A review of biophysical differences between aquatic exercise and land-based
The following research the aquatic exercise is from W. Matt Denning and Matthew Seeley are with Brigham Young University, Department of Exercise Sciences, Provo, UT. Eadric Bressel is with Utah State University, Biomechanics Laboratory and Sports Medicine Research Center, Logan UT. Dennis Dolny is with Utah State University, Sports Medicine Research Center, Logan UT. Megan Bressel is with Mountain West Physical Therapy in Logan, UT; and was published in 2012 by International Journal of Aquatic Research and Education.
Four of the most popular modes of aquatic exercise are deep water (DW) exercise, shallow water (SW) exercise, water calisthenics (WC), and underwater treadmill (UT) exercise.
The mechanical requirements of each aquatic exercise mode may elicit different physiological and biomechanical responses.
The purpose of this descriptive literature review was to evaluate some biophysical differences between aquatic and land-based exercises.
The biophysical variables reviewed included oxygen consumption (VO2), heart rate (HR), rating of perceived exertion (RPE), stride length, stride frequency, pain, and measures of functional gain.
Based on the studies reviewed, when compared with similar land-based exercises, VO2 and HR maximum values were lower during DW and SW exercise, but, depending on water depth and exercise performed, may be greater during WC and UT exercise.
RPE during DW exercise was generally similar to land exercise during max effort.
Stride frequency tended to be lower for all four aquatic exercises, relative to on land counterparts.
Pain levels tended to be similar between WC and land exercise, yet may decrease after UT exercise.
The popularity of aquatic exercise is increasing for various reasons, including increased accessibility of pool facilities and improved understanding of health-related benefits; this increase in popularity is particularly true for special populations.
Individuals who suffer from various orthopedic dysfunctions (e.g., arthritis) and have difficulty performing on-land exercise may benefit from aquatic exercise.
Aquatic physical therapy may facilitate ease of movement, swelling reduction, and pain relief due to the pressure and warmth of water.
In addition, the effects of water resistance (i.e., drag forces) may increase energy expenditure and decrease mechanical loads on lower extremity joints.
It is important for rehabilitative clinicians to understand physiological and biomechanical responses (hereafter referred to as biophysical variables) that are related to aquatic exercise.
Clinicians should also understand how the aforementioned responses may differ, relative to land-based exercise. Deep water (DW) and shallow water (SW) exercise, water calisthenics (WC), and underwater treadmill (UT) exercise are some of the most popular forms of aquatic exercise.
The mechanical requirements of each aquatic exercise may elicit different physiological and biomechanical responses.
For example, DW exercise does not include ground contact, while SW exercise, UT exercise, and WC do include a ground contact.
This difference may partially explain why oxygen consumption (VO2) is typically lower during DW running, relative to SW running.
The mechanical requirements of each aquatic exercise also make it difficult for clinicians to know which aquatic exercise will achieve desired therapeutic goals.
For instance, if the goal of the clinician is to prescribe an aquatic exercise that most closely mimics the oxygen consumption demands of land-based exercise, then an understanding of the physiological responses of each type of aquatic exercise is imperative.
If the goal of the clinician is to prescribe an aquatic exercise that increases functionality while decreasing pain, then an understanding of the biomechanical and pain responses is equally important.
Knowledge of different biophysical responses during aquatic exercise will help clinicians prescribe the most beneficial form of aquatic treatment for their patients.
The purpose of this paper was to provide a descriptive literature review of acute and chronic biophysical differences between aquatic and land-based exercise.
The biophysical variables examined in our review included VO2, heart rate (HR), rating of perceived exertion (RPE), stride length and stride frequency, pain level, and functional gains.
The practical aim of this paper was to help clinicians better understand which exercise environment (i.e., land or water) may be most advantageous for their patients.
Not all biophysical variables have been included in this review.
For example, blood lactate, blood pressure, ground reaction forces, muscle electromyography, and joint kinematics were not included mainly because of limited data for comparisons between modes.
In this paper, we reviewed (a) four modes of aquatic exercise, (b) physiological responses (VO2, HR, and RPE) for aquatic exercise compared with land-based counterparts, and (c) biomechanical and pain responses (i.e., stride frequency, stride length, pain, and functional gains) for aquatic exercise, relative to land-based exercise.
To accomplish our stated purpose, searches were performed using several databases and search engines that included SPORTDiscus, Academic Search Premier, CINAHL, Google Scholar, PubMed, and a university library catalog.
We chose to cite studies that involved the aforementioned biophysical variables (a) during and/ or after aquatic and land-based exercise (walking or running with the exception of a few WC studies), (b) on young and elderly able-bodied subjects or young and elderly special populations, and (c) without the use of equipment (e.g., drag materials).
Modes of aquatic exercise
Although there are many different modes of aquatic exercise and therapy, this descriptive literature review included only deep water (DW) exercise, shallow water (SW) exercise, water calisthenics (WC), and underwater treadmill (UT) exercise, as these modes are the four most commonly cited in the literature.
Deep water exercise is performed when individuals walk or run in the water with no contact with the pool floor.
Typically, DW exercise involves minimal translation through the water (the participants stay in the same place).
Flotation aids (e.g., a buoyancy vest or belt) are often used to suspend the participant, so that no ground contact occurs during the exercise.
Shallow water exercise is performed in a depth typically at the xiphoid level, where participants may run or walk propelling themselves through the water.
Participants are able to contact the pool floor, therefore, eliminating the need for flotation devices.
Water calisthenics are achieved by performing a variety of aerobic conditioning and resistance training exercises usually in the shallow end of a pool.
This mode of aquatic exercise includes any type of exercise except continuous walking and running. Underwater treadmill exercise uses a treadmill submerged in water.
Some underwater treadmills include adjustable water jets.
These jets allow the therapist to alter the horizontal forces of water resistance.
In addition, some underwater treadmills permit the therapist to adjust water depth and regulate the vertical ground reaction forces that are applied to the participant by the treadmill.
By systematically controlling the horizontal resistive forces and vertical ground reaction forces that are applied to the participants, a therapist can better control exercise intensity.
This level of control is not found in other forms of aquatic exercise.
Each mode of aquatic exercise results in different physiological responses.
The studies reviewed in this section met the criteria for investigating VO2, HR, or RPE during comparable aquatic and land-based modes.
Oxygen consumption is the product of cardiac output (stoke volume × heart rate) and arterial-venous oxygen difference (a-v O2 diff), and is linearly related to caloric energy expenditure (see Table 1).
Oxygen consumption is frequently used to indicate the level of aerobic intensity for a certain individual or activity.
A comparison of VO2 values allows for an objective comparison of intensity between modes of exercise.
Researchers have indicated that maximum oxygen consumption (VO2max) during DW exercise is lower than VO2max values during over-ground treadmill running (Table 1).
There is a large variability in results that range from a 10% decrease to a 27% decrease.
Although some females obtained a VO2max lower than males, both genders display lower values in the water compared with land.
This would indicate that gender is not a contributor to the lower VO2max values during DW running exercise.
Lower VO2max values are not attributable to age, even though Nakanishi et al. (1999a) indicated that younger males had a lower percent decrease, 21% compared with the 27% decrease in older males.
Multiple factors may contribute to lower VO2 response during DW exercise; however, it is believed that water temperature, cardiovascular responses to hydrostatic pressure, and different muscle activity during DW exercise play some role in lowering VO2 values.
Other factors may also include the lack of a ground support phase and different DW exercise styles.
Although few studies have examined VO2 responses during SW exercise, these studies have indicated a 16.3% and 10% decrease in VO2max response compared with land treadmill running (Table 1).
This relatively small difference between SW running and land treadmill running indicates that SW running may elicit metabolic responses similar to land treadmill running.
Dowzer et al. (1999) and Town and Bradley (1991) compared VO2 values between SW running and DW running and indicated that VO2 is greater during SW running than during DW running.
Shallow water running VO2 values more closely resembles land treadmill running VO2 values.
This may be because SW running involves buoyant forces and a ground support phase that are more similar to land treadmill running.
Perhaps a more compelling reason for the VO2 difference between DW running and SW running is the greater relative velocity of the water during SW.
As relative velocity increases, water resistance also increases and may counteract the effects of buoyancy, i.e., greater water resistance equals greater energy expended when all else is held constant.
Researchers have examined VO2 during WC and have reported conflicting results (Table 1).
Some researchers report greater VO2 values than land based exercise, while other researchers report lower VO2 values during WC, compared with similar land exercises or land treadmill VO2max tests.
These contradictory results may be partially explained by the large variation in the type of calisthenics that were studied.
For example, Barbosa et al. (2007) required participants to perform a “rocking horse” exercise, moving the arms and legs at the same time, while Johnson et al. (1977) examined exercises that involved the arms and legs separately.
Darby and Yaekle (2000) used both leg only and arm/leg exercises separately but changed the cadence of the exercises according to the participant’s heart rate. In addition, we believe that it is difficult to control for exercise intensity during WC and, depending on intensity and exercise type, WC may elicit different VO2 responses.
Oxygen consumption for UT and land treadmill exercise has been extensively studied and is highly dependent on treadmill speed and water depth.
Relative to land treadmills, it is easier to control exercise intensity using underwater treadmills due the control of treadmill speed and water depth.
Speed and depth are two vital variables when considering UT exercise.
For example, Hall et al. (1998) found that when treadmill speeds were 0.97 m/s, VO2 values were similar between aquatic and land conditions in healthy females.
When speeds were 1.25 and 1.53 m/s, however, VO2 values were greater during UT walking compared with land treadmill walking.
Another study by Hall, Grant, Blake, Taylor, and Garbutt (2004) indicated that VO2 was significantly lower in patients with rheumatoid arthritis when speeds were lower than 0.97 m/s.
In contrast, Masumoto et al. (2008) reported greater VO2 values during 0.67 m/s walking in xiphoid deep water on UT versus land.
The UT used in this study employed a water current that matched the speed of walking and likely accounted for the greater VO2 observed during UT.
Water depth may also influence UT VO2 values.
Alkurdi et al. (2010) compared VO2 values in females during land and UT walking at six speeds (0.67–1.78 m/s) and three water depths (xiphoid, 10 cm below, and 10 cm above xiphoid).
Regardless of walking speed, VO2 was significantly greater in the lowest water depth compared with all other conditions, while land VO2 was similar to the 10 cm above xiphoid depth.
These results demonstrate that relatively minor changes in water depth near the xiphoid process influence exercise VO2.
In support of these findings, Pohl and McNaughton (2003) reported the highest VO2 values for UT walking and running occurred during thigh-deep water levels, followed by waist-deep water levels.
Land treadmill walking and running elicited the lowest VO2 values.
At ankle and knee depths, Gleim and Nicholas (1989) reported that the lowest VO2 values occur during land treadmill walking, with greater values at ankle depth and even greater values at the water depth just below the knee.
It would seem that as UT speed increases, water resistance elicits greater VO2 values, and as water depth increases above the pelvis, water buoyancy produces lower VO2 values.
Whether the VO2 response would be lower, higher, or equal to similar land-based running responses may depend on the combination of both treadmill speed and water depth.
One combination that seems to produce similar VO2 values in an arthritic population is to set the water depth to the xiphoid and to set the treadmill speed to approximately 1.04 m/s for water and land modes.
Because HR is a component of cardiac output (i.e., stoke volume × HR) and hence VO2, the trends in HR reported in the literature tended to follow those for VO2 when comparing modes between environments.
In comparison with measuring VO2, however, HR is a more clinically-friendly measure and therefore a description of the HR trends for each mode between environments follows.
Numerous researchers have investigated differences in HR response between DW running and land treadmill exercises.
Due to the large number of studies reporting similar results, it may be concluded with some confidence that DW running elicits lower maximal heart rate (HRmax) values than land running.
For instance, HRmax values during DW running are nearly 15% less than land-based running.
This difference is thought to be independent of age and gender and has even been observed in trained runners.
Two studies were included in this review that reported HR during SW exercise and both reported similar decreases in HR during SW exercise when compared with land running exercise.
These same researchers were also in agreement that SW exercise elicited higher HR values than DW exercise.
The greater HR values observed may in part be due to the presence of a ground reaction force and the higher relative fluid velocity that occurs during SW than DW exercise.
Mixed results have been reported for HR responses during WC compared with land-based exercises. Hoeger et al. (1995) reported significantly lower HR values (≈ 10 beats/min. lower) while Johnson et al. (1977) reported greater HR values, (≈ 15 beats/min. greater) during WC than land-based exercise.
The differences in results are likely due to differing methods.
Participants in the Johnson et al. (1977) study performed the same exercises under both conditions (water and land), whereas the Hoeger et al. (1995) study compared various water exercises to a maximal treadmill running test.
As with oxygen consumption, HR responses during UT exercise depend on the treadmill speed. Hall et al. (1998) found that when treadmill speeds were 1.25 and 1.53 m/s, HR was greater during UT running compared with land treadmill running.
At lower speeds (0.69 m/s and 0.97 m/s), HR was less or equal to the HR values achieved on land.
Accordingly, UT exercise at speeds above 0.97 m/s may result in a HR that is greater than what would be produced on a land treadmill, and any speed below 0.97 m/s may elicit lower HR values.
This may only be true, however, if the water is set at the xiphoid level.
Masumoto et al. (2008) compared walking at 2.4 kph on land with walking in xiphoid-depth water on a UT with a current that matched the walking pace.
HR was greater in UT vs. land; however, this was likely due to the ∼48% greater VO2 due to walking against a current.
When the UT walking pace was adjusted (1.8 kph) to yield a VO2 comparable to land-based walking, HR values were the same.
Rating of Perceived Exertion
Borg’s rating of perceived exertion (RPE) is based on a subjective feeling of exertion and fatigue during exercise and is used to assess and regulate exercise intensity (see Table 1).
The theoretical premise of RPE is that a person rates her/his exercise whole body exertion using a numerical value on a scale from 6 to 20 (or 1–10), representing a verbal expression of effort during exercise.
There have been a variety of studies investigating RPE during various aquatic exercises, including the DW exercise (Table 1).
Results of these studies revealed that during maximal effort, no differences in RPE between DW running and land-based running occur.
Matthews and Airey (2001) measured RPE at a submaximal effort using 60, 70, and 80% of heart rate reserve to which reported RPE scores were 1.4, 2.3, and 2.8 points greater during DW running, relative to land RPE.
To the knowledge of the authors, no peer-reviewed research has compared RPE between SW exercise and land-based exercise.
Two studies examining RPE during WC have produced mixed results (Table 1). Barbosa et al. (2007) investigated RPE at two different water depths and reported that RPE at hip depth was greater than RPE at breast depth and on land.
There was no significant difference between breast depth and land exercise.
Conversely, Hoeger et al. (1995) reported lower RPE levels during WC with participants immersed to the arm pits (axilla) versus land.
These mixed results may be partially accounted for by the differences in the exercises performed and data collection procedures, all of which likely influenced RPE.
With many varieties of WC, it is difficult to compare RPE outcomes for aquatic and land-based calisthenics.
Researchers have reported that gait speed influences RPE during UT exercise.
Rutledge et al. (2007) studied RPE during UT exercise at three speeds (2.9, 3.35, and 3.8 m/s) and three water jet resistance levels (0%, 50%, 75%).
They reported that RPE was greater for land treadmill than UT exercise with 50% and 75% jet resistance.
Hall et al. (2004) reported that at speeds greater than 0.7 m/s, RPE in the legs was greater in water than on land.
Below 0.7 m/s, there was no significant difference.
This contradicts Denning et al. (2010) who reported no difference for RPE for speeds greater than 0.7 m/s. This difference is likely related to how the RPE scale was directed, regionally at legs or globally at the whole body.
Another likely factor is water depth.
When the water level is at the xiphoid level, RPE seems to decrease.
Alkurdi et al. (2010) compared RPE in females during land and UT walking at 6 speeds (0.67–1.78 m/s) and 3 water depths (xiphoid, 10 cm below and 10 cm above xiphoid).
Regardless of walking speed, RPE was significantly greater in the lowest water depth compared with all other conditions, while land RPE was similar to the 10 cm above xiphoid depth.
These results demonstrate that relatively minor changes in water depth can influence a person’s perception of effort.
Masumoto et al. (2008) compared walking at 2.4 kph on land with walking in xiphoid-depth water at the same speed on a UT with a water current resistance that matched the walking pace.
Separate RPE values focusing on breathing and legs were greater in SW vs. land; however, this was likely due to the ∼48% greater VO2 due to walking against the current.
When the UT walking pace was adjusted (1.8 kph) to yield a VO2 comparable to land-based walking, RPE values were the same.
Biomechanical and pain responses
In this section, we discuss studies regarding biomechanical and pain responses conducted using the four aquatic modes, compared with a similar land-based mode.
Stride frequency and length and pain and functional gains in special populations (i.e., rheumatoid arthritis, osteoarthritis, fibromyalgia, and lower back pain) are included (Table 2) as they are frequent biomechanical or pain response dependent variables used in this line of research.
One biomechanical dependent variable is the rate at which strides occur during exercise.
Lower extremity kinematics during DW running are different from kinematics during land running.
Studies examining stride frequency during DW exercise are presented in Table 3.
Stride frequency during DW running is close to half of the stride frequency for running on land examined two different styles of DW running and observed that both styles, a scissors-type task (cross country style) and running-type task (high-knee style), elicited lower stride frequencies.
The cross-country style of DW running, however, was more similar to land running than the high-knee style.
The lack of ground support and water resistance during DW exercise may account for the decreased stride frequency.
Few scientists have investigated stride frequency during SW exercise.
Stride frequency is decreased in SW walking for adults and elderly individuals.
Town and Bradley (1991) compared stride frequency during DW and SW running and reported that stride frequency was 108.2 strides*min-1 during SW running and 83.9 strides*min-1 during DW running.
To the knowledge of the authors, no peer-reviewed research has compared stride frequency between WC exercise and land-based exercise.
There is limited research on the biomechanical characteristics of UT exercise (Table 3).
As with DW exercise, stride frequency is nearly 50% less during UT walking than during land treadmill walking.
Hall et al. (1998) reported a 27 strides*min-1 deficit during UT walking when compared with land treadmill walking in healthy females.
A common finding among many UT studies is lower stride frequencies regardless of the speeds used.
One researcher contended that the main difference in stride frequency for UT exercise occurs during running and not during walking.
The length of a stride typically is measured as the absolute distance from one foot contact (e.g., toe or heel) to the next for the same foot.
To the knowledge of the authors, no peer-reviewed research has compared stride length between DW exercise and land-based exercise.
This finding may not be surprising given the lack of foot contact occurring during DW exercise.
Only two studies have compared stride length differences between SW exercise and land-based exercise, and these studies are somewhat contradictory (Table 3).
Barela and Duarte (2008) indicated lower stride lengths occur during SW walking with elderly individuals (approximately 70 years of age).
Barela et al. (2006), however, reported no difference in stride length in healthy adults (i.e., approximately 29 years of age).
This may indicate age affects stride length during SW and land-based exercise.
To the knowledge of the authors, no peer-reviewed research has compared stride length between WC exercise and land-based exercise.
Researchers who have studied stride or step length during UT exercise reported longer strides or steps, compared with walking on land at the same speed.
These results are probably due to buoyant forces that cause participants to “float” for an extended period of time, similar to the gait of astronauts walking in a microgravity environment.
Due to the lower stride frequencies and the mixed reports regarding stride length, it appears that during SW and UT exercise, the principle of specificity is not met; stride frequency and stride length during aquatic exercise are not similar to land-based exercise.
Even though stride frequency and stride length may be different during aquatic exercise, the therapeutic effect related to functional gains may still be positive. In reviewing studies regarding functional gains, a quantitative mobility measurement (e.g., time up & go test (TUG), 1-mile walk time, 100 m walk time) had to be present for us to include the study within our review.
DW and SW Exercise:
There is a lack of research measuring functional gains after DW and SW exercise.
To our knowledge, no study has compared functional gains that result from DW or SW exercise to functional gains resulting from land based exercises.
Researchers investigating the effects of WC on functional gains can be found in Table 2.
Jentoft et al. (2001) tested functional gains in women with fibromyalgia with a 100 m walk time test and reported no difference in walk time between the aquatic and land-based interventions, although both groups improved.
The improved walking times remained after a 6-month follow up.
Similar results have been reported in subjects with chronic low back pain.
Although WC does not improve functional gains more than land-based exercise, WC does appear to improve functional gains in special populations as effectively as land-based treatments.
This idea is supported by researchers who used different functional gain tests to study the therapeutic effect of WC on various pathological populations.
There is a lack of research measuring functional gains after UT exercise.
Denning et al. (2010), the only study found comparing functional gains after underwater and land treadmill treatment, measured functional gains using TUG scores before and after the aquatic and land interventions in individuals with osteoarthritis.
TUG scores were 240% greater (i.e., time was longer) after land treatment when compared with UT treatment.
This indicates a significant improvement in functional gains after UT walking.
This section presents results of research examining pain responses in pathological populations (i.e., rheumatoid arthritis, osteoarthritis, fibromyalgia, and lower back pain) and does not include healthy subjects.
Only two studies have examined pain during DW exercise.
Both studies used a visual analog scale (graded from 0–10) and reported similar results.
For example, Assis et al. (2006) reported no significant difference in pain levels between aquatic and land-based groups with an average decrease in pain of 36% for both groups.
Melton-Rogers et al. (1996) reported no difference in pain levels between aquatic and land-based groups when measured at peak VO2 and at 60% of peak VO2.
We are not aware of any studies that have investigated pain during or after SW exercise and land-based exercise.
Numerous researchers have investigated the effects of WC exercise on pain for special populations (Table 2).
Most researchers concluded that there is no difference in pain between the aquatic and land-based mode when measured after a training period.
Wyatt et al. (2001) and Evcik et al. (2008), however, did find a significant reduction in pain levels after aquatic treatment.
Evcik et al. (2008) used a 10 cm visual analog scale and reported a 40% decrease in pain after the aquatic treatment and only a 21% decrease after the land-based treatment.
Each paper regarding WC and pain reported improved pain levels after aquatic treatment, indicating WC as an adequate option to reduce pain in special populations.
Clinicians should be aware, however, that this notion may not be fully supported by research, as some studies that did not compare the aquatic mode to a land-based mode found contradicting results.
In addition, some of the studies included different modes of aquatic exercise (i.e., shallow water walking) in their training programs.
Denning et al. (2010), the only study investigating pain during UT treatment for participants with osteoarthritis, reported significant pain reduction after the aquatic intervention (using a 10 cm visual analog scale).
No difference in pain was reported after the land-based intervention.
The purpose of this paper was to provide a descriptive literature review of some acute or chronic biophysical differences between aquatic and land-based exercise.
The following key points may be drawn from our targeted review of the literature:
- Underwater treadmill exercise can elicit lower, equal, or greater values than land exercise. The variability in VO2 and even HR between environments is most likely related to the nonlinear effect of fluid resistance with changes in treadmill speed and the unloading that occurs with changes in water depth.
- Generally, maximal effort DW exercise is capable of eliciting RPE responses that are similar to RPE responses during maximal effort land exercise.
- Stride frequency tends to be lower in all aquatic exercise modes compared with equivalent land exercise.
- Researchers tend to report no difference in pain levels during water calisthenics compared with land exercise in special populations. There is, however, a significant decrease in pain using both mediums.
- Exercising on a UT may improve functional gains (e.g., TUG) more than land treadmill exercise. It should be noted that this observation is based on only one available study suggesting there is a need for future research using this mode of aquatic exercise.
This study was supported by a grant from the National Swimming Pool Foundation.
We have the conviction that Kaenz, will improve your health and well-being, being a great investment, to reduce expenses for medications, hospitalizations and avoid surgeries, we invite you to experience this first virtual aquatic therapist, which will change your habits, recharging energy, toning your body, keeping you young and positively impacting your quality of life, sharing with your loved ones.