We performed a prospective observational cohort study that included all consecutive adult ESRD patients (≥18 years) that underwent KT at the University Hospital “12 de Octubre” (Madrid, Spain) between November 2014 and December 2022. We also included combined KT procedures (i.e., pancreas-kidney, liver-kidney and heart-kidney). Patients experiencing primary graft non-function, death or graft loss within the first week were excluded. The study was performed in accordance with the ethical standards laid down in the Declarations of Helsinki and Istanbul. The local Ethics Committee approved the study protocol (CEIC number: 14/030) and written informed consent was obtained from all participants at the time of recruitment to the institutional cohort.

The study cohort has been described elsewhere [24–26]. Participants were enrolled at the time of transplantation and followed-up for at least 1 year or, alternatively, until graft loss or death if occurred earlier. Patients were seen regularly at the outpatient clinic at scheduled follow-up visits or whenever clinically indicated. Demographics, clinical, laboratory and microbiological variables and post-transplant events were prospectively collected in a standardized case report form.

The primary study outcome was the occurrence of overall post-transplant infection, with recipient age at transplantation as the independent variable of interest. Bacterial and opportunistic infection were considered secondary outcomes. As an exploratory approach to immunosenescence, we analyzed the association between recipient age and peripheral blood lymphocyte subpopulation counts (CD3⁺, CD4⁺ and CD8⁺ T-cells) and the CD4+/CD8+ ratio throughout the first post-transplant year in patients that did induction therapy with T-cell-depleting agents such as anti-thymocyte globulin (ATG).

All recipients of organs from donors after circulatory death underwent induction therapy with ATG (either Thymoglobulin® at a dose of 1–1.5 mg/kg/day for 5–7 days or Grafalon® at a dose of 3 mg/kg/day for 3 days), with delayed introduction of a calcineurin inhibitor (CNI) from day 6. Recipients at high immunological risk also received ATG induction with early CNI initiation from post-transplant day 0. Basiliximab (20 mg on days 0 and 4) with delayed CNI introduction from day 5 was reserved to patients at high risk for CNI-related nephrotoxicity (i.e., older age or comorbidities). Maintenance immunosuppression consisted of tacrolimus (0.1 mg/kg daily, adjusted to a target trough level of 10–15 ng/mL during the first month and 5–10 ng/mL thereafter); mycophenolate mofetil (1 g twice daily) or enteric-coated mycophenolate sodium (360 mg twice daily); and prednisone (1 mg/kg daily with progressive tapering). Conversion to mammalian target of rapamycin (mTOR) inhibitor-based regimens with reduced-dose tacrolimus (target trough level of 3–6 ng/mL) was performed on an individual basis for recipients experiencing severe CNI-related adverse effects, difficult-to-treat cytomegalovirus (CMV) infection, or malignancy.

All patients received preoperatively a single dose of intravenous (IV) cefazolin (or ciprofloxacin in case of ß-lactam hypersensitivity). Prophylaxis for Pneumocystis jirovecii pneumonia was administered with trimethoprim-sulfamethoxazole (160/800 mg three times weekly) or monthly intravenous pentamidine for 9 months. In patients at high-risk for CMV infection, universal prophylaxis with oral valganciclovir (900 mg daily) was given for 3 (seropositive recipients [R+] that received ATG induction) or 6 months (serology mismatch [donor positive/recipient negative (D+/R-)]). Intermediate-risk patients (R+ without ATG induction) were managed by polymerase chain reaction (PCR)-guided pre-emptive therapy, and IV ganciclovir (5 mg/kg/12 h) or oral valganciclovir (900 mg/12 h) for at least 2 weeks was initiated in the presence of high level or rapidly increasing CMV DNAemia. Valganciclovir doses were adjusted according to renal function as necessary [27].

The diagnosis of post-transplant infection was established by at least one of the following criteria: a) positive culture of an unequivocally pathogenic microorganism from any sample; b) isolation of any microorganism from a sample obtained under sterile conditions; c) isolation of a potentially pathogenic microorganism from any sample obtained from sterile or non-sterile sites (where urine was always considered a non-sterile specimen irrespective of whether the sample had been collected by the midstream clean-catch technique or catheterization), accompanied by clinical manifestations compatible with the different infectious syndromes (e.g., fever, chills, lumbar pain, graft pain and/or irritative voiding symptoms in the case of pyelonephritis); and/or d) clinical data suggestive of infection without microbiological isolation and complete resolution under empirical antimicrobial treatment. Episodes of asymptomatic bacteriuria, cystitis, asymptomatic CMV DNAemia and low-level BK polyomavirus DNAemia were excluded. Opportunistic infection included CMV disease, infections due to intracellular bacteria (e.g., Listeria monocytogenes or mycobacteria), other herpesviruses (herpes simplex virus and varicella-zoster virus), yeasts (Candida spp. and Cryptococcus spp.), molds, P. jirovecii and parasites (Cryptosporidium, Toxoplasma gondii and Leishmania spp.) [28]. Proven or probable invasive fungal disease was defined based on the criteria proposed by the European Organization for Research and Treatment of Cancer and the Mycoses Study Group [29]. CMV disease comprised viral syndrome or end-organ disease, as detailed in Supplementary Methods [30]. Further definitions are also provided as Supplementary Material.

Peripheral blood lymphocyte subpopulation (CD3⁺, CD4⁺ and CD8⁺ T-cells) counts were measured at post-transplant months 1, 3, 6 and 12 by means of an automated multicolor flow cytometry system (BD Multitest™ six-color TBNK reagent with acquisition on the BD FACSCanto II instrument using BD FACSCanto clinical software, all from BD Biosciences, San Jose, CA).

Quantitative data were summarized by the mean ± standard deviation (SD) or the median with interquartile range (IQR). Qualitative variables were expressed by absolute and relative frequencies. Categorical variables were compared using the χ² test. Student’s t-test or U Mann-Whitney test were applied for normally distributed continuous variables, and the Kruskal-Wallis test was used for comparing T-cell counts across increasing age groups.

Univariable and multivariable Fine and Gray’s competing risk regression models were fitted to assess the impact of recipient age at transplantation on the occurrence of overall, bacterial and opportunistic infection, with death from any cause as the competing event. The variable “recipient age” was analyzed in different ways: as a continuous manner by 10-year strata (age groups), and dichotomized according to the cut-off values used in the literature to define “older KT recipient” (≥60, ≥70, ≥75 and ≥80 years) [23]. We selected all demographic and clinical factors with a P-value < 0.05 in the univariable analysis to be entered in the multivariable models as independent (explanatory) variables. Collinearity was assessed by the variance inflation factor (VIF), with values <2 considered as acceptable. Since recipient age was our main variable of interest, it was always retained in the model in the presence of significant collinearity with other factors (e.g., donor age).

Associations were expressed as subdistribution hazard ratios (SHRs) with 95% confidence intervals (CIs). Statistical analysis was performed with SPSS version 29.0.1.0 (IBM Corp., Armonk, NY) and StataNow version 19.5 (StataCorp College Station, TX).

We included 712 KT recipients, whose demographics and clinical features are shown in Table 1. Median and mean age at transplantation were 56.6 (IQR: 43.2–68.5) and 55.4 ± 15.6 years, respectively. According to pre-established age thresholds, 300 (42.1%) patients were ≥60 years, 155 (21.8%) were ≥70 years, 84 (11.8%) were ≥75 years, and 27 (3.8%) were ≥80 years.

The distribution of pre-transplant comorbidities and transplant-related variables in the different age strata is detailed in Supplementary Table S1. The prevalence of some conditions (non-coronary heart disease or cerebrovascular disease) and pre-transplant solid organ malignancy showed a clear gradient across increasing age strata. This association was less evident for other comorbidities, which peaked either at 60–70 (coronary heart disease, chronic pulmonary disease and peripheral arterial disease) or 70–80 years (hypertension and diabetes). The proportion of patients at the high-risk category for CMV infection (D+/R-) decreased with age. Regarding the type of donor, the proportion of living donation and donation after circulatory death also decreased with age.

After a median follow-up period of 728 days (IQR: 544.8–1,071), graft loss and all-cause death occurred in 33 (4.6%) and 68 patients (9.6%), respectively. One- and two-year survival rates were 95.3% and 91.9%, whereas the corresponding estimates for death-censored graft survival were 96.9% and 94.7%, respectively.

There were 1,088 distinct episodes of infection in the entire cohort. As shown in Supplementary Table S2, most common clinical syndromes were acute pyelonephritis (306 out of 1,088 episodes [28.1%]), viral syndrome (115 [10.5%]), digestive tract infection (114 [10.5%]), upper respiratory tract infection (95 [8.7%]), pneumonia (91 [8.4%]) and skin and soft tissue infection (90 [8.3%]). Overall, 155 episodes (14.2%) were associated to bacteriemia.

The predominant bacterial pathogens were Escherichia coli (136 out of 1,088 episodes [12.5%]), Klebsiella pneumoniae (92 [8.5%]), Pseudomonas aeruginosa (55 [5.1%]), Enterococcus spp (50 [4.6%]) and Clostridioides difficile (44 [4.0%]). Regarding antimicrobial susceptibility profiles, 77.2% (71/92) and 22.1% (30/136) of K. pneumoniae and E. coli isolates, respectively, were extended-spectrum β-lactamase (ESBL) producers, whereas one third of P. aeruginosa isolates (29.1% [16/55]) exhibited a multidrug-resistant phenotype (i.e., non-susceptibility to at least one agent in ≥3 antibiotic classes). Herpesviruses accounted for 163 episodes of infection (14.9%), with predominance of CMV (102 [9.4%]).

Within the subgroup of microbiologically documented episodes of urinary tract infection, E. coli (40.7% [116/285]), K. pneumoniae (28.4% [81/285]) and P. aeruginosa (8.8% [25/285]) were the most commonly isolated uropathogens (Supplementary Figure S1).

The cumulative incidence of overall post-transplant infection was 58.1% (95% CI: 54.5–61.7). In detail, 414 patients developed at least one episode of infection (incidence rate of 1.61 episodes per 1,000 transplant-days). The median interval from transplantation to the first episode was 57 days (IQR: 17.3–234.5). The cumulative incidence of the secondary outcomes of bacterial and opportunistic infection were 44.9% (95% CI: 41.3–48.6) and 22.8% (95% CI: 19.8–25.9), respectively.

First, we assessed whether recipient age exerted an effect on the risk of post-transplant infection in unadjusted analyses. Each 10-year increase was associated with the occurrence of overall infection (SHR: 1.18; 95% CI: 1.11–1.26; P-value < 0.001). Regarding secondary outcomes, the risks of bacterial (SHR: 1.17; 95% CI: 1.09–1.26; P-value < 0.001) and opportunistic infection (SHR: 1.26; 96% CI: 1.13–1.40; P-value < 0.001) were also increased with each 10-year increase. Using the lowest age group (<30 years) as the reference category, all the groups above 50 years had a significantly increased risk of overall infection, whereas the risk increase for bacterial and opportunistic infection achieved statistical significance for the age strata beyond 60 years (Figure 1).

The variable was further dichotomized according to cut-off values conventionally used to define “older KT recipient”. Recipients ≥60 and ≥70 years have significantly increased risks for the primary and secondary outcomes as compared to those below these thresholds. For recipients ≥75 years these associations were only significant for the secondary outcomes, whereas none of the comparisons achieved statistical significance for the age category ≥80 years as compared to those below (Figure 2).

As detailed in Supplementary Table S3, a number of variables were identified as predictive factors for the primary outcome of overall infection on the basis of their univariable P-values < 0.05. As expected, we found significant collinearity between receptor and donor age, between pre-transplant diabetes and diabetic nephropathy, and between living donor and cold ischemia time. Taking into account their clinical relevance, diabetes and living donor were kept in the model, in addition to recipient age. Therefore, the following covariates were included in the multivariable Fine and Gray’s regression model for the primary outcome of overall post-transplant infection: diabetes mellitus, chronic pulmonary disease, solid organ cancer, cerebrovascular disease, peripheral arterial disease, reflux nephropathy, CMV serology mismatch, pre-transplant dialysis, living donor, requirement of ICU admission, delayed graft function [DGF], early re-intervention and new-onset diabetes. After multivariable adjustment, each 10-year increase in recipient age continued to be independently associated with the occurrence of overall (adjusted SHR: 1.09; 95% CI: 1.02–1.18; P-value = 0.011).

Univariable analyses for the secondary outcomes are shown in Supplementary Tables S4, S5. After adjusting for the clinical covariates that achieved a P-value < 0.05 (diabetes mellitus, chronic pulmonary disease, cerebrovascular disease, peripheral arterial disease, reflux nephropathy, pre-transplant dialysis, type of donor, number of HLA mismatches, DGF, early re-intervention and new-onset diabetes), each 10-year increase in recipient age was independently associated with the occurrence of bacterial infection (adjusted SHR: 1.09; 95% CI: 1.00–1.18; P-value = 0.038). On the other hand, the association for opportunistic infection was not longer significant (adjusted SHR: 1.09; 95% CI: 0.96–1.25; P-value = 0.165) following multivariable adjustment (diabetes mellitus, coronary heart disease, reflux nephropathy, CMV serology mismatch, living donor, basiliximab induction, new-onset diabetes and acute graft rejection).

There were no significant associations across increasing age groups with regards to the reference category (<30 years) for either the primary or secondary outcomes (Figure 3).

When age was dichotomized according to conventional thresholds, recipients ≥60 years exhibited a significantly higher risk of overall infection in the multivariable model (adjusted SHR: 1.25; 95% CI: 1.00–1.54; P-value = 0.045). As for secondary outcomes, the risk of opportunistic infection was significantly increased for recipients ≥70 years (adjusted SHR: 1.54; 95% CI: 1.02–2.32; P-value = 0.038) (Supplementary Figure S2).

Finally, we compared across 10-year strata peripheral blood lymphocyte subpopulations counts assessed at different time points among KT recipients that did not receive ATG induction therapy (n = 371 patients with available data). There was a progressive decrease with increasing age for CD3⁺ and CD4⁺ T-cell counts at months 1, 3, 6 and 12 after transplantation (P-values for all comparisons <0.001). For the CD8⁺ T-cell count, the association was only significant at months 1 and 3 (P-values < 0.001), whereas the trend was less clear at month 12 (P-value = 0.008). No significant differences were found for the CD4+/C8+ ratio (Figure 4).