Introduction

Lung transplantation is the only effective treatment option for patients with end-stage lung disease. Lung donations come mainly from patients who have suffered brain death (i.e., donation after brain death, DBD) and donation after cardiac death (DCD) [1]. Infection is one of the leading causes of both short-term and long-term death in lung transplantation recipients [1, 2]. Donor-derived infection (DDI) is an important infection in lung transplant recipients [3]. Although the epidemiology varies markedly among different geographical regions, risk factors of prolonged (> 7 days) ICU stay, vasopressor support, and cardiopulmonary resuscitation increases colonization with multidrug-resistant organisms (MDRO) including extended-spectrum β-lactamase -producing Enterobacteriaceae (ESBL-E) and carbapenem-resistant organisms (CRO) in donors [4]. The current study suggests that MDRO or CRO may significantly contribute to the burden of bacterial DDI [5, 6]. Lung transplant recipients appear to be disproportionately affected, potentially due to the high frequency of MDRO colonization of the donor respiratory tract [7, 8]. MDRO-associated DDI or CRO-associated DDI may play a role in the development of early posttransplant infections in transplant recipients [9], leading to several complications, including stomal leak, bleeding, graft loss, and even death [7, 10, 11]. If donor colonization or infection with CRO is known before transplantation, a risk–benefit evaluation should be made based on the organ to be transplanted and the site of the positive donor cultures. Clinicians suggest that acquiring the lung from a donor experiencing CRO bacteremia or respiratory colonization for transplantation should be avoided [3].

Traditionally, CRO screening of donor lungs is based on bacterial culture followed by phenotypic antimicrobial susceptibility testing. However, there are two main problems with culture-based screening: (i) the time to results, which requires at least 72 h, is too slow to meet the needs of donor lung acquisition; and (ii) the requirements for trained personnel, equipment, sites, etc., which are inconsistent with the organ procurement organizations (OPOs) [12]. The culture method has a result reporting time delay, which may lead to occult CRO infection in the donor lungs during the period before culture and susceptibility test results are available. Therefore, a rapid, simple, and accurate technique for CRO screening of donor organs is urgently needed at the point of care. This could effectively reduce latent donor CRO infection, reduce the incidence of postoperative DDI and infection-related adverse events of lung transplantation, and improve the prognosis of patients who receive lung transplantation.

The Xpert® Carba-R assay (Cepheid, Sunnyvale, CA, USA) is an on-demand PCR test specifically designed for the qualitative detection and differentiation of five common carbapenemase gene families, including blaKPC, blaNDM, blaVIM, blaIMP, and blaOXA-48 with results available in less than one hour [13]. Due to the portability of the instrument, the donor lung acquisition team can test the donor lung for CRO at the OPO and decide whether to acquire the lung and perform the transplantation after obtaining the CRO results in real time. This approach would help avoid the acquisition of a CRO-infected donor lung and reduce the occurrence of CRO-associated DDI and infection-related complications after lung transplantation caused by using CRO-infected lungs. This study used the GeneXpert II instrument combined with the Xpert Carba-R test on-site to detect CRO in donor bronchoalveolar fluid (BALF) as an off-label specimen before donor lung acquisition.

Materials and methods

Patients and study design

This historically controlled prospective trial was approved by the Ethics Committee of Shanghai Chest Hospital, Shanghai Jiao Tong University (No. IS21101) and registered on chictr.org.cn (ChiCTR2100053687). Written informed consent was obtained from the recipients. We obtained the authorization of the donor's next of kin and the Ethics Committee to use donor-derived biological samples for clinical and research purposes. All recipients who enrolled in the study accepted lung transplant treatment at Shanghai Chest Hospital. The screening group included patients who underwent lung transplantation from November 2021 to September 2022 with negative CRO results of the donor lung by the Xpert Carba-R test. The historically controlled group included patients who underwent lung transplantation from January 2016 to December 2020. All patients who were included in the Xpert Carba-R screening group and control group met the same inclusion criteria of this study.

The study design is shown in Fig. 1. From November 2021 to September 2022, twenty-nine consecutive recipients were observed in the study. Among them, three cases died during the waiting period for donor lung supply, and one of these cases had an MDRO infection in the lung. The remaining twenty-six recipients eligible for inclusion criteria were included in the Xpert Carba-R screening group. During the same period, forty donor lungs were available for transplant. Thirty-five donor lungs met the inclusion criteria and five dropped out for various reasons. The thirty-five donor lungs were defined as potential donor lungs for Xpert Carba-R testing. If one or more carbapenemase gene (including blaKPC, blaNDM, blaVIM, blaIMP, and blaOXA-48) was positive in BALF by Xpert Carba-R, the potential donor lungs were not acquired and would be declined for the subsequent lung transplantation. Only the potential donor lungs of carbapenemase gene negative were used as donor lungs in the Xpert Carba-R screening group. For the historic controls, twenty-three cases of donor lungs and twenty-two recipients were consecutively observed from January 2016 to December 2020. According to the inclusion criteria, nineteen recipients and the matching donor lungs were included.

Fig. 1
figure 1

Study flowchart. MDRO, multidrug-resistant organisms

Full size image

Donor inclusion criteria

Donor meeting the following criteria were considered for inclusion in the study: legal and ethical instruments of organ donation can be obtained; DBD; same blood type or different but compatible with recipient blood type; age ≥ 18 and ≤ 70 years; oxygenation index (P/F) > 230 mmHg (FiO2 = 1.0, PEEP = 5 cm H2O); chest radiograph showing a clear lung field or a mild to moderate exudation; appropriate size-match in lung donor and recipient, or a poor size-match but with satisfactory matching after volume reduction to donor lung; absence of chest trauma; absence of aspiration or slight aspiration improved after treatment; absence or a small amount of purulent secretions in the airway; cold ischemia time < 8 h; and no evidence of MDRO or CRO infection and colonization.

Recipient inclusion criteria

The patients who suffered non-neoplastic pulmonary disease accepted for lung transplantation showed: normal blood sugar levels or fasting blood glucose of diabetic patients that were controlled below 10 mmol/L; normal or mildly abnormal liver function (Child–Pugh grade A); normal or mildly abnormal renal function (GFR > 59 ml/min); age ≤ 75 years; a postoperative survival time was more than 1 week; no evidence of MDRO (including CRO, methicillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant enterococci (VRE), extended-spectrum β-lactamases (ESBLs)-producing bacteria, or multidrug-resistant tuberculosis (MDR-TB)) infection or colonization were noted before lung transplantation. For MDRO detection, BALF or sputum specimens of recipients were obtained every three days for microbial culture and antimicrobial susceptibility testing before lung transplantation.

Xpert Carba-R CRO detection in donor lung

For the BALF sample, we used the Xpert Carba-R test with the off-label following procedure: approximately 15 ml of sterile normal saline was injected into the left and right main bronchus of the donor lung by bronchoscopy. At least 20 ml BALF was recovered by aspiration. One ml of the BALF suspension was added to a 5 mL Xpert Carba-R sample reagent vial and mixed for 10 s. Using the supplied pipette, 1.7 ml of sample reagent was transferred into the sample chamber of the Xpert Carba-R cartridge, the lid was closed, and the cartridge was placed in the Cepheid GeneXpert platform for testing.

Bacterial culture of BALF samples

10 µl of BALF from the donor lung was inoculated onto a Columbia blood agar blood plate, a vancomycin-containing chocolate agar plate, and a crystal violet-containing MacConkey agar plate, and incubated at 35 °C for 24–48 h. When the BALF samples grew more than 104 CFU/mL, colonies were selected and identified using matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF–MS). Antimicrobial susceptibility testing (AST) of these identified bacteria was performed by using the disc diffusion method and the broth microdilution method according to the 2022 Clinical Laboratory Standard Institute (CLSI) guidelines [14].

Clinical outcomes

DDI was defined as any infection present in the donor that is transmitted to one or more recipients during or after transplantation [15]. At present, there is still a lack of effective methods for the identification of DDI in donor infections, especially the lack of consensus on the identification of DDI using molecular diagnostic tests. In this study, BALF or sputum specimens from recipients were obtained every three days for microbial culture and antimicrobial susceptibility testing before lung transplantation. Cultures growing MDRO, especially those indicating CRO -infected or colonized- recipients, were used to exclude those patients from the study cohort. Therefore, in this study, we defined CRO-associated DDI as an infection in a recipient within 1 week after lung transplantation, if the CRO-containing organism recovered from the recipient is the same as the CRO-containing organism carried by the donor lung and the organisms has the same antimicrobial susceptibility profile. Death of recipients related by infection with any donor- or recipient-derived pathogen within 60 days after lung transplant was described as infection-related death within 60 days, which includes CRO-associated DDI-related death within 60 days.

In this study, we collected the clinical characteristics of the donors and recipients to compare the baselines between the two groups. The outcomes and incidence of DDI, CRO-associated DDI, and infection-or CRO-associated DDI-related death within 60 days were mainly focused on the effect of Xpert Carba-R donor lung screening on the prognosis of recipients.

Statistical analysis

We analyzed variable distribution by the D’Agostino-Pearson test. The continuous variables of normal distributions were expressed as mean ± standard deviation, and the continuous variables of non-normal distribution were presented as medians and quartiles. The classification variables were presented as frequencies and percentages. The continuous variables were compared by an independent sample t-test or rank-sum test, and the comparison of classification variables was conducted by Fisher's exact test. Logistic regression was used for the univariate analysis and multivariate analysis. Penalized likelihood logistic regression was used when quasi-complete separation happened. A p-value < 0.05 was considered statistically significant. SPSS 26.0 software (IBM, Chicago, USA) plus Normaltest software package and Statistical Analysis System (SAS) statistical software package, version 9.4 were used for statistical analysis.

Results

In the thirty-five potential donors, nine (25.71%) with positive Xpert Carba-R results in BALF were declined for lung transplantation before the acquisition. Finally, the twenty-six recipients and matching CRO-negative donor lungs (74.29%) were included in the Xpert Carba-R screening group.

The baseline characteristics of donors and recipients in both groups were summarized in Table 1. The factors of donors, including age, sex, death mechanism, duration of brain death declaration (BDD) to lung transplantation, mechanical ventilation support time, and cold ischemia time were not statistically significant between the screening and control groups. In the Xpert Carba-R screening group, the donor lungs of carbapenemase gene negative were used for lung transplantation.

Table 1 Baseline characteristics of donors and recipients in the Xpert Carba-R screening group and control group
Full size table

In this study, microbial culture and antimicrobial susceptibility testing of BALF specimens from donor lungs were routinely cultured prior to lung transplantation regardless of the Xpert Carba-R results. The BALF culture reports of donors in the Xpert Carba-R screening group suggested that in addition to one case of methicillin-resistant Staphylococcus epidermidis (MRSE) infection, there was one case of carbapenem-resistant Klebsiella pneumoniae (CRKP) infection, and one case carbapenem-resistant Acinetobacter baumannii (CRAB) infection.

The rate of CRO isolation from donor lung specimens was significantly lower in the Xpert Carba-R screening group (2/26, 7.69%) compared with that in the control group (7/19, 36.84%) (p = 0.02) (Table 1). For specimens from recipients, there was a significant difference between the two groups in CRO positivity rate of postoperative lung grafts (9/26, 34.62% vs. 15/19, 78.94%, p < 0.01) (Table 1).

Whether the presence of CRO infection in the donor lung has effects on DDI and outcomes of recipients is unclear. In this study, compared with non-CRO infected-donor lungs, the use of donor lungs with CRO infection significantly increased DDI (p < 0.01) and infection-related death within 60 days (p < 0.01) in recipients after lung transplantation (Table 2).

Table 2 The effict of CRO isolation in preoperative donor lungs on outcomes in lung transplant recipients
Full size table

Among forty-five lung transplantation patients (twenty-six in the Xpert Carba-R screening group and nineteen in the control group), DDI and CRO-associated DDI occurred in 19.23% and 7.69% in the screening group, and 36.84% and 31.58% in the control group, respectively (Table 3). We observed several infection-related deaths within 60 days. There were two (7.69%) and seven (36.84%) deaths in the two groups, respectively (OR = 0.14, 95% CI 0.03–0.80, p = 0.02), including one CRO-related death due to DDI (3.85%) in the screening group and six (31.58%) in the control group (OR = 0.09, 95% CI 0.01–0.80, p = 0.03) (Tables 3 and 4). Adjusted for the potential confounders of BMI and diabetes mellitus in recipients suggested by the results of univariate analysis, multivariate analysis showed that Xpert Carba-R screening for donor lungs significantly reduced the risk of CRO-associated DDI relating death within 60 days in recipients (OR = 0.05, 95% CI 0.003–0.74, p = 0.03) (Table 4). In this study, the Xpert Carba-R screening group received a lower dose of therapeutic antibiotics than the control group (p = 0.02) (Table 3).

Table 3 The effect of Xpert Carba-R screening on outcomes in lung transplant recipients
Full size table
Table 4 Effect of Xpert Carba-R screening and other clinical characteristics on CRO-associated DDI relates death within 60 days in recipients, univariable and multivariable analyses
Full size table

Discussion

Donor-derived infections, especially those that are CRO-related, are an important risk factor leading to poor outcomes of lung transplantation. Occult CRO infections in donor lungs often result in high mortality, which with carbapenem-resistant Klebsiella pneumoniae infection can reach 72.7% [16]. Current CRO screening methods, mainly based on bacterial culture followed by AST, are too slow to provide information on infection in donor lungs before transplantation. Thus, infected lungs may be transplanted, leading to postoperative infections requiring antimicrobial therapy. In addition, CRO-related infections, especially those caused by metallo-β-lactamases (MBLs) producing organisms, are very difficult to treat, leading to higher mortality rates after transplantation. The Xpert Carba-R test, which provides qualitative detection and differentiation of five commonly encountered carbapenemase gene families with a turnaround time of less than one hour, can provide real-time screening for donor organs.

The Xpert Carba-R test has been reported to show excellent performance detecting CRO in a variety of clinical specimens, including 95% sensitivity and 99% specificity for rectal swabs [17], 92.9% sensitivity, and 86.7% specificity for sputum samples [18], and 95% sensitivity and 95% specificity for bronchial aspirates [19], although the latter two specimen types are considered off-label. In this study, the lung acquisition team was able to bring the GeneXpert II instrument and cartridges to the OPO site to enable rapid, point-of-care testing for CRO in bronchoalveolar lavage specimens taken from the donors. This enabled clinicians to recognize CRO infection in the donor lung before acquisition and transplantation, improving the postoperative prognosis of recipients.

In this study, cumulative DDDs were significantly different among patients in the Xpert Carba-R screening group and those in the Control group. There may be a cost-effect in the screening process using the Xpert Carba-R test, although this will require additional studies for confirmation. CRO infections have become a major international public health problem due to inadequate treatment options and the historically slow pace of developing novel antimicrobial drugs. In China, there has been a sharp increase in infections caused by CRO and the emergence of new resistance genotypes in multiple bacterial species has been observed. Some bacterial species have been documented to carry two or more carbapenemase genes [13]. Due to the poor outcome associated with transplanting CRO-positive organs, it is critical to screen organs from DBD and DCD before transplantation. The quick, on-demand, real-time capacity of the Xpert Carba-R test provides enough time for the acquisition team to screen potential donor lungs before transplantation, which is not possible with traditional culture and AST methods. This enables the securing of donor organs more cost-effectively.

To our knowledge, this is the first study demonstrating that accurate screening for CRO in donor lungs in a real-time manner raises the potential to reduce CRO-associated DDI mortality. However, there are several limitations to this study. First, this study used a group of historical patients as controls when the real-time molecular screening method was not yet available. These historical controls may have given results that could be biased by a variety of confounding factors. Such as the low average age of recipients in historical controls. Second, the study was performed in a single center with relatively small numbers of patients recruited. A larger, multi-center clinical trial is being planned with a focus on healthcare economics.

In conclusion, the capacity for quick and accurate detection and characterization of CRO by Xpert Carba-R provides an ideal screening tool in donor lungs before transplantation. This enabled clinicians to know the CRO infection status of donor lungs in a real-time manner and to formulate a more precise plan for acquiring the lung from the donor for transplantation.