The DNA fragmentation of MDCK cells induced by diatrizoate is related to its hypertonicity in this in vitro model of radiocontrast cytotoxicity. Nuclear disintegration with subsequent cell death may contribute to the pathophysiology of radiocontrast-induced nephropathy, particularly in the hypertonic/hypoxic environment of the renal medulla. The present results underscore the importance of avoiding hyperosmolal urine states in patients at high risk of radiocontrast-induced nephropathy.

Physicochemical factors contribute to the cytotoxicity of radiocontrast agents in vitro. The redistribution of tight-junction-associated membrane proteins by the ionic radiocontrast agents corresponds with the loss of the barrier function of the epithelial cell monolayer, which is a major pathophysiological mechanism in acute renal failure. The radiocontrast agents with reduced osmolality are less cytotoxic than diatrizoate, independent of their ionicity. Hyperosmolality appears to be a more important determinant of the cytotoxicity of diatrizoate than ionic strength.

Diatrizoate is more toxic than iopamidol, which is partly related to its higher osmolality. The cytotoxicity of radiocontrast agents induces a redistribution of polarized membrane proteins which could contribute to the pathophysiology of radiocontrast-induced nephropathy.

Direct cytotoxic effects of radiocontrast (RC) agents have been implicated in radiocontrast nephropathy (RCIN). The interaction between extracellular calcium, which plays a central role in intercellular contacts, and the in vitro toxicity of RC was tested in Madin-Darby canine kidney (MDCK) cell monolayers grown on permeable supports. Cell viability was determined by trypan blue exclusion. The function of intercellular junctions was assessed by measuring the electrical transmonolayer resistance (TMR). The cell contacts were examined with indirect immunofluorescence microscopy using antibodies against the junctional proteins E-cadherin, ZO-1 and occludin. The ionic RC agents diatrizoate and ioxaglate (74 mg iodine/ml), but not the nonionic compounds iohexol or iodixanol, decreased ionized calcium (Ca²⁺) in the incubation media from 1.48 ± 0.04 mM (control) to 0.89 ± 0.06 mM (diatrizoate), respectively to 1.05 ± 0.08 mM (ioxaglate). Diatrizoate, and to a lesser extent ioxaglate, reduced the number of viable MDCK cells and showed a redistribution of the E-cadherin, ZO-1 and occludin immunofluorescence signal with a parallel decrease of the TMR indicating an impaired monolayer integrity. A similar reduction of extracellular Ca²⁺ through EGTA failed to reproduce these effects. Conversely, raising Ca²⁺ in diatrizoate-containing media to control levels did not abrogate its toxicity. In conclusion, the ionic RC agents diatrizoate and ioxaglate, but not the nonionic compounds iohexol or iodixanol, reduce extracellular Ca²⁺ in vitro. However, this reduction of Ca²⁺ does not explain their cytotoxic effects which could contribute to the pathogenesis of RCIN in vivo by opening intercellular junctions.