We read with great interest the “Mexican consensus on the approach to and treatment of acute variceal bleeding”, by Higuera de la Tijera et al.,1 whom we congratulate for their valuable work. We would like to make a comment on statement 3, related to fluid management.

The authors recommend replacing intravascular volume with balanced crystalloid solutions, arguing that there is a lower association with adverse events and mortality, based on a single randomized clinical trial (RCT) by Semler et al. in 2018 (n = 7,942). However, better evidence is currently available.

The largest RCT at present was recently published that compared the use of lactated Ringer’s solution versus saline solution 0.9% in intravenous fluid administration (n = 43,626).2 The primary outcome was a composite of death or hospital readmission within 90 days of the index admission. Incidence of the composite was 20.3% in the lactated Ringer’s solution group and 21.4% in the saline solution group, with no statistically significant difference (adjusted difference: –0.53 percentage points; 95% CI: –1.85−0.79; p = 0.35). No serious adverse events were reported. The conclusion was that a hospital-wide policy of preferential use of lactated Ringer’s solution, compared with saline solution, did not significantly reduce mortality or hospital readmission.

In addition, a systematic review and meta-analysis of individual patient data3 that included 6 RCTs (n = 34,685) evaluated whether the use of balanced crystalloid solutions, compared with saline solution 0.9%, reduced in-hospital mortality in critically ill adults admitted to the intensive care unit. Mortality was 16.8% in the balanced crystalloid group and 17.3% in the saline group (OR: 0.962; 95% CI: 0.909–1.019; absolute difference: –0.4 percentage points). The posterior probability that balanced crystalloids would reduce mortality was 89.5%. Nevertheless, in patients with traumatic brain injury, mortality was higher in the balanced crystalloid group (19.1% vs. 14.7%; OR: 1.424; 95% CI: 1.100–1.818; absolute difference: 3.2 percentage points), with a 97.5% probability that balanced crystalloids would increase mortality in that subgroup.

That finding could be explained by the higher risk of cerebral edema when using intravenous solutions with a lower sodium concentration (130 mEq/L in Hartmann’s solution and 140 mEq/L in Plasma-Lyte), compared with normal saline solution (154 mEq/L). Considering that patients with acute variceal bleeding are at high risk for developing hepatic encephalopathy,4 we believe that the routine use of balanced crystalloid solutions could result in more risks than benefits in that context, but this must be confirmed through RCTs, given that none of the previous studies are specific for acute variceal bleeding.

Another point to be taken into account is that the acquisition and administration of large quantities of different types of solutions has an ecologic and economic impact. The approximate cost of 100 mL of sodium chloride solution 0.9% is $0.60 USD, whereas the cost of balanced crystalloid solutions is higher, with Plasma-Lyte being the most expensive, at between $3.00 and $3.80 USD.

In conclusion, we consider that, at present, the available evidence does not justify recommending one intravenous solution over another, in a generalized manner, for patients with variceal bleeding.