Aspergillus After Lung Transplantation: Prophylaxis, Risk Factors, and the Impact on Chronic Lung Allograft Dysfunction

ABSTRACT

Background

Invasive pulmonary aspergillosis (IA) poses significant challenges for lung transplant (LTx) patients, with unclear risk factors and preventive strategies. The effectiveness of nebulized amphotericin B (AmB) or statins for IA prevention and the effect of IA on chronic lung allograft dysfunction (CLAD) and mortality remain questionable.

Methods

Data were collected from all LTx patients transplanted between December 1, 2013 and January 1, 2022 at the University Medical Center Groningen. IA, was defined according to published criteria. Prespecified risk factors were compared between patients with and without IA post‐LTx and were entered in a logistic regression model. Two additional logistic regression models were built with factors that might be associated with statin or AmB prophylaxis and IA. A matched case‐control study was conducted for the association between statins and IA, with matching based on follow‐up time.

Results

Aspergillus was cultured in 110 /274 (40%) patients post‐LTx and 89/110 (81%) were classified as probable IA. MMF use, airway stenosis, Aspergillus cultured pre‐LTx, CLAD, and acute rejection (AR), were significantly associated with IA. Statin use was associated with a lower incidence of IA, while AmB prophylaxis showed no significant effect. A significant statin effect could not be confirmed by the case control analysis. There was no significant difference in all‐cause mortality between patients with and without IA (34% vs. 29%).

Conclusions

The high incidence of IA post‐LTx necessitates more effective strategies. Key targets for intervention include prior positive cultures, airway stenosis, AR, and the use of MMF. The role of statins remains unclear and requires further research.

Out of 274 lung transplant (LTx) patients, 89 (32%) developed invasive aspergillosis IA. Factors associated with IA were mycophenolate mofetil use, airway stenosis, Aspergillus cultured pre‐LTx, chronic lung allograft dysfunction (CLAD), and acute rejection. The role of nebulized amphotericin B and statins for prevention of IA remains unclear.

Abbreviations

AmB

amphotericin B

AR

acute rejection

CLAD

chronic lung allograft dysfunction

COPD

chronic obstructive pulmonary disease

FEV1

forced expiratory volume in 1 second

FVC

forced vital capacity

IA

invasive aspergillosis

LTx

lung transplant/lung transplantation

1. Introduction

Invasive aspergillosis (IA) poses significant challenges for lung transplant (LTx) recipients, frequently leading to increased rates of morbidity and mortality [1, 2, 3]. The reported incidence of fungal infection in LTx patients ranges from 15% to 35% [4]. Managing IA in the LTx context is particularly complex due to potential drug–drug interactions between antifungal therapy and immunosuppressants. Recognizing the importance of prevention over treatment, there is an urgent need to delve deeper into the factors that are associated with Aspergillus infection.

There is considerable variation among various LTx centers in prophylactic therapy for IA, with differences in regimen and treatment duration [5]. Variable outcomes have been described regarding the efficacy of nebulized amphotericin B (AmB) and azoles as preventive measures. A systematic review and meta‐analysis from 2013 including 22 studies did not find a significant reduction in IA or Aspergillus colonization with universal anti‐Aspergillus prophylaxis [6]. However, more recent studies suggested that prophylaxis does indeed result in fewer cases of aspergillosis following LTx [7, 8, 9]. Targeted prophylaxis, for patients at high risk of IA, may be preferred over universal prophylaxis due to comparable IA rates and lower rates of adverse events [10]. Alternatively, a preemptive therapy approach, based on bronchoalveolar lavage (BAL) culture and galactomannan (GM), has shown a 50% reduction in antifungal exposure compared to universal prophylaxis [11]. Another potential strategy for preventing aspergillosis could be statin therapy. Statins inhibit ergosterol synthesis, a critical component of the fungal cell membrane, and therefore may play a role in fungal prophylaxis [12].

In addition, it remains unclear whether Aspergillus colonization and IA are associated with chronic lung allograft dysfunction (CLAD) [4, 8, 13–16]. Conversely, progression of CLAD probably also leads to an increased risk of Aspergillus colonization and IA. However, there is limited reporting on the course of lung function in LTx patients with Aspergillus colonization or IA.

To address the problem of IA following LTx, the primary goals of this study were to identify factors associated with IA following LTx and to explore potential protective modalities, including nebulized AmB and statin therapy. In addition, we evaluated the impact of IA on lung function, the development of CLAD, and mortality.

2. Material and Methods

2.1. Patients

The study was waived by the Ethical Committee based on its observational nature using routine care data (METc number 2024/292). The study complies with the ISHLT Ethics Statement. All patients provided written informed consent upon entry into the program. Data were retrospectively collected at the University Medical Center Groningen (UMCG), the Netherlands. All adult patients who underwent unilateral or bilateral LTx between December 2013 and January 2022 in the UMCG were eligible for inclusion.

2.2. Transplant Care

Follow‐up of all LTx patients took place at least every 3 months for monitoring of transplant function. A protocolized bronchoscopy is performed 1–3 months after transplantation. Subsequent bronchoscopies are only performed on clinical indication. All patients received standard induction therapy with basiliximab at Day 1 and Day 4 and maintenance immunosuppression most with tacrolimus, prednisolone, and mycophenolate mofetil (MMF). Alternative immunosuppressants were cyclosporine, azathioprine, everolimus, or sirolimus. Post‐LTx prophylactic therapy included: nebulized AmB 5 mg twice daily for Aspergillus in the cohort from December 2017 to January 2022, cotrimoxazole for Pneumocystis pneumonia, and valganciclovir (900 mg OD) or acyclovir 1000 mg for herpes viruses, depending on CMV (mis)match. Nebulized AmB was administered during hospitalization immediately post‐LTx (for the entire length of hospital stay, depending on tolerance and/or toxicity).

In our LTx program, statins were prescribed exclusively for hypercholesterolemia management, not as prophylaxis against aspergillosis.

2.3. Aspergillus Infection Criteria and Definitions

In accordance with prior studies and guidelines, the diverse manifestations of aspergillosis in LTx patients were delineated for the objectives of this study as follows [4, 17–19]:

Aspergillus airway colonization is characterized by the detection of Aspergillus in respiratory secretions (sputum or BAL) through culture, polymerase chain reaction (PCR), or GM testing, in the absence of symptoms, radiologic, and endobronchial changes [19].

Invasive pulmonary aspergillosis (IPA) is defined according to the European Organization for Research and Treatment of Cancer (EORTC) criteria [17, 19]. Proven pulmonary IA necessitates the presence of a host factor (LTx patients), a mycological criterion (Aspergillus in sputum or BAL via culture, PCR, or GM), and a clinical criterion (symptoms, radiological abnormalities, or endobronchial changes). For a diagnosis of proven extrapulmonary IA, histological evidence consistent with fungal tissue invasion is also required. Probable aspergillosis requires a host factor, a clinical criterion, and a mycological criterion. Possible aspergillosis is diagnosed when cases fulfil the criteria for a host factor and clinical criterion but lack a mycological criterion.

Tracheobronchial aspergillosis, a subtype of IA, is characterized by the presence of a host factor accompanied by a mycological criterion and a clinical criterion, which includes endobronchial changes observed during bronchoscopy or symptomatic presentation along with a decline in lung function.

2.4. Treatment Response Definitions

Treatment responses were defined as follows: “Complete response” clinical recovery and recovery of lung function, “Intermediate response,” not fully clinically recovered and/or not fully recovered lung function, and “Treatment failure,” lung function decline without any recovery or death.

2.5. Pulmonary Function

Spirometry pre‐IA was performed as part of standard care 0–3 months before infection. Baseline lung function is computed as the mean of the two best postoperative forced expiratory volume in 1 second (FEV1) values, measured > 3 weeks apart. Follow‐up measurements were performed at the routine visits of outpatients with IA. Follow‐up measurements took place 3 and 6 months after the onset of IA. Spirometry was performed according to ATS/ERS guidelines [20]. CLAD was defined as a persistent decline of 20% or more in FEV1 value from baseline value posttransplantation, independent of a change in forced vital capacity (FVC), according to the most recent ISHLT criteria [21].

Acute rejection (AR) was clinically defined as an acute deterioration of allograft function without an obvious other cause and/or histological confirmation.

2.6. Statistical Analysis

Clinical parameters were compared between patients with and without IA post‐LTx using Mann–Whitney U test or chi‐squared test, as appropriate. Prespecified risk factors and protective factors were entered into a multivariable logistic regression analysis model. Two additional logistic regression models were built to adjust for potential confounders regarding the use of statins and AmB prophylaxis. In the adjusted analysis for AmB, the following parameters were entered: AmB prophylaxis, ICU length of stay, pre‐LTx Aspergillus cultured, stenosis, AR, MMF, and age. In the adjusted analysis for statins, the following variables were included: statin use, low‐density lipoprotein (LDL), diabetes mellitus, age, airway stenosis, ICU length of stay, AR, MMF use, and the presence of cardiovascular risk factors.

We performed a before–after analysis of patients receiving nebulized AmB prophylaxis. Here, we compared the cohort from 2013 to December 2017, in which no AmB prophylaxis was used, with the cohort from December 2017 to January 2022, in which AmB prophylaxis was used.

Due to statins being initiated at various times post‐LTx, we also conducted a sensitivity analysis using a matched case‐control study, where the cases were matched based on follow‐up time from LTx to IA.

In addition, to assess differences in outcomes between patients with IA according to EORTC criteria or ISHLT criteria, we also conducted a sensitivity analysis only including patients with IA classified by the ISHLT criteria.

Paired sample T‐tests were used for within‐group analyses to compare FEV1 and FVC before, 3 months, and 6 months after start of IA. A Cox proportional hazards model with IA as a time‐dependent variable was used to compare post‐LTx survival and CLAD development between patients with and without IA.

A p value of less than 0.05 was considered significant. All analyses were carried out using IBM SPSS for Mac, version 24.0.

3. Results

3.1. Patient Characteristics

In total, 274 patients received an LTx between December 2013 and January 2022 and were included in the study. The median follow‐up time was 56 (IQR 32–88) months. Baseline characteristics and clinical parameters for all LTx patients with and without IA are shown in Table 1.

TABLE 1.

Baseline characteristics and clinical parameters of lung transplant patients with and without Aspergillus.

Variable	All Patients	IA	Non‐IA	p value *
Recipients, n (%)	274 (100)	89 (32)	185 (68)
Age at LTx, years	56 (46–61)	55 (45–61)	56 (48–61)	0.319
Gender, male (%)	137 (50)	44 (49)	93 (50)	0.897
Transplant indication, n (%)				0.183
COPD	122 (45)	44 (49)	78 (42)
Fibrosis	72 (26)	17 (19)	55 (30)
Pulmonary hypertension	35 (13)	11 (12)	24 (13)
Cystic fibrosis	19 (7)	10 (11)	9 (5)
Non‐CF bronchiectasis	1 (0.4)	0 (0)	1 (0.5)
Other	25 (9)	7 (8)	18 (9.7)
Bilateral LTx, n (%)	254 (93)	81 (91)	173 (94)	0.360
DM, n (%)	71 (26)	23 (26)	48 (26)	0.985
Body mass index, kg/m2	24 (21–28)	23 (21–26)	24 (21–28)	0.190
Statins, n (%)	89 (33)	19 (21)	70 (38)	0.006
Smoking, pack years	13 (0–30)	13 (0–30)	13 (0–30)	0.625
Aspergillus pre‐LTx, n (%)	58 (21)	27 (30)	31 (17)	0.010
AmB prophylaxes, n (%)	138 (50)	46 (52)	92 (50)	0.684
MMF	220 (80)	82 (92)	138 (75)	< 0.001
Acute rejection, n (%)	147 (54)	61 (69)	86 (47)	< 0.001
CLAD, n (%)	52 (19)3	27 (30)	25 (14)	< 0.001
Airway stenosis, n (%)	23 (8)	15 (17)	8 (4)	< 0.001

Note: Continuous variables are expressed as median (interquartile range).

Abbreviations: AmB, amphotericin B; CLAD, chronic allograft dysfunction; COPD, chronic obstructive pulmonary disease; DM, diabetes mellitus; IA, invasive aspergillosis; LTx, lung transplantation; MMF, mycophenolate mofetil.

^* p value for the difference between patients with invasive aspergillosis and patients without invasive aspergillosis post‐LTx.

3.2. Aspergillus Colonization and Classification of IA

Of the 274 patients in the study cohort, 110 patients had a positive Aspergillus culture. Of those patients, 89 (81%) were classified as having probable IA, according to the EORTC criteria (Table 1) and 21 (19%) were only colonized with Aspergillus and did not have evidence of invasive disease. There were no cases of possible or proven IA in our cohort.

Of the 89 patients with probable IA, 35% were diagnosed based on a combination of host factors, culture results, and radiological findings and 62% were diagnosed based on host factors, culture results, and evidence (based on bronchoscopy or pulmonary function) of Aspergillus tracheobronchitis. Three patients (3%) underwent treatment because of deep wound infections with IA.

3.3. IA

The median age of patients with IA was 55 years (IQR 45–61) and 44 (49%) were male. The median time between LTx and IA was 6 months (IQR 1–18). Prior to LTx, Aspergillus had been cultured from sputum in 27/89 (30%) of patients. Prior to Aspergillus infection, 23 (26%) of the patients had diabetes and 19 (21%) were on statins. During hospitalization for LTx, 46 (52%) of patients who developed IA received nebulized AmB prophylaxis, MMF was used in 82 (92%) of the patients as part of their immunosuppressive regimen. Total 10 of the 46 patients (22%) with AmB prophylaxis experienced a breakthrough infection while on nebulized AmB prophylaxis, while the remaining 36 (78 %) developed IA after prophylaxis was discontinued.

Positive Aspergillus culture occurred in 34 of the 89 patients with IA after fungal treatment (38%). This could involve a recurrence of aspergillosis, colonization, or treatment failure.

3.4. Aspergillus Species

The predominant cultured Aspergillus species was Aspergillus fumigatus, comprising 95 isolates (35%), followed by A. niger with 9 isolates (3%), and A. flavus with 4 isolates (2%) (Figure S1). Aspergillus was isolated from BAL specimens in 51% of cases, from sputum cultures in 46%, and from wound infections in 3%. GM testing in BAL samples was performed in 33 patients (30%), yielding positive results in 20 patients (61%). Azole resistance prevalence was 15%.

3.5. Antifungal Treatment

Antifungal treatment was administered in all patients with IA. The therapeutic modalities are shown in Table S1. Predominantly, 61 patients (69%) received monotherapy with voriconazole, while 9 patients (10%) received combination therapy consisting of voriconazole and intravenous liposomal‐AmB. In addition, six patients (7%) received nebulized AmB monotherapy. The median duration of treatment was 12 weeks (IQR 6–14). Adverse events are shown in Table S2. Notably, the most frequently reported adverse effects comprised of nausea/vomiting (16%), visual disturbances (9%), diarrhoea (8%), rash (8%), and hepatotoxicity (8%). Thirteen patients (15%) prematurely discontinued therapy due to adverse reactions.

Complete treatment response occurred in 52/89 (58%), intermediate response in 20/89 (23%), and treatment failure in 17/89 (19%).

3.6. Risk Factors for IA

We compared predefined risk factors for IA between patients with and without IA post‐LTx (Table 1). In multivariate analysis, MMF use (OR = 7.02; 95% CI [2.57–19.19]), airway stenosis (OR = 6.20; 95% CI [2.16–18.00]), Aspergillus cultured pre‐LTx (OR = 2.71; 95% CI [1.24–5.96]), CLAD (OR = 3.08; 95% CI [1.37–6.89]), and AR (OR = 2.42; 95% CI [1.29–4.52]), were significantly associated with IA (Table 2).

TABLE 2.

Clinical parameters associated with invasive aspergillosis post lung transplantation.

	OR	95% CI	p value
Unilateral	1.69	0.52–5.52	0.401
Cystic fibrosis	1.80	0.47–7.05	0.401
Age	1.02	0.99–1.04	0.184
Pre‐LTx Aspergillus	2.71	1.24–5.96	0.013
AmB prophylaxis	1.01	0.55–1.88	0.968
MMF	7.02	2.57–19.19	< 0.001
Statins	0.38	0.19–0.76	0.006
Stenosis	6.2	2.16–18.00	< 0.001
Acute rejection	2.42	1.29–4.52	0.006
CLAD	3.08	1.37–6.89	0.006

Note: data derived by multivariable logistic regression.

Abbreviations: AmB, amphotericin B; CLAD, chronic lung allograft dysfunction; LTx, lung transplantation; MMF, mycophenolate mofetil.

The additional sensitivity analysis of patients with IA classified according to the ISHLT criteria included 51 of the original 89 patients with IA. The difference was due to the absence of bronchoscopy in these patients. The analysis showed consistent findings: pre‐LTx Aspergillus colonization, AR, CLAD, airway stenosis, MMF use, and statin use were all associated with post‐LTx IA. However, the associations of statins and AR with IA were no longer significant, likely due to reduced statistical power as a result of the smaller patient sample size in this analysis (Table S3).

3.7. Nebulized AmB Prophylaxis

In total, 138 (50%) of all LTx patients received prophylaxis with nebulized AmB. In the IA group, 46 patients (52%) had received prophylaxis with nebulized AmB, and in the non‐IA group 92 patients (50%). We performed a before–after analysis of patients receiving nebulized AmB prophylaxis. The median duration of AmB prophylaxis was 26 days (IQR 18–36). Nebulized AmB was not associated with a decreased risk of IA (Table 1) in both the crude and adjusted analyses (adjusted OR = 0.96; 95% CI [0.54–1.70]).

3.8. Statins

Eighty‐nine (33%) of all patients received statins. Of those, 53 (60%) were started on statins pre‐LTx. The median time between LTx and the start of statins was 20 months (IQR 7.0–37.5).

Statin use was associated with a decreased risk of IA in both the crude and adjusted analyses (adjusted OR = 0.43; 95% CI [0.22–0.86]). However, the same association studied in the matched case‐control analysis for post‐LTx follow‐up time, showed no significant association between statin use and IA (adjusted OR = 0.95, 95% CI 0.46–1.96) (Table S4).

3.9. LTx Function in Patients With IA

Spirometry data were available in 77 (87%), 83 (93%), and 79 (89%) patients before they developed IA, 3 months after IA, and 6 months after IA, respectively. Spirometry results are shown in Figure 1 and Table S5. There was a significant decline in FEV1 (p < 0.001) but not in FVC (p = 0.948) at the time of IA. After antifungal treatment FEV1 and FVC increased at first follow‐up within 3 months and even more at 6 months follow‐up (Figure 1 and Table S5).

FIGURE 1.

(a) FEV1 before, during, and after IA; (b) FVC before, during, and after IA.

Of the 89 patients with IA, 61 (69%) had at least one episode of AR. Of those with AR, 34 (56%) had an episode of AR prior to IA and 27 (44%) after IA. By comparison, in the 185 patients without IA the AR incidence was 47%.

3.10. CLAD

During the study period, 52 out of 274 patients (19%) developed CLAD. Total 25 (14%) of the patient without IA developed CLAD, whereas 27 (30%) of the patients with IA developed CLAD during the study period (Table 1). Of those 27 patients with IA and CLAD, 7 developed CLAD before IA, and 20 developed CLAD after IA. Of those seven patients who had CLAD before IA, four developed progression of their CLAD after IA (Figure 2).

FIGURE 2.

Chronic lung allograft syndrome pre‐ and postinvasive aspergillosis.

In the patients with IA, the median time between the infection and CLAD was 17 (IQR 4–37) months.

A Cox proportional hazards model with IA as a time‐dependent covariate showed that IA was significantly associated with the development of CLAD (HR 2.712, p < 0.001).

3.11. All‐Cause Mortality

All‐cause mortality, during the median follow‐up period of 4.6 year, was 30% (83/274). In patients with IA, the mortality rate was 34 % (30/89), compared to 29 % (53/184) in those without IA. One‐ and two‐year survival rates in LTx patients with Aspergillus infection were 87% and 73%, respectively. In the Cox proportional hazards model with consideration for the time between LTx and IA, no statistically significant difference in mortality was observed between LTx patients with IA and those without IA (HR 1.547; p = 0.067).

3.12. Aspergillus Colonization Subgroup Analysis

Of the 274 included patients, 21 (19%) were classified as having Aspergillus colonization. A subgroup analysis was conducted to juxtapose clinical parameters, AmB prophylaxis, and the utilization of statins between patients with Aspergillus colonization and those with negative sputum cultures. Notably, within the Aspergillus colonization subgroup, a higher incidence of CLAD was observed (6/15; 40%) in comparison to the culture‐negative group (19/145; 13%) (p = 0.032). Nonetheless, upon multivariable analysis, no significant disparities in CLAD occurrence or other parameters were determined between patients with Aspergillus colonization and patients who were culture‐negative.

4. Discussion

In this single‐center, observational study with a long follow‐up time, we show that IA remains an important complication of LTx. We confirm the use of MMF, AR, airway stenosis, and a positive Aspergillus culture pre‐LTx as important risk factors of IA. CLAD was strongly associated with IA, but CLAD mostly occurred after IA. Furthermore, we also found a possible protective effect of statin therapy in the prevention of IA, although residual confounding cannot be completely ruled out. Aspergillus colonization and IA are associated with lung function decline and CLAD, while all‐cause mortality does not seem to be directly related.

Of the previously reported risk factors, we could not confirm single LTx as a risk factor for IA, probably due to the low percentage of single LTx procedures (< 10%) in our center. MMF was associated independently with IA, consistent with the literature [22]. A possible explanation could be that MMF enhances A. fumigatus‐induced oxidative burst of polymorphonuclear neutrophils, but without a corresponding increase in fungal killing [22]. AR is also associated with IA. This association could potentially be attributed to the impact of AR on the airways of LTx patients, which might favor easy colonization and growth of Aspergillus. However, it is also plausible that it may stem from the intensity of immunosuppression and the administration of high‐dose methylprednisolone (1000 mg for 3 days in our center) as treatment for AR. From a pathophysiological perspective, high‐dose methylprednisolone inhibits the release of pro‐inflammatory cytokines and autophagy by alveolar macrophages. Furthermore, the use of high‐dose corticosteroids is a well‐established risk factor for invasive fungal disease [23]. Next, airway stenosis following LTx seems to be associated with IA. However, this observation should be interpreted with caution as it involves small number of patients. A correlation between Aspergillus infection and the subsequent emergence of clinically significant endobronchial abnormalities has been previously reported in a cohort of LTx patients [24].

An interesting observation from our study is the possible protective effect of statins against the development of IA, although not confirmed in our sensitivity analysis. This corresponds with another observational study which suggested that statin use was independently associated with a reduced risk of IA [12]. Statins function as competitive inhibitors targeting 3‐hydroxy‐3‐methylglutaryl coenzyme A reductase. This enzyme catalyzes the conversion of HMG‐CoA to mevalonate, a crucial step in the synthesis of cholesterol in humans and ergosterol in fungi. Ergosterol is a crucial component of the cell membrane in fungi, and they cannot survive without it [12, 25]. However, the effect of statins on IA may be influenced by timing, as statins were initiated at different points during posttransplant follow‐up. Moreover, the median duration between LTx and Aspergillus infection was 6 months. Therefore, it could be that these patients did not have sufficient time to receive statins when compared to patients without IA (confounding by indication). To account for this, we conducted a matched case‐control analysis in which we could not confirm the protective effect of statins on IA. Yet, the drawback of this approach is that it reduces the overall follow‐up duration. Consequently, we cannot conclusively determine from this study whether statins truly protect against IA. Still, considering the variable outcomes of AmB prophylaxis, the potential drug interactions of azoles with immunosuppressants, and the rationale behind the mechanism of action of statins, administering statins as antifungal prophylaxis to post‐LTx patients remains an interesting hypothesis, and could be tested in a randomized controlled trial [6].

In our center, we have a low threshold for treating IA with antifungal treatment for 3 months. It is possible that we may be overtreating patients. However, with this treatment strategy, we do not observe a substantial difference in mortality between patients with and without aspergillosis. Our 1‐year survival rates in LTx patients with IA (87%) were comparable to other studies with 1‐year survival rates between 81% and 84 % [4, 13].

Despite the recovery of lung function in the first months following the IA diagnosis in LTx patients, we observed a significant increase in the development of CLAD compared to patients without IA. Of course, it is not possible to exclude other contributing factors to the development of CLAD, given the long median duration between LTx and IA (17 months).

Aspergillosis has been demonstrated as a risk factor for CLAD in multiple studies, suggesting that the presence of Aspergillus in the bronchi may lead to epithelial injury, followed by dysregulation of repair mechanisms ultimately responsible for chronic fibroblast proliferation and progressive allograft dysfunction [4, 8, 13–15]. In our study, most CLAD cases (74%) occurred after the IA diagnosis instead of before. This suggests that IA might lead to CLAD, rather than CLAD leading to more IA, also described before [23]. We were unable to fully assess CLAD as a risk factor in our study because the model included patients who developed CLAD both before and after IA. Only seven patients had CLAD prior to IA, which is expected given that the median time from LTx to IA and CLAD were 6 and 31 months, respectively.

Our study has several limitations. First, it remains challenging to differentiate between Aspergillus colonization and invasive disease. In previous studies, Aspergillus colonization was reported much more frequently [4, 7, 19]. It could be that in other transplant centers more posttransplant surveillance bronchoscopies with fungal cultures were performed, thereby increasing the likelihood of diagnosing Aspergillus colonization. However, given that Aspergillus colonization may lead to CLAD, it suggests the presence of some form of invasive disease alongside colonization [14, 15]. For clinical practice, the distinction has to be based on clinical symptoms, lung function, microbiological, and imaging findings. Since bronchoscopy may not always be feasible, clinicians may need to rely on sputum culture or antigen testing. When symptoms are significant or lung function declines without another clear cause, it may be justified to initiate treatment in these immunocompromised patients. This decision always involves balancing the anticipated benefits against the potential toxicity and side effects.

It is also possible that Dutch LTx patients have a higher risk of developing IA due to the high population density and intensive Dutch agricultural sector, which is associated with a high prevalence of azole resistance (in our study 15%). In this study, we primarily used the EORTC criteria, instead of the ISHLT criteria, to define IA, relying on lung function decline as a surrogate marker for endobronchial involvement. This approach is practical given the high prevalence of positive Aspergillus cultures (40% in LTx patients in our center) and the challenges of applying ISHLT criteria, which necessitates bronchoscopy but lack clear evidence mandating it for diagnosis [19]. The referenced study underpinning these criteria do not explicitly mandate bronchoscopy for the diagnosis of IA [26]. Importantly, bronchoscopy often does not alter management, as patients with abnormal CT findings, declining lung function, and positive sputum cultures are typically treated without BAL confirmation.

This may explain the higher incidence of IA in our study, as other centers may rely on ISHLT criteria. To evaluate whether there is a difference in outcomes between patients classified by EORTC or ISHLT criteria, we conducted a sensitivity analysis selecting the patients who fulfilled ISHLT criteria for IA. No differences were found in the factors associated with IA (Table S3).

A second limitation is that we compared the group of patients with IA with those without IA, because we base our decision to treat on this diagnosis. The latter group also included the 21 patients with Aspergillus colonization. We chose not to exclude these patients with Aspergillus colonization from analysis and to include them in the group without IA because this analysis best aligns with our primary study question and is most appropriate for daily clinical practice, where a distinction is made between treated and untreated patients. In addition, in our subgroup analysis, we observe that the group comprising patients with Aspergillus colonization closely resembles the group with negative cultures. Third, it is challenging to investigate the causative effects of AmB and statins on IA in an observational study, due to the possibility of residual confounding (e.g. significant variability in the duration of prophylactic AmB use, severity of IA, and confounding by indication for statin use). Finally, not all patients underwent chest CT scans or bronchoscopies for diagnosis, potentially leading to some patients being erroneously classified as having Aspergillus colonization.

5. Conclusions

The observed incidence of IA highlights the need for more effective strategies to prevent IA, and thus, the development of CLAD. Potential intervention targets include prior positive cultures, airway stenosis, the use of MMF, and possibly antifungal prophylaxis with statins. Prospective trials are urgently needed to address the effectiveness of available preventive therapies in reducing the burden of IA in LTx patients.

Author Contributions

Johanna P. van Gemert: conceptualization, investigation, writing – original draft, methodology, writing – review and editing, formal analysis. Ger Jan Fleurke: investigation, conceptualization, methodology, software. Onno W. Akkerman: conceptualization, investigation, supervision. C.Tji. Gan: conceptualization, investigation, supervision. Willie N. Steenhuis: investigation, software. Huib A.M. Kerstjens: conceptualization, investigation, supervision, methodology. Erik A.M. Verschuuren: conceptualization, investigation, methodology, supervision. Douwe F. Postma: formal analysis, conceptualization, investigation, methodology, writing – original draft, supervision.

Supporting information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Supporting Information

Gemert J. P. van, Fleurke G. J., Akkerman O. W., et al. “ Aspergillus After Lung Transplantation: Prophylaxis, Risk Factors, and the Impact on Chronic Lung Allograft Dysfunction.” Transplant Infectious Disease 27, no. 4 (2025): 27, e70020. 10.1111/tid.70020

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.