Renal involvement in visceral leishmaniasis has been recognized for many years, with several studies reporting the occurrence of urinary abnormalities6,50–52, tubulointerstitial nephritis, GN, and AKI51–58. Nonetheless, the mechanisms by which VL can impair kidney function are not fully understood. This study prospectively evaluated the involvement of anti-Leishmania Igs and CIC in the occurrence of AKI in a cohort of hospitalized patients with VL under treatment from Minas Gerais.
Previous studies in Ethiopia52 and Brazil8,56,57,61 addressing AKI in VL identified several risk factors such as male sex, advanced age, jaundice, hyperglobulinemia, hypoalbuminemia, and HIV or other secondary infections8,52,61. The use of amphotericin B or other nephrotoxic drugs for VL treatment, alongside pre-renal factors such as hypotension and fluid loss, as well as the presence of parasites in renal tissue, have been implicated in renal damage in VL42,52,54,57,61,66–69,77. However, no consensus on the definitive factors involved in the occurrence of AKI in VL remains.
Similarly, data on AKI prevalence are inconsistent. A previous Brazilian study in Ceará reported 26.3% (15 of 57 patients) of AKI in patients admitted at a reference hospital56. Subsequently, a prevalence of 33.9% (76 of 224 patients)61 and 46% (23 of 82 patients)57 was reported in adult cohorts recruited at the same hospital. A similar rate (45.9%; 67 of 146 patients) of AKI was also demonstrated in children with VL8, and 17.4% (52 of 298 patients) of AKI was reported in an Ethiopian cohort52. Herein, we equally demonstrated a high prevalence of kidney injury in patients with VL, accounting for 85.7% (12 of 14) of patients. This prevalence was greater than that reported previously by other groups, which may be attributed to different criteria used to determine AKI, sample size, or even severity of patients, as kidney failure is an important criterion to predict VL death5.
Notably, these studies evaluated renal function using sCr levels, since this is commonly the endogenous marker used in clinical practice and for staging AKI according to KDIGO guidelines. Nonetheless, the aforementioned levels can be affected by diet, muscular mass, age, sex, drugs, and other endogenous substances such as bilirubin80–82. Therefore, cystatin C has been considered an alternative and important renal injury biomarker as it is not affected by the factors mentioned80–84. Additionally, cystatin C presents higher sensitivity when compared to sCr, since its levels exhibit elevation even in cases of mild renal function impairment80. Here, we concurrently evaluated the renal function through plasmatic levels of cystatin C and sCr in patients with VL, which indicated an increase in cystatin C from early treatment visit, while sCr, and in turn, eGFR, remained unchanged in the same period. Thus, our data reinforced previous reports53,85 that cystatin C is a superior marker of renal function than sCr due to better sensitivity and consistency. Despite some reports of changing inflammation86,87, this cohort did not exhibit a positive relationship between cystatin C and inflammatory markers, such as IL-1β, IL-6, TNF, and IFN-γ in bivariate (data not shown) and multivariate analysis, supporting the use of cystatin C as kidney injury biomarker even in infectious diseases88–90.
Additionally, the long-term assessment of cystatin C levels displays that kidney injury in VL context has a transitory feature with mild to moderate intensity, in accordance with previous reports52. This underscores the necessity of more sensitive biomarkers to assess renal function in patients with VL, which could lead to early detection of kidney impairment. In this context, elevated levels of serum neutrophil gelatinase-associated lipocalin were also observed in patients with VL who developed AKI57, thus emphasizing the potential of these molecules as promising biomarkers to guide the clinical management of VL.
Nonetheless, only a limited number of studies have sought to understand whether the immune response to VL and its immunopathogenic consequences might be implicated in this acute renal impairment. Considering the extent of immune activation in VL, particularly the humoral response, we posit that an underlying cause of kidney damage could plausibly be of immunological origin. Consequently, the excessive production of Ig, coupled with the subsequent formation of CIC that can aggregate and preferentially accumulate on endothelium cells within high-pressure vessels, assumes a pivotal role in kidney inflammation through activation of the complement system, and destruction of nephron units91.
As anticipated, patients with VL had elevated levels of CIC containing IgG, IgG1, and IgG3. Interestingly, CIC was positively correlated with the levels of specific immunoglobulins, indicating the presence of parasite antigens in the CIC composition, as previously reported35,36. A positive correlation between the levels of cystatin C and CIC (IgG-, IgG1- and IgG3- containing) or Leishmania-specific Igs (IgG, IgG1, and IgG3) was observed in patients with VL. Additionally, the multivariate analysis revealed that CIC containing IgG or CIC-IgG and anti-Leishmania IgG3 were predictors for cystatin C levels, suggesting an immunological nature of VL renal damage, which corroborates with the findings reported by Elshafie et al 62.
Concomitantly, our findings align with the established knowledge regarding the relationship between CIC and tissue injury in models of autoimmunity and other infectious diseases92,93. In VL, earlier reports underscored the occurrence of GN with Igs deposits in both animal models and human disease 41,48,49,78. Although the direct assessment of in situ CIC was not feasible in this study, a significant proportion of patients whose urinalysis data were evaluated exhibited indications of glomerular inflammation, as evidenced by hematuria94,95, pyuria, proteinuria, and granular casts– indicative of an immune complex-mediated GN.
Emphasizing CIC formation as an inherent and constant process, typically accompanied by subsequent mechanisms of inactivation and elimination, in which the activation of the complement system assumes a central role is important37. Notably, in our cohort, high serum levels of C3a and C5a were observed. Interestingly, CIC containing IgG, IgG1, or IgG3, and specific Igs levels are positively correlated with C5a levels, which reinforced that high levels of CIC are associated with activation of the complement system. The activation of the complement system stands as a key mechanism involved in GN mediated by immune complexes92. Considering this and the fact that VL itself may facilitate CIC deposits in several ways, the positive correlation between C5a and cystatin C levels implies the potential involvement of the complement system activation in renal damage among patients with VL. Furthermore, the importance of C5a and its receptor (C5aR) in kidney diseases has been described in animal models of lethal C3 glomerulopathy96, ischemia-reperfusion97,98, lupus nephritis99, immunoglobulin A nephropathy100, rhabdomyolysis-induced AKI101, diabetic kidney disease102 and in human mesangial proliferative GN103. This involvement primarily manifests through a C5a-mediated proinflammatory response and apoptosis and may be occurring in VL.
In concurrence with this, amphotericin B, the first-line treatment for VL according to the Brazilian Minister of Health guidelines, is usually associated with severe side effects, such as nephrotoxicity and AKI8,56,61. Nonetheless, the severity of these side effects depends on the formulation used during treatment. Lipidic formulations typically have more tolerability and provide a safer and more effective treatment, albeit still warranting a careful monitoring75,76. Corroborating with the aforementioned data, five patients in this study initially received D-Amph. B, while only one patient received meglumine antimoniate as the initial choice for VL treatment. The majority of the patients (4 out of 5 with D-Amph. B and meglumine antimoniate) presented complications during VL treatment, predominantly of renal or hepatic nature. Given this, L-Amph. B was prioritized as a replacement owing to its safety. It may explain the higher prevalence of AKI in patients with VL observed in our study as compared to the data reported by other groups.
Furthermore, because of L-Amph. B is the preferred treatment for patients with VL with severe conditions, such as HIV coinfection or kidney impairment, the occurrence of AKI could be mistakenly overestimated when associated with this treatment. Thus, patients with VL concomitantly infected with HIV or with a previous impaired kidney function were excluded from this evaluation. In addition, the patients were stratified according to daily and cumulative doses of amphotericin B and days of treatment to assess their influence on renal injury. Interestingly, by previous reports8,104, no significant changes in sCr, eGFR, or cystatin C levels were observed during the amphotericin B use in our cohort. Nevertheless, this does not eliminate the potential impact of these factors on the observed phenomenon, particularly since we lack a comprehensive study design for this purpose, and the relationship between amphotericin B formulations and AKI was not investigated in this study.
Here, we seek to highlight that other factors, such as parameters of the immune response in VL potentially contribute to renal injury. Thus, AKI in patients with VL was associated with high levels of specific Igs, CIC, and C5a. Concomitantly, despite no relationship between amphotericin B use and renal function impairment in our cohort, we believe that the treatment may be an important cofactor since it can release parasitic antigens into the circulation53,105, favoring the formation of immune complexes and complement system activation.
Despite the limitations of our study, including its relatively small sample size, absence of urine markers evaluation, or the deposition of immunocomplexes in situ, the findings presented herein underscore an additional mechanism to join other evidence for the full understanding of VL-related renal dysfunction. Our findings demonstrated a significant issue that warrants comprehensive exploration in a subsequent multicentric study involving a larger number of patients, intrinsic differences to the parasite, population, clinical severity, and therapeutic interventions.
In summary, AKI is an important issue in VL, being associated with morbidity and mortality56,57. The occurrence of AKI has a multifactorial feature, in which comorbidities, hemodynamic abnormalities, VL treatment, or use of nephrotoxic drugs as well as levels of immunocomplexes and immunoglobulins and complement system activation may be associated with nephrotoxicity8,57,61,66–69. Given this scenario, the need for a multicenter study is evident, with standardized and well-controlled clinical and immunological analyses to determine the actual occurrence of AKI in patients with VL. This approach will facilitate a more precise definition of the factors involved in kidney disease associated with VL. Moreover, our data reinforces the importance of careful monitoring of renal functions in patients with VL, employing markers with high sensitivity. This early detection of VL-associated kidney injury is essential for averting complications and guiding the clinical management of the disease.