Trimetazidine Decreases Risk of Contrast-Induced Nephropathy in Patients With Chronic Kidney Disease

Abstract

Objectives:

We sought to synthesize and analyze the available data from randomized controlled trials (RCTs) for trimetazidine (TMZ) in the prevention of contrast-induced nephropathy (CIN).

Background:

Contrast-induced nephropathy after coronary angiography is associated with poor outcomes. Trimetazidine is an anti-ischemic drug that might reduce incidence of CIN, but current data are inconclusive.

Methods:

We searched MEDLINE/PubMed, EMBASE, Scopus, Cochrane Library, Web of Science, and ScienceDirect electronic databases for RCTs comparing intravenous hydration with normal saline (NS) and/or N-acetyl cysteine (NAC) versus TMZ plus NS ± NAC for prevention of CIN. We used RevMan 5.2 for statistical analysis with the fixed effects model.

Results:

Of the 808 studies, 3 RCTs met criteria with 290 patients in the TMZ plus NS ± NAC group and 292 patients in the NS ± NAC group. The mean age of patients was 59.5 years, and baseline serum creatinine ranged from 1.3 to 2 mg/dL. Trimetazidine significantly reduced the incidence of CIN by 11% (risk difference 0.11; 95% confidence interval, 0.16-0.06; P < .01). There was no significant heterogeneity between the studies (I² statistic = 0). The number needed to treat to prevent 1 episode of CIN was 9.

Conclusions:

The addition of TMZ to NS ± NAC significantly decreased the incidence of CIN in patients undergoing coronary angiography. In conclusion, TMZ could be considered as a potential tool for prevention of CIN in patients with renal dysfunction.

Introduction

Contrast-induced nephropathy (CIN) is the third most common cause of hospital-acquired acute kidney injury. ¹ It is defined as an elevation of serum creatinine (SCr) by ≥0.5 mg/dL or 25% within 48 hours after contrast administration. The pathophysiology of CIN is thought to involve medullary hypoxia secondary to renal vasoconstriction and acute tubular necrosis (ATN) due to direct cytotoxic effects from the contrast agent. Its incidence varies widely and depends on patients’ risk factors, preexisting chronic kidney disease (CKD) being the most important of these; other predisposing conditions include diabetes, congestive heart failure, older age, hypotension, and increased contrast volume. ^{2 –5} Although CIN after coronary angiography and percutaneous cardiovascular intervention (PCI) is usually self-limited, it is associated with prolonged hospitalization, more adverse cardiovascular events, and increased in-hospital and long-term mortality. ^6,7 Prognosis is worse for patients with CIN requiring dialysis. ^3,6 The number of patients at risk of CIN is expected to increase due to aging population as well as the escalating prevalence of CKD and diabetes. ⁷ Since the treatment of CIN is mostly supportive, the cornerstone of management is prevention especially in patients at high risk.

Available options for prevention include minimizing the contrast dose used, avoiding volume depletion, cessation of nephrotoxic medications, administration of intravenous (IV) saline or sodium bicarbonate, and use of oral N-acetyl cysteine (NAC). ⁸ The 2014 European Society of Cardiology (ESC) and the European Association for Cardio-Thoracic Surgery (ESC/EACTS) guidelines recommend the use of IV hydration (class I), reducing contrast dose, and using iso-osmolar contrast (class I and IIa, respectively) and short-term statin therapy (class IIa) for prevention of CIN in mild to moderate CKD. ⁹ Although these guidelines do not recommend use of NAC (class III), it is still used since it is well tolerated, potentially beneficial, and relatively inexpensive. ¹⁰

A cellular anti-ischemic agent trimetazidine (TMZ), a piperazine derivative, has been extensively used in Europe as an antianginal agent. ¹¹ Trimetazidine is protective against free radical damage, likely because of an antioxidant effect, and can minimize renal ischemic–reperfusion injury (IRI). ^12,13 A recent animal model study demonstrated that administration of TMZ is a protective tool against the histopathologic changes in CIN. ¹⁴

The purpose of this meta-analysis was to synthesize data regarding the efficacy of TMZ in addition to NS ± NAC versus NS ± NAC alone for preventing CIN in coronary angiography and percutaneous coronary interventions.

Material and Method

Search Strategy and Study Selection

Two authors (GNN and IK) independently and in duplicate searched MEDLINE/PubMed, EMBASE, Scopus, Cochrane Library, Web of Science, and ScienceDirect electronic databases from inception until May 2014 for RCTs comparing TMZ plus NS ± NAC versus NS ± NAC only for the prevention of CIN after coronary angiography and percutaneous coronary interventions. Search key words were “trimetazidine,” “prevention,” “coronary angiography,” and “percutaneous coronary intervention” (Supplemental Table 1). We also reviewed prior meta-analyses and the references of the selected studies. We did not set any search limitations by publication dates or language.

Data Extraction

For each of the selected trials, we extracted information regarding the number of patients enrolled, inclusion and exclusion criteria, procedure performed, definition of CIN, number of patients developing CIN in each group, control therapy used, and dose of TMZ. We also extracted the baseline characteristics of the study population including the presence of diabetes or hypertension among patients and SCr at randomization.

Statistical Analysis

We performed analyses in concordance with Cochrane Collaboration and the Quality of Reporting of Meta-analyses (QUORUM) ¹⁵ guidelines using Review Manager 5.2 version (Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark). The primary pooled outcome measure estimated was the Mantel-Haenszel-weighted risk difference (RD) using a random effects model. We conducted separate sensitivity analyses by excluding either the study with the largest number of patients or the study that included diabetic patients and then analyzing the remaining 2 studies. Two authors (GNN and IK) independently assessed the risk of bias of the included studies using the Cochrane Collaboration’s tool for assessing risk of bias. ¹⁶ Publication bias was estimated visually by funnel plots and the weighted regression test of Egger. ¹⁷ Additionally, in order to ascertain study quality indicators beyond the quality assessment scoring system, we conducted a meta-regression analysis in 2 steps. Meta-regression is a tool used in meta-analysis to examine the impact of moderator variables on study effect size using regression-based techniques. ¹⁸ The overall quality of clinical recommendations was rated using the Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) approach.

Results

The search strategy retrieved a total of 808 citations until May 2014. Through a thorough review of titles and abstracts, 750 studies were rejected for lack of relevance. Finally, 3 studies ^{19 –21} met the criteria and were included in this analysis (Figure 1). The main characteristics of individual studies are illustrated in Table 1.

Figure 1.

Trial flowchart showing selection of studies.

Table 1.

Included RCT Studies with Baseline Characteristics.^a

Study, Year, n	Procedure	Mean Age (SD)	% Male	Baseline Creatinine, Mean (SD), mg/dL	% Hypertension	% Diabetes Mellitus	Definition of Chronic Kidney Disease	Left Ventricular Ejection Fraction	Contrast Volume Used	Control	Dose of TMZ
Onbasili et al, 19 2007 (n = 82)	Diagnostic angio/PCI	60 (11)	69.5	1.31 (0.25) for TMZ and 1.26 (0.16) for control	100% for TMZ and 100% for control	20% for TMZ and 26% for control	Serum creatinine level >1.2 mg/dL or creatinine clearance of <50 mL/min	48% for TMZ and 47% for control	225 mL in TMZ and 240 mL in control group	Normal saline intravenously at 1 mL/kg/h for 12 hours before and 12 hours after procedure	20 mg 3 times a day for 72 hours starting 48 hours before procedure
Rahman et al, 20 2012 (n = 400)	Diagnostic angio/PCI	56.5 (10.4)	90	1.39 (0.23) for TMZ and 1.40 (0.24) for control	72% for TMZ and 86% for control	None	Serum creatinine level between 1.2 and 2.5 mg/dL	NM	95 mL in TMZ group and 97 mL in control group	Normal saline intravenously at 1 mL/kg/h for 12 hours before and 12 hours after procedure	35 mg twice daily for 96 hours starting 48 hours before procedure
Shehata, 21 2014 (n = 100)	Elective PCI	59 (6)	68	2 (0.5) for TMZ and 2 (0.4) for control	46% for TMZ and 50% for control	100% for TMZ and 100% for control	Creatinine clearance of 30-90 mL/min with mean creatinine clearance being 48.5 mL/min	54% for TMZ and 55% for control	270 mL in TMZ and 280 mL in control group	Normal saline intravenously at 1 mL/kg/h for 12 hours before and 12 hours after and N-acetyl cysteine 1200 mg orally before and after procedure	35 mg twice daily for 72 hours starting 48 hours before procedure

Abbreviations: angio, angiography; CIN, contrast-induced nephropathy; NM, not mentioned; PCI, percutaneous cardiovascular intervention; RCT, randomized controlled trial; TMZ, trimetazidin; SCr, serum creatinine; SD, standard deviation.

^aCIN was defined as an absolute increase in SCr by 0.5 mg/dL at 24 hours procedure and/or relative increase in SCr by 25% at 48 hours after procedure.

The overall analysis contained 582 patients (292 for the SOC group and 290 for the TMZ plus NS ± NAC group) with a mean age of 59.5 years. One study included patients undergoing elective PCI, ²¹ while 2 included patients undergoing either diagnostic angiography or elective PCI. ^19,20 The baseline SCr ranged from 1.26 to 2 mg/dL. One study included patients with both diabetes and hypertension, ²¹ whereas the remaining 2 studies excluded patients with diabetes. ^19,20

In the TMZ plus NS ± NAC group, 15 (5.2%) of the 290 patients developed CIN compared to 49 (16.8%) of the 292 patients in the control group. All studies individually concluded that the TMZ plus NS ± NAC groups had a lower incidence of CIN compared to control groups. Upon pooling the RDs, TMZ plus NS ± NAC was associated with a significant reduction in the incidence of CIN (RD 0.11; 95% confidence interval [CI], −0.16 to −0.06; P < .01; Figure 2). The number needed to treat to prevent 1 episode of CIN was 9 patients.

Figure 2.

Forest plot showing contrast-induced nephropathy (CIN) risk difference between trimetazidine plus NS ± NAC versus NS ± NAC alone. NAC indicates N-acetyl cysteine; NS, normal saline.

In separate sensitivity analyses, the RD between TMZ plus NS ± NAC versus control groups persisted after excluding the largest study ²⁰ (RD 0.15; 95% CI, 0.24-0.05; P < .01; Figure 3) and after excluding the study that included diabetic patients ²¹ (RD 0.11; 95% CI, 0.16-0.06; P < .01; Figure 4). The funnel plot of intervention effects of individual studies against each study’s size showed an asymmetric scatter plot; however, assessment of publication bias was not optimal due to the small number of studies (Figure 5). Table 2 shows the risk of bias assessment of studies as scored by the Cochrane Collaboration tool. ¹⁶

Figure 3.

Forest plot showing contrast-induced nephropathy (CIN) risk difference between trimetazidine plus NS ± NAC versus NS ± NAC alone after excluding the largest study. NAC indicates N-acetyl cysteine; NS, normal saline.

Figure 4.

Forest plot showing contrast-induced nephropathy (CIN) risk difference between trimetazidine plus NS ± NAC versus NS ± NAC alone after excluding diabetic patients. NAC indicates N-acetyl cysteine; NS, normal saline.

Figure 5.

Funnel plot showing intervention effects of individual studies against a measure of EACH study’s size.

Table 2.

Cochrane Collaboration Tool for Assessing Risk of Bias.

Study, Year	Sequence Generation	Allocation Concealment	Blinding	Incomplete Outcome Data	Selective Outcome Reporting	Other Sources of Bias
Onbasili et al, 19 2007	Unclear	Unclear	Unclear	Low	High	Low
Rahman et al, 20 2012	Unclear	Unclear	Unclear	Low	Low	Low
Shehata, 21 2014	Low	Low	Unclear	Low	Low	Low

Additionally, in order to ascertain study quality indicators beyond the quality assessment scoring system, we conducted a meta-regression analysis in 2 steps. All studies reported age, gender, body mass index, baseline SCr, and contrast volume. Only 2 studies reported history of hypertension and estimated ejection fraction, while diuretics, renin–angiotensin system blockers, metformin use, history of smoking, diabetes mellitus, and glucose control were reported in 1 study only. We were able to perform meta-regression only on variables reported in all studies. First, we evaluated the heterogeneity across the studies and then the differences in basic characteristics between treatment and control groups. No differences were found in both steps (Supplementary Table 2).

Discussion

Our meta-analysis based on currently available data demonstrates a significant decrease in incidence of CIN by means of TMZ administration in conjunction with NS ± NAC when compared to NS ± NAC alone in patients undergoing coronary angiography.

Contrast-induced nephropathy is a serious potential complication after coronary angiography procedures sometimes requiring renal replacement therapy, extended hospital stay, and permanent renal dysfunction and thus leading to an increased morbidity and mortality. ^3,6 Its incidence correlates strongly with the preprocedure renal function and ranges from 2% in general population to up to 50% in patients with advanced kidney disease. ^22,23 In this article, we present a meta-analysis of RCTs evaluating the use of the anti-ischemic drug TMZ as an additive measure and conclude that periprocedural use of TMZ decreases the incidence of postcoronary angiography CIN.

In these studies, CIN was defined as an elevation in SCr of >0.5 mg/dL or >25% from baseline. Shehata ²¹ evaluated a high-risk population (diabetic patients with CKD) and showed a 16% decrease in incidence of CIN in the TMZ group compared to the control group. The other 2 studies included in this analysis evaluated nondiabetic ²⁰ or a mixture of diabetic and nondiabetic ¹⁹ patients with CKD and found similar results. The TMZ was given orally (20 mg 3 times a day ¹⁹ or 35 mg twice a day ^20,21 ) for 72 hours starting 48 hours before the procedure. Of these studies, the largest was Rahman et al with 200 patients in each arm. This was well randomized at baseline with similar patterns of medication usage and kidney function in both arms. Also, as shown in the sensitivity analysis, the RD persisted even after excluding this study. Additionally, the meta-regression analysis showed no differences either across the studies or between groups indicating randomization to be successful.

The historically reported incidence of CIN is between 75% and 90% in diabetic patients with advanced kidney disease. ²⁴ Subsequently, preventive measures including hydration, cessation of diuretics and other nephrotoxic agents, the use of decreased contrast volume and low osmolar, nonionic contrast agents and possibly NAC among other therapeutic agents have been implemented with varying degrees of success. Authors of the articles included in this analysis enrolled patients with stable coronary artery disease undergoing elective coronary angiogram and randomized them in 1:1 fashion to receive either placebo or TMZ. Patients with acute coronary syndrome or urgent/emergent indication for angiography were excluded and only medically stable patients were enrolled in the studies. Both groups received NS at a rate 1 mL/kg 12 hours before the procedure and up to 12 hours ^20,21 to 24 hours ¹⁹ thereafter. Oral administration of NAC was used in 1 study. ²¹

Based on the assumption that reactive oxygen species (ROS) and renal medullar ischemia are involved in the pathogenesis of CIN, the authors of the above-mentioned RTCs hypothesized that the anti-ischemic and antioxidant properties of TMZ might exert a protective effect against CIN. The formation of ROS, renal medullar ischemia, vasoconstriction, and direct tubular damage from contrast agents is thought to play a major role in the mechanism of CIN. This insult usually happens immediately after the use of contrast agents and is reversible with no long-term sequelae in most cases. The pathology of CIN often involves ATN caused by the direct cytotoxic effects of the contrast media and/or renal vasoconstriction resulting in medullary hypoxia. The renal injury may be mediated by alterations in nitric oxide, endothelin, and/or adenosine. ^{25 –28} The vasoconstriction and decreased oxygenation during ischemia lead to anaerobic metabolism, depletion in adenosine triphosphate (ATP), and altering the membrane ionic ATP-dependent pumps favoring the entry of calcium, sodium, and water into the cell.

Trimetazidine exerts renoprotective effects through several different mechanisms. These effects have been experimentally assessed in various models including cell cultures, isolated tissues, and different perfused organs. It was found to decrease IRI in an animal model of acute renal failure, ^29,30 gentamicin-induced nephrotoxicity, ³¹ and CIN in the studies included in this analysis. With regard to IRI, TMZ was shown to prevent the mitochondrial membrane dissipation, restore nicotinamide adenine dinucleotide phosphate levels, decrease Ca²⁺-induced mitochondrial swelling, restore glutathione peroxidase levels, ³² and enhance the expression of hypoxia-inducible factor 1α. ²⁹ Another study found that TMZ protects against dephosphorylation of nucleotides leading to an increase in ATP concentration thereby preventing ischemia. ³⁰ Its beneficial effects extend from the time of initial injury to the recovery phase. The TMZ promotes renal tubular repair by increasing the expression of stathmin (increases the concentration of tubulin available in the cytoplasm for microtubule assembly to initiate the mitotic phase of the cell cycle and repair) and reduces the tubulointerstitial fibrosis. ²⁹ However, these late effects on tubulointerstitial fibrosis might be due to the effects of TMZ on CD4- and CD8-positive immune cells. ³³ During ischemia, inflammatory cells adhere to activated endothelium in the peritubular capillaries of the outer renal medulla resulting in further hypoxic injury to the proximal tubule. In autotransplanted pig kidneys, TMZ decreased inflammatory cell infiltration and decreased expression of vascular cell adhesion molecule 1, ³⁴ which might be another mechanism by which TMZ prevents CIN.

The therapeutic potential of TMZ in preventing CIN is still debated. Most of the data demonstrating the anti-ischemic antioxidant effect of TMZ come from preclinical studies. We present an analysis of 3 RCTs, with similar intervention and definition of outcomes. Our meta-analysis emphasizes the findings of these RTCs and suggests that TMZ may be used as an additive measure to decrease the incidence of CIN after cardiac catheterization in the appropriate clinical context. However, due to the small sample size in the selected studies, these findings need to be confirmed in a large, multicenter RCT. The selected studies included patients with moderate CKD limiting the ability of each study to generalize its findings to patients with severe renal disease. The inclusion of patients with severe renal disease might not have been feasible due to safety issues. The TMZ can cause extrapyramidal symptoms and other movement disorders and is contraindicated in patients with estimated glomerular filtration rate <30 mL/min. However, a recent international multicenter retrospective cohort study of patients with heart failure showed that TMZ is safe, and the addition of the drug to conventional optimal heart failure therapy can improve mortality and morbidity. ³⁵

Limitations

Our meta-analysis has limitations that should be taken into account. First, limited number of studies limits effectiveness of assessment of publication bias. Second, meta-regression relies on aggregated data from studies rather than data from individual patients. Therefore, the power to detect a difference in aggregate or to identify explanatory variables using meta-regression is greatly diminished compared with large primary trials with patient-level data. Additionally, in the present studies, the diagnosis of CIN was primarily based on the absolute or relative change in plasma creatinine concentration. Contrast media themselves may decrease tubular creatinine secretion and thereby leading to a small transient increase in plasma creatinine level, independent of changes in GFR. However, at present, SCr is the cheapest and most broadly accepted marker of kidney function. Also, the 3 studies utilized different dosing and duration of preprocedural TMZ and this is an important limitation. In addition, compliance to IV hydration/NAC cannot be assessed; however, it can be assumed to be high considering the inpatient nature of the RCTs. Additionally, the risk of bias was unclear across 3 of the 6 domains and based on the comorbidities of the patients studied, the applicability of the findings may not be generalizable. As a result, the level of evidence provided by this analysis is 1C. However, in spite of significant limitations, this analysis is potentially the best current evidence (1C) of the utilization of TMZ as an adjunct for prevention of CIN in high-risk patients as well the rationale for a large, multicenter clinical trial for evaluation of standardized dosing and duration of preprocedural TMZ.

Conclusion

In conclusion, this meta-analysis showed that TMZ could be considered as a potential tool for CIN prevention, especially in patients with CKD. However, considering the risk of bias inherent in the studies, the small sample size and the level of evidence being 1C, decision making should be individualized to each patient and each clinical context. A large, well-designed trial that incorporates the evaluation of clinically relevant outcomes in participants with different underlying risks of CIN is required to more adequately assess the role for TMZ in CIN prevention.