In the department of sports medicine at the Institute for applied Training Science (IAT) in Leipzig, mainly high-performance athletes, from different sports, are seen mostly for preparticipation screenings. A retrospective analysis of our database from 2010 to 2019 for exercise testing on a bicycle ergometer identified 4,899 athletes. After excluding athletes with an age of >20 years, 2,217 datasets remained. Since the exercise-induced blood pressure regulation was intended to be analyzed in contrast with echocardiographic findings, only exercise tests of athletes were included who had a complete echocardiographic examination within the same year. After data adjustments for doublings, 739 datasets of young athletes remained. Since most of the echocardiographic studies were performed on the day of exercise testing or within a few weeks relevant breaks from sport after injury or illness between exercise testing and echocardiography can ruled out in general. All athletes were highly trained elite athletes of their sports and members of the national team squad according to their age. Non of the tested pupils suffered from any kind of cardiovascular desease nor took any kind of medication that might have an effect on the intrinsic blood pressure reaction

Height and weight were measured using a scale with integrated straightedge (Seca scale, Model 701 with telescope measuring stick Model 220, Seca GmbH & Co.Kg. Hamburg, Germany), and body mass index (BMI) was calculated by dividing weight in kilograms through the square of height in meters. Body fat was determined by skinfold thickness using the ten-fold model introduced by Parizkova (16). Fat percentage was then used to partition total body mass into fat mass and fat-free mass. Resting and systolic blood pressure (rSBP) and diastolic blood pressure (rDBP) was measured with a standard sphygmomanometer (Erka, Bad Tölz, Germany) adjusted to the individual's arm circumference on both arms in a sitting position after at least 5 min of resting; the average was calculated for the present study. Exercise systolic blood pressure (eSBP) and diastolic blood pressure (eDBP) was measured with with the similar sphygmomanometer (Erka, Bad Tölz, Germany).

Tests were performed on cycle ergometers (Lode sport Excalibur, Lode, Groningen, the Netherlands). The standard protocol for all females started with an initial workload of 50 Watt (W), increment 30 W; stage duration was 3 min. Only petite and lightweight girls started with an initial workload of 25 W. The initial workload for all male athletes was 100 W (increment 30 W, stage duration 3 min). Boys of little height or weight (<50 kg) could also be tested with the girl's protocol. The supervising physician decided the test protocol. All tests were supervised by experienced staff and were performed until subjective exhaustion. All participants were motivated to reach their maximal limits of work capacity by oral motivation. Sine all participants were experienced athletes, all participants were familiar with voluntary maximal exhaustion. Heart rates and BP were measured at the end of each stage, including maximal SBP and DBP (mSBP; mDBP). Heart rates were obtained automatically from 12-channel ECG (cardio 300, Custo med GmbH, Ottobrunn, Germany) but were verified and corrected (if necessary) manually afterward. BP was measured manually using standard sphygmomanometers adjusted to the individual's arm circumference.

Experienced echocardiographers performed two-dimensional echocardiography according to guidelines of the german society of cardiology valid for the study period (Compare: Hagendorff et al.; Manual zur Indikation und Durchführung der Echokardiographie). From 2010 to 2014, Philips IE 33 system (Philips Healthcare, Hamburg, Germany) and after 2014 a Philips Epiq, both with a 3.5 MHz transducer. Left atrial diameters (anteroposterior) were assessed from either B-mode or M-mode parasternal long-axis views. LV end-diastolic diameter and septal wall thickness were measured in the parasternal long-axis at the level of the LV minor axis, approximately at the mitral valve leaflet tips. The absolute and relative mass of the heart was calculated using the formula established by (17) Pulsed Doppler profiles at the distal margins of the mitral valve leaflets, deceleration time of the E wave and, tissue-Doppler-derived mitral annular velocities (average of septal and lateral E′) were assessed as indices of diastolic function. Movement in the M-mode (TAPSE and MAPSE) further assessed the function of the right and the left ventricle. For the relevant echocardiographic of this study z-scores were calculated. During the time of the study interval overall 3 physicians where involved in echocardiographic data gathering. one of the examiners was a specialist for sports medicine with more than 20 years experience in echocardiography, one was cardiologist and sports cardiologist with the highest certified level of training in transthoracal echokardiograhy and one was resident physician in training for internal medicine and sports medicine who was trained and observerd in echocardiography by the cardiologist. Complete echocardiographic data were available for all athletes of the presented study.

The classification of sports by Schnabel and Thieß (18) subdivides disciplines in technical-acrobatic sports, double-fight, endurance, sprint power/ strength and game sports. Based on this, all athletes were classified in one of these clusters to compare sport related differences in exercise related BP reaction. Furthermore, all athletes were clustered for their individual training intensity (by training hours per week) 4 clusters of training intensity were defined: 1–5, 6–10, 11–15 and 16–20 h/week. Given that not only the training intensity but also the age has an impact on the regulation of the BP during exercise, the component of age was considered separately. Athletes aged 10–12 years, 13–15 years, 16–18 years, and 18–20 years were clustered for age-dependent examinations.

The data obtained from the stress tests and echocardiographic examinations were processed in Microsoft Office Excel 2016 for Windows from Microsoft Corporation (Redmond, USA) and statistically calculated using IBM SPSS Statistics 25.

Except for age, all data fulfilled criteria for normal statistical distribution as verified by previous authors (19) thus consistent descriptive presentation as meanSD was chosen for all variables including age. Mean comparisons were calculated with a single factorial variance analysis. Due to the literatur (20), the robust Welch test was also tested for corresponding significance. In significant differences, a posthoc test and Bonferroni correction were calculated. The significance level was = 0.05. According to Cohen (1988), the effect sizes were interpreted as follows:.01 small effect,.06 medium effect, and.14 large effect. Correlation effects are described by the correlation coefficient |r|. According to Cohen Guidelines (1988) data was interpreted as: |r| = 0.10 a weak |r| = 0.30 a moderate |r| = 0.50 describe a strong correlation. Associations of variables were analyzed using Pearson's correlation coefficients. Normative ranges were defined as the central 95% of all observations, and lower and upper limits were estimated assuming normality as means 2-fold SD. Additionally, 95% confidence intervals for limits of normative ranges are presented in squared brackets [e.g., (1234)].

A total of n = 739 athletes (male = 442, female = 297) with a median age of 15.8 years (±2,3 years) were analyzed. There was no significant difference in age and duration of training years between male and female athletes. The demographic and clinical characteristics of this population are summarized in Table 1. Male athletes were significantly heavier and taller than their female counterparts, had less body fat, a higher fat-free mass (FFM), and a higher resting blood pressure rSBP and rDBP (Compare Table 1)

An overview of the echocardiographic data is shown in Table 2 below. Z-scores of the relevant echocardiographic results are given in Table 3. The absolute cardiac size in male athletes was significantly higher (815.4 ± 188.2 g) than in female athletes (637.7 ± 115.4 g). In relation to the body weight, the male heart volume was also higher 11.85 ± 1.54 g/kg body weight than that of the female 10.79 ± 1.56 g/ kg body weight. There were no significant differences in both the medial and lateral E wave and the right ventricular function in the form of TAPSE (tricuspid annular plane systolic excursion). The diameter values of the left ventricular posterior wall during the diastole is within the normal range for both female and male post-growth athletes. The male subjects have significantly higher wall thicknesses of 0.99 ± 0.15 cm than the female subjects with 0.88 ± 0.13 cm (p < 0.001). The diameter of the left atrium 3.42 ± 0.43 (♂), bzw. 3.2 ± 0.41 cm (♀) was within the standard range. The values of male athletes were significantly higher than those of female athletes [0.221; 95% CI (0.158. 0.283)]. The differences between the sexes in the left ventricular diameter's values during the diastole (LVEDD) and the interventricular septum thickness were significantly different, too (both with p.001). All subjects were in the standard range with their respective values. All relevant echocardiographid data are presented in relation to athletes age and sex in Figures 4–11.

Table 4 summarizes the findings related to exercise testing. SBP increased significantly (p < 0.001 for both sexes). whereas DBP remained almost unchanged in femal and male athletes. Male athletes absolved more exercise stages, started with a higher workload, attained higher peak workloads, and showed a more pronounced SBP response than female athletes

In the cohort of 10–12 year old junior athletes, the SBP at maximal effort was 159 mmHg (±18 mmHg). In the progression of the age of the athletes, the SBP increased continuously. 13–15 year old athletes had an average SBP of 182 ± 20 mmHg, 16–17 year old athletes 192 ± 20 mmHg, and 18–20 year old athletes 196 ± 22 mmHg. There was a significant difference between male and female athletes in which male athletes gained higher values. Within the different age cohorts, the SBP was also significantly different between sexes. (see Figure 1)

The influence of the training frequency on the maxSBP is shown in Figure 2. The statistical calculation showed that there were significant differences between the frequency of the training and the blood pressure development of the subjects (F = 11.823, p < 0.001). The post-hoc comparison of the categories of these significant differences between the lowest training frequency of 1–5 h/week and 6–10 h/week (−11.161; 95% CI [−16.43; −5.89] and between a training effort of 1–5 h and 11–15 h per week (−14.754; 95% CI [−23.07; −6.44]. The maximum systolic blood pressure of children and adolescents is initially linear. From a training range of 6–10 h, the values flatten and the blood pressure increases only minimally in both sexes. The maximum values are reached with a training range of 16–20 h.

To investigate the influence of the sporting type, the athletes were divided into sporting type groups according to Schnabel (1993) in technical-acrobatic sports, double-fight, endurance, sprint power/strength and game sports. The maximum systolic blood pressure values differed significantly in terms of the respective sports (F (4;732) = 4.850, p = 0.001). The highest mSBP values were observed in the sprint/ strength athletes with an average value of 190 (±23) mmHg. The lowest values were recorded by athletes from technical-acrobatic sports with an average mSBP value of 178 (±21) mmHg. Significant differences were found in the maximum systolic blood pressure values between the technical-acrobatic and endurance sports (−9.657; 95%-CI [−17.12; 1.04] as well as the technical-acrobatic and rapid-power or strength athletes (−12.653, 95%-CI [−21.147; −3.83]. In both cases, the technical-acrobatic athletes had lower mSBD values. Compare Table 5.

The Pearson correlation coefficient shows a moderate positive correlation between the relative heart volume (g/kg BM) and the mSBD (r = 0.171). The thicker the interventricular septum during the junior athletes' diastole, the higher the systolic rest values were (r = 0.323). The effect can be interpreted as moderate. If one combines above two echo parameter with the maximum systolic blood pressure value, the effect (r = 0.490) strengthens and indicates that the higher the echocardiographic diameters, the higher the mSBD values during the load.

Finally, it was investigated whether those young athletes who had significantly higher systolic blood pressure levels during the stress test also showed corresponding abnormalities in cardiac ultrasound. (see Figure 3) Significant differences of the mean SBP were found in the statistical calculation of the left ventricular diameter during the diastole, F (3; 738) = 27.552, p.001 effect strength high (η²= 0.101) (Cohen, 1988). It is noticeable that with increasing deviation from the mean, the diameters in the left ventricle also increase during the diastole. Those young athletes who had a maximum systolic blood pressure value >229 mmHg also showed the highest end-diastolic diameter. The differences were highly significant (F (3.738) = 28.134, p.001) with also high effect strength (η²= 0.103). Both between the single and the double SD (−0.0983; 95%-CI [−0.136; −0.061]) and between the simple and the more than double SD, i.e., a systolic blood pressure development of more than 229 mmHg (−0.1376; 95%-CI [−0.193; −0.082]), the calculations showed highly significant results. Equally significant were the calculations of the left ventricular posterior wall thickness (p.001). Similar to the above results, significant differences between the single and the double SD (−0.1099; 95% CI [−0.152; −0.068]) and between the simple SD and the classification >229 mmHg (−0.1374, 95% CI [−0.2; −0.075]) were found. Concerning right ventricular function in the form of TAPSE, the groups do not differ significantly (p = 0.089). The left atrium diameter showed significant differences between the mSBD classifications, F (3. 727) = 12.308, p.001. The post-hoc testing significant differences between the group of athletes with single SD and double SD (−0.236; 95%-CI [−0.365; −0.108]) and between single SD and the classification group “blood pressure >229 mg” (−0.288; 95%-CI [−0.478; −0.096]).

Of the 739 studied young athletes, percentiles of blood pressure levels were calculated. These were initially calculated jointly for both sexes and all age groups. Within the 95th percentile are blood pressure values diastolic of 95 mmHg and systolic of 230 mmHg. Across all age groups and separated by gender, 95% of young male athletes aged 10–20 years have a blood pressure below 230/95 mmHg. For female athletes, the 95th percentile is 206/90 mmHg.

In addition to distinguishing blood pressure limits by sex, age-dependent distinctions are of particular interest. Thus, all subjects were additionally assigned to a corresponding age group. For each age group, blood pressure limits have now been determined by calculating the percentiles, which can quickly provide information about possible exercise-induced hypertonia (see Tables 6–8).

However, since this work is only a cross-sectional analysis, the long-term effects of stress-induced hypertension are neglected. Mainly the long-standing competitive sport with numerous intensive training sessions and competitions provokes functional and structural changes to the musculoskeletal system and the cardiovascular system (30). To rule out pathological changes due to intensive training, further, regular examinations must take place for each athlete during his/her career. Only in this way, can the influence of a long-term high training workload reveal individual cardiovascular risks and provide information about how the body deals with high blood pressure provoked by training in the long term.

The presented blood pressure percentiles for children and adolescents are related to sex and age to simplify the graphic representation. The aspect of body hight, which is known to effect the age dependent resting blood pressure was not taken into account and might lead to misinterpreting the exercise blood pressure children and adoslescents.

Another limitation of this work is that a pre-test activity, the nutritional status of the athletes, but also temporary emotional and psychological stress, such as in school or family, were not taken into account. However, all these factors can influence athletic performance and resulting blood pressure. The extent to which they have an impact remains debatable.

To define cluster of paediatric subjects by age may not be appropriate when investigating the influence of physical maturity on parameters. Unfortunatly no data of puberty was gathered during the regular clinical assessment. Therefore the influence of pre- or late puberty on exercise related blood pressure regulation cannot be assessed in this study.