REF-1547 · Arcoverde-Mello et al., 2025 · Brazilian Journal of Medical and Biological Research · Paper · PubMed · Free full text

Does a 2-h gap beat 15 h for the lactate curve?

In 14 active men over 3 weeks the curve shifted right in both arms, no clear difference. Very likely Ghiarone 2019's men, not a new trial. Not doubles against singles.

Main caveat: Not an independent trial. It very likely shares its men with (our inference from the same project, protocol and count of 15).
limited, mixed evidenceMeasured on recreational people

This analysis comes from the project that produced Ghiarone 2019 (REF-1535) and Andrade-Souza 2020 (REF-1536). In Study 1, five randomly chosen participants did three trials in randomised order.

A 100-min ride plus six 1-min sprints came 15 h before an interval session (once-daily), about 2 h before it (twice-a-day), or not at all (control); the interval session was the same in all, ten 2-min bouts with a low-carbohydrate breakfast beforehand. MCT1 mRNA at 3 h was about 80% higher after twice-a-day than once-daily (P = .038).

It also rose in the control (P = .046), and twice-a-day was not significantly above the control. MCT4 did not change. In Study 2, 14 men trained for 3 weeks, the same two sessions as a pair, 2 h apart on three days a week or 15 h apart across six days a week.

A graded test before and after showed the lactate curve further right in both groups: slope index 0.60 ± 0.59 to 0.11 ± 0.07 and 0.56 ± 0.75 to 0.10 ± 0.07 mmol/L/min, time effect P = .007, group × time P = .609. Our reading: the gene signal differed acutely and the lactate curve did not clearly differ after training.

The title's "does not change the lactate curve" does not match the results: the curve changed in both arms; what did not clearly differ was the change between them.

Protocol

Who: healthy physically active men recruited March 2015 to October 2016. Study 1: 33.2 ± 1.9 y, 78.6 ± 9.9 kg. Study 2: 29.1 ± 5.1 and 24.4 ± 9.3 y, 72.5 ± 7.0 and 81.1 ± 12.8 kg.
First session: 100 min of cycling at the midpoint between the first and second lactate thresholds, 8 min of rest, then six 1-min bouts at 125% of peak power with 1 min of rest. The authors report it lowers muscle glycogen by about 45%, from earlier work.
Second session: 5-min warm-up, then ten 2-min bouts at a work rate 20% of the way from the second lactate threshold to peak power, with 1 min of passive rest.
Study 2 schedule: twice-a-day: 7:00-9:00 and 11:00 on three days a week. Once-daily: 20:00 and 11:00 the next day, six days a week. That makes three pairs a week in both (our arithmetic), so the twice-a-day arm also has more days without training.
Diet: a strict plan, detailed in Ghiarone 2019; figure captions give a low-carbohydrate diet of about 7% carbohydrate and a high-carbohydrate diet of 75%. Neither arm is a normal-glycogen comparison at the second session, so fuel was meant to be held, not tested. Glycogen is not reported in this paper.
Tests: graded test from 50 W, +25 W every 4 min with 1-min breaks, capillary lactate; an exponential fit gives the slope index. Biopsies in Study 1 before, right after and 3 h after intervals.
Analysis: two-way ANOVA with Bonferroni; one participant excluded by Grubbs' test on lactate parameters.
Funding and interests: CAPES (Brazil), the Brazilian National Council for Scientific and Technological Development and an Australian Research Council grant to D.J. Bishop. The open full text carries no conflict statement.

Key Results

Study 2 measureOnce-daily (n = 7): pre → postTwice-a-day (n = 7): pre → postGroup × time P (ηp2)Time P
Lactate slope index (mmol/L/min)0.60 ± 0.59 → 0.11 ± 0.070.56 ± 0.75 → 0.10 ± 0.07.609 (0.02).007
Exercise duration in the graded test (min)41.6 ± 4.4 → 47.5 ± 4.942.4 ± 7.6 → 49.5 ± 8.7.189 (0.14)< .01
Peak plasma lactate (mmol/L)10.4 ± 1.3 → 11.0 ± 1.59.2 ± 2.7 → 10.6 ± 2.6.276 (0.10)< .01
Study 1 measure (n = 5)ResultP
MCT1 mRNA, condition × moment.046 (ηp2 0.44)
MCT1, 3 h, twice-a-day vs once-dailyabout 80% higher.038
MCT1, 3 h vs pre, controlhigher (about 106% in the discussion).046
MCT1, 3 h, twice-a-day vs controlabout 50% higher, not significant
MCT4 mRNA, condition × momentno change.252 (ηp2 0.27)
No clear difference, not "similar." There is no equivalence test and only 7 men an arm. Post-training means and SDs are close, but the pre-training SDs are as large as the means. The authors report considerable individual variation and do not show it.
Abstract against results. The abstract says MCT1 rose only in the twice-a-day condition; the results also show a rise in the control at 3 h.
Exercise duration. An interaction of ηp2 = 0.14 at P = .189 cannot rule out a difference in this sample, as the authors acknowledge.
Our arithmetic: the slope falls by the printed −0.49 and −0.46 from the means above.

Mechanisms

Measured: acute MCT1 mRNA, which does not mean protein; a whole-body lactate curve, which does not mean lactate transport.
Authors' interpretation: close sessions may prolong exposure to raised plasma lactate, and lactate may signal MCT1 transcription. Their own text calls part of this speculative, and plasma lactate during these sessions is expected from earlier work rather than measured here.
Authors' conclusion: MCT1 mRNA alone did not map to changes in plasma lactate. Three weeks may be too short for protein to differ.
Not measured: MCT1 or MCT4 protein before and after training, muscle lactate, lactate flux by tracer.

Practical Application

Does a 2-h gap beat 15 h for the lactate curve? In 14 active men over 3 weeks the curve shifted right in both arms, no clear difference. Very likely Ghiarone 2019's men, not a new trial. Not doubles against singles.
Good for: one more reason not to read a bigger gene signal after close sessions as a bigger training gain. Here the mRNA signal differed acutely in 5 men and the curve did not clearly differ after training in 14; no mRNA was followed through training.
Good for: saying that the lactate curve moved in both arms over 3 weeks in active non-athletes. With no non-training arm this is a before-and-after change, not a null on the training itself, and among the curve's fitted parameters only the slope index moved.
Not good for: threshold advice to a trained athlete. No threshold session was done, no threshold is reported, and these men had a peak oxygen uptake near 37 mL/kg/min.
Not good for: two sessions against one. Both arms did the same two sessions; only the gap and the rest days changed.
Not good for: counting as extra evidence beside Ghiarone 2019 or Andrade-Souza 2020. It is the same project; the 3-week men are very likely Ghiarone's, and whether the 5 in Study 1 were also 3-week men is not stated.
Not good for: fuel. Both arms were meant to start the second session low; no well-fed arm trained.

Limitations

Not an independent trial. It very likely shares its men with REF-1535 (our inference from the same project, protocol and count of 15); a 15th participant is excluded here as an outlier, leaving 7 and 7.
A posteriori. Sample size was not set for these outcomes, and 3 weeks of data serve a question the study was not designed around.
Marker, not performance. A slope index from a curve fit, a graded test duration, no time trial.
Gap, rest days and time of day move together: evening plus morning across six days, or two mornings on three days. The authors call it a gap comparison; a day-off difference comes with it.
The acute n is 5, nine biopsies per person, with wide variability and a control arm that also rose.
Reporting. Peak lactate pre-training in the twice-a-day group is 9.3 ± 2.6 in the table and 9.2 ± 2.7 in the text; the gap is 15 h here and 14 h in the Ghiarone abstract, and the printed Study 2 clock times (ride from 20:00, about 2 h long, intervals at 11:00) give about 13 h end to start, the way the 2-h gap is measured (our arithmetic).
Conflict statement absent from the open text.
How this entry was checked and changed
Why this level

a parallel RCT under 20 earns B (rubric), lowered one step: the data are a posteriori with sample size set for other outcomes, the outcome is a lactate curve parameter rather than performance, and the acute part has 5 men. Read from the full text (CC BY). Not independent of REF-1535 and REF-1536: the paper cites both as reports of its umbrella project. It also recruited 15 men on the protocol of REF-1535, whose 8 and 7 fit 7 and 7 after one exclusion, so they are very likely the same men (our inference; no sentence says so). The three count as one project, not three independent tests