To examine possible alterations in acid-base parameters in patients switching from lanthanum carbonate (LanC) to sucroferric oxyhydroxide (SFOH).

Fifteen stable hemodialysis patients were switched from LanC to SFOH. Only nine continued on SFOH, three returned to LanC and the other three switched to sevelamer carbonate. The later six patients served as a control group to the SFOH group of nine patients. Blood was sampled on the 3-d and the last 2-d interval of the week prior to switching and six weeks after. Bicarbonate levels (HCO₃^-), pH, pO₂, pCO₂ were measured, and the mean of the two measurements (3-d and 2-d interval) was calculated.

Comparing pre-switching to post-switching measurements in the SFOH group, no statistically significant differences were found in any of the parameters studied. The mean pre-switching HCO₃^- was 22.41 ± 1.66 mmol/L and the mean post-switching was 22.62 ± 2.25 mmol/L (P = 0.889). Respectively, the mean pH= 7.38 ± 0.03 vs 7.39 ± 0.03 (P = 0.635), mean pCO₂= 38.41 ± 3.29 vs 38.37 ± 3.62 mmHg (P = 0.767), and Phosphate = 1.57 ± 0.27 vs 1.36 ± 0.38mmol/L (P = 0.214). There were not any significant differences when we performed the same analyses in the control group or between the SFOH group and control group. No correlations were found, either between pre-switching LanC daily dose or between post-switching daily dose of the new binder and the measured parameters.

In our small study, switching from LanC to SFOH did not have any significant effect on blood bicarbonate levels and gas analysis, indicating that there is no need to change hemodialysis prescription regarding these parameters.

Metabolic acidosis is a characteristic complication of end stage kidney disease (ESKD) and is associated with major systemic effects and increased mortality[1,2]. Hyperphosphatemia, another common complication of ESKD, partially contributes to metabolic acidosis and affects the acid-base status of the patients[1]. Phosphate binders used for control of hyperphoshatemia contribute to acid-base balance through their effect on serum phosphate and through the effect of the binders’ constituents that have alkaline or acidic properties[3]. The most characteristic is the case of sevelamer hydrochloride (SevH), which is associated with dose-dependent aggravation of metabolic acidosis in ESKD patients due to its hydrochloride content[4,5]. This acidosis is ameliorated after switching to sevelamer carbonate (SevC) or other phosphate binders with alkaline content, such as calcium carbonate (CalC) or lanthanum carbonate (LanC)[6-9]. However, despite the above mentioned studies, a recent meta-analysis suggests that the effect of phosphate binders in clinically important outcomes, such as metabolic acidosis, remains understudied[10].

Sucroferric oxyhydroxide (SFOH) is a polynuclear iron(III)-oxyhydroxide-based phosphate binder recently approved for the treatment of hyperphosphatemia in patients with ESKD. It is a safe and potent phosphate binder with increasing use among hemodialysis patients, both as an initial treatment choice or as an alternative when other binders fail[11,12]. However, little is known regarding the influence of this novel binder on the acid-base status of hemodialysis patients. Aim of our study was to examine possible alterations in acid-base parameters in hemodialysis patients switching from LanC to SFOH.

Patients who switched to SFOH (SFOH group) had an effective phosphate control (Table 2), without increasing the pill burden, [number of LanC pills: 3 (1-4) vs SFOH pills: 2 (1-6), P = 0.849]. Comparing the mean values of the measured acid-base balance parameters, pre- and post-switching, no difference was found in any of them (Table 2). No statistically significant differences were found even when we analyzed the data for the 3-d and the 2-d interdialytic intervals separately (Supplementary Tables 1 and 2), or when we performed the same analyses in the control group or between the two groups (data not shown).

No correlations were found, either between pre-switching daily LanC dose and the measured parameters, or between post-switching daily dose of the new binder and the measured parameters (data not shown).

The only significant differences that we found were between the 3-d and 2-d measurements. HCO₃^-, BE, and pH were significantly lower and K⁺ was higher at the 3-d vs 2-d interdialytic interval, as expected. For example, in the whole group of patients, baseline 3-d HCO₃^- were 21.2 ± 2.17 mmol/L, BE: -3.5 (-4.8 to-2.3) mmol/L, pH: 7.373 ± 0.031, and K⁺: 4.69 ± 0.46 mmol/L. Respective values for the baseline 2-d interval were: HCO₃^-: 22.53 ± 2.68 mmol/L (P = 0.005, comparing to 3-d), BE: -2.2 (-3.3 to -0.4) mmol/L (P = 0.001), pH 7.389 ± 0.029 (P = 0.003) and K⁺ 4.29 ± 0.42 mmol/L (P = 0.04). Accordingly, 3-d pCO₂ was lower (37.2 ± 3.34 mmHg) vs 2-d (38.19 ± 4.51 mmHg), however the difference did not reach statistical significance (P = 0.061). Phosphate levels were the same, 1.42 ± 0.4 mmol/L vs 1.46 ± 0.26 mmol/L (P = 0.801), indicating that phosphate binders are effective in maintaining stable phosphate levels all days. When we performed the same analyses in the post-switching measurements or in the different groups, the findings were similar (data not shown).

In our study, we found that when hemodialysis patients are switching from LanC to SFOH, despite the loss of carbonate provided by LanC, there is no significant consequence on acid-base balance. Several explanations may exist for this finding. First, the iron oxide hydroxide binds phosphate in the gastrointestinal tract through a direct ionic interaction between the negatively charged oxygen ions on the phosphate and the ferric ions in the ferric oxide, forming FePO₄[13]. During this interaction, hydroxyl groups are released[13], indicating that SFOH acts as a base. An additional explanation could be that the effect of the alkali contained in LanC is small and its loss is compensated by endogenous adaption from blood and bone buffering and residual renal function, when present. This may be particularly true when doses of LanC are not very high or when there is no severe acidosis, like in our patients. There was only one prospective study found where LanC was introduced in phosphate binder-naïve patients. There, serum bicarbonate levels remained stable after 9 mo of LanC in 28 patients with moderate CKD (21.9 ± 2.9 mmol/L at baseline vs 21.8 ± 2.4 mmol/L at 9 mo), indicating that the effect of the carbonate is small[14]. Furthermore, it is well known that switching to LanC can ameliorate the acidosis caused by SevH[7,8]. However, this may not be the result of the alkalinizing agent of LanC, but rather the consequence of the withdrawal of the hydrochloride of SevH. Treatment with SevH results in a significant increase in dietary acid load, and this is the main cause of the observed acidosis[4]. In favor of this is the observation that patients who switch from SevH to bixalomer, a non-hydrochloride and non-carbonate phosphate binder, improve their metabolic acidosis. This indicates that the withdrawal of chloride without the additional effect of carbonate is sufficient to ameliorate the acidosis[15]. Lastly, in our study, no correlation was found between LanC dose and acid-base parameters. The above support the hypothesis that the effect of the carbonate content of LanC on acid-base status is rather small. Moreover, concomitant use of CalC could compensate for the effects of loss of carbonate provided by the LanC on acid-base status. However, only two of our patients were on CalC, and the doses remained stable throughout the study. Finally, due to the small sample size of our study, a type II error cannot be excluded. Nevertheless, the harmful effect of SevH on acid-base balance was shown even in studies with 8 to 16 patients[7,15,16]. Thus, the effects of SFOH switching would probably be detected in our study if they were significant. In addition, finding a significant deference between the acid-base status of 2-d and 3-d interdialytic interval could be considered a supporting finding that despite the small sample size, differences were detected when they were significant.

In our study, SFOH was effective in controlling phosphate without increasing the pill burden, in accordance with previous studies[17,18]. However, some of our patients could not tolerate the new binder, as treatment with SFOH appears to be predominantly complicated by gastrointestinal disturbances[17,19].

Finally, we observed that patients were more acidotic in the long interdialytic interval. This has been already shown in previous studies[20], and this is why we decided to perform measurements both in the long and the short interdialytic interval in an effort to increase the possibility of detecting potential effects of SFOH switching on acid-base status.

The main limitation of the present study is the small sample size, as possible effects of switching on acid-base parameters may not be detected, even if present. Yet, the small number of patients allowed for stable conditions and better compliance, and this, together with the prospective nature, can be considered a strength of the study. Another limitation is that we did not measure the protein dietary intake that can affect acid-base balance. However, nutritional instructions remained the same during the study, and albumin and cholesterol levels did not change, supporting that there were no major changes in dietary habits. Finally, we cannot generalize our findings in patients with severe acidosis or in patients with ESKD on peritoneal dialysis.

To the best of our knowledge, this is the first study that evaluates acid-base status in patients receiving the new phosphate binder SFOH. Therefore, it can serve as a pilot for further studies with a larger number of patients, different binders and extended durations to better understand their effect in clinically important outcomes, such as metabolic acidosis.

In conclusion, switching from LanC to SFOH did not have any significant effect on blood bicarbonate levels and gas analysis, indicating that there is no need to change hemodialysis prescription regarding these parameters. However, when selecting a phosphate binder, potential consequences on acid-base balance should be considered, and monitoring of serum bicarbonate levels is part of good clinical practice.