Advertisement

Caffeine intake and its influences on heart rate variability recovery in healthy active adults after exercise: A systematic review and meta-analysis

Published:January 21, 2022DOI:https://doi.org/10.1016/j.numecd.2022.01.015

      Highlights

      • This is the first systematic review with meta-analysis that investigated CAF influences on HRV.
      • RMSSD and HF indexes not demonstrated slower recovery after exercise.
      • Caffeine did not affect HRV recovery following exercise.

      Abstract

      Aims

      Results regarding the effects of caffeine (CAF) intake on the autonomic control of heart rate recovery exercise remain inconclusive. Thus far, no study has used effect measures to pool the results of different experiments. We aim to assess the acute effect of CAF intake before exercise on the recovery of heart rate variability (HRV) after exercise through a systematic review and meta-analysis.

      Data synthesis

      Randomized controlled clinical trials were included; sample composed of physically active or trained adults; CAF should be offered/ingested before exercise, with dosage between 100 and 400 mg or between 2 and 6 mg/kg and administration/ingestion route analogous in the protocols; studies required to present results of HRV indices before and after exercise. Bias risk analysis and meta-analysis were performed. Twelve studies were included in the qualitative synthesis and five studies were encompassed in the quantitative synthesis (meta-analysis). For the Root-mean-square standard deviation (RMSSD) index we revealed p = 0.67, Total 95% confidence interval (95% CI) ranged from −0.45 to 0.29, and 66.7% for heterogeneity between groups were reported. Concerning the High Frequency (HF) index, we observed p = 0.22, Total 95% CI that ranged from −0.34 to 0.30, and 44% for heterogeneity between groups.

      Conclusions

      CAF intake did not affect heart rate variability recovery after exercise.

      Keywords

      1. Introduction

      Caffeine (CAF) is a methylxanthine alkaloid capable of causing obstruction of adenosine receptors (A1, A2A and A2B) and thus raise the activity of the central nervous system (Barcelos et al., 2020) [
      • Barcelos R.P.
      • Lima F.D.
      • Carvalho N.R.
      • Bresciani G.
      • Royes L.F.
      Caffeine effects on systemic metabolism, oxidative-inflammatory pathways, and exercise performance.
      ]. For these motives, CAF is a substance vastly consumed for different reasons globally, including improvement of sports performances, cognitive or thermogenic purposes [
      • Guest N.S.
      • VanDusseldorp T.A.
      • Nelson M.T.
      • et al.
      International society of sports nutrition position stand: caffeine and exercise performance.
      ,
      • Maughan R.J.
      • Burke L.M.
      • Dvorak J.
      • et al.
      IOC consensus statement: dietary supplements and the high-performance athlete.
      ,
      • Grgic J.
      • Trexler E.T.
      • Lazinica B.
      • Pedisic Z.
      Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis.
      ,
      • Shen J.G.
      • Brooks M.B.
      • Cincotta J.
      • Manjourides J.D.
      Establishing a relationship between the effect of caffeine and duration of endurance athletic time trial events: a systematic review and meta-analysis.
      ]. The U.S. Food and Drug Administration (FDA) and the European Food Safety Authority expect that the daily consumption of equal to 400 mg of CAF is not associated with adverse cardiovascular effects or bodily toxicity [
      • Turnbull D.
      • Rodricks J.V.
      • Mariano G.F.
      • Chowdhury F.
      Caffeine and cardiovascular health.
      ] (see Table 1, Table 2).
      Table 1Description of the characteristics of the study population of articles by author and year, sample, age (years), weight (kg), height (cm), BMI (kg/m2) (mean ± SD), exercise, average peak oxygen (ml/kg/min), exercise intensity.
      Author/yearsStudy DesignSampleAge (years)Weight (kg)Height (cm)BMI (kg/m2)ExerciseAverage Peak Oxygen (ml/kg/min)Exercise Intensity
      Benjamim et al. (2021)RCT30 healthy men23.33 ± 3.1571.14 ± 12.31163 ± 4.5123.00 ± 2.75Resistance (Leg press, squat, abductor and extensor chair)Not reported75% 1RM
      Blake et al. (2020)RCT12 resistance-trained men22.75 ± 4.5191.05 ± 17.77183.4 ± 7.3Not calculatedResistance (Elbow flexion and extension)Not reportedMaximal peak isometric force at 0° degrees and 60° for both peak elbow flexion and extension
      Bunsawat et al. (2015)RCT18 healthy individuals26 ± 1.0Not reportedNot reported23.9 ± 0,8TreadmillNot reportedMaximum oxygen consumption (VO2max), followed by 2 min active recovery at a speed of 3.5kmh and grade of 0%.
      Gonzaga et al. (2017)RCT32 healthy and physically active men23,59 ± 3,4578.87 ± 12.14179 ± 7.1424.40 ± 2.82Treadmill44.00 ± 12.2560% on VO2max during 30 min
      Gonzaga et al. (2019)RCT32 healthy physically active men23.69 ± 3.7578.38 ± 6.92180 ± 4.024.33 ± 2.04TreadmillGroup: High VO2: 53.32 ± 8.79

      Low VO2: 34.69 ± 6.92
      60% on VO2max during 30 min
      Karayigit et al. (2021)RCT17 resistance-trained female23.0 ± 2.064.0 ± 4.0168.0 ± 3.0Not calculatedResistance (Squat and Bench press)Not reportedSquat and bench press repetitions to failure at 40% 1RM
      Lopes-Silva et al. (2016)RCT10 men athletes (experienced in taekwondo)21 ± 4,071.0 ± 12.9180 ± 8,0Not calculatedFightNot reportedEach combat consisted of Three 2-min rounds with 1-min intervals
      Nishijima et al. (2002)RCT8 healthy men25.5 ± 0.868.3 ± 1.7172.2 ± 4.3Not calculatedCycle ergometerNot reportedBetween 40 and 50% ventilatory thresholds
      Rolim et al. (2018)RCT21 young physically active men22.3 ± 2.9Not reportedNot reported25.2 ± 2.7 mg/kg2TreadmillNot reported3 min of warm-up at 3 km/h and 2.5% of slope. In sequence, increments 1 km/h each minute until reach 85% of Maximal HR, test was interrupted.
      Sarshin et al. (2020)RCT20 recreationally active males24.0 ± 2.074.70 ± 7.07178.8 ± 4.6Not calculatedCycle ergometerNot reported4-min standardized warm-up against a fixed load of 1 kilopond and three separate 2s sprints were performed against a load of 0.075 kp·kg−1 of body mass were interspersed with 45-s of active recovery.
      Thomas et al. (2017)RCT21 (13 males) healthy subjects25.5 ± 3.5Not reportedNot reported26.2 ± 4.6Cycle ergometer32.2 ± 6.4Cycled workload at 75% VO2max during 15 min.
      Yeragani et al. (2005)RCT11 (6 males) heathy subjectsM: 29.0 ± 7.0

      F: 23.05 ± 7.0
      Not reportedNot reportedNot calculatedCycle ergometerNot reportedExercise began with pedaling intensity initiated at 25 W, and increased by 25 W increments at every 3 min stage.
      HR = heart rate; RCT = randomized controlled trial; RM = repetition maximum; W= watts; VO2 = oxygen volume consumed during exercise; VO2Max = maximum oxygen volume consumed during effort test.
      Table 2Description of the selected articles by author and year, time ingestion CAF, CAF dose, placebo, analysis time, HRV index and main conclusions.
      Author/yearsIngestion CAF (before exercise)CAF dosePlaceboAnalysis

      Time
      HRV IndexMain conclusions
      Benjamim et al. (2021)45 min(∼4.21 mg/kg)

      300 mg
      Capsule (starch)0-5, 5–10, 10–15, 15–20, 25-25 and 25–30 min post-exerciseSDNN, RMSSD, RRTri, TINN, SD1, SD2, LF and HF(ms2)CAF protocol delayed HRV recovery after strength exercise.
      Blake et al. (2020)30 min(∼3.84 mg/kg)

      350 mg
      Capsule (starch)Immediately (<1 min) and 60 min post-exerciseRMSSDCAF protocol did not affect HRV recovery after exercise.
      Bunsawat et al. (2015)45 min(∼5.06 mg/kg) 400 mgPills5, 15 and 30 min post-exerciseLF/HFCAF protocol delayed vagal reactivation after exercise.
      Gonzaga et al. (2017)45 min(∼3.80 mg/kg)

      300 mg
      Capsule (starch)0-5, 5–10, 10–15, 25–30, 35–40, 45–50 and 55–60 min post-exerciseRMSSD, SD1, SDNN, HF, LF and LF/HFIntake CAF before exercise delayed the parasympathetic recovery following moderate aerobic exercise.
      Gonzaga et al. (2019)45 min(∼3.80 mg/kg)

      300 mg
      Capsule (starch)0-5, 5–10, 10–15, 25–30, 35–40, 45–50, 55–60 min post-exerciseSDNN, RMSSD, SD1, SD2, LF(n.u.), HF(n.u.) and LF/HFCAF slows parasympathetic recovery from exercise in individuals with lower cardiorespiratory capacity (VO2peak<42.46 mL/kg/min).
      Karayigit et al. (2021)60 min(3 mg/kg)

      192 ± 12 mg and (6 mg/kg)

      384 ± 24 mg
      Decaffeinated coffeeImmediately post-latest exercise (<1 min)SDNN, RMSSD, TP, HF (ms2) LF (ms2), LF/HF, HF(n.u.) and LF(n.u.)CAF protocols did not demonstrate changes in HRV recovery between interventions.
      Lopes-Silva et al. (2016)60 min(5 mg/kg)

      355 ± 64,5 mg
      Capsule (Cellulose)180, 360 s post-exerciseRMSSDCAF did not delay vagal reactivation between 180 and 360 s following exercise.
      Nishijima et al. (2002)120 min(∼4.39 mg/kg)

      300 mg
      Capsule5-10, 15–20 and 25–30 min post-exerciseTP, HF and LFCAF protocol demonstrated significant changes in HRV indexes (LF and power total) at 30 min after exercise when compared to placebo.
      Rolim et al. (2018)60 min3 mg/kgCapsule60–300 s post-exercise, with the treadmill velocity reduced to 2.4 km/h.SD1 and SD2CAF accelerated parasympathetic reactivation in the submaximal exercise test in the first 3 min.
      Sarshin et al. (2020)45 min(3 mg/kg)

      224.1 ± 21.2 and (6 mg/kg)

      448.2 ± 42.4
      Capsule (dextrose)0-5 and 30–35 min post-exerciseSDNN and RMSSDCAF improved HRV recovery post- exercise.
      Thomas et al. (2017)10 min300 mgGum0–5 min and 5–10 min after exerciseSDNN, RMSSD, LF(nu), HF (nu), LF/HF, SD2 (ms), ApEnCAF did not impaired HRV indexes recovery after exercise.
      Yeragani et al. (2005)60 min5 mg/kgCola drink0–5 min after exerciseLF (ms2), HF (ms2), LF/HF, ApEnCAF delayed HRV recovery after exercise.
      The increased release of catecholamines in the bloodstream promoted by CAF intake results in adjustments to the cardiac autonomic modulation that induce consequent tachycardias [
      • Banerjee P.
      • Ali Z.
      • Levine B.
      • Fowler D.R.
      Fatal caffeine intoxication: a series of eight cases from 1999 to 2009.
      ]; which directed quite a few studies to explore the safety of CAF use before the practice of physical exertions. Koenig et al. [
      • Koenig J.
      • Jarczok M.N.
      • Kuhn W.
      • et al.
      Impact of caffeine on heart rate variability: a systematic review.
      ] in their systematic review about caffeine and heart rate variability (HRV) revealed an increase in vagal flow via frequency domain indexes, but that result does not allow for exercise interventions.
      The recovery of HRV after exercise is considered one of the most reliable methods to measure and predict adverse cardiovascular events and subsequent mortality [
      • Cole C.R.
      • Blackstone E.H.
      • Pashkow F.J.
      • Snader C.E.
      • Lauer M.S.
      Heart-rate recovery immediately after exercise as a predictor of mortality.
      ], and CAF may disturb this response [
      • Turnbull D.
      • Rodricks J.V.
      • Mariano G.F.
      • Chowdhury F.
      Caffeine and cardiovascular health.
      ]. Through the consecutive heartbeat intervals (RR-intervals) it is likely to indirectly investigate the behavior of the autonomic nervous system (parasympathetic and sympathetic) on the heart [
      • Vanderlei L.C.M.
      • Pastre C.M.
      • Hoshi R.A.
      • Carvalho TD de
      • Godoy MF de
      Noções básicas de variabilidade da frequência cardíaca e sua aplicabilidade clínica.
      ,
      • Shaffer F.
      • Ginsberg J.P.
      An overview of heart rate variability metrics and norms.
      ]. Accordingly, delayed vagal (parasympathetic) reactivation after exercise might be considered an undesirable cardiac outcome [
      • Cole C.R.
      • Blackstone E.H.
      • Pashkow F.J.
      • Snader C.E.
      • Lauer M.S.
      Heart-rate recovery immediately after exercise as a predictor of mortality.
      ,
      • Peçanha T.
      • Bartels R.
      • Brito L.C.
      • Paula-Ribeiro M.
      • Oliveira R.S.
      • Goldberger J.J.
      Methods of assessment of the post-exercise cardiac autonomic recovery: a methodological review.
      ].
      Others studies have assessed the influence of CAF on HRV recovery after exercise and its effects remain inconclusive. While CAF has been recognized in a similar population (e.g., physically active and/or trained athletes) the significances of these studies remain contradictory. For this reason, a recent systematic review was unable to refute or support the safety of CAF evaluated through HRV recovery [
      • Benjamim C.J.R.
      • Kliszczewicz B.
      • Garner D.M.
      • et al.
      Is caffeine recommended before exercise? A systematic review to investigate its impact on cardiac autonomic control via heart rate and its variability.
      ]. Furthermore, this study had methodological limitations (e.g., years of included studies, absence of quantitative analysis and inclusion of studies with the use of substances mixed with CAF) that may have led to erroneous results [
      • Benjamim C.J.R.
      • Kliszczewicz B.
      • Garner D.M.
      • et al.
      Is caffeine recommended before exercise? A systematic review to investigate its impact on cardiac autonomic control via heart rate and its variability.
      ]. Up till now, no study has enforced effect measures to pool the outcomes of different studies and provide a global estimate of the influences of CAF on HRV recovery.
      With the objective of clarifying the previously mentioned queries, our study intended to execute a systematic review and meta-analysis of randomized and controlled clinical trials to evaluate the effects of CAF intake on HRV recovery after exercise.

      2. Methods

      2.1 Registration

      The review was pronounced according to the recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes [
      • Page M.J.
      • McKenzie J.E.
      • Bossuyt P.M.
      • et al.
      The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.
      ] and is registered in the PROSPERO database (CRD42021273706).

      2.2 Search strategy and study selection

      The searches were accomplished using five databases (EMBASE, Cochrane Library, MEDLINE (via PubMed), Web of Science, and SCOPUS) and the application of the keywords “Autonomic Nervous System” OR “Heart Rate Variability” AND “Caffeine” AND “Exercise”.
      After distinguishing the appropriate studies, the articles were exported to the Rayyan QCRI program (Qatar Computing Research Institute, Qatar) to reject the duplicates revealed in the search. Then, the studies were screened in the Rayyan program by adding the filter: “study in humans” and, then, the studies were later screened by reading the title and abstract. The eligibility stage was the final selection phase for inclusion of articles, and was performed by two independent reviewers (AAP and CJRB) fully reading the articles. If at this stage there was a dispute about a study, an additional reviewer was invited to give a verdict (VEV).
      The studies needed to be sourced from peer-reviewed journals, published from the beginning of the database until March 2021. For inclusion, the articles needed to attain all the criteria described below: randomized clinical trial (RCT) design; participants greater than 18 years old, healthy, physically active, and trained. CAF needed to be consumed before exercise, in any format (e.g., capsules, pills, drink, gum, coffee, glue or spray) and its dosage needed to vary between 100 and 400 mg or between 2 and 6 mg/kg, in conformity with the limits established by the FDA. For the placebo protocol, the studies necessitated offering an intervention physically and visually comparable to the CAF format (e.g., dextrose capsule, opaque gelatin, starch, maltodextrin or cellulose, CAF-free coffee, bottled mineral water with artificial refreshment) and with an identical taste. These studies all necessitated reporting the linear values (time or frequency domain), geometric values and/or non-linear HRV indices logged before and after exercise.

      2.3 Data extraction

      Information about the author, study design, features of the study participants, intervention and the results of the respective studies were stated. After resolving the identified inconsistencies via a discussion, missing data were requested by contacting the corresponding study authors. When the authors’ correspondent did not reply, the Web Plot Digitizer® was enforced to extract data presented in charts. This stage was finalized by two reviewers (CJRB and LAG) from the research group self-reliantly.

      2.4 Assessment of the risk of bias

      The risk of bias analysis was initiated via the Cochrane organization guidelines [
      ], using the Review Manager program (RevMan 5.4.1). The authorization for calculating the risks of bias in randomized clinical trials is an instrument based on the domain assessment, specifically, a critical assessment achieved distinctly to different aspects of the risk of bias [
      • Carvalho A.P.V.
      • Silva V.
      • Grande A.J.
      Avaliação do risco de viés de ensaios clínicos randomizados pela ferramenta da colaboração Cochrane.
      ]. The assessment was split into seven cohorts: “Random sequence generation”, “Allocation concealment”, “Blinding of participants and personnel”, “Blinding of outcome assessment”, “Incomplete outcome data”, “Selective reporting” and “Other Bias”. The classification was split into three responses, “Low risk” for low risk of bias, “Unclear risk” for questions not explained by the author of the original article and “High risk” for high risk of bias. Our decisions were based on the “Reviewer's judgment and criteria for judgment” table [
      • Carvalho A.P.V.
      • Silva V.
      • Grande A.J.
      Avaliação do risco de viés de ensaios clínicos randomizados pela ferramenta da colaboração Cochrane.
      ]. The risk of bias analysis was completed by two independent authors (AAP and LAG), and another (VEV) was consulted if there were any disagreements in their choices.

      2.5 GRADE analysis

      The Grades of Recommendation, Assessment, Development, and Evaluation (GRADE) Working Group [
      • Atkins D.
      • Best D.
      • Briss P.A.
      • et al.
      Grading quality of evidence and strength of recommendations.
      ] was followed so as to analyze the certainty of evidence, including study design, broadly categorized as observational studies (low evidence) and randomized trials (high evidence). Study quality (detailed study methods and execution) and significant limitations secondarily were measured using strength of evidence analysis [
      • Meader N.
      • King K.
      • Llewellyn A.
      • et al.
      A checklist designed to aid consistency and reproducibility of GRADE assessments: development and pilot validation.
      ]. Summary of Findings was formed using GRADEpro GDT® version 4 (McMaster University, ON, Canada).

      2.6 Data analysis: systematic review

      A descriptive synthesis was implemented to describe detailed data about how each study was conducted. The key details for each study were described in texts and tables. The outcome of the individual qualitative analysis for each study was achieved by studying HRV indices at rest versus the recovery values after exercise with and without CAF.
      For the HRV indices estimations, all indices measured were specified by the studies: RMSSD = square root of the mean of the square of the differences between adjacent normal RR intervals; SDNN = standard deviation of all normal RR intervals recorded in a time interval, expressed in ms. HF = high frequency component of power spectral analysis, which ranged from 0.15 Hz to 0.44 Hz in absolute units (ms2) and normalized units (n.u.) (vagal component); LF = low frequency component, which ranged between 0.04 and 0.15 Hz in absolute units (ms2) and normalized units (n.u.) (sympathetic and parasympathetic influence); LF/HF = low and high frequency ratio; SD1 (poincare plot) = standard deviation of the instantaneous variability of the beat-to-beat heart rate; SD2 ((poincare plot) = standard deviation of long-term continuous RR interval variability; and, ApEn = Approximate Entropy.

      2.7 Data analysis: meta-analysis

      The insertion of the HRV values in the meta-analysis was the performance of the RMSSD and HF indices. The effects of the CAF interventions on each index were measured using the mean and standard deviation (MSD). MSD were assessed as the difference between the intervention and control groups before (baseline) and after exercise. The data required to construct the meta-analysis was the first period recorded following the exercise. The datasets were arranged in Microsoft Excel, and then prepared in the Review Manager Program (RevMan 5.4.1) to perform the meta-analysis.
      Heterogeneity was assessed using the I2 statistic, wherein a number greater than 50% was recognized as indicating substantial heterogeneity between the tests [
      • Higgins J.P.T.
      • Thompson S.G.
      Quantifying heterogeneity in a meta-analysis.
      ].
      For the values of “CI” and “Test for size of the general effect”, values of p < 0.05 (or, <5%) were considered significant. Similarly, we enforced a random effects model, in preference to a fixed effect model, because this was a more conventional technique that allows that the heterogeneity of the study may deviate beyond chance, offering extra generalizable results [
      • Lotufo P.A.
      • Valiengo L.
      • Benseñor I.M.
      • Brunoni A.R.
      A systematic review and meta-analysis of heart rate variability in epilepsy and antiepileptic drugs: HRV in Epilepsy.
      ]. As a result of the low number of articles, no meta-regression was undertaken. All data was generated by the Review Manager Program (RevMan 5.4.1).
      For HRV analysis; linear methods were computed in the time (RMSSD - square root of the square mean of the differences between adjacent normal RR-interval in a time interval, expressed in ms) and frequency domains (HF – high frequency, 0.15–0.4 Hz) [
      • Vanderlei L.C.M.
      • Pastre C.M.
      • Hoshi R.A.
      • Carvalho TD de
      • Godoy MF de
      Noções básicas de variabilidade da frequência cardíaca e sua aplicabilidade clínica.
      ].

      3. Results

      A total of 1026 studies were recognized by searches in the five databases. After removal of duplicates (n = 454), 586 publications were screened for inclusion. Amongst them, 475 records were dismissed after reviewing the title and/or abstract. The remaining 111 papers were selected for full text reading and 99 were excluded. Finally, 12 studies were included in the qualitative synthesis (systematic review) and 5 studies were used for the quantitative synthesis (meta-analysis). The search process and selection stage details are illustrated in the Flow Diagram of the PRISMA protocol (Fig. 1).

      3.1 Analysis of the risk of bias

      All studies presented the “High Risk” or “Unclear Risk” bias consistent with the authors' judgment (Fig. 2). The majority of the problems arose from the area “Blinding of outcome assessment” (12 out of 13 studies) or “Random sequence Generation”, wherein most authors were unclear about the procedure adopted for this field. The “Other Bias” domain asked questions regarding: populations with different body mass indexes (BMI) [
      • Blake M.S.
      • Johnson N.R.
      • Trautman K.A.
      • Grier J.W.
      • Stastny S.N.
      • Hackney K.J.
      Neither a multi-ingredient pre-workout supplement nor caffeine were effective at improving markers of blood flow or upper-body resistance exercise performance.
      ,
      • Benjamim C.J.R.
      • Monteiro L.R.L.
      • Pontes Y.M.M.
      • et al.
      Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
      ] underestimation of VO2 values because of the lack of verbal encouragement from the researcher throughout the test [
      • Karapetian G.K.
      • Engels H.J.
      • Gretebeck K.A.
      • Gretebeck R.J.
      Effect of caffeine on LT, VT and HRVT.
      ], controlled breathing during the recovery period, inducing an unnatural recovery [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ], lack of information about the menstrual period for the female subjects [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ,
      • Thomas R.M.
      • Algrain H.A.
      • Ryan E.J.
      • et al.
      Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
      ] and heterogeneous groups [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ]. In the article by Lopes-Silva et al. [
      • Lopes-Silva J.P.
      • Silva Santos JF da
      • Branco B.H.M.
      • et al.
      Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
      ], it was necessary to study the effects of CAF in a fight situation, but, since the apparatus for HRV evaluation was located on the athletes' torso, the fight protocol was adapted, greatly restricting the athletes' movements during the fight simulation. This made the findings in a real competitive situation ambiguous [
      • Lopes-Silva J.P.
      • Silva Santos JF da
      • Branco B.H.M.
      • et al.
      Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
      ].
      Figure 2
      Figure 2Bias risk analysis judgment by study and bias domains.

      3.2 CAF and its impact on HRV recovery parameters

      The studies included in our systematic review were published between 2002 and 2021. The total sample size is 232 participants (Table S1). The studies evaluated HRV via linear (time and frequency domain), geometric and non-linear indices (Table S2). Benjamim et al. [
      • Benjamim C.J.R.
      • Monteiro L.R.L.
      • Pontes Y.M.M.
      • et al.
      Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
      ] stated that CAF delayed the recovery of the HRV parameters after exercise in healthy males. Bunsawat et al. [
      • Bunsawat K.
      • White D.W.
      • Kappus R.M.
      • Baynard T.
      Caffeine delays autonomic recovery following acute exercise.
      ] detected a delay in post-exercise vagal reactivation, Gonzaga et al. [
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Garner D.M.
      • Valenti V.E.
      Involvement of cardiorespiratory capacity on the acute effects of caffeine on autonomic recovery.
      ] explained that CAF delayed parasympathetic recovery in physiologically healthy subjects and subjects with Vo2peak<42.46 mL/kg/min [
      • Rolim P.
      • da Costa Matos R.A.
      • Von Koenig Soares E. de MK.
      • Molina G.E.
      • da Cruz C.J.G.
      Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
      ]. Nishijima et al. [
      • Nishijima Y.
      • Ikeda T.
      • Takamatsu M.
      • et al.
      Influence of caffeine ingestion on autonomic nervous activity during endurance exercise in humans.
      ] established a significant increase in the LF and Total power indices in the post-exercise period, and Yeragami et al. [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ] explained the increase in the LF index in the group that consumed CAF in the same period. Thomas et al. [
      • Thomas R.M.
      • Algrain H.A.
      • Ryan E.J.
      • et al.
      Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
      ] detected that CAF intake induced a decrease in cardiac vagal control linked to the placebo group via the ApEn index. Nevertheless, Rolim et al. [
      • Rolim P.
      • da Costa Matos R.A.
      • Von Koenig Soares E. de MK.
      • Molina G.E.
      • da Cruz C.J.G.
      Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
      ] stated that CAF accelerated parasympathetic reactivation after submaximal exercise in physically active males. Blake et al. [
      • Blake M.S.
      • Johnson N.R.
      • Trautman K.A.
      • Grier J.W.
      • Stastny S.N.
      • Hackney K.J.
      Neither a multi-ingredient pre-workout supplement nor caffeine were effective at improving markers of blood flow or upper-body resistance exercise performance.
      ], Karayigit et al. [
      • Karayigit R.
      • Naderi A.
      • Akca F.
      • et al.
      Effects of different doses of caffeinated coffee on muscular endurance, cognitive performance, and cardiac autonomic modulation in caffeine naive female athletes.
      ], Lopes-Silva et al. [
      • Lopes-Silva J.P.
      • Silva Santos JF da
      • Branco B.H.M.
      • et al.
      Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
      ], and Sarshin et al. [
      • Sarshin A.
      • Naderi A.
      • da Cruz C.J.G.
      • et al.
      The effects of varying doses of caffeine on cardiac parasympathetic reactivation following an acute bout of anaerobic exercise in recreational athletes.
      ] revealed no outcome of CAF on HRV indices after exercise.
      Five studies encountered the inclusion criteria for our meta-analysis [
      • Benjamim C.J.R.
      • Monteiro L.R.L.
      • Pontes Y.M.M.
      • et al.
      Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
      ,
      • Thomas R.M.
      • Algrain H.A.
      • Ryan E.J.
      • et al.
      Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Valenti V.E.
      Caffeine affects autonomic control of heart rate and blood pressure recovery after aerobic exercise in young adults: a crossover study.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Garner D.M.
      • Valenti V.E.
      Involvement of cardiorespiratory capacity on the acute effects of caffeine on autonomic recovery.
      ,
      • Karayigit R.
      • Naderi A.
      • Akca F.
      • et al.
      Effects of different doses of caffeinated coffee on muscular endurance, cognitive performance, and cardiac autonomic modulation in caffeine naive female athletes.
      ]. Gonzaga et al. [
      • Rolim P.
      • da Costa Matos R.A.
      • Von Koenig Soares E. de MK.
      • Molina G.E.
      • da Cruz C.J.G.
      Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
      ] and Karayigit et al. [
      • Karayigit R.
      • Naderi A.
      • Akca F.
      • et al.
      Effects of different doses of caffeinated coffee on muscular endurance, cognitive performance, and cardiac autonomic modulation in caffeine naive female athletes.
      ] inspected two experimental groups: the first study with different peak VO2 values and the second with changed CAF doses, and a control group for the two protocols, all groups were included. During the creation of the meta-analysis graph, the groups were divided into subgroups to verify whether the type of exercise could influence HRV recovery. The first group “CAF in resistance exercise” represents studies whose protocol included resistance exercises. The second group “CAF in aerobic exercise” denoted studies whose protocol included aerobic exercises. The remaining deviations and 95% confidence intervals (CI) constant with the HRV indices for each assay are illustrated in Fig. 3A and B.
      Figure 3
      Figure 3A. Effects of CAF intake before exercise on HF indices recovery following effort. B. Effects of CAF intake before exercise on RMSSD indices recovery following effort.
      Figure 3A shows the forest plot graphs to compare the time domain index (RMSSD) between individuals who consumed CAF and individuals that consumed placebo. We enforced random effects model and standard mean difference to measure the effect size (black diamonds); the proportion of the center of the diamonds represents the 95% CI. A negative effect specifies impact on the group that ingested CAF related to controls. No significant change was noticed for the RMSSD, for “Test for overall effect” as we revealed p = 0.19 and a heterogeneity of 63%. In the subgroup “CAF in resistance exercise” the subtotal (CI) was −0.39 (95% CI: −0.98, 0.20)”. The subgroup “CAF in aerobic exercise” offered for “Test for overall effect”, as p = 0.22 and a heterogeneity of 58%. The subtotal (CI) was 0.20 (95% CI: −0.12 to 0.51) for this subgroup. The total outcome of the meta-analysis for this index revealed p = 0.67, a “Total (CI)” was −0.08 (95% CI: 0.45 to 0.29), and 66.7% for heterogeneity between groups.
      Figure 3B shows the forest plot graphs to link the frequency domain (HF) index between individuals who ingested CAF and individuals that consumed placebo. We imposed a random effects and standard mean difference model (Std. Mean Difference) to quantify the effect size (black diamonds); the quantity for the center of the diamonds represents the 95% CI. A negative effect designates impact on the group that consumed CAF compared to controls. No significant change was documented for HF, for “Test for overall effect”, we discovered p = 0.36 and a heterogeneity of 45%. The subtotal (CI) was −0.22 (95% CI: −0.70 to 0.26) for the “CAF in resistance exercise” subgroup. The subgroup “CAF in aerobic exercise” offered for “Test for overall effect”, as p = 0.40 and a heterogeneity of 20%. The subtotal (CI) was 0.15 (95% CI: −0.20 to 0.51) for this subgroup. The total outcome of the meta-analysis for this index revealed p = 0.22, “Total (CI)” was 0.02 (95% CI: −0.34 to 0.30), and 44% for heterogeneity between groups.

      4. Discussion

      Our study was commenced to evaluate the acute effects of pre-exercise CAF intake on post-exercise HRV recovery through a systematic review and meta-analysis. The present meta-analysis delivers evidence that CAF intake did not significantly influence HRV post-exercise recovery compared to the placebo based on the time (RMSSD) and frequency domain (HF) indices. The “total (CI)” values were close to zero, assumed to be the line of nullity of the statistic.
      CAF is often ingested in sports for improvement in cognitive performance [
      • McLellan T.M.
      • Caldwell J.A.
      • Lieberman H.R.
      A review of caffeine's effects on cognitive, physical and occupational performance.
      ], its effects are mostly related to the adenosine receptors blockade and increased central nervous system activity. This core mechanism justifies that CAF be considered as an ergogenic resource as it can enhance physical performance. CAF effectiveness in enhancing the athletes' performance in diverse sports training profiles (e.g., endurance, strength) has made it the most prescribed substance in sports [
      • Maughan R.J.
      • Burke L.M.
      • Dvorak J.
      • et al.
      IOC consensus statement: dietary supplements and the high-performance athlete.
      ,
      • Grgic J.
      • Trexler E.T.
      • Lazinica B.
      • Pedisic Z.
      Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis.
      ]. CAF use can be prescribed in low (<3 mg/kg), moderate (≥3 mg/kg up to 6 mg/kg) and high (9 mg/kg) doses [
      • Grgic J.
      • Trexler E.T.
      • Lazinica B.
      • Pedisic Z.
      Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis.
      ]. Its use is recommended by the International Society for Sports Nutrition (ISSN) [
      • Guest N.S.
      • VanDusseldorp T.A.
      • Nelson M.T.
      • et al.
      International society of sports nutrition position stand: caffeine and exercise performance.
      ] and the International Olympic Committee (IOC), and permitted by World Anti-Doping Agency (WADA) [
      • Maughan R.J.
      • Burke L.M.
      • Dvorak J.
      • et al.
      IOC consensus statement: dietary supplements and the high-performance athlete.
      ]. Moreover, the safety of CAF estimated though cardiac autonomic control remains uncertain and presents controversial data, as it might be influenced by many variables such as the dose consumed and the body's’ physiological sensitivities [
      • Beaumont R.
      • Cordery P.
      • Funnell M.
      • Mears S.
      • James L.
      • Watson P.
      Chronic ingestion of a low dose of caffeine induces tolerance to the performance benefits of caffeine.
      ,
      • Caliskan S.G.
      • Kilic M.A.
      • Bilgin M.D.
      Acute effects of energy drink on hemodynamic and electrophysiologic parameters in habitual and non-habitual caffeine consumers.
      ].
      Consistent with the FDA, consumption of up to 400 mg of CAF per day (≈6 mg of CAF per kilogram of body weight) is not related to the adverse effects of CAF e.g., anxiety, headaches, nausea and restlessness in healthy adults [
      • Spriet L.L.
      Exercise and sport performance with low doses of caffeine.
      ]. CAF dosage in the chosen studies in this meta-analysis imposed doses ranging from 3.0 to 6.0 mg/kg, values considered harmless consistent with the FDA [
      • Turnbull D.
      • Rodricks J.V.
      • Mariano G.F.
      • Chowdhury F.
      Caffeine and cardiovascular health.
      ]. Yet, there is an extensive range with regards the degree of sensitivity to the CAF effects, as chronic CAF consumption in regular consumers can generate tolerance [
      • Beaumont R.
      • Cordery P.
      • Funnell M.
      • Mears S.
      • James L.
      • Watson P.
      Chronic ingestion of a low dose of caffeine induces tolerance to the performance benefits of caffeine.
      ].
      In non-habitual consumers, one of the key effects of CAF are the HR increase and peaks in systolic blood pressure. All the same, these effects have a tendency to be gradually reduced with continued CAF intake [
      • van Dam R.M.
      • Hu F.B.
      • Willett W.C.
      Coffee, caffeine, and health.
      ]. In regular coffee drinkers, CAF ingested at concentrations of 100 mg and 200 mg had no significant effects on HRV [
      • Rauh R.
      • Burkert M.
      • Siepmann M.
      • Mueck-Weymann M.
      Acute effects of caffeine on heart rate variability in habitual caffeine consumers.
      ] in healthy young persons. Yet, when evaluated at the same concentrations in young healthy non-habitual drinkers, CAF promoted a reduction in the parasympathetic indices related to modulation of the heart (RMSSD, HF, SDNN, pNN50) [
      • Sondermeijer H.P.
      • van Marle A.G.J.
      • Kamen P.
      • Krum H.
      Acute effects of caffeine on heart rate variability.
      ].
      The different behavior of CAF responses to intake about its typical consumption may be connected to an adjustment of the adenosine receptors [
      • Sondermeijer H.P.
      • van Marle A.G.J.
      • Kamen P.
      • Krum H.
      Acute effects of caffeine on heart rate variability.
      ]. In 1979, Fredholm [
      • Fredholm B.B.
      Are methylxanthine effects due to antagonism of endogenous adenosine?.
      ] revealed an increase in the number of adenosine binding sites in the cerebral cortex of rats treated for two weeks with 10 mg/kg/day of CAF. This response supports the theory that chronic CAF ingestion might result in newly created adenosine receptors, partly reducing the blocking action of CAF and thus lessening its stimulant effects [
      • Fredholm B.B.
      Are methylxanthine effects due to antagonism of endogenous adenosine?.
      ].
      Thus, the studies in this meta-analysis offered heterogeneous criteria and assessments of habitual CAF consumption, which is an imperative variable that needs to be controlled, as it impacts the size of the effects of CAF on the autonomic nervous system as reported above. In the study by Bunsawat et al. [
      • Bunsawat K.
      • White D.W.
      • Kappus R.M.
      • Baynard T.
      Caffeine delays autonomic recovery following acute exercise.
      ] they considered as an exclusion factor consumption of beverages with CAF greater than the equivalent of three cups of coffee per day (>285 mg/day). Nevertheless, in the study by Thomas et al. [
      • Thomas R.M.
      • Algrain H.A.
      • Ryan E.J.
      • et al.
      Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
      ] all participants consumed low to moderate in CAF intake (≤300 mg/day).
      Some studies measured higher values for the usual consumption of CAF; Nishitjima et al. [
      • Nishijima Y.
      • Ikeda T.
      • Takamatsu M.
      • et al.
      Influence of caffeine ingestion on autonomic nervous activity during endurance exercise in humans.
      ] studied regular CAF users (≤500 mg per day). Whereas, in the study directed by Blake et al. [
      • Blake M.S.
      • Johnson N.R.
      • Trautman K.A.
      • Grier J.W.
      • Stastny S.N.
      • Hackney K.J.
      Neither a multi-ingredient pre-workout supplement nor caffeine were effective at improving markers of blood flow or upper-body resistance exercise performance.
      ] participants were included who consumed CAF between ≥100 mg and ≤500 mg per day.
      In research by Rolim et al. [
      • Rolim P.
      • da Costa Matos R.A.
      • Von Koenig Soares E. de MK.
      • Molina G.E.
      • da Cruz C.J.G.
      Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
      ] and Sarshin et al. [
      • Sarshin A.
      • Naderi A.
      • da Cruz C.J.G.
      • et al.
      The effects of varying doses of caffeine on cardiac parasympathetic reactivation following an acute bout of anaerobic exercise in recreational athletes.
      ], they involved subjects who did not routinely consume CAF (<120 mg/day; <3 times/week). Lopes-Silva et al. [
      • Lopes-Silva J.P.
      • Silva Santos JF da
      • Branco B.H.M.
      • et al.
      Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
      ] studied the level of habitual consumption of the subjects (n = 10) and they revealed that two subjects were considered to have a high level of CAF consumption (greater than six cups of coffee each day), while others reported a low-level consumption (less than two cups of coffee each day).
      Members in the study completed by Yearagani et al. [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ] have a low level of habitual CAF intake specified as 62 ± 14 mg per day. Further research did not provide the required information for their subjects [
      • Benjamim C.J.R.
      • Monteiro L.R.L.
      • Pontes Y.M.M.
      • et al.
      Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Valenti V.E.
      Caffeine affects autonomic control of heart rate and blood pressure recovery after aerobic exercise in young adults: a crossover study.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Garner D.M.
      • Valenti V.E.
      Involvement of cardiorespiratory capacity on the acute effects of caffeine on autonomic recovery.
      ].
      Also, the modification in CAF metabolism can be influenced by gender as a result of hormonal differences, for instance in females during the menstrual cycle. In the study led by Skinner et al. [
      • Skinner T.L.
      • Jenkins D.G.
      • Leveritt M.D.
      • et al.
      Factors influencing serum caffeine concentrations following caffeine ingestion.
      ], no alteration was documented in the extent of the ergogenic effect in males and females, after consuming 3 mg/kg of CAF 90 min before an exercise assessment. Still, higher post-exercise CAF concentrations were noted in females, suggesting a greater physiological CAF durability.
      It is notable that the majority of studies involving CAF supplementation are conducted in males. One of the reasons is the sampling complexity caused by the menstrual cycle and the use of contraceptives that adjust CAF metabolism [
      • Pickering C.
      • Grgic J.
      Caffeine and exercise: what next?.
      ].
      This scenario was noticed in our review, where the majority of the nominated articles were conducted solely in males [
      • Blake M.S.
      • Johnson N.R.
      • Trautman K.A.
      • Grier J.W.
      • Stastny S.N.
      • Hackney K.J.
      Neither a multi-ingredient pre-workout supplement nor caffeine were effective at improving markers of blood flow or upper-body resistance exercise performance.
      ,
      • Benjamim C.J.R.
      • Monteiro L.R.L.
      • Pontes Y.M.M.
      • et al.
      Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
      ,
      • Nishijima Y.
      • Ikeda T.
      • Takamatsu M.
      • et al.
      Influence of caffeine ingestion on autonomic nervous activity during endurance exercise in humans.
      ,
      • Lopes-Silva J.P.
      • Silva Santos JF da
      • Branco B.H.M.
      • et al.
      Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Garner D.M.
      • Valenti V.E.
      Involvement of cardiorespiratory capacity on the acute effects of caffeine on autonomic recovery.
      ,
      • Rolim P.
      • da Costa Matos R.A.
      • Von Koenig Soares E. de MK.
      • Molina G.E.
      • da Cruz C.J.G.
      Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
      ,
      • Gonzaga L.A.
      • Vanderlei L.C.M.
      • Gomes R.L.
      • Valenti V.E.
      Caffeine affects autonomic control of heart rate and blood pressure recovery after aerobic exercise in young adults: a crossover study.
      ,
      • Sarshin A.
      • Naderi A.
      • da Cruz C.J.G.
      • et al.
      The effects of varying doses of caffeine on cardiac parasympathetic reactivation following an acute bout of anaerobic exercise in recreational athletes.
      ]. In the study by Bunsawat et al. [
      • Bunsawat K.
      • White D.W.
      • Kappus R.M.
      • Baynard T.
      Caffeine delays autonomic recovery following acute exercise.
      ], experiments were completed in the first seven days of the female subjects’ menstrual cycle to balance the effects of sexual hormones on HRV. Still, the studies by Thomas et al. [
      • Thomas R.M.
      • Algrain H.A.
      • Ryan E.J.
      • et al.
      Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
      ] and Yeragani et al. [
      • Yeragani V.K.
      • Krishnan S.
      • Engels H.J.
      • Gretebeck R.
      Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
      ] did not present evidence about the menstrual period of their subjects. Karayigit et al. [
      • Karayigit R.
      • Naderi A.
      • Akca F.
      • et al.
      Effects of different doses of caffeinated coffee on muscular endurance, cognitive performance, and cardiac autonomic modulation in caffeine naive female athletes.
      ] completed a single study that started its tests entirely in females.
      Considering that CAF has its peak concentration between 30 and 60 min after ingestion [
      • Smits P.
      • Thien T.
      • van't Laar A.
      Circulatory effects of coffee in relation to the pharmacokinetics of caffeine.
      ], the study of Nishijima et al. [
      • Nishijima Y.
      • Ikeda T.
      • Takamatsu M.
      • et al.
      Influence of caffeine ingestion on autonomic nervous activity during endurance exercise in humans.
      ] brings important pharmacokinetics aspects, namely the time of CAF supply. CAF was offered 120 min before exercise; despite this, HRV was slowly recovered in the group that ingested the caffeine in this study. Smith et al. [
      • Benowitz N.L.
      Clinical pharmacology of caffeine.
      ] ensure that the half-life of caffeine in healthy individuals occurs between 2 h and 8.5 h after ingestion. This fact may be enough to interfere with recovery from HRV after exercise.
      While the CAF stimulant effect has been recognized, there are still many practical aspects similar to the influence of CAF on post-exercise autonomic recovery, as some of the nuances about its use remains unstandardized. We anticipate that the features identified in this meta-analysis will offer some guidelines for further studies, to improve comprehension of the effects of CAF on the autonomic nervous system. The results conclude that CAF is a benign supplement for the cardiovascular system and can be used in competitive sports to enhance performance.
      The result limitations are that the studies included in this review investigated the effects of CAF in a physically active and apparently healthy young population and, so, these results cannot be extrapolated to other populations. Adolescent athletes can be routinely approached with nutritional strategies to use CAF as ergogenic support; but we do not have evidence to guarantee its safety by assessing HRV after exercise. We cannot neglect CAF is one of the fat burners most used in clinical practice so as to help with weight loss in combination with physical exercise [
      • Tabrizi R.
      • Saneei P.
      • Lankarani K.B.
      • et al.
      The effects of caffeine intake on weight loss: a systematic review and dos-response meta-analysis of randomized controlled trials.
      ]. Consequently, we highlight those studies evaluating the effects of caffeine on cardiac recovery in overweight and obese individuals are still required to clarify the impacts of CAF on the cardiovascular health of this population.
      In conclusion, the current meta-analysis provides statistical consequences that CAF intake did not significantly affect HRV post-exercise recovery compared to placebo in healthy or physically active individuals based on HRV indices that reflect the parasympathetic cardiac regulation.

      Funding sources

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Appendix A. Supplementary data

      The following are the supplementary data to this article:

      References

        • Barcelos R.P.
        • Lima F.D.
        • Carvalho N.R.
        • Bresciani G.
        • Royes L.F.
        Caffeine effects on systemic metabolism, oxidative-inflammatory pathways, and exercise performance.
        Nutr Res. 2020; 80: 1-17
        • Guest N.S.
        • VanDusseldorp T.A.
        • Nelson M.T.
        • et al.
        International society of sports nutrition position stand: caffeine and exercise performance.
        J Int Soc Sports Nutr. 2021; 18: 1
        • Maughan R.J.
        • Burke L.M.
        • Dvorak J.
        • et al.
        IOC consensus statement: dietary supplements and the high-performance athlete.
        Br J Sports Med. 2018; 52: 439-455
        • Grgic J.
        • Trexler E.T.
        • Lazinica B.
        • Pedisic Z.
        Effects of caffeine intake on muscle strength and power: a systematic review and meta-analysis.
        J Int Soc Sports Nutr. 2018; 15
        • Shen J.G.
        • Brooks M.B.
        • Cincotta J.
        • Manjourides J.D.
        Establishing a relationship between the effect of caffeine and duration of endurance athletic time trial events: a systematic review and meta-analysis.
        J Sci Med Sport. 2019; 22: 232-238
        • Turnbull D.
        • Rodricks J.V.
        • Mariano G.F.
        • Chowdhury F.
        Caffeine and cardiovascular health.
        Regul Toxicol Pharmacol. 2017; 89: 165-185
        • Banerjee P.
        • Ali Z.
        • Levine B.
        • Fowler D.R.
        Fatal caffeine intoxication: a series of eight cases from 1999 to 2009.
        J Forensic Sci. 2014; 59: 865-868
        • Koenig J.
        • Jarczok M.N.
        • Kuhn W.
        • et al.
        Impact of caffeine on heart rate variability: a systematic review.
        J Caffeine Res. 2013; 3: 22-37https://doi.org/10.1089/jcr.2013.00098
        • Cole C.R.
        • Blackstone E.H.
        • Pashkow F.J.
        • Snader C.E.
        • Lauer M.S.
        Heart-rate recovery immediately after exercise as a predictor of mortality.
        N Engl J Med. 1999; 341: 1351-1357
        • Vanderlei L.C.M.
        • Pastre C.M.
        • Hoshi R.A.
        • Carvalho TD de
        • Godoy MF de
        Noções básicas de variabilidade da frequência cardíaca e sua aplicabilidade clínica.
        Rev Bras Cir Cardiovasc. 2009; 24: 205-217
        • Shaffer F.
        • Ginsberg J.P.
        An overview of heart rate variability metrics and norms.
        Front Public Health. 2017; 5: 258
        • Peçanha T.
        • Bartels R.
        • Brito L.C.
        • Paula-Ribeiro M.
        • Oliveira R.S.
        • Goldberger J.J.
        Methods of assessment of the post-exercise cardiac autonomic recovery: a methodological review.
        Int J Cardiol. 2017; 227: 795-802
        • Benjamim C.J.R.
        • Kliszczewicz B.
        • Garner D.M.
        • et al.
        Is caffeine recommended before exercise? A systematic review to investigate its impact on cardiac autonomic control via heart rate and its variability.
        J Am Coll Nutr. 2020; 39: 563-573
        • Page M.J.
        • McKenzie J.E.
        • Bossuyt P.M.
        • et al.
        The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.
        BMJ. 2021; 372: n71
      1. Higgins J. Thomas J. Cochrane handbook for systematic reviews of interventions. second ed. Standards Information Network, 2019
        • Carvalho A.P.V.
        • Silva V.
        • Grande A.J.
        Avaliação do risco de viés de ensaios clínicos randomizados pela ferramenta da colaboração Cochrane.
        Diagn Treat. 2013; 18
        • Atkins D.
        • Best D.
        • Briss P.A.
        • et al.
        Grading quality of evidence and strength of recommendations.
        BMJ. 2004; 328: 1490
        • Meader N.
        • King K.
        • Llewellyn A.
        • et al.
        A checklist designed to aid consistency and reproducibility of GRADE assessments: development and pilot validation.
        Syst Rev. 2014; 3: 82
        • Higgins J.P.T.
        • Thompson S.G.
        Quantifying heterogeneity in a meta-analysis.
        Stat Med. 2002; 21: 1539-1558
        • Lotufo P.A.
        • Valiengo L.
        • Benseñor I.M.
        • Brunoni A.R.
        A systematic review and meta-analysis of heart rate variability in epilepsy and antiepileptic drugs: HRV in Epilepsy.
        Epilepsia. 2012; 53: 272-282
        • Blake M.S.
        • Johnson N.R.
        • Trautman K.A.
        • Grier J.W.
        • Stastny S.N.
        • Hackney K.J.
        Neither a multi-ingredient pre-workout supplement nor caffeine were effective at improving markers of blood flow or upper-body resistance exercise performance.
        Int J Exerc Sci. 2020; 13: 167-182
        • Benjamim C.J.R.
        • Monteiro L.R.L.
        • Pontes Y.M.M.
        • et al.
        Caffeine slows heart rate autonomic recovery following strength exercise in healthy subjects.
        Rev Port Cardiol. 2021; 40: 399-406
        • Karapetian G.K.
        • Engels H.J.
        • Gretebeck K.A.
        • Gretebeck R.J.
        Effect of caffeine on LT, VT and HRVT.
        Int J Sports Med. 2012; 33: 507-513
        • Yeragani V.K.
        • Krishnan S.
        • Engels H.J.
        • Gretebeck R.
        Effects of caffeine on linear and nonlinear measures of heart rate variability before and after exercise.
        Depress Anxiety. 2005; 21: 130-134
        • Thomas R.M.
        • Algrain H.A.
        • Ryan E.J.
        • et al.
        Influence of a CYP1A2 polymorphism on post-exercise heart rate variability in response to caffeine intake: a double-blind, placebo-controlled trial.
        Ir J Med Sci. 2017; 186: 285-291
        • Lopes-Silva J.P.
        • Silva Santos JF da
        • Branco B.H.M.
        • et al.
        Caffeine ingestion increases estimated glycolytic metabolism during taekwondo combat simulation but does not improve performance or parasympathetic reactivation.
        PLoS One. 2015; 10e0142078
        • Bunsawat K.
        • White D.W.
        • Kappus R.M.
        • Baynard T.
        Caffeine delays autonomic recovery following acute exercise.
        Eur J Prev Cardiol. 2015; 22: 1473-1479
        • Gonzaga L.A.
        • Vanderlei L.C.M.
        • Gomes R.L.
        • Garner D.M.
        • Valenti V.E.
        Involvement of cardiorespiratory capacity on the acute effects of caffeine on autonomic recovery.
        Medicina. 2019; 55: 196
        • Rolim P.
        • da Costa Matos R.A.
        • Von Koenig Soares E. de MK.
        • Molina G.E.
        • da Cruz C.J.G.
        Caffeine increases parasympathetic reactivation without altering resting and exercise cardiac parasympathetic modulation: a balanced placebo design.
        EJSS (Champaign). 2019; 19: 490-498
        • Nishijima Y.
        • Ikeda T.
        • Takamatsu M.
        • et al.
        Influence of caffeine ingestion on autonomic nervous activity during endurance exercise in humans.
        Eur J Appl Physiol. 2002; 87: 475-480
        • Karayigit R.
        • Naderi A.
        • Akca F.
        • et al.
        Effects of different doses of caffeinated coffee on muscular endurance, cognitive performance, and cardiac autonomic modulation in caffeine naive female athletes.
        Nutrients. 2020; 13: 2
        • Sarshin A.
        • Naderi A.
        • da Cruz C.J.G.
        • et al.
        The effects of varying doses of caffeine on cardiac parasympathetic reactivation following an acute bout of anaerobic exercise in recreational athletes.
        J Int Soc Sports Nutr. 2020; 17: 44
        • Gonzaga L.A.
        • Vanderlei L.C.M.
        • Gomes R.L.
        • Valenti V.E.
        Caffeine affects autonomic control of heart rate and blood pressure recovery after aerobic exercise in young adults: a crossover study.
        Sci Rep. 2017; 7: 14091
        • McLellan T.M.
        • Caldwell J.A.
        • Lieberman H.R.
        A review of caffeine's effects on cognitive, physical and occupational performance.
        Neurosci Biobehav Rev. 2016; 71: 294-312
        • Beaumont R.
        • Cordery P.
        • Funnell M.
        • Mears S.
        • James L.
        • Watson P.
        Chronic ingestion of a low dose of caffeine induces tolerance to the performance benefits of caffeine.
        J Sports Sci. 2017; 35: 1920-1927
        • Caliskan S.G.
        • Kilic M.A.
        • Bilgin M.D.
        Acute effects of energy drink on hemodynamic and electrophysiologic parameters in habitual and non-habitual caffeine consumers.
        Clin Nutr ESPEN. 2021; 42: 333-338
        • Spriet L.L.
        Exercise and sport performance with low doses of caffeine.
        Sports Med. 2014; 44 (S2): S175-S184
        • van Dam R.M.
        • Hu F.B.
        • Willett W.C.
        Coffee, caffeine, and health.
        N Engl J Med. 2020; 383: 369-378
        • Rauh R.
        • Burkert M.
        • Siepmann M.
        • Mueck-Weymann M.
        Acute effects of caffeine on heart rate variability in habitual caffeine consumers.
        Clin Physiol Funct Imag. 2006; 26: 163-166
        • Sondermeijer H.P.
        • van Marle A.G.J.
        • Kamen P.
        • Krum H.
        Acute effects of caffeine on heart rate variability.
        Am J Cardiol. 2002; 90: 906-907
        • Fredholm B.B.
        Are methylxanthine effects due to antagonism of endogenous adenosine?.
        Trends Pharmacol Sci. 1979; 1: 129-132
        • Skinner T.L.
        • Jenkins D.G.
        • Leveritt M.D.
        • et al.
        Factors influencing serum caffeine concentrations following caffeine ingestion.
        J Sci Med Sport. 2014; 17: 516-520
        • Pickering C.
        • Grgic J.
        Caffeine and exercise: what next?.
        Sports Med. 2019; 49: 1007-1030
        • Smits P.
        • Thien T.
        • van't Laar A.
        Circulatory effects of coffee in relation to the pharmacokinetics of caffeine.
        Am J Cardiol. 1985 Dec 1; 56: 958-963
        • Benowitz N.L.
        Clinical pharmacology of caffeine.
        Annu Rev Med. 1990; 41: 277-288
        • Tabrizi R.
        • Saneei P.
        • Lankarani K.B.
        • et al.
        The effects of caffeine intake on weight loss: a systematic review and dos-response meta-analysis of randomized controlled trials.
        Crit Rev Food Sci Nutr. 2019; 59: 2688-2696