Epidemiology of Invasive Mold Infections in Lung Transplant Recipients

Abstract

Invasive mold infections (IMIs) are a major source of morbidity and mortality among lung transplant recipients (LTR) yet information regarding the epidemiology of IMI in this population are limited. From 2001–2006, multicenter prospective surveillance for IMIs among LTR was conducted by the Transplant-Associated Infection Surveillance Network. The epidemiology of IMI among all LTR in the cohort is reported. Twelve percent (143/1173) of LTRs under surveillance at 15 U.S. centers developed IMI infections. The 12-month cumulative incidence of IMIs was 5.5%; 3-month all-cause mortality was 21.7%. s caused the majority (70%)of IMIs; non-infections (39; 27%) included: (5), mucormycosis (3), and “unspecified” or “other” mold infections (31). Late-onset IMI was common: 52% occurred within one year post-transplant (median 11 months, range 0–162 months). IMIs are common late-onset complications with substantial mortality in LTRs. LTRs should be monitored for late-onset IMIs and prophylactic agents should be optimized based on likely pathogen.

Background

Lung transplantation is a life-saving intervention for patients with end stage lung disease. In the United States in 2011, 1,849 lung and heart/lung transplant procedures were performed (1). Despite important advances in surgical technique, immunosuppressive regimens, and the development of novel antifungal agents in recent years, lung transplant recipients remain at substantial risk for development of invasive fungal infections (IFIs) (2). However, few data are available on the overall burden of IFIs in this population. Historical studies have been primarily limited to small retrospective investigations (2–5) from which conflicting results have been reported (6, 7) and limited conclusions can be drawn.

To better understand the burden of IFIs and their associated outcomes among transplant recipients, the Centers for Disease Control and Prevention and partners formed the Transplant-Associated Infection Surveillance Network (TRANSNET), a multicenter consortium designed to perform prospective surveillance for IFIs among selected major transplant centers in the United States. TRANSNET provides the most comprehensive epidemiologic investigation of IFIs in the solid organ transplant population to date (8). Per overall TRANSNET analysis, the 12-month cumulative incidence of invasive mycoses in lung and heart-lung transplant recipients was 8.6% and invasive mold infections (IMIs) accounted for 70% of all IFIs in this transplant population. This was in contrast to the heart, kidney and liver solid organ transplant populations for which 35%, 21%, and 18% of all IFIs respectively were caused by molds (8). Given the importance of IMIs in the lung transplant population, we used the TRANSNET prospective data to further describe the epidemiology of IMIs in lung transplant recipients.

Methods

Study design

The study was conducted in accordance with U.S. Good Clinical Practice regulations and guidelines; human subject approval or waiver was obtained at each institution from which data was reported to TRANSNET.

Surveillance was conducted prospectively among 15 solid organ transplant centers in the United States from March 2001 through March 2006 (8). A standardized case report form was used to collect data on all cases that developed an IFI during the surveillance period, regardless of when the transplant occurred. Data collected on cases included demographic information, transplant date and type, method of diagnosis, comorbid conditions and co-infections, immunosuppressive and antifungal use, and 3-month follow-up status. An IMI was defined as proven or probable by the European Organization for Research and Treatment of Cancer/Mycoses Study Group (EORTC/MSG) criteria (9). Demographic data, transplant information, and limited follow-up data were also collected on all patients who underwent transplantation at study sites during the surveillance period (incidence cohort). Because lung transplant patients are generally considered at higher risk for IMIs than other organ transplant recipients (8), any combination of solid organ transplants that included lung were included in this analysis (for example, a patient who was a recipient of both kidney and lung allografts). For patients who developed more than one IMI, only the first IMI was used for incidence calculations.

Statistics

All analyses were conducted using SAS version 9.3 (SAS Institute Inc., Cary, NC). Twelve-month cumulative incidence of first IMI for lung transplant patients was estimated accounting for the competing risks of death, relapse, and re-transplantation; CI estimates were calculated using the ‘cmprsk’ package, v. 2-2-2, in R (v. 2-14-1). Bivariate statistics were calculated using student’s t-test, Chi-square or Fisher’s exact tests, as appropriate. P-values < 0.05 were considered statistically significant.

Results

During the five year surveillance period, we identified 1173 lung transplant recipients under surveillance at 11 of 15 participating transplant centers, and 143 (12.2%) developed IMIs. Most patients (107, 74.8%) developed one IMI; 28 (19.5%) developed 2 IMIs, 5 (3.5%) developed 3 IMIs, and 3 (3.5%) developed 4 or more. Fifty-three (37.1%) IMIs were classified as “proven” according to EORTC/MSG criteria. The most common pathogen identified was (104; 73%); non-infections (39; 27%) included: (5; 3.5%), mucormycosis (3; 2%), and “unspecified” or “other” mold infections (31; 21.7%) (Figure 1). infections were most commonly due to (n=54), followed by (n=10), (n=9), (n=4), (n=1), unknown species (n=8), and multiple unidentified species (n=18). The 12-month CI of invasive aspergillosis (IA) and non-IA infections was 4.13% and 1.35%, respectively (Figure 2).

Figure 1.

Infecting mold pathogens in 143 episodes of invasive mold infections occurring in lung transplant recipients.

Figure 2.

Twelve month cumulative incidence by type of first invasive mold infection (top line invasive aspergillosis; bottom line non-aspergillosis invasive mold infection).

Median age at time of IMI diagnosis was 55 years; most patients were white (125, 87.4%) and first-time transplant recipients (134, 93.7%). Chronic obstructive pulmonary disease was the most common indication for transplantation (59, 41%), followed by pulmonary fibrosis (25, 17.5%) and cystic fibrosis (21, 14.7%). The majority of patients (117, 81.8%) had IMI limited to the pulmonary system; disseminated infection (8, 5.6%) and skin (5, 3.7%), combined pulmonary/sinus (3, 2%) and sinus (2, 1.4%) involvement occurred less frequently. Fungemia and bone involvement were rare, with one case of each reported. Dyspnea (84, 58.8%), cough (75, 52.5%), and increased sputum production (55, 38.5%) were the most common respiratory symptoms within the first 7 days of IMI diagnosis; less than one third of patients presented with fever (44, 30.8%). Extrapulmonary symptoms were less common, with less than 5% of IMI-infected patients experiencing central nervous system, sino-nasal, or cutaneous symptoms.

However, there were significant differences in site of involvement and presentation between IA and non-IA mold infections. Compared with IA infections (Table 1), non-IA mold infections occurred more frequently in men (65.8% vs. 44.7%, p=0.036) and presented more commonly as a cutaneous infection (14.3% vs. 0%, p<0.05) while IA was more likely to be limited to the pulmonary system (93.1% vs. 62.9% p <0.001). Patients with IA were more likely to have dyspnea (65.4% vs. 41%, p<0.01) and cough (57.7% vs. 38.5%, p<0.05) while non-IA trended to have more papular or skin nodules (15.4% vs. 0%, p=NS).

Table 1.

Characteristics of TRANSNET lung transplant patients who developed invasive mold infection

Characteristic, n (%)	IA (n = 104)	Non-IA* (n = 39)	P value
Age in years, median (range)	55 (20–70)	54 (18–68)	0.464
Male sex	46 (44.7)	25 (65.8)	0.036
Race			0.231
White	95 (95.0)	30 (88.2)
Black/African-American	5 (5.0)	4 (11.8)
Number of months post-transplant to IMI diagnosis, median (range)	10.5 (0–162)	16 (0–83)	0.205
Underlying disease prompting transplant
COPD/emphysema	41 (39.4)	18 (46.2)	0.467
Cystic Fibrosis	17 (16.4)	4 (10.3)	0.360
Pulmonary Fibrosis	19 (18.3)	6 (15.4)	0.686
Sarcoidosis	4 (3.8)	0 (0)	0.575
Other†	23 (22.3)	11 (28.9)	0.446
Prior transplant	7 (6.7)	2 (5.1)	1.000
Combination transplant**	3 (2.8)	2 (5.1)	0.614
Anatomical site of IMI involvement
Pulmonary only	95 (93.1)	22 (62.9)	<0.001
Sinus only	0 (0)	2 (5.7)	0.073
Sinus and pulmonary	2 (1.9)	1 (2.9)	1.000
Disseminated	4 (3.9)	4 (11.4)	0.214
Skin	0 (0)	5 (14.3)	0.001
Blood only	0 (0)	1 (2.9)	0.273
Bone	1 (1)	0 (0)	1.000
Symptoms present within 7 days of IMI diagnosis
Dyspnea	68 (65.4)	16 (41.0)	0.008
Cough	60 (57.7)	15 (38.5)	0.040
Increased sputum production	44 (42.3)	11 (28.2)	0.123
Fever	34 (32.7)	10 (25.6)	0.542
Chest pain	15 (14.4)	3 (7.7)	0.399
Hemoptysis	12 (11.5)	2 (5.1)	0.351
Sino-nasal congestion/pain	3 (2.9)	5 (12.8)	0.035
Papular or nodular skin lesions	0 (0)	6 (15.4)	<0.001
Central nervous system signs/symptoms	6 (5.8)	2 (5.1)	1.000
Three-month mortality	23 (22.1)	8 (20.5)	0.836
IMI listed as contributing cause of death	12 (52.2)	4 (50)	0.647
Assessed at the time of IMI diagnosis
Renal insufficiency	34 (32.7)	16 (42.1)	0.431
Diabetes	45 (43.3)	12 (30.8)	0.017
Neutropenia	2 (1.9)	0 (0.0)	1.000
Assessed within 90 days prior to IMI
Organ rejection	22 (21.2)	7 (18.0)	0.817
Prophylactic antifungal therapy
Amphotericin B, inhaled	8 (7.7)	4 (10.3)	0.736
Fluconazole	9 (8.7)	1 (2.6)	0.286
Itraconazole, oral	21 (20.2)	4 (10.3)	0.164
Itraconazole, intravenous	2 (1.9)	2 (5.1)	0.300
Voriconazole	2 (1.9)	4 (10.3)	0.047
Ketoconazole	2 (1.9)	0 (0)	1.000
Empiric antifungal therapy
Amphotericin B, intravenous	2 (1.9)	2 (5.1)	0.560
Fluconazole	5 (4.8)	0 (0)	0.396
Itraconazole, oral	4 (3.9)	0 (0)	0.551
Voriconazole	4 (3.9)	0 (0)	0.551
Caspofungin	1 (1)	0 (0)	1.000
Antifungal treatment for IMI
Amphotericin B, intravenous	28 (26.9)	9 (23.1)	0.640
Amphotericin B, inhaled	19 (18.3)	5 (12.8)	0.438
Fluconazole	0 (0)	1 (2.6)	0.273
Itraconazole, oral	25 (24.0)	7 (18.0)	0.437
Itraconazole, intravenous	3 (2.9)	2 (5.1)	0.614
Voriconazole	58 (55.8)	14 (35.9)	0.034
Caspofungin	30 (28.9)	6 (15.4)	0.099

IA, invasive aspergillosis; COPD, Chronic Obstructive Pulmonary Disease

^*scedosporiosis, fusariosis, mucormycosis, other mold infections, and unspecified mold infections (fungal organisms observed on histopathology but not cultured)

^†Kartagener’s Syndrome, alpha-1 antitrypsin deficiency, pulmonary hypertension, Eisenmenger syndrome, and idiopathic lung failure

^**4 lung/heart and 1 lung/liver

The 3-month mortality for all IMI patients was 21.7%; IMI was noted to be a contributing cause of death in 52% of cases. There was no significant difference in mortality among the IA group compared with the non-IA group (22% vs 21%, p=0.84).

IMIs occurred a median of 11 months post-transplant (range: 0–162 months)(Figure 3). Early IMI (≤90 days) accounted for 25% of IMIs; 65% of IMIs occurred within 2 years of transplantation. Median time to IA was slightly less than non-IA IMI (10.5 months vs. 16 months respectively, p=0.2). Considerable variability in time to onset was seen among non-IA IMIs: “unspecified” mold infections (mold visualized on histopathology but not grown in culture) occurred a median of 3 months post-transplant, while infections (median 12 months post-transplant), “other” mold infections (median 16 months post-transplant), and mucormycosis (median 26 months post-transplant) occurred later.

Figure 3.

Frequency of invasive mold infection diagnosis by months post-transplant (n=143)

Discussion

This report describes the first and the largest prospective surveillance for IMI among lung transplant recipients and underscores the importance of IMIs in the lung transplant population. Notably, we confirm findings from prior retrospective epidemiologic investigations of IMI among lung transplant recipients, including: IMIs typically occur late (>90 days) in the post-transplant course, the lung as the most common anatomical site of involvement, and dyspnea and cough as the most common presenting symptoms. spp. were the most common molds isolated, comprising over 70% of all IMIs. This is also comparable to other studies which found IA to account for 63% to 69% of all IFIs among lung transplant recipients (10, 11). However, our data also confirm the importance of non-IA mold infections in this population as 30% of IMIs were caused by molds other than . While there was no significant difference in the frequency of disseminated disease in the non-IA versus IA mold groups, skin infection was more common in the non-IA IMI group, likely explained by the 26% of “other” IMIs caused by dematiaceous fungi.

IMI was associated with a high all-cause 3-month mortality of over 20% and was considered a contributor in the cause of death in over half of all patients with IMI that died. Our results are similar to those reported in 2010 by Arthurs et al, who reported 16% all-cause mortality associated with IFIs and a significant association with decreased survival (10). While these contemporary rates are lower than previously reported (7), the decrease may reflect differences in the type of IMIs studied and/or advances in recognition and treatment of IMIs over the years. Regardless, mortality still remains unacceptably high, and preventing IMIs remains a worthy goal.

When evaluating the presenting signs and symptoms of IMI, only 30% of patients experienced fever within 7 days of IMI diagnosis. This low rate of pyrexia is similar among other SOT recipients with IA (12), but is substantially lower than pyrexia rates seen among HSCT recipients, which have been reported at nearly 50% (18), as well as among critically ill immunocompetent patients with IA, which have been reported at 55–85% (13). The EORTC/MSG definitions for IFI do not utilize fever as criteria for diagnosis of IMI; our findings support this and suggest that absence of fever should not exclude IMI from the clinical differential diagnosis.

The majority of IMIs occurred after the first three months, but within the first post-lung transplant year (median 11 months). Late IMI onset was found for both IA and non-IA mold infections. Although there was no statistically significant difference in the time to onset of IA versus non-IA molds when grouped together, considerable variability was seen in time to onset among different non-IA mold infections, with a median time to onset of 26 months for mucormycosis. Thus, the majority of all IMIs tend to be late onset (>90 days post-transplant) with certain non-IA mold infections occurring very late (> 1 year) in the post-transplant course. One possible explanation for this shifting epidemiology may be the routine early use of mold-active prophylaxis in lung transplant recipients and development of infection after prophylaxis is stopped. The 2004 American Society of Transplantation (AST) guidelines recommended continuing prophylaxis after lung transplantation at least until bronchial anastomosis remodeling is complete, which takes 4 to 8 months (14). A 2006 survey of 43 international lung transplant centers found that 69% used antifungal prophylaxis during the immediate post-transplant period as the anastomosis was healing, most commonly an aerosolized formulation of amphotericin B alone or in combination with itraconazole (15). While the advent of newer triazole antifungals has altered some center’s prophylaxis strategy, more recent data demonstrates a continued heavy reliance on itraconazole and aerosolized amphotericin B and significant discontinuation rates of newer triazoles due to toxicity (16). Thus, use of mold active prophylaxis during the early post-transplant period may be influencing the timing of IMIs, resulting in late onset disease when prophylaxis is no longer routinely employed. Alternatively, the shift to later IMI may also be related to increased exposures, as patients resume normal activities of living following the transplant procedure.

Consistent with the 2009 AST guidelines (17), prevention strategies should give consideration to the known epidemiology of IMIs in this population, including the infecting pathogens, the mode of transmission/initial site of infection, and the usual timing of IMIs following lung transplantation. This approach allows prophylactic therapy to be targeted during windows of highest risk. The majority of experts agree that the historical risk for IFI is substantially high enough during the immediate post lung transplant period to warrant universal antifungal prophylaxis until the anastomosis is healed (18). Additionally, based on our data, expert opinion holds that upon completion of initial prophylaxis, patients and caregivers should be particularly vigilant in monitoring patients for late onset IMIs and avoiding environmental exposures that may lead to inoculation with these pathogens (19). Finally, prophylactic agents should be individualized based on the type of lung transplant performed (e.g. use of systemic rather than aerosolized agents in single lung transplant recipients) and therapeutic drug monitoring considered to ensure adequate absorption of orally administered triazoles (20).

In summary, IMIs, which tend to appear in the late and very late post-lung transplant period, are associated with high mortality. remains the most common mold pathogen; however, non-IA molds are also an important cause of IMIs. Appreciation of the epidemiology of IMIs and assessment of each patient’s individual risks should be used to refine prevention strategies in the lung transplant population.