Adverse effects of hyperchloraemic solutions
I read with interest the recently published article by Serrano and colleagues.1 The authors interpret their findings as showing that preoperative intravenous hydration with 0.9% normal saline before major abdominal surgery did not prevent postoperative acute kidney injury. However, numerous factors may have modified or even obscured any possible treatment effect of preoperative intravenous hydration per se.
One of my concerns relates to the use of 0.9% saline as the study fluid. The authors considerably downplay the potential impact of the use of 0.9% saline on the findings. I disagree with their general statement that clinical evidence regarding the potential adverse effects of chloride-rich solutions is scarce or lacking. Surprisingly, the authors cite two publications in support of their statement when, in fact, those publications support the opposite view.2,3 Shaw et al.2 concluded ‘The use of a calcium-free balanced crystalloid for replacement of fluid losses on the day of major surgery was associated with less postoperative morbidity than 0.9% saline’. Lobo and Awad3 concluded ‘Nevertheless, there is evidence to suggest that balanced crystalloids cause less detriment to renal function than 0.9% saline, with perhaps better clinical outcome. Hence, we argue that chloride-rich crystalloids such as 0.9% saline should be replaced with balanced crystalloids as the mainstay of fluid resuscitation to prevent ’prerenal’ acute kidney injury’.
I similarly disagree with the authors’ statement that hyperchloraemia and acidosis associated with 0.9% saline resuscitation do not usually lead to clinical consequences. The association between the administration of chloride-rich fluids and adverse outcomes seems robust.4,5 The adverse effects of isotonic saline are probably related to the accompanying acute increase in serum chloride concentration. Unfortunately, neither absolute baseline and postoperative serum chloride concentrations nor the differences between preoperative and postoperative serum chloride concentrations are reported by Serrano and colleagues.1 No definition of hyperchloraemia is provided. Without knowing the preoperative incidences of hyperchloraemia, the finding of no statistically significant difference in the postoperative incidence of hyperchloraemia between groups is meaningless. Interpretation of the data is further made difficult because the timing of the postoperative measurements of serum chloride concentrations and the duration of postoperative hyperchloraemia are not clearly spelled out.
The incidences of postoperative hyperchloraemia of 23 and 19% in the 0.9% saline and control groups, respectively, are rather high and worrisome. Possibly, the preoperative hydration with isotonic saline contributed to this adverse outcome. However, it is more likely that intra-operative intravenous administration of unbalanced, hyperchloraemic solutions had been responsible. Unfortunately, the type of administered fluid was neither specified by protocol nor listed in the results. This presents another serious limitation of the study. Intraoperatively, patients received mean amounts of intravenous fluids of approximately 3.5 l within a mean time of approximately 3.5 h. The large standard deviations indicate that individual patients may have received up to 5 l. If these fluids had been mostly hyperchloraemic solutions, their adverse effect on renal function could have been considerable. Infusion of 2 l of 0.9% NaCl in healthy awake individuals caused an increase in serum chloride concentration from 103 to 108.5 mmol l−1 within 60 min, which was associated with a 40% decrease of renal blood flow velocity and renal cortical tissue perfusion.6
As the adverse effects of hyperchloraemic crystalloid and colloid solutions might have obscured any potential treatment effects, it is essential to know the perioperative serum chloride concentrations and the properties of the intraoperatively administered fluids. In general, the overall evidence clearly provides a harm signal associated with hyperchloraemic crystalloid solutions without a comparable benefit signal. There is thus no physiological rationale for continued use of ‘normal’, but nonphysiologic saline for perioperative volume therapy when better, safer and more physiological solutions are readily available.