Comparison of Normal Saline, Ringer's Lactate, and Sodium Bicarbonate for Prevention of Contrast-induced Nephropathy in Patients with Coronary Angiography: A Randomized Double-blind Clinical Trial

Introduction

Contrast-induced nephropathy (CIN) is one of the major causes of prolonged hospitalization in patients who have received iodine contrast medium, resulting in excessive cost, mortality, and morbidity.[1234] CIN is a kind of reversible acute kidney injury (AKI) that is defined as an increase in serum creatinine (Cr) by ≥0.3 mg/dL within 48 h or ≥1.5 times baseline, which is presumed to have occurred within the prior 7 days; or urine volume <0.5 mL/kg/h for 6 h.[5] Traditionally, CIN is defined as an AKI that takes place 24–72 h after iodine radiographic contrast medium injection and is characterized by serum Cr rise by 0.5 or more; or its increase by 25% or more from its baseline.[2] However, in recent studies, any amount of Cr rise is considered important for diagnosis for CIN with exclusion of other causes of AKI.[25]

Incidence of CIN as a whole in general population is 0.6%–2.3%, but it is notably higher among different disease subgroups.[2] The accurate mechanism of contrast nephrotoxicity is not fully understood due to its multifactorial characteristics. Perhaps a combination of reduced medullary blood flow which leads to tissue hypoxia and a direct tubular injury due to toxic effects of contrast medium plays major roles in disease pathophysiology.[2]

Various strategies have been adapted for preventing CIN as vasoconstriction reduction, reestablishment of sufficient blood flow in renal capillaries, decreasing hypoxia in renal medulla, and anti-oxidant effect of some drugs.[67]

Despite heterogeneity in the results of different studies, European Renal Best Practice suggests that volume expansion with solutions such isotonic sodium chloride or sodium bicarbonate (Bi) is more beneficial rather than no volume expansion.[8] The rationale of administration of Bi or prescription of acetazolamide is that urine and medulla acidification is reduced with increased Bi excretion.

Since oxygen free radicals are produced in acidic environment and their production is withheld or their effects are neutralized in alkaline settings, urine alkalization can be considered as a preventive method for CIN.[91011121314] One can propose that administration of Ringer's lactate (RL) as a hydration solution to the patients can be beneficial. Since this solution contains less chloride that normal saline (NLS), it draws on less hyperchloremic acidosis. In addition, other studies have shown that restriction of chloride-rich solutions, although decreasing the risk of acidosis and renal failure, may give rise to incidence of metabolic alkalosis.[12] Considering the fact that not many studies have focused on hydration with RL in prevention of CIN, this study was aimed to evaluate the effect of this solution in urine alkalization and also suggesting the more effective solution prevention of CIN.

Materials and Methods

Results

In this study, 300 patients including 150 men (50%) with a mean age of 59.1 ± 10.6 years were studied. Patients were divided into three groups: NLS group with 96 cases, RL group with 103 patients, and Bi group with 101 participants. Table 1 describes some of the demographic and clinical data of our patients; there was no significant difference in baseline data of three groups at the beginning of the study. The overall incidence of CIN in our study was 10% (30 patients) [8.3% (8 patients) in NLS, 16.5% (17 patients) in RL, and 5% (5 patients) in Bi group, P = 0.018]. Pairwise comparison showed that difference of CIN incidence was only significant among Bi and RL groups (P = 0.012).

Table 2 compares the characteristics between who developed CIN and those who did not. There was not significant difference in baseline GFR (78.4 ± 26.0 with CIN vs. 69.8 ± 21.6 without CIN, P = 0.154). Although baseline urine pH had no significant difference at baseline (5.1 ± 1.3 in CIN vs. 4.9 ± 1.1 in without CIN, P = 0.212), urine pH after trial in CIN group was lower than in who did not develop CIN (5.5 ± 1.4 vs. 6.3 ± 1.8, P = 0.024). In addition, delta of urine pH was higher in those who did not develop CIN (1.4 ± 1.8 vs. 0.4 ± 1.5, P < 0.001).

Table 3 presents comparisons of laboratory data before and after contrast administration among our patients. It revealed significant Cr level reduction (1.0 ± 0.2 vs. 0.9 ± 0.2, P = 0.007) and increase in eGFR (67.4 ± 21.0 vs. 72.2 ± 22.5, P = 0.003) in Bi group. The K level after trial was significantly different among three groups (P = 0.012); post hoc analysis showed significance level among RL and NLS (P < 0.001) as well as RL and Bi (P = 0.033). The K level was significantly decreased after injection of Bi solutions (4.4 ± 0.5 before trial vs. 4.2 ± 0.4, P = 0.030). The urine pH level after trial was significantly increased in all study population (4.9 ± 1.2 vs. 6.2 ± 1.8, P < 0.001) and also in each group (4.9 ± 1.3 vs. 5.4 ± 1.3 in NLS, 4.8 ± 1.0 vs. 5.7 ± 1.3 in RL, and 4.9 ± 1.1 vs. 7.5 ± 1.9 in Bi; P < 0.001). Additionally, after trail urine pH was significantly different among three groups (P < 0.001); post hoc analysis revealed that this significance was only between Bi and NLS (P < 0.001) and between RL and Bi (P < 0.001). Calculation of urine pH difference prior and after trial was significantly different in three groups (P < 0.001); the greatest delta urine pH was in Bi group; hence, higher urine pH and higher urine delta pH after administration were seen in Bi group. Delta GFR pre and post hydration was statistically significant only in Bi group and showed greater increase in GFR after hydration with this solution (P = 0.043).

The results of regression analysis revealed use of RL versus Bi solution (OR: 5.9, confidence interval: 1.8–14.2, P = 0.004) as effective factor in prevention of CIN. At the end of this study, no patient needed renal replacement therapy and mortality was not seen in any group.

Discussion

CIN is one of the major complications of contrast injection, so its prevention methods have always been considered.[1516] Several studies have been conducted to prevent CIN.[17181920212223] Our study showed that the overall incidence of CIN was 10%, which was slightly higher than similar studies.[2] This could be due to the fact that we set the criteria of CIN definition as 0.3 or more increase in Cr level, whereas most studies have set these criteria as 0.5. The largest number of patients had CIN in RL group and the smallest one was in Bi group. Numerous studies have reported the effectiveness of Bi in prevention of CIN.[172425] The suggested mechanism is that alkalization of renal tubular fluid may decrease the damages caused by free radicals.[1726] Other researches have stated that administration of NLS may increase the risk of AKI incidence by mechanism of inducing acidosis and have suggested RL as an alternative in prevention of AKI.[2728] Yunos et al. evaluated the effect of 0.9% saline and balanced crystalloid in 1644 patients in intensive care unit (ICU) and concluded that 0.9% saline causes more AKI compared with crystalloid solutions.[27] There was a very high proportion of CIN with RL. We know that one of the major theories in pathogenesis of contrast nephropathy is medullary hypoxia; otherwise lactate is involved in maintaining the acid–base equilibrium under aerobic conditions. Studies by Hussmann et al. and Khajavi et al. clearly demonstrate that RL is toxic in severe hemorrhagic shock.[2930] In an analogy performed by Chowdhury et al. between 0.9% saline and RL on 12 patients, 0.9% saline caused 13% reduction in renal and cortical blood flow compared with other solutions.[31] In contrast, some studies report that 0.9% saline although causing acidosis in patients does not have significant adverse effects.[32] The other two studies compared the effects of different solutions used after kidney transplantation and showed that NLS although causing hyperchloremic acidosis is no different from RL in outcome.[3334]

In SPLIT randomized clinical trial in 2278 ICU-admitted patients, buffered crystalloid compared with NLS did not reduce the risk of AKI.[35]

Waters et al. in a study performed on patients undergoing aortic reconstructive therapy showed that excessive administration of NLS versus RL, despite saline-induced hyperchloremic acidosis, made no difference in complications, ICU stay, and hospital stay.[36] Since few studies have focused on the effects of RL in prevention of CIN, we decided to evaluate the effect of this solution on prevention of CIN, which showed inferiority for RL in prevention of CIN. On the other hand, NLS showed more desirable results in prevention of CIN compared with RL; however, they were not statistically significant. We have reported previously that both Bi and AZ reduce the risk of CIN-related AKI;[17] also in this study we showed that Bi has more protective effect rather than NSL or RL on prevention of CIN. No significant difference in prevention of CIN was seen in comparison of RL versus NLS and NLS versus Bi. It seems likely that hydration with Bi is more effective in prevention of CIN and our study could not prove superiority of RL over the two other solutions.

In previous trials, urine alkalization has been proved to prevent CIN.[1726] Our study confirmed it as well. All the solutions in this study significantly increased urine pH. Therefore, considering the fact that delta urine pH in Bi group was remarkably greater than the others, the incidence of CIN was the least in this group. It is likely that more alkalinized urine comes with less incidence of CIN.

Hypokalemia is one of the complications of Bi's injection.[15] This phenomenon took place in our Bi group but it was not clinically significant. In this setting, RL brings about less hypokalemia rather than other solutions.

Conclusion

Bicarbonate solution is more effective than RL in prevention of CIN in patients undergoing coronary angiography. There was no significant difference between Bi group and NLS group.