Background
Contrast-induced nephropathy (CIN) is the third leading cause of acute renal failure in hospitalized patients as a result of the expanded use of iodinated contrast media (CM). Consequently, this iatrogenic complication is associated with prolonged hospitalization, significant morbidity and mortality, and increased health care costs [
1,
2]. The incidence of CIN is higher in patients with predisposing risk factors such as renal impairment, diabetes mellitus, advanced age, congestive heart failure, simultaneous use of nephrotoxic drugs, hypovolemia or large amounts of CM [
3].
The pathophysiology of CIN is complex and not fully understood. Yet, evidence from numerous studies suggests that a combination of several mechanisms is responsible for the development of CIN [
4]. Local renal ischemia is a direct result of CM induced prolonged vasoconstriction, which primarily affects the outer medulla. Hypoxic injury to this region is aggravated by an increased tubular cell oxygen demand after administration of CM. Consequently, oxidative stress, which enhances the production of reactive oxygen species (ROS) and triggers a local inflammatory response, may cause additional cell injuries during the reperfusion phase [
5]. The second important effect of CM involves direct cytotoxic damage to renal tubular cells.
Mannose-binding lectin (MBL) is an innate pattern-recognition, multimeric protein of the complement system that is primarily synthesized in the liver. Several mutations in the
MBL2 gene negatively influence the concentration of circulating functional MBL multimers. Hence, MBL deficiency can be observed in up to 30% of the Caucasian population [
6]. Binding of MBL to certain sugar residues on pathogens leads to the activation of the lectin pathway of the complement cascade and subsequently to killing or phagocytosis of microorganisms. Furthermore, MBL is also involved in the binding and removal of dying cells, e.g. after ischemic injury [
7,
8].
With regard to ischemia and oxidative stress several studies have highlighted the crucial role of MBL in aggravating the inflammatory response and tissue damage during ischemia/reperfusion (I/R) injury of the heart [
9‐
11], the brain [
12,
13], as well as the kidney [
14‐
16]. However, comparable data in human CIN are scarce. In a recent study Wang et al. have shown that MBL was strongly upregulated in the urine proteome profiles after application of CM [
17]. Moreover, in patients with CIN urine levels of MBL significantly increased after the procedure whereas MBL levels remained stable in patients who did not develop CIN. These results suggest that the lectin pathway might be involved in the pathogenesis of human contrast-medium-induced kidney injury. Hence, the current study tested the hypothesis that MBL deficiency is associated with a reduced incidence of CIN.
Methods
Study design
Post-hoc analysis of a multicenter, randomized, open-label, controlled trial (Clinicaltrials.gov Identifier: NCT00130598) that compared three different prevention procedures of CIN between March 2005 and December 2009 [
18]. The primary endpoint of the study was the maximum change in estimated glomerular filtration rate (eGFR) within 48 hours.
Study population
In the original trial 273 admitted patients with renal dysfunction, who were scheduled to receive radiographic CM during diagnostic or therapeutic procedures within the next 24 hours, were randomized to three prevention regimens. Subsequently, 258 patients were included in the final analysis. The study was conducted according to the principles of the revised Declaration of Helsinki, had been approved by the local ethical committees (Ethikkommission beider Basel, Basel/Liestal, Switzerland; Comitato Etico Centro Cardiologico Monzino, Milano, Italy), and all participants gave written informed consent for the study. The aim of the original trial was to investigate two regimens of sodium bicarbonate vs. standard volume supplementation with isotonic sodium chloride in the prevention of CIN.
Renal dysfunction was defined as serum creatinine level >93 μmol/L for women and >117 μmol/L for men or eGFR <60 mL/min/1.73 m2 as assessed 24 hours before the radiographic procedure. Patients were excluded if they were <18 years old, pregnant, or allergic to radiographic contrast, were undergoing dialysis, had severe heart failure (NYHA III-IV), had taken N-acetylcysteine ≤24 hours before CM administration, or had a clinical vulnerable condition requiring continuous fluid therapy (e.g. severe sepsis).
Patients were randomly assigned to one of three preventive regimens which are outlined in more detail in the original study [
18] (Group A: Standard 24 h sodium chloride; group B: 7 h sodium bicarbonate; group C: Short-term sodium bicarbonate).
Definition of endpoints
The primary endpoint for this study was the development of CIN. Multiple definitions have been used to quantify renal damage after CM. Creatinine-based CIN was defined as an increase of ≥44 μmol/L or ≥25% in serum creatinine concentration within 48 hours after exposure to CM. Cystatin C is an evolving marker for the prediction and diagnosis of CIN. Several studies suggest that serum cystatin C is a more sensitive and rapid indicator of changes in GFR than serum creatinine [
19]. In a recent study, a cystatin C increase of ≥10% 24 hours after exposure to CM was found to reliably predict the occurrence of CIN and future adverse events (death from any cause and chronic dialysis) [
20]. Therefore, we included this cut-off for the definition of cystatin-based CIN in our secondary analysis. Serial serum cystatin C samples were available from 241 of 258 patients. Furthermore, time to hospital discharge, in-hospital morbidity and mortality, renal replacement therapy and mortality at 90 days were included as secondary endpoints.
Assessment of renal function
Peripheral venous blood samples for the measurement of creatinine and cystatin C were obtained on the day preceding the scheduled administration of CM and in the morning of the subsequent two days after the contrast exposure. The measurement of these parameters is outlined in more detail in the original study [
18].
The eGFR was calculated by means of the abbreviate
Modification of Diet in Renal Disease (MDRD) equation [
21].
Determination of mannose-binding lectin levels
For the MBL level analysis baseline serum samples were available from 246 of 258 patients. MBL serum concentrations were measured on the same day blinded to any patient data using a commercially available Sandwich-ELISA Kit (
MBL Oligomer ELISA KIT 029, Lucerna Chem, Luzern, Switzerland) as described previously [
22].
Depending on the clinical setting several cut-off values for MBL deficiency have been proposed ranging from 100 to 1000 ng/mL [
6]. Thus, we first analyzed the predefined endpoints in relation to MBL levels, and only in a secondary analysis chose a threshold of ≤500 ng/mL for the definition of MBL deficiency, as this cut-off has been shown to reliably predict low-producing
MBL2 genotypes [
23].
Statistical analysis
Continuous data are expressed as means (standard deviation) or median (interquartile range (IQR)) where appropriate, whereas categorical variables are summarized as frequency (percentage). Differences in patient characteristics and outcome measures according to MBL serostatus were analyzed using the Fisher`s exact test or the Mann–Whitney-U-Test where appropriate. Due to the non-Gaussian distribution of human MBL levels (as determined by Q-Q plot and normality tests) two-group comparison of serum MBL levels were performed using a Mann–Whitney-U-test, whereas a Kruskal-Wallis one-way analysis of variance or Friedman test was used for multigroup comparison where appropriate. Dunn’s post test was used to correct p-values for multiple comparisons.
Stepwise logistic regression models were used to estimate the effect of MBL levels on predefined endpoints in multivariate analyses after adjustment for covariables with univariate P values less than 0.1. Covariables tested in univariate analysis included patient age, sex, baseline renal function (as assessed by serum creatinine or cystatin C), history of vascular risk factors, acute myocardial infarction or congestive heart failure, amount of administered CM, medication use, prevention regimen (B or C versus A) and baseline blood pressure. All testing was two-tailed, and p values less than 0.05 were considered to be statistically significant. All statistical analyses were performed with the use of SPSS for Windows, version 15.0 (SPSS), and Prism for Windows, version 4 (GraphPad).
Discussion
The lectin pathway of the complement system has been shown to considerably contribute to tissue damage during ischemia/reperfusion injury of several organs, including the kidneys [
14,
16]. This is the first study to assess the importance of serum MBL with respect to the development of CIN in a clinical trial. MBL deficiency did not influence the occurrence of CIN as defined by a commonly quoted serum creatinine increment. However, it was associated with a limited increase in Cystatin C after the administration of CM.
To our knowledge the role of MBL has not been investigated in a rodent model of CIN. However, the fact that local ischemia and reperfusion is at least in part responsible for the development of CIN [
4], and that MBL aggravates tissue damage during I/R injury [
9,
11,
12,
15], suggests a role of MBL in the pathogenesis of contrast-induced acute kidney injury.
With regard to the primary endpoint, the incidence of creatinine-based CIN, several reasons might account for the absent effect of MBL deficiency in this study. According to the baseline characteristics, the study cohort included only a rather small percentage of very high-risk patients (as reflected by 29/246 (11.7%) with GFR <30 mL/min/1.73 m
2, 52/246 (21.1%) with coronary angiography and ad hoc PCI, 69/246 (28.0%) with contrast media volume ≥140 mL [
24], 90/246 (36.6%) with diabetes mellitus, and the exclusion of patients with advanced congestive heart failure), and CIN was diagnosed infrequently, especially in the sodium chloride prevention group (incidence 1.2%). Due to the latter fact the analysis of smaller differences between MBL sufficient and deficient patients is limited. Furthermore, in this study cohort low to moderate CM-induced damage is probably not reflected by a decline in kidney function as measured by the serum creatinine concentration, a parameter with limited sensitivity to detect an acute deterioration in renal function within 48 hours [
25]. Besides underestimating the true change in GFR, the increase in serum creatinine after CM exposure is delayed achieving a maximum two to five days after CM exposure as compared to cystatin C, a more sensitive marker, which was shown to rise earlier, to peak as early as 24 hours after CM administration, and to detect even subtle changes in GFR after acute kidney injury including CIN [
19,
26‐
30]. Indeed, when we analyzed the influence of MBL deficiency on the cystatin C course after CM exposure we observed a remarkable association: As compared to patients with MBL levels >500 ng/mL subjects with MBL deficiency were almost two-times less likely to develop a cystatin C increase ≥10% after administration of CM, a cut-off that has recently been proposed as an independent diagnostic and prognostic tool with respect to the occurrence of CIN and future adverse events [
20]. This suggests that deficiency of MBL might attenuate some of the detrimental effects of CM. Though being small in magnitude this increase in cystatin C of ≥10% might have important consequences for the patients as even apparently minor decreases in renal function have been shown to be associated with excessive mortality rates independent of other known risk factors [
20,
31]. However, we were not able to demonstrate a consistent association of MBL deficiency with superior clinical outcomes, which the study was not powered for.
On the other hand, the importance of renal I/R injury in pathogenesis of CIN might be overestimated as several other factors (including direct toxic effect on renal tubular cells, increased urinary viscosity, and tubular obstruction) influence kidney function after exposure to CM [
32]. Moreover, I/R models that demonstrated a pivotal role of MBL in the context of renal ischemia (i.e. transient or permanent occlusion of a major renal artery) are most likely not comparable to the complex events occurring after administration of CM. The duration and extent of ischemia might be crucial as well. In a rodent model of renal I/R activation of the MBL-pathway was not induced by warm ischemia of less than 30 minutes [
14]. Data on the duration and significance of renal vasoconstriction and hypoxia after CM exposure in humans remain controversial [
4]. However, it has been demonstrated that MBL and its associated protease mannose-binding protein-associated serine protease 2 are strongly upregulated in the post-procedural urinary proteome profiles after application of CM, and that urine levels of MBL significantly increase in patients who developed CIN whereas MBL levels remain stable in non-CIN patients [
17]. Finally, the alternative pathway has been implicated in experimental renal ischemia/reperfusion injury and might play a dominant role in humans as compared to the MBL pathway [
33].
Despite the randomized controlled trial design of the original study and the precisely characterized cohort the present study has important limitations including the post hoc analysis of MBL serum levels and the use of a surrogate marker as primary endpoint. Furthermore, limiting the measurement of serum creatinine to 48 hours after CM exposure might result in an underestimation of creatinine-based CIN. As genetic material was not available, analysis of MBL deficiency solely relied on MBL phenotype. However, MBL serum levels show little variation throughout life, and correlate well with the functional activity of the MBL pathway
in vivo[
34]. As individuals with the same genotype may vary up to tenfold in MBL serum levels, measurement of MBL serum levels by ELISA might in fact represent a more reliable approach than determination of genotypes [
35,
36]. MBL levels (as measured by sandwich ELISA) have been shown to be significantly increased in Asian pre-dialysis and dialysis patients compared to healthy controls [
37,
38]. To the best of our knowledge comparable studies in a Caucasian population with moderate renal impairment (similar to our study patients) are lacking.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
MO, VP, MT, and CM were the principal investigators and take primary responsibility for the paper. CM and MT designed the study. TK, AC, IM, SH, and TB recruited the patients and collected data. MO and VP performed the laboratory work for this study, analyzed and interpreted data, and wrote the article. TK, AC, IM, SH, TB, GM, MT, and CM provided intellectual content, revised the article, and gave final approval.