Renal injury in TDT patients is being observed more frequently with prolonged patient survival, and has recently become the object of extensive research.(3–5) Since chronic anemia, iron overload and ICT coexist in virtually all patients, it is hard to isolate and study a specific contributing factor. Our study aimed at examining the tubular and glomerular functions of TDT patients treated with two different ICT; we then compared our results to those of the same patients examined 10 years earlier.(6) We evaluated 36 TDT patients; all of them had normal Cr, serum electrolytes and eGFR.
Tubular Injury
Hypercalciuria was found in 28% of the patients, increased Fe K in 33%, high UAE in 64%, and high levels of TmP/GFR in 25%. Those results are in accordance with previous reports(5,7). No significant difference was found between the two ICT groups in those parameters.
Hypercalciuria is a major concern in TDT patients because of its association with osteoporosis(39) and renal stones(40,41). In our study, hypercalciuria was significantly negatively correlated with mean serum ferritin for the prior 10 years, and with the 10-year amount of transfused iron (Figure 6). Those correlations were stronger in the group of patients treated with DFO+/-DFP. While the frequency of hypercalciuria in our patients was similar to that in Quinn et al.,(42) in his study, unlike ours, it was associated with a higher intensity of transfusion. This discrepancy might be explained by the inclusion of non-transfusion-dependent patients in Quinn et al.'s study, and the use of different ICT. Wong et al(35) reported hypercalciuria in 90% of TDT patients, correlated to DFX dose. Hypercalciuria in our cohort was much less common, but the negative correlation with ferritin level might have been a consequence of the ICT treatment or over chelation. The percentage of patients with hypercalciuria increased significantly in the present study, especially for patients in the DFX group. We did not know if this increment is related to the change to DFX as the principal ICT used by the patients in our study or a tubular damage over the years.
A positive correlation was found between urinary Ca/Cr ratio and Hb levels, and it was stronger in the group of patients treated with DFO+/-DFP. Further studies are needed to clearly define the factors associated with hypercalciuria and groups at risk, who might require a stricter bone density scan program.
Abnormally high uNAG levels, reflecting tubular endothelial injury, were significantly more prevalent and higher in the DFX group compared to the DFO+/-DFP group. Our results are in accordance with prior reports of DFX associated renal tubular dysfunction.(6,29,32,33,43) Recently Annayev and Karakas (44) reported a higher prevalence of tubular injury, expressed in higher 2-microglobulin levels, in patients treated with a high dose of DFX. A negative correlation (r = -0.35, p = 0.03) was found between uNAG levels and the mean serum ferritin for the prior 10 years. This might also be attributed to over chelation, direct tubular iron toxicity, direct toxic effect of DFX or the cumulative dose of DFX, similar to the trend described by Wong et al.(35) in DFX dose related hypercalciuria.
In patients treated with DFO+/-DFP, a strong positive correlation was found between uNAG levels and the 10-year amount of transfused iron. Such a correlation was not found in patients treated with DFX. Iron renalא toxicity has been well described in TDT patients, and Koliakos et al.(23) reported a similar correlation between ferritin levels and uNAG in DFO-treated patients. Michelakakis et al.(25) also reported similar findings, and found that the effect can be reversible in patients well-chelated with DFO.
Heart and liver T2* MRI studies are becoming the standard of care in evaluating tissue iron load in TDT patients, almost completely replacing liver biopsy.(45) In our study, uNAG levels were negatively associated with heart iron content as evaluated by heart T2* MRI; such a correlation was not found with liver iron content. Several previous studies have reported a lack of correlation between heart and liver iron contents.(46,47) DFX and DFP's efficiency at reducing heart iron content has been demonstrated.(31,46) In light of this, our results can be explained by chelation toxicity, with patients who are better chelated as determined by low heart iron content having higher ICT-related tubulopathies. This observation correlates well with our results of a negative correlation between uNAG and ferritin levels.
As already noted, the complex interplay between the factors contributing to renal tubular injury is difficult to untangle, but there appears to be two different trends in the different ICT groups. In the DFO+/-DFP groups, tubular injury seems to be related to iron toxicity or related to the DFX treatment. The use of DFX over years can be the dominant factor in kidney injury. Long term continuous renal function follow up in patients treated continuously with the same chelator can resolve this question. Since usually patients received different chelators during their lives, this comparison seems difficult to perform. In addition, correlations differed in terms of uNAG levels and hypercalciuria, possibly due to the different injury localization inflicted on the renal tubule, and inconsistency in the degree of injury as reflected by iron overload markers, ferritin levels, and heart MRI; one should therefore make conclusions with caution.
Twenty nine of the patients in the current study were evaluated in our previous study.(6) Serum Cr significantly increased from the previous study in patients treated with DFX but not in patients treated with DFO+/-DFP. The eGFR was significantly lower than in the previous study in patients treated with DFX, but not in patients treated with DFO+/-DFP. These findings are worrisome, especially in children who will likely undergo ICT for life. Although all ICTs have been reported to cause renal injury, DFX has been shown to specifically affect eGFR and increase serum Cr.(29,48,49)
A ten year follow-up study of renal function in TDT patients treated exclusively with DFO(16) reported a mild decrease in eGFR, with most patients remaining within the normal range, similar to our cohort. Lai et al.17 reported a significant decline in eGFR in patients with an existing tubular injury; in our cohort, the decrease in eGFR was significantly more common in the DFX group, (p = 0.0093), in agreement with previous studies. However, these finding should be viewed in light of the well-reported hyperfiltration in TDT patients(42) and its effect on the interpretation of eGFR, precluding any definite conclusions; more sensitive tools for estimating GFR are warranted. In summary, tubular injury is common in TDT patients, with higher uNAG levels in DFX-treated vs. DFO+/-DFP-treated patients. Chronic use of DFX was also associated with a non-significant decrease in eGFR (P = 0.06). Larger studies are needed to understand the complex mechanism of glomerular and tubular injury in thalassemia, but awareness and close monitoring of underlying renal dysfunction is warranted, especially in patients treated with DFX.