16.6) 141 (51.9)13 (41.9)28 (58.4) >117 (21.5)5 (16.2)12 (25.0) Amount of maximum hemoptysis, mL10 (1–50); n = 875 (0-17.5); n = 3220 (5–90); n = 550.002Fever14 (15.9)6 (18.8)8 (14.3)0.582Shortness of breath16 (18.2)7 (21.9)9 (16.1)0.497Chest distress3 (3.4)0 (0.0)3 (5.4)0.183Pain in the chest8 (9.1)2 (6.3)6 (10.7)0.483Microbiological or serological evidence of PTB-CCPA Sputum fungal culture10 (11.4)5 (15.6)5 (8.9)0.456Bronchoalveolar lavage fluid fungal culture7 (8.0)1 (3.1)6 (10.7)0.205Pathology60 (68.2)7 (21.9)53 (94.6)<0.001Positive Aspergillus antigen in bronchoalveolar lavage fluid26/33 (78.8)13/18 (72.2)13/15 (86.7)0.312Positive Aspergillus galactomannan in serum10/45 (22.2)5/21 (23.8)5/24 (20.8)0.810Positive Aspergillus IgG antibody31/42 (73.8)12/15 (80.0)19/27 (70.4)0.496Positive 1,3-beta-D-glucan (BDG) in serum3/17 (17.6)0/7 (0.0)3/10 (30.0)0.110Chest CT findingsn = 81 (92.0)n = 32 (100.0)n = 49 (87.5) Cavity Number 0.040161 (75.3)28 (87.5)33 (67.4) 2–420 (24.7)4 (12.5)16 (26.5) Bilateral13 (16.0)3 (9.4)10 (20.4)0.186Located in the lung segment 0.157Right upper63 (64.9)28 (77.8)35 (57.3) Left upper30 (30.9)8 (22.2)22 (36.1) Right dorsal2 (2.1)0 (0.0)2 (3.3) Left dorsal2 (2.1)0 (0.0)2 (3.3) Size (mm)38.0 (25.0–57.5)49.0 (27.8–75.0)33.0 (24.0–45.0)0.016Aspergillomas Number of fungal balls 0.153174 (91.4)31 (96.9)43 (87.8) 27 (8.6)1 (3.1)6 (12.2) Bilateral4 (4.9)0 (0.0)4 (8.2)0.011Located in the lung segment 0.092Right upper56 (65.9)25 (78.1)31 (58.5) Left upper28 (32.9)7 (21.9)21 (39.6) Right dorsal1 (1.2)0 (0.0)1 (1.9) Size (mm)19.5 (10.0–27.8)19.0 (10.0–26.0)20.0 (10.0–29.0)0.749The average CT density of pulmonary aspergilloma13.5 (−7.3-26.0)10.0 (−24.3-20.8)15.3 (−3.7-28.3)0.154Aspergilloma size/cavity size0.63 (0.33–0.87)0.49 (0.23–0.79)0.71 (0.35–0.90)0.062Pleural thickening77 (95.1)31 (96.9)46 (93.9)0.543Fibrosis74 (91.4)25 (78.1)49 (100)0.001Consolidation23 (28.4)12 (37.5)11 (22.4)0.117Treatment Drug use duration (months)8.0 (6.0–11.0)9.1 (6.4–12.8)8.0 (6.4–9.4)0.265≤10 months62 (70.5)17 (53.1)45 (80.4)0.007>10 months26 (29.5)15 (46.9)11 (19.6) Bronchial artery embolization12 (13.6)3 (9.4)9 (16.1)0.379Data are presented as median (interquartile range) or n (%); PTB-CCPA, post-tuberculosis chronic cavitary pulmonary aspergillosis; AF, antifungal therapy; AF+BRA, antifungal therapy combined with bronchoscopic removal of aspergilloma. *Two individuals have a history of tuberculosis, but the number of occurrences is unclear. Comparison of characteristics of patients with Post-Tuberculosis chronic Cavitary pulmonary aspergillosis according to the treatment methods. Data are presented as median (interquartile range) or n (%); PTB-CCPA, post-tuberculosis chronic cavitary pulmonary aspergillosis; AF, antifungal therapy; AF+BRA, antifungal therapy combined with bronchoscopic removal of aspergilloma. *Two individuals have a history of tuberculosis, but the number of occurrences is unclear. Characteristics of BRA The characteristics of the BRA treatment are listed in Supplementary Table 1. Bronchoscopy revealed that the aspergillomas or fungal masses were completely cleared in these patients as shown in Supplementary Figure 2. Supplementary Figure 3 shows the pathological results before and after BRA treatment. Intra-procedural bleeding was the most common complication, occurring in 33 patients (58.9%). The median blood loss was 10 mL (IQR 7.5–20.0). In all cases, bleeding was controlled with local instillation of ice-cold saline and/or topical application of adrenaline (0.1 mg/mL) via the bronchoscope. No patient experienced hemodynamic compromise (defined as a sustained drop in systolic blood pressure >20 mmHg from baseline or requiring vasopressor support) or required transfusion. There were no procedure-related deaths, pneumothorax, or respiratory failure. Treatment outcomes of PTB-CCPA The post-treatment CT findings and follow-up data regarding recurrence and mortality are shown in Table 2. It was observed that the aspergilloma clearance rate in the AF+BRA group was 100%, which was significantly higher than AF group (14.8%, p < 0.001). The proportion of patients with thinner cavity walls in the AF+BRA group (55.3%) was higher than that in the AF group (3.7%) (p < 0.001) after treatment. The median follow-up duration was 34.9 months (IQR 19.0–54.0) for the AF group and 55.5 months (IQR 23.3–77.3) for the AF+BRA group (p = 0.029). Seven patients were lost to follow-up. Six patients (6.8%) had died of PTB-CCPA, with 4 deaths in the AF group (12.5%) and a significantly lower cumulative mortality rate in the AF+BRA group (p = 0.034, Figure 2A). During the follow-up period, the recurrence rate of PTB-CCPA was 48.1%, with recurrence detected in 39 patients. The median time to recurrence was 22.3 months (IQR 16.6–42.1), with eight patients experiencing recurrence twice. There were no significant differences in recurrence-free survival rates between the two groups (Supplementary Figure 4A). When the antifungal treatment duration was ≤6 months, ≤8 months, ≤10 months, the recurrence rates in the AF+BRA group consistently showed lower recurrence rates. Kaplan-Meier survival curve analysis showed that in the subgroups with antifungal treatment durations less than 6 months or 8 months, the AF+BRA group had higher recurrence-free survival rates than the AF group (p = 0.018, Supplementary Figure 4B-C; p = 0.036, Figure 2(B)). We further divided the patients into four subgroups based on antifungal treatment duration (< 8 and ≥8 months for both AF and AF+BRA groups), the recurren·ce rates for these groups were 66.67%, 38.46%, 40.62%, and 45.83%, respectively. Kaplan–Meier survival curve analysis showed that the cumulative recurrence-free survival rate in the AF < 8 group was lower than that in the AF > 8 group (p = 0.024), while there were no significant differences between AF+BRA < 8 and AF+BRA > 8, or between AF > 8 and AF+BRA < 8 (Supplementary Figure 4D-F). The patient with the longest follow-up (7 years post-treatment) showed no recurrence of aspergillomas on CT scans (Supplementary Figure 5). Kaplan-Meier Estimates of Cumulative Survival and Recurrence-Free Survival of Patients with Post-Tuberculosis Chronic Cavitary Pulmonary Aspergillosis. Survival rate of all patients and two groups (AF+BRA) (A), recurrence-free survival in the subgroup with antifungal treatment duration < 8 months for all patients and two groups (B). AF, antifungal therapy; AF+BRA, antifungal therapy combined with bronchoscopic removal of aspergilloma. Table 2.Comparison of outcomes of patients with Post-Tuberculosis chronic Cavitary pulmonary aspergillosis according to the treatment methods. All patients (N = 88)AF (N = 32)AF+BRA (N = 56)P-valueHigh-resolution CT chest finding after treatmentn = 74 (84.1)n = 27 (84.4)n = 47(83.9) Cavity No changes41 (55.4)19 (70.4)22 (46.8)0.093Disappeared1 (1.3)1 (3.7)0 (0.0) Smaller9 (12.2)2 (7.4)7 (14.9) Larger23 (31.1)5 (18.5)18 (38.3) Cavity wall <0.001No changes46 (62.1)25 (92.6)21 (44.7) Disappear1 (1.4)1 (3.7)0 (0.0) Thinning26 (35.1)1 (3.7)25 (53.2) Thickening1 (1.4)0 (0.0)1 (2.1) Aspergillomas <0.001No changes9 (12.2)9 (33.3)0 (0.0) Disappear51 (68.9)4 (14.8)47 (100.0) Smaller14 (18.9)14 (51.9)0 (0.0) Follow-upn = 81 (92.0)n = 31 (96.9)n = 50 (89.3) Follow-up duration (months)47.6 (22.3–72.3)34.9 (19.0–54.0)55.5 (23.3–77.3)0.029No. of deaths from PTB-CCPA6 (8.3)4 (13.8)2 (4.7)0.137No. of deaths from other3 (4.2)1 (3.4)2 (4.7)0.858Recurrence rate39 (48.1)15 (48.4)24 (48.0)0.973Recurrence time (months)22.3 (16.6–42.1)22.7 (16.8–32.7)22.0 (16.0–44.8) Cumulative recurrence rates of 1 year4/81 (4.9)2/31 (6.5)2/50 (4.0)0.6212 years22/81 (27.2)9/31 (29.0)13/50 (26.0)0.7663 years28/81 (34.6)12/31 (38.7)16/50 (32.0)0.5374 years32/81 (39.5)13/31 (41.9)19/50 (38.0)0.725Recurrence rates by subgroups of drug-use duration 6 M8/18 (44.4)5/8 (62.5)3/10 (30.0)0.1688 M23/41 (56.1)10/15 (66.7)13/26 (50.0)0.30010 M32/56 (57.1)11/17 (64.7)21/39 (53.8)0.45012 M35/63 (55.5)13/21 (61.9)22/42 (52.4)0.473Recurrence times =131 (38.3)14 (45.2)17 (34.0)0.238≥ 28 (9.9)1 (3.2)7 (14.0) Surgical treatment after relapse1 (1.2)0 (0.0)1 (2.0)0.200Data are presented as median (interquartile range) or n (%); PTB-CCPA, post-tuberculosis chronic cavitary pulmonary aspergillosis; AF, antifungal therapy; AF+BRA, antifungal therapy combined with bronchoscopic removal of aspergilloma. Comparison of outcomes of patients with Post-Tuberculosis chronic Cavitary pulmonary aspergillosis according to the treatment methods. Data are presented as median (interquartile range) or n (%); PTB-CCPA, post-tuberculosis chronic cavitary pulmonary aspergillosis; AF, antifungal therapy; AF+BRA, antifungal therapy combined with bronchoscopic removal of aspergilloma. To evaluate the association of the characteristics on recurrence, we performed subgroup analyses based on these variables in 81 patients with follow-up data (Supplementary Table 2). It was determined the recurrence rate in the AF+BRA group (0.0%) was significantly lower than in the AF group (54.5%) (p = 0.037) for patients with severe sputum production. Moreover, for patients aged ≥ 60 years, the recurrence rate in the AF+BRA group (16.7%) was lower than in the AF group (40.0%), and for patients with hemoptysis > 10 mL, the recurrence rate in the AF+BRA group (42.9%) was lower than in the AF group (62.5%). Notably, there were no significant differences in other parameters in other subgroup analyses. When the antifungal treatment duration was ≤6 months, ≤8 months, the recurrence rates in the AF group (62.5%, 66.7%) were higher than AF+BRA group (30.0%, 50.0%). Results of the Kaplan-Meier survival curve analysis showed that in the subgroups with antifungal treatment durations less than 6 months or 8 months, the AF+BRA group had higher recurrence-free survival rates than the AF group (p = 0.018, p = 0.036) (Supplementary Figure 4B and 2B). Subsequently, the patients were divided into four subgroups based on antifungal treatment duration (< 8 and ≥8 months for both AF and AF+BRA groups), Kaplan-Meier survival curve analyses showed that the cumulative recurrence-free survival rate in the AF < 8 group was lower than that in the AF > 8 group (p = 0.024), but no significant differences were recorded between AF+BRA < 8 and AF+BRA > 8, or between AF > 8 and AF+BRA < 8 (Supplementary Figure 4). Identification of recurrence risk factors in PTB-CCPA patients In further analyses, we grouped the patients based on whether they had recurrence, and compared the characteristics (Supplementary Table 3). Among the patients who experienced recurrence after treatment, the proportion of patients under 60 years of age (p = 0.018), PTB-CCPA patients with more than one pulmonary cavity (p = 0.032), patients with larger cavities (p = 0.095), patients with a treatment duration of less than 10 months (p = 0.027) and patients with no reduction in size of the aspergilloma (p = 0.066) was higher than in the non-recurrent group. Other factors were not significantly different (p > 0.1) between recurrent and non-recurrent patients. Based on the above results, a binary logistic regression model was established containing the five variables. The model was used for predicting the risk factors of recurrence in PTB-CCPA patients (Figure 3). Notably, older age (≥60 years) and reduction or disappearance of the aspergilloma were identified as protective factors against recurrence. Conversely, PTB-CCPA patients with more than one pulmonary cavities had a significantly higher risk of recurrence after initial successful treatment, compared to patients with only a single cavity. Binary Logistic Regression Analysis of Risk Factors for Recurrence in Patients with Post-Tuberculosis Chronic Cavitary Pulmonary Aspergillosis after Treatment. HR, Hazard Ratio. Discussion In our retrospective analysis, AF+BRA achieved 100% aspergilloma clearance vs. 14.8% with AF alone. Defining cure as complete Aspergillus eradication, AF+BRA achieved 100% cure. The clearance or reduction in size of the aspergillomas was a protective factor for PTB-CCPA recurrence. This is consistent with findings from a retrospective study conducted in 2021, which identified aspergillomas in the cavity as a risk factor for poor outcomes after CCPA treatment [25]. AF+BRA reduces recurrence risk through aspergilloma clearance and faster symptom relief, although this is a retrospective study and did not specifically compare the timing of symptom improvement, the results demonstrated that the proportion patients who exhibited cavity wall thinning after treatment was higher in the AF+BRA. We hypothesize that BRA may prevent Aspergillus proliferation by removing the aspergilloma, reducing secondary inflammation in the surrounding lung tissue and secretions, which further alleviates symptoms, thereby accelerating the healing process. This may also partially explain the finding that AF+BRA significantly reduced the recurrence rates in patients with severe sputum production and hemoptysis. The overall recurrence rate (48.1%) in the entire study patient cohort is consistent with that reported previously [11,26,27]. In our study, we found that age ≥60 years is a protective factor for PTB-CCPA recurrence, which is consistent with the results of a previous study which uncovered that the age of recurrent patients was lower than that of non-recurrent patients [p = 0.050] [28]. Further examination showed that the AF+BRA group had higher proportion of patients under the age of 60, higher proportion of PTB-CCPA patients with more than one pulmonary cavities, higher rates of sputum production, hemoptysis, number of hemoptysis episodes, and maximum bleeding volume compared to the AF group, which implies a more severe disease and more recurrence. The AF+BRA group had a higher proportion of patients who received AF treatment for durations shorter than 10 months compared to the AF group. Evidence from numerous prior studies has shown that a short duration of AF treatment is a risk factor for recurrence in CCPA patients, which suggests that the AF+BRA group may have a higher recurrence risk due to shorter treatment times [11,28–30]. Thus, patients in the AF+BRA group achieved satisfactory treatment outcomes with a shorter duration of AF. Given that the follow-up time was proportional to the recurrence rate, the AF+BRA group had a longer follow-up time. Despite higher theoretical risk factors, AF+BRA showed equivalent recurrence rates to AF alone (48.0% vs 48.4%), suggesting that the BRA combined with AF reduced the recurrence risk in PTB-CCPA. Our analysis uncovered a significant difference in treatment duration between the two groups, and thus, we performed a subgroup analysis based on the duration of AF. Kaplan-Meier survival curve analysis revealed that when the treatment duration was less than 6 or 8 months, the cumulative recurrence-free survival rate in the AF+BRA group exceeding that in the AF group. This suggests that the combination of BRA with AF can significantly prolong the time to recurrence in PTB-CCPA for patients with a treatment duration of less than 8 months. However, when patients with treatment durations longer than 8 months were included, the difference in recurrence-free survival curves between the two groups were not significant. Prospective studies have shown that extending AF treatment duration to 12 months can significantly prolong the recurrence-free survival time in CCPA patients [p < 0.0001] [10]. Therefore, we speculate that combining BRA with AF may effectively shorten the AF treatment duration in PTB-CCPA patients without affecting recurrence rates. To verify this hypothesis, we divided the patients into four subgroups based on whether the AF treatment duration was less than 8 months, and compared their recurrence-free survival curves. The results were consistent with those reported previously: the AF group exhibited a significantly higher cumulative recurrence-free survival rate with a longer treatment duration. However, for the AF+BRA group, prolonging the treatment duration failed to confer additional benefit regarding the recurrence rates. Of note, no significant difference was observed in the cumulative recurrence-free survival rate between the AF+BRA group with a treatment duration of less than 8 months and the AF group with a treatment duration greater than 8 months. To eliminate potential influencing factors, we analyzed the baseline data and other treatment processes across the four subgroups, and found no statistically significant differences. This further supports our hypothesis that the combination of AF and BRA therapy can effectively shorten the AF treatment duration in PTB-CCPA patients without altering the recurrence rates. Furthermore, although our study was not designed to perform a formal cost-effectiveness analysis, the significantly reduced duration of antifungal treatment in the AF+BRA group suggests a substantial potential for long-term cost savings, primarily by avoiding the expenses associated with lifelong antifungal medication and monitoring. In our cohort, the BRA procedure demonstrated favorable patient acceptability, evidenced by high tolerability and compliance rates. This can be explained by patients’ preference for a treatment strategy offering a prospect of definitive cure and a shorter duration of systemic medication. It is noteworthy that pre-procedural patient concerns, particularly regarding hemoptysis risk, were effectively mitigated by our standardized protocol which included a period of initial antifungal therapy and consideration of bronchial artery embolization for high-risk cases. For now, BRA has not been widely evaluated or applied globally, which primarily due to the high risk of massive bleeding occurring during the procedure, which can potentially lead to death, suggesting that appropriate patient selection and operator’s technical skills are crucial factors that determine the treatment outcome. The inflammation and secondary vascular malformation surrounding the cavity wall, induced by the aspergilloma, are recognized as the key pathological basis for massive hemoptysis [31,32]. Based on our experience, the management of bleeding risks during bronchoscopic intervention for pulmonary tuberculous cavities complicated with aspergilloma must be systematic: First, high-risk factors should be identified, including a history of massive hemoptysis, enhanced CT indicating rich percavity vascularity, active bleeding or obvious mucosal hyperplasia observed during bronchoscopy, and a platelet count < 60 × 109/L; Adequate preoperative preparation is essential, involving adjustment of anticoagulant medications (e.g. discontinuing clopidogrel 5–7 days in advance) and conducting coagulation function tests. For patients showing severe local inflammation and congestion under bronchoscopy, it is recommended to administer systemic antifungal therapy for 1–2 months first and proceed with the intervention only after the inflammation subsides. For extremely high-risk patients, prophylactic BAE may be considered(Supplementary Figure 6). Among the 56 patients in the AF+BRA group, nine patients (16.1%) underwent pre-procedural bronchial artery embolization (BAE) due to a high risk of bleeding (e.g. history of significant hemoptysis, visible vascular malformations on imaging). These BAE procedures were all performed prior to the first BRA session. To minimize intraoperative complications and reduce the risk of mortality, the duration of each BRA session was shortened, and multiple sessions were conducted instead of attempting to remove the fungal ball in a single procedure. Intraoperative haemostasis follows a stepwise strategy: for minor bleeding, ice-cold saline can be injected via the bronchoscope along with suction to promptly clear the blood and prevent airway obstruction; for active bleeding, local spraying of diluted epinephrine or thrombin is employed, combined with systemic haemostatic agents, and for significant bleeding, blood transfusion and fluid infusion may also be administered; in the event of massive haemorrhage, the patient should be immediately placed in the lateral decubitus position on the affected side to maintain airway patency, and a combination of local agents, endobronchial balloon compression, and systemic medication should be used. If these measures prove ineffective, emergency BAE or surgical haemostasis is required (Supplementary Figure 6). Overall, the bleeding risk associated with this treatment was well controlled, and 41.1% of patients had fewer than 3 sessions, indicating its great potential for widespread adoption. Except for the few patients who were not suitable for bronchoscopic intervention due to physical reasons or when the cavities were located in distal airways preventing bronchoscope access, the operation was successful in most patients, achieving complete Aspergillus removal from the cavities, effectively improving the symptoms. Therefore, it must be stressed that the successful and safe application of this technique hinges on its performance by rigorously trained and highly skilled endoscopists in tertiary-grade hospitals with full resuscitation capabilities. Our findings delineate reduction or disappearance of the aspergilloma were identified as protective factors against recurrence, and BRA can effectively eliminate the aspergillomas from cavities. Therefore, we strongly believe that BRA has great potential to shorten the duration of AF therapy, improve survival rates, and reduce recurrence rates in PTB-CCPA patients. To our knowledge, this is the first systematic evaluation of the efficacy of BRA in the treatment of PTB-CCPA. However, there are some limitations which need to be discussed. Firstly, this was a retrospective cohort study conducted at a specialized hospital, the baseline data and drug treatment durations were not well-controlled or standardized, which may have interfered with the analysis of recurrence and mortality rates. The AF+BRA group exhibited markers of more severe disease at baseline and shorter course of drug therapy. Consequently, their superior outcomes may powerfully underscore the efficacy of the combined intervention, but we cannot rule out that they may also reflect a clinical tendency to select higher-risk patients for more aggressive management. Secondly, although we performed a retrospective analysis of all patients treated over an 8-year period at the largest tuberculosis specialty hospital in southern China, the number of patients with complete treatment data was relatively small, which affected the statistical power of our analyses and the accuracy of the results. Thirdly, the long-term follow-up part of this study relied on telephone interviews for initial screening and patient recall. Although a considerable proportion of patients (e.g. 21 out of 32 non-recurring patients in the AF+BRA group) provided subsequent CT re-examination data, not all patients underwent uniform, planned imaging follow-up. This model, which depends on symptom triggers or patient willingness for re-examination, may fail to identify asymptomatic radiological recurrences, potentially leading to some overestimation of the long-term recurrence-free survival rate. Future prospective studies should establish fixed time points for imaging re-evaluations. Therefore, further multi-center, prospective studies should be performed to optimize the clinical utility of this treatment approach. Conclusions Compared with single AF therapy, a combination of AF and BRA therapy can rapidly and completely clear aspergillomas from the cavities of PTB-CCPA patients, reduce inflammation in the cavity walls, alleviate patient suffering, improve the quality of life, prevent massive hemoptysis caused by PTB-CCPA, shorten AF treatment duration, lower recurrence rates, and improve survival rates. In our hospital, bronchoscopy for aspergilloma removal did not cause significant adverse effects on patients. Therefore, we anticipate that this technique will gain broad application by many clinicians to improve patient outcomes. Supplementary Material Supplemental Material