Introduction

Handcycling is becoming increasingly popular in individuals with a spinal cord injury (SCI) for both mobility and sports.1, 2 The popularity can be explained by its ease of use together with a relative low physiological3 and mechanical strain.4

To achieve the best handcycle performance in daily life or sports, a good training program is important. Valent et al.1, 5 concluded that an 8- to 12-week handcycle interval training program is effective in improving the physical capacity of people with an SCI during and after rehabilitation. However, the question arises as to what the best training regime is for handcyclists who want to compete in a mountain time trial. The distribution of exercise intensity zones for mountainous handcycling races is more than likely to be different from that seen in other handcycling races as the race duration is longer than the majority of recreational or competitive rides. Quantifying data on the intensity of such a race is needed to prescribe appropriate training regimes. Besides, the effect of the level of the SCI, due to differences in among others autonomic control,6 on the exercise intensity and endurance performance is interesting as well.

Currently, there are two studies that have described the exercise profile during handcycling races.7, 8 Janssen et al.7 investigated the exercise intensity and predictors of a 10-km handcycling race on a flat surface. This race was performed in a wheelchair with a handcycle system attached to the wheelchair instead of a fixed frame handcycle, which is normally used in competition and is needed when handcycling up a mountain, that is, the front wheel of the attachable handcycle system will slip due to the mass distribution when cycling up a mountain. Abel et al.8 performed a case study describing the exercise profile of an ultra-long handcycling race (540 km), which is an interesting study but based on only one person in a very extreme situation.

Besides analysis of the exercise intensity during a race, analyzing predictors (for example, peak power output (POpeak), lesion level, body mass and/or mass of the handcycle) of the race time might also be helpful to understand and optimize performance. Previous studies indicated POpeak (R=0.9),7, 9 peak oxygen uptake (VO2peak) (R=0.7–0.9)7, 9 and the exercise intensity during the race as best predictors of a 20-km time trial performance or 10 km race. The question is whether the same factors determine the race performance of a mountain time trial in handcycling. Besides the above-mentioned factors, the classification and mass of the participant and/or handcycle might be very important factors as well in a mountain time trial. Therefore, the purpose of this pilot study was (1) to analyze the exercise intensity of mountain handcycling race and to study the influence of SCI level; and (2) to determine predictors of the handcycling race performance.

Materials and methods

The handbike battle

The handcyclists participated in a mountain time trial among teams from eight Dutch rehabilitation centers. The 20.2-km race was held in June 2013 on the Kaunertaler Gletscherstrasse in Austria, starting at the toll gate (1287 m) and ending at the Ochsenalm (2150 m).

Participants

Forty (mainly recreational) handcyclist participated in the race and, due to the available heart rate (HR) monitors, the HR was monitored during the race in a subgroup (N=17). Characteristics of the (sub)group are described in Table 1. Inclusion criteria for participating in the race and the study were SCI and no contraindications for exercise. Before testing, all participants were screened by a rehabilitation physician for contraindications. Time since injury varied between 1 and 29 years (median: 6 years).

Table 1 Personal and lesion characteristics of the participants
Full size table

The study was approved by the local ethics committee of the Center for Human Movement Sciences, University Medical Center Groningen, University of Groningen. All participants signed an informed consent form before the start of the study.

Design

In the 2 weeks before the race, participants attended the laboratory (across rehabilitation centers) to determine physical capacity during a graded exercise test and personal and handcycle characteristics.

Personal and handcycle characteristics

Personal characteristics such as lesion level and completeness,10 age and gender were registered and are shown in Table 1. To study the effect of autonomic control on the performance, participants were divided into two subgroups: (1) individuals with cervical or T1-T5 (high) lesions; and (2) lesions at T6 or below (low). In the cervical lesion group supraspinal control to sympathetic pathways to both the heart and the vascular system can be impaired; the high thoracic group can have damage to some sympathetic pathways to the heart and the main components of the vascular system in terms of blood pressure control; the low injury level group will have spared control of the sympathetic pathways to the heart and major vascular beds controlling blood pressure.11, 12

The body mass and height of the participants were measured in the centers. The mass of the participant and wheelchair (total mass) were measured on a wheelchair weighing scale and, thereafter, the mass of the wheelchair was subtracted from the total mass. The mass of the handcycle was measured the day before the race.

All participants were classified according to the 2002 handcycle classification rules of the Union Cycliste Internationale into four classifications: H1 (tetraplegic with impairments corresponding to a complete cervical lesion at C8 or above), H2 (paraplegic with impairments corresponding to a complete lesion from Th1 to Th10), H3 (paraplegic with impairments corresponding to a complete lesion from Th11 or lower) and H4 (paraplegic with impairments corresponding to a complete lesion from Th11 or below).

Physical capacity

The physical capacity was determined in the rehabilitation centers on an arm crank ergometer (Lode Angio, Groningen, The Netherlands), with the personal handcycle on a roller system (Tacx, Terneuzen, The Netherlands) or on the Cyclus2 ergometer (RBM elektronik-automation GmbH, Leipzig, Germany). The protocol used was similar between centers. The exercise test was preceded by a 5-min resting period in which the oxygen uptake and the HR in rest were continuously measured by an Oxycon delta (CareFusion, San Diego, CA, USA) and a Polar heart rate monitor (Polar Electro Oy, Kempele, Finland). Participants were asked to empty their bladder before testing.

After a warm-up, the graded exercise test was performed. The starting workload of the graded exercise test was dependent on the classification and (estimated) handcycling level of the participants and was set between 20 and 30 W for the lower level handcyclist and between 40 and 60 W for the higher level handcyclists. The workload was increased every minute by 8–10 W (lower level) and 15–20 W (higher level).

HR, VO2 and respiratory exchange ratio were averaged over each 30-s period. The highest value during the test was taken as the peak value. POpeak was the highest power output that was maintained for at least 30 s.

HR monitoring

The HR was recorded continuously during the race with a Polar RS400 (5 s storing interval; Polar Electro Oy) or Suunto T6 (10-s storing interval; Suunto, Vantaa, Finland). Thereafter, the average over each 10-s period was calculated for the Polar data. The percentage heart rate reserve (%HRR)13 was calculated by using the HRrest and HRpeak measured before and during the graded exercise test. When during the race a higher HR was found than the HRpeak, this value was used (N=4).

Finally, the exercise intensities were determined and classified according to the American College of Sports Medicine guidelines: very light intensity (<30%HRR), light intensity (30–39%HRR), moderate intensity (40–59%HRR), vigorous intensity (60–89%HRR) and near maximal to maximal intensity (90–100%HRR).14 The vigorous intensity classification was further divided into light endurance training (60–69%HRR), intensive endurance training (70–79%HRR) and maximal lactate steady-state training (80–89%HRR), which are often used as guidelines in cycling training.

Statistics

Descriptives were calculated for all variables. Mann–Whitney U-tests and chi-squared tests were used to investigate whether the groups with a high and low lesion level were different regarding personal and handcycle characteristics, physical capacity, race time and exercise intensities during the race.

An univariate multi-level regression analysis was performed with race time (in minutes) as a dependent variable. A multi-level regression analyis was performed to be able to correct for possible differences between the centers, that is, different testers and different equipments. Independent variables were personal characteristics (age, gender, lesion level (high=0; low=1) classification (H1/H2=0; H3/H4=1), body mass, body mass index (BMI) and waist circumference), handcycle mass, hours per week involved in sports or handbike training, fitness level (POpeak and VO2peak) and HR parameters during the race (%HRRmean and %time in different intensity zones). Level of significance was set at P<0.05.

Results

Descriptives

Table 2 shows the participant characteristics. Unfortunately, not all requested tests were performed in all centers, leading to missing values for some parameters. The high lesion level groups had a significantly lower VO2peak compared to those with a low lesion level (P=0.03).

Table 2 Participant characteristics for the total study group (total group), the group that participated in the heart rate monitoring during the race (HR group) and the groups with a high and low lesion level
Full size table

Exercise intensity

Table 3 shows the exercise intensities during the race. The average %HRR during the race was 70±7%, ranging between 60 and 82%. The participants exercised most of the time at a vigorous (73%) intensity, especially in the 80–89%HRR zone (29% of the total time) and 70–79%HRR zone (29% of the total time). The only significant difference between the high and low lesion group was the time spent in the very light intensity zone (high: 2.2±1.6%; low: 0.7±1.1%, P=0.03).

Table 3 Descriptives of the heart rate monitoring data during the race of the total group that was monitored (HR group, N=17), the high lesion group (N=5) and the low lesion group (N=12)
Full size table

Predictors of race performance

Results of the regression analyses are shown in Table 4. Waist circumference, handbike mass and fitness level (POpeak and VO2peak) explained each between 26 and 39% of the race time. Persons with a smaller waist circumference, lighter bike and better fitness level had a better race performance. Figure 1 shows the relationship between POpeak and race time.

Table 4 Predictors of race time (in minutes)
Full size table
Figure 1
figure 1

Relationship between peak power output (POpeak) and the race time (N=33).

Full size image

The exercise intensity during the race (%HRRmean) explained 57% of the race time and the percentage time in different intensity zones explained 15–70% of the race time. The race time was better when the average exercise intensity was higher, the percentage time in the very light to vigorous intensity (in the 60–69%HRR) zone was lower or the percentage time in the 80–89%HRR zone was higher. Figure 2 shows the relationship between %HRRmean and the race time.

Figure 2
figure 2

Relationship between mean percentage heart rate reserve during the race and the race time (N=17).

Full size image

Discussion

Exercise intensity

The present pilot study is the first that looked at the exercise intensity and predictors of a mountain time trial in handcycling. The average %HRR during the 20-km mountain time trial was lower (70±7%HRR) compared with what was found during a 10-km handcyling race on a flat terrain (83±9%HRR).7 Furthermore, the fitness level of the participants in the present study (VO2peak: 33.2±10.2 ml kg−1 min−1; POpeak: 2.0±0.6 W kg−1) was on average better than in the participants of Janssen et al.7 (VO2peak: 14.2±3.8 ml kg−1 min−1 and POpeak: 0.72±0.30 W kg−1 for the tetraplegic classification; VO2peak: 27.1±6.0 ml kg−1 min−1 and POpeak: 1.64±0.32 W kg−1 for the paraplegic classification), which would probably lead to a lower exercise intensity in our participants when performing in the same race at comparable velocity. The difference in fitness level between studies might be explained by the training period of our participants before the mountain time trial or because the trial is a bigger physical challenge and might attract, therefore, fitter participants and those with a lower lesion level. While 10 of the 16 (63%) participants of Janssen et al.7 had a tetraplegia, only 15% of the participants in the mountain time trial had a tetraplegia.

The lower mean %HRR in our study might be due to a different pacing technique since it is a longer race with more resistance due to the slope than on a flat terrain. During the 10-km race participants probably did not stop (average race time: 37.5 min) while the recreational handcyclist during the mountain race had to stop frequently (average race time: 238.7 min). The stops are visualized by the higher drops in %HRR and the lumpier altitude line in Figure 3 (slow handcyclist) compared with Figure 4 (fast handcyclist). The fast handcyclist only showed large drops in %HRR when going downhill or cycling on the almost flat part along the reservoir (Figure 4). As a consequence of the stops, the HR drops and the average %HRR over the time trial will be lower. This was also indicated by the strong association between percentage time performing in the light intensity zone and the race time (R2=70%), that is, the handcyclist who stops frequently will spent more time in the (very) light intensity zone and will have a lower %HRRmean during the race, which were both related to a longer race time. Unfortunately, we did not register the reasons for stopping and whether this was due to local (arms) or central (cardiorespiratory) fatigue or both.

Figure 3
figure 3

Altitude (bold solid line, right y axis) and heart rate response (%HRR, left y axis) of one of the recreational handcyclist (high lesion level group). The horizontal lines indicate the %HRR zones. The lumpier altitude line compared with Figure 4 indicates the multiple stops for this recreational handcyclist during the race. Average %HRR during the race of this participant is 62%.

Full size image
Figure 4
figure 4

Altitude (bold solid line, right y axis) and heart rate response (%HRR, left y axis) of one of the faster handcyclists (low lesion level group). The horizontal lines indicate the %HRR zones. Average %HRR during the race of this participant is 83%.

Full size image

Predictors of race time

The best predictors of the race time were waist circumference, mass of the handcycle, fitness level (both POpeak and VO2peak) and exercise intensity during the race. Surprisingly, lesion level and classification were not predictors of the race time. The expectation was that loss of motor control below the level of the lesion might influence the sports performance. Furthermore, the autonomic control might be important as well since it has an effect on cardiovascular function (blood pressure and HR) and subsequently the performance.6 Therefore, our group was divided into two, based on an injury level above or below T6. Injury below T6 tends to spare cardiac and most blood vessel control.6 Unexpectedly, only a difference in VO2peak and time spent in the very light intensity zone was found between the high and low lesion group. The high lesion group showed a lower VO2peak and a longer percentage of time spent in the very light intensity zone. However, these differences did not lead to a difference in race time between groups. This lack of difference in race time between the high and low lesion groups might be explained by the small and heterogeneous groups regarding, among others, physical capacity. Furthermore, only 6 of the 11 persons with a high lesion level had complete loss of autonomic control (AIS A). The role of the autonomic control on sports performance should be subject for future studies.

The effect of exercise intensity on race time was discussed above and can probably be explained by the number of stops and the ability to cycle at a high intensity (80–90%HRR) during a substantial part of the race. A moderate correlation (R=0.66, P<0.01) between exercise intensity and race velocity was found before during the 10-km race on flat terrain; those with a higher average exercise intensity were faster.7 This indicates that better handcyclists in general are able to perform at a higher average %HRR, which is also visualized in Figure 2.

The association between race time and fitness level was found before in a 10-km and 20-km race.7, 9 The correlation coefficients between the mountain time trial and POpeak (R=0.6) and VO2peak (R=0.6) were somewhat lower compared with these previous studies (POpeak: R=0.9; VO2peak: R=0.7–0.9).7, 9 Other factors such as the mass of the participant and his handcycle will be other important factors when riding up a mountain. The external power output during steady-state cycling can be expressed as the sum of the external energy losses to rolling resistance (Froll), air drag (Fair), gravitation force when propelling up an incline (Fincl) and internal friction of the handcycle (Fint).15 Froll and Fint will be the same on a flat or hilly terrain, Fair might be lower in the mountain due to the lower velocity but Fincl will be much higher. Fincl (calculated by m·g·sinα) is dependent on the mass of the handcycle and its user. In contrast to the mass and BMI of the participants, waist circumference and mass of the handcycle were significant predictors of race performance. The body mass can be high but very functional for sports performance when the fat mass is low and the muscle mass is high. It was found before that the waist circumference correlated stronger with percentage fat mass (R=0.83), measured by bioelectrical impedance analysis, than the BMI (R=0.51) in people with SCI.16 This might explain why waist circumference is a better predictor of race performance compared with body mass or BMI.

Personal characteristics such as gender, age and classification did not predict the race performance. Within these groups, the variety in training status and fitness levels was wide, which may explain this result.

Limitations

Due to the multi-center design, different testers were involved and different equipments were used. Furthermore, not all centers collected all data as asked for in this pilot study, leading to missing values. However, by using a multi-level regression analysis a correction was made for the possible differences between centers. A future study should standardize the test execution in all centers.

Since we only had 17 HR monitors available, it was not possible to measure the HR of all participants during the race. Nevertheless, it was shown that the race performance was mostly at a vigorous intensity and a clear relationship was found between %HRR during the race and the race time.

Conclusion

A 20-km mountain time trial in a handcycle is intensive. Since a large majority of a mountain time trial is performed for long durations at a vigorous exercise intensity (70–89%HRR), it is advisable to incorporate this in the training program. The results of the present study showed that to optimize the race performance it is important to be fit, with a small waist circumference and to have a light-weight handbike. Level of SCI was not significantly associated with race time.

Data Archiving

There were no data to deposit.