Comparison of infectious complications following early versus late pulse steroid therapy in renal transplant recipients

Abstract

BACKGROUND

Renal transplant recipients are at increased risk of infections due to immunosuppression. Pulse steroid therapy, used to treat acute rejection, may further heighten this risk. However, the impact of the timing of pulse steroid therapy on infectious complications remains unclear.

AIM

To compare infectious complications following early vs late pulse steroid therapy in renal transplant recipients.

METHODS

This study included adult renal transplant recipients who received pulse steroid therapy either within 90 days (early) or after 90 days (late) post-transplantation between January 2016 and December 2020. Patients receiving both early and late pulse therapy were excluded. Infectious complications requiring hospitalization were recorded from the first pulse steroid dose until death or December 31, 2021. Infection burden was expressed as episodes per 100 patient-years, and incidence rate ratios (IRRs) with 95% confidence intervals (CI) were calculated.

RESULTS

A total of 128 patients were included (early: n = 84; late: n = 44). The overall infection rate was lower in the early group compared with the late group (42.64 episodes per 100 patient-years vs 68.68 episodes per 100 patient-years; IRR = 0.62, 95%CI: 0.46-0.83, P < 0.01). Rates of pneumonia (IRR = 0.44, 95%CI: 0.22-0.86, P = 0.02), sepsis (IRR = 0.35, 95%CI: 0.13-0.96, P = 0.04), and viral infections (IRR = 0.42, 95%CI: 0.19-0.91, P = 0.03) were also lower in the early group. Mortality was significantly higher in patients receiving late pulse steroid therapy (18.2% vs 4.8%, P = 0.02).

CONCLUSION

Late pulse steroid therapy in renal transplant recipients is associated with a significantly higher burden of infectious complications and mortality compared with early therapy. These findings highlight the need for enhanced surveillance and tailored infection prevention strategies following late intensification of immunosuppression.

Key Words: Renal transplantation; Pulse steroid therapy; Infectious complications; Pneumonia; Sepsis; Immunosuppression; Incidence rate ratio

Core Tip: The infectious risk associated with pulse steroid therapy in renal transplant recipients may depend on its timing after transplantation. In this study, patients receiving late pulse steroid therapy (after 90 days post-transplant) experienced significantly higher rates of overall infections, pneumonia, sepsis, and viral infections compared with those treated early. These findings suggest that late intensification of immunosuppression carries a greater infectious burden and highlight the need for enhanced surveillance and reconsideration of prophylactic strategies when pulse steroids are administered in the late post-transplant period.

INTRODUCTION

Renal transplantation is a life-saving procedure for patients with end-stage renal disease, offering improved quality of life and survival compared to dialysis. However, the necessity of immunosuppressive therapies to prevent graft rejection places recipients at a significantly elevated risk of infectious complications. These infections vary in nature and timing post-transplant. Several studies have demonstrated a time-dependent pattern of infections following renal transplantation, with opportunistic infections predominating early and community-acquired infections occurring later as immunosuppression is modified[1-3]. In the early post-transplant period, typically within the first 90 days, opportunistic infections such as cytomegalovirus (CMV) disease and BK virus infections predominate due to the intense immunosuppressive burden[1]. In contrast, the late post-transplant period, after 90 days, is characterized by a higher incidence of community-acquired infections, such as urinary tract infections (UTIs) and pneumonia, as the immunosuppressive regimen is often tapered[1]. This temporal variation in infection profiles is closely linked to the dynamic interplay of immunosuppressive drugs, patient immune status, and environmental exposures.

Pulse steroid therapy, involving high-dose intravenous corticosteroids like methylprednisolone, is a cornerstone treatment for acute rejection episodes, including T-cell-mediated, antibody-mediated, or mixed rejection. This therapy temporarily heightens immunosuppression, potentially increasing susceptibility to infections. The safety profile of pulse steroids with respect to infectious complications in renal transplant recipients remains poorly defined. Notably, it is unclear whether the timing of pulse steroid administration - early vs late post-transplant - differentially impacts the risk of infections. Recent studies suggest that late post-transplant corticosteroid intensification may be associated with increased risk of severe infections, particularly sepsis and viral reactivation, compared with early postoperative exposure[4-6]. Early post-transplant, the immune system is heavily suppressed due to induction and maintenance therapies, which may amplify the infectious risk of additional immunosuppression. Conversely, in the late period, reduced prophylaxis and less frequent monitoring may exacerbate vulnerability to infections following pulse steroids. Understanding these risks is critical for optimizing post-transplant care, including tailoring monitoring protocols and prophylaxis strategies to mitigate infectious complications. To date, no studies have directly compared infectious outcomes following early vs late pulse steroid therapy in renal transplant recipients. This study aims to address this gap by evaluating infection burden, infection types, graft outcomes, and mortality in a retrospective cohort from a tertiary care centre in India.

The primary objective of this study was to investigate the impact of early vs late pulse steroid therapy on the incidence and types of infectious complications in renal transplant recipients. Specifically, the study aimed to compare the rates of overall infections, specific infection types (e.g., pneumonia, sepsis, viral infections), and outcomes such as graft loss and mortality between patients receiving pulse steroids within 90 days (early) vs after 90 days (late) post-transplant. By delineating the differential effects of timing, the study sought to provide insights into optimizing immunosuppressive strategies and infection prevention protocols. Secondary objectives included characterizing the baseline demographics, immunosuppressive regimens, and indications for pulse steroid therapy to contextualize the infection risk profiles in both groups.

MATERIALS AND METHODS

A descriptive comparative study was conducted on renal transplant recipients at the Department of Nephrology, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, from January 1, 2016, to December 31, 2020. Patients receiving pulse doses of methylprednisolone in either the early (within 90 days post-transplant) or late (after 90 days) period were included. Inclusion criteria were age ≥ 18 years at transplant and a minimum follow-up of 90 days. Exclusion criteria included receiving pulse steroids in periods, repeat or additional organ transplants, incomplete records, or loss to follow-up. This classification was chosen because pulse steroid is usually administered after transplant to treat acute rejection, and rejection has been categorized as early or late depending on whether it occurs within or after 90 days from the date of transplant, respectively[2]. Pulse steroid therapy was analyzed as a single treatment course per patient. Only infections occurring after the first pulse steroid course were included in the analysis. Patients who received both early and late pulse steroid therapy during follow-up were excluded to avoid overlapping exposure periods.

Data collected included age, sex, transplant type (living or cadaveric donor), human leukocyte antigen mismatch, desensitization therapy, induction regimen, maintenance immunosuppression, antimicrobial prophylaxis, infectious complications, graft loss, and mortality. Maintenance immunosuppressive drugs were considered if administered for over 7 days. The entire burden of immunosuppression given prior to, during and after the administration of pulse steroids from the time of transplantation to the end of follow up period was included in the study for both early and late pulse steroid groups.

All infectious episodes requiring hospital admission, whether community- or hospital-acquired, were included, provided they occurred at least 48 hours after the first pulse steroid dose. Follow-up time for infection analysis accrued from the date of first pulse steroid administration until death or December 31, 2021, whichever occurred earlier. Infection burden was expressed as episodes per 100 patient-years to account for unequal follow-up duration between individuals and between early and late pulse steroid groups. A 48-hour attribution window was used to allow for the biological onset of steroid-induced immunosuppression and has been employed in prior transplant infection studies evaluating treatment-associated infectious risk. Data sources included hospital admission, discharge, and investigation records. Patients were followed until December 31, 2021, or death, whichever occurred first.

Statistical analysis

Incidence rate ratios (IRRs) were estimated using Poisson regression with follow-up time incorporated as an offset to account for varying observation periods. IRRs are presented with 95% confidence intervals (CI). Baseline characteristics were compared using the χ² or Fisher’s exact test for categorical variables and the Mann-Whitney U test for non-normally distributed continuous variables. A two-sided P-value < 0.05 was considered statistically significant. Data analysis used SPSS version 20.

Given the higher mortality observed in the late pulse steroid group, a sensitivity analysis was performed by truncating follow-up at the time of death to assess the robustness of infection rate estimates. The direction and magnitude of IRRs remained unchanged, indicating that the observed differences in infection burden were not solely driven by differential mortality between groups. Although death may act as a competing event for infection observation, incidence rate–based analyses incorporating follow-up duration partially address this limitation; formal competing-risk modeling could not be performed due to data constraints.

Definitions

Pulse steroid therapy: Discontinuous intravenous administration of > 250 mg prednisone equivalent per day for one or more days[3].

Early pulse therapy: Pulse corticosteroids within 90 days post-transplant.

Late pulse therapy: Pulse corticosteroids after 90 days post-transplant.

UTI: Symptomatic infection in any urinary system component (kidneys, ureters, bladder, urethra).

Diarrhoea: Loose, watery stools ≥ 3 times in 24 hours, of suspected or confirmed infectious aetiology.

Sepsis: Life-threatening organ dysfunction due to dysregulated host response to infection.

Bacterial infection: Isolation of a pathogenic bacterium in culture contributing to clinical symptoms.

Viral infection: Diagnosed by serological evidence, positive nucleic acid amplification test, viral inclusion bodies in histopathology, or specific clinical signs.

CMV disease: Detectable CMV in a clinical specimen with accompanying clinical manifestations.

BK viraemia: Detectable plasma BK virus DNA.

BK virus nephropathy: Viral inclusions or positive SV-40 stain on allograft biopsy, or plasma BK viral load > 10000 copies/mL.

Tubercular infection: Diagnosed by microbiological (microscopy, culture, nucleic acid amplification), radiological, pathological (e.g., caseating granuloma), or laboratory findings (e.g., elevated adenosine deaminase in pleural fluid) in an appropriate clinical context.

Intestinal parasitic infections were diagnosed by stool microscopic examination: Infective episodes were distinct if they occurred during separate hospital admissions or involved different organ systems. Recurrent infections in the same organ system (e.g., UTI) were counted as separate episodes only if there was documented clinical resolution followed by a new hospital admission. Chronic or disseminated infections were counted as a single episode.

Two reviewers independently extracted data on pulse steroid therapy (dosage, number of pulses) and infectious complications.

RESULTS

Of 516 renal transplant recipients from January 1, 2016, to December 31, 2020, 128 (24.8%) met the inclusion criteria. The mean age was 32.1 years (SD: 9.0 years). Haemodialysis was performed in 127 (99.2%) patients, with 1 (0.8%) receiving pre-emptive transplantation. Donors were live in 127 (99.2%) cases and cadaveric in 1 (0.8%) case. Males comprised 92.2% of recipients. Early pulse steroid therapy was administered to 84 (65.6%) patients and late therapy to 44 (34.4%). Mean times to pulse steroid administration were 15 days (early) and 550 days (late). The primary indication was empirical therapy for suspected rejection (59.5% early, 45.5% late). Among patients treated empirically for rejection, biopsy confirmation was obtained where clinically feasible. Other indications included mixed rejection (8.3% early, 27.3% late), acute antibody-mediated rejection (14.3% early, 13.6% late), borderline cellular rejection (10.7% early, 11.4% late), acute T cell-mediated rejection (4.8% early, 0% late), and chronic antibody-mediated rejection (0% early, 2.3% late).

Baseline characteristics (Table 1) showed no differences in age, gender, or diabetes status between groups. The median number of pulses was 3, with a median dose of 1500 mg in both groups. Immunosuppression details (Table 2) indicated no differences in induction agent use (interleukin-2 receptor antagonist, lymphocyte-depleting agent, or none), use of lymphocyte-depleting agent, or plasmapheresis. A trend toward greater bortezomib use was noted in the late group (P = 0.05), though usage was low. Plasmapheresis for ABO-incompatible transplant desensitization occurred in 8 (9.5%) early and 4 (9.1%) late group patients. Post-transplant plasmapheresis indications included antibody-mediated rejection (13.1% early, 13.6% late), mixed rejection (4.8% early, 20.5% late), and chronic antibody-mediated rejection (0% early, 2.3% late). All patients received tacrolimus and mycophenolate mofetil/sodium, with no differences in cyclosporine, azathioprine, or mechanistic target of rapamycin inhibitor use. Infection prophylaxis included daily single-strength cotrimoxazole and topical clotrimazole (1% w/v) in both groups.

Table 1 Characteristics of patients receiving early and late steroids, n (%).

Variables	Early (n = 84)	Late (n = 44)	P value (df for χ2)
Age in years [median (IQR)]	30 (25-38)	30 (26-37)	0.94
Gender (male)	78 (92.9)	40 (90.0)	0.74 (1)
Dose of pulse steroid in mg [median (IQR)]	1500 (1500-1500)	1500 (1500-1500)	0.48
HLA mismatch [median (IQR)]	3 (3-5)	3 (3-4)	0.94
ABO compatible	77 (91.7)	40 (90.9)	1.0 (1)
Diabetes status			0.52 (2)
Pre-existing DM	1 (1.2)	2 (4.5)
PTDM	9 (10.7)	4 (9.1)
No diabetes	74 (88.1)	38 (86.4)
CMV prophylaxis	40 (47.6)	19 (43.2)	0.63 (1)
Graft loss	2 (2.4)	1 (2.3)	1.0 (1)
Survival status			0.02 (1)
Alive	80 (95.2)	36 (81.8)
Died	4 (4.8)	8 (18.2)

df: Degree of freedom; IQR: Interquartile range (25 percentile to 75 percentile); PTDM: Post-transplant diabetes mellitus; DM: Diabetes mellitus; CMV: Cytomegalovirus.

Table 2 Immunosuppression in patients receiving early and late steroids, n (%).

Variables	Early (n = 84)	Late (n = 44)	P value (df for χ2)
Induction agent			0.54 (2)
Basiliximab	53 (63.1)	25 (56.8)
ATG	21 (25.0)	15 (34.1)
None	10 (11.9)	4 (9.1)
Plasmapheresis	28 (33.3)	21 (47.7)	0.11 (1)
Overall ATG use	27 (32.1)	18 (40.9)	0.32 (1)
Bortezomib1	1 (1.2)	4 (9.1)	0.05 (1)
Mtor inhibitor	2 (2.4)	0 (0.0)	0.54 (1)
Cyclosporine	2 (2.4)	1 (2.3)	1.00 (1)
Azathioprine	2 (2.4)	3 (6.8)	0.34 (1)

¹Bortezomib use was limited to patients treated for antibody-mediated or mixed rejection.

df: Degree of freedom; ATG: Antithymocyte globulin.

Infectious complications (Table 3) showed overall infection rates of 42.64 (early) and 68.68 (late) episodes per 100 patient-years (IRR = 0.62, 95%CI: 0.46-0.83, P < 0.01). Pneumonia rates were 6.46 (early) and 14.79 (late) per 100 patient-years (IRR = 0.44, 95%CI: 0.22-0.86, P = 0.02). Sepsis rates were 2.58 (early) and 7.40 (late) per 100 patient-years (IRR = 0.35, 95%CI: 0.13-0.96, P = 0.04). UTI and diarrhoea rates were similar between groups. Viral infection rates (Table 4) were 4.84 (early) and 11.62 (late) per 100 patient-years (IRR = 0.42, 95%CI: 0.19-0.91, P = 0.03). Rates of CMV disease, BK viraemia, tubercular, bacterial, fungal, and parasitic infections were not significantly different (Table 4). Parasitic infections in the early group (7.1%) included cryptosporidiosis, microsporidiosis, and giardiasis (two patients each); the late group (6.8%) had cryptosporidiosis (three patients). Diarrhoea demonstrated a non-significant trend toward higher incidence in the late group (P = 0.06). No BK virus nephropathy cases were observed. Graft loss rates were similar between groups. Overall mortality was significantly higher in the late pulse steroid group compared with the early group (18.2% vs 4.8%, P = 0.02). Review of available clinical records indicated that most deaths in the late group occurred during hospitalization for severe infections, predominantly sepsis and pneumonia, whereas deaths in the early group were infrequent (Table 1).

Table 3 Infectious complications in patients receiving early and late steroids.

Infection	Early (n = 84) (no of episodes per 100 patient years)	Late (n = 44) (no of episodes per 100 patient years)	Incidence rate ratio (95%CI)	P value
Overall infections	42.64	68.68	0.62 (0.46-0.83)	< 0.01
Pneumonia	6.46	14.79	0.44 (0.22-0.86)	0.02
Urinary tract infection	12.27	19.02	0.64 (0.37-1.13)	0.13
Diarrhoea	10.98	19.02	0.58 (0.33-1.02)	0.06
Sepsis	2.58	7.40	0.35 (0.13-0.96)	0.04

CI: Confidence interval.

Table 4 Infections due to specific etiological agents in patients receiving early and late steroids.

Infection	Early (n = 84) (total infection episodes)	Early (n = 84) (total infection episodes per 100 patient years)	Late (n = 44) (total infection episodes)	Late (n = 44) (total infection episodes per 100 patient years)	Incidence rate ratio (95%CI)	P value
Bacterial infection	41	13.24	13	13.74	0.96 (0.52-1.80)	0.91
Viral infection	15	4.84	11	11.62	0.42 (0.19-0.91)	0.03
CMV disease	6	1.94	3	3.17	0.61 (0.15-2.44)	0.49
BK viremia	3	0.97	1	1.06	0.92 (0.09-8.82)	0.94
Tuberculosis	4	1.29	2	2.11	0.61 (0.11-3.34)	0.57
Fungal infection	9	2.91	7	7.40	0.39 (0.15-1.05)	0.06
Aspergillosis	3	0.97	2	2.11	0.46 (0.08-2.74)	0.39
Mucor	1	0.32	1	1.06	0.31 (0.02-4.89)	0.40
Parasitic	8	2.58	3	3.17	0.81 (0.22-3.07)	0.76

CI: Confidence interval; CMV: Cytomegalovirus.

This study relied on incidence rate-based analyses rather than time-to-event methods because detailed time-to-first-infection data were not consistently available in this retrospective cohort. IRRs are particularly well-suited for evaluating infectious complications in transplant recipients, where recurrent events and heterogeneous follow-up are common. This approach allowed meaningful comparison of infection burden between early and late pulse steroid groups while accounting for differential observation periods. Complete time-to-first infection data were not consistently retrievable from retrospective records, precluding Kaplan-Meier or log-rank survival analyses.

DISCUSSION

The relationship between pulse steroid therapy and infectious complications in renal transplant recipients is complex and not fully elucidated. A meta-analysis by Edel et al[7] found no increased risk of serious infections with pulse steroids, but it excluded renal transplant patients, limiting its applicability. In other contexts, such as Antineutrophil Cytoplasmic Antibody-associated vasculitis, high-dose intravenous methylprednisolone was associated with elevated infection risk within the first three months[8]. In transplant settings, the interplay of multiple immunosuppressive agents complicates isolating the effect of pulse steroids. Historical data from the 1970s indicated that high-dose steroids for rejection led to infections in 20% of cases[9]. However, a paediatric study comparing high-dose intravenous methylprednisolone to low-dose oral prednisolone found no difference in septicaemia or pneumonia rates[10]. Specific infections, such as BK virus replication[11], nocardiosis[12], pneumocystis jiroveci pneumonia[13], mucormycosis during coronavirus disease-2019[14], and herpes zoster[15], have been linked to pulse steroids in various cohorts.

The timing of pulse steroid administration may modulate infection risk due to differences in the net state of immunosuppression. Early post-transplant, the high baseline immunosuppressive burden from induction and maintenance therapies may amplify the effect of pulse steroids, though routine prophylaxis (e.g., cotrimoxazole) may mitigate some risks. A Brazilian study found that pulse steroid was a risk factor for early but not late invasive fungal infections[16]. In the late period, reduced prophylaxis and less frequent monitoring may increase susceptibility. A study by Joachim et al[17] reported a six-fold higher pneumonia risk with pulse steroids for chronic antibody-mediated rejection after three months, implicating pulse steroids as a key contributor. Our findings align with this, showing higher rates of overall infections, pneumonia, sepsis, and viral infections in the late pulse steroid group. It is also plausible that patients requiring late pulse steroid therapy represent a clinically higher-risk subgroup with chronic or refractory rejection, which itself may be associated with adverse outcomes. However, the parallel increase in infection-specific outcomes - including pneumonia, sepsis, and viral infections - supports the interpretation that heightened infectious susceptibility following late immunosuppressive escalation plays a central role in the observed mortality difference. Future prospective studies incorporating cause-specific mortality and standardized rejection severity scoring would help disentangle the overlapping risks.

An important secondary finding of this study was the significantly higher mortality observed among recipients receiving late pulse steroid therapy. Although late rejection episodes may reflect greater immunological severity and may necessitate more aggressive immunosuppressive interventions, the excess mortality in the late group appeared to be temporally associated with severe infectious complications, particularly sepsis and pneumonia. This observation is consistent with the substantially higher infection burden observed in this group and suggests that infectious complications likely contributed meaningfully to mortality rather than rejection severity alone. These findings underscore the vulnerability of late post-transplant recipients to fatal infections following intensification of immunosuppression.

Several factors may explain the higher infection rates in the late group. T-cell susceptibility to glucocorticoids may vary over time. Early post-transplant, high maintenance steroid levels may downregulate glucocorticoid receptors on T-cells, reducing their sensitivity to pulse steroids, as demonstrated by Berki et al[18]. In contrast, late post-transplant, lower maintenance immunosuppression may result in greater T-cell suppression by pulse steroids, increasing infection risk. Additionally, latent infections acquired early may reactivate during heightened immunosuppression in the late period, as supported by the RESITRA study, which identified relapsing viral infections as a risk factor for late infections[19]. The cessation of prophylaxis in the late period may further exacerbate this risk.

Study strengths include its large cohort from a high-volume tertiary centre, exclusion of patients receiving both early and late pulse steroids to avoid confounding, and adjustment for follow-up duration using infection rates per 100 patient-years. The similarity in baseline characteristics and immunosuppressive regimens between groups enhances comparability. This is the first study to specifically compare early and late pulse steroid effects on infections in renal transplant recipients. This study has a few limitations. Cause-specific mortality could not be systematically adjudicated due to the retrospective nature of the dataset and reliance on existing medical records. Additionally, formal competing-risk modelling to account for death as a censoring event for infection outcomes could not be performed. Nevertheless, the consistent direction and magnitude of infection-related findings across multiple outcomes support the robustness of the observed association between late pulse steroid therapy and adverse infectious and survival outcomes. The study was not powered to detect differences in rare infectious subtypes, resulting in wide confidence intervals for certain outcomes, such as fungal infections. The higher rate of bortezomib use in the late group (9.1% vs 1.2%, P = 0.05) represents a potential unmeasured confounder that may have influenced the observed infection rates. Multivariable adjustment for potential confounders such as bortezomib use, plasmapheresis, rejection phenotype, and post-transplant diabetes mellitus was not feasible due to the limited sample sizeand risk of model over-fitting. Therefore, the reported IRRs account for follow-up duration but are not fully adjusted for all potential clinical covariates. Furthermore, aggregate patient-years of follow-up after pulse steroid therapy could not be separately retrieved for each group due to the retrospective nature of the dataset; however, the use of incidence rates and rate ratios incorporating individual follow-up time mitigates bias arising from unequal observation periods. The absence of time-to-event survival analysis limits visualization of infection-free survival differences between groups.

CONCLUSION

This study demonstrates that late pulse steroid therapy is associated with significantly higher rates of overall infections, pneumonia, sepsis, and viral infections compared to early therapy in renal transplant recipients. These findings underscore the need for heightened vigilance and tailored infection prevention strategies following late pulse steroid administration. Clinicians should consider increasing monitoring frequency and potentially reinstituting prophylactic agents like cotrimoxazole to reduce infection risk. Future prospective studies are needed to confirm these findings and explore the mechanisms underlying the differential infectious risk, such as T-cell dynamics and latent infection reactivation. These insights could guide personalized immunosuppressive and prophylaxis protocols to optimize outcomes in renal transplant recipients.