Context:

Sunscreen lotions are important to protect the skin during outdoor exercise, but they may interfere with sweating.

Objective:

To measure the effect of 2 water-resistant sunscreens on local sweat production in men and women exercising in the heat and to compare those effects with the expected inhibition resulting from the use of an antiperspirant.

Design:

Randomized crossover study.

Setting:

Exercise in the heat (ambient temperature = 30.2°C ± 0.4°C dry bulb and 58% ± 4.3% relative humidity) in a controlled-environment laboratory.

Patients or Other Participants:

Twenty physically active, apparently healthy college students, 10 men (age = 22.5 ± 2.8 years, height = 1.771 ± 0.069 m, mass = 70.2 ± 11.0 kg) and 10 women (age = 22.2 ± 3.2 years, height = 1.625 ± 0.075 m, mass = 57.7 ± 7.9 kg).

Intervention(s):

With sweat-collection patches applied to their right and left scapular regions, the participants performed 2 exercise sessions on consecutive days. We assigned skin treatments (antiperspirant; organic chemical sun filter, sunscreen A; inorganic physical sun block, sunscreen B; no lotion) randomly to side and session. Participants pedaled at 79% ± 1% of maximum heart rate for 20 minutes in the heat.

Main Outcome Measure(s):

Scapular localized sweat rate.

Results:

No baseline, environmental, or exercise condition was different among skin treatments. Scapular localized sweat rate was lower for the antiperspirant treatment (88.3 μL/min·dm2 ; 95% confidence interval [CI] = 82.0, 94.7 μL/min·dm2) and the inorganic physical sun block (sunscreen B) treatment (99.3 μL/min·dm2 ; 95% CI = 93.1, 105.5 μL/min·dm2) than for the organic chemical sun filter (sunscreen A) treatment (114.8 μL/min·dm2 ; 95% CI = 108.8, 120.6 μL/min·dm2) or the no-lotion treatment (122.6 μL/min·dm2 ; 95% CI = 116.2, 129.0 μL/min·dm2; P < .01).

Conclusions:

The inorganic physical sun block, sunscreen B, hindered effective sweating to the same extent as the antiperspirant, whereas the treatment with the organic chemical sun filter, sunscreen A, was not different from the control treatment. At this stage, it is not possible to identify the specific ingredient responsible for the effect.

Key Points
  • We identified 1 sunscreen formulation that reduced effective sweating.

  • Athletes should use a sunscreen that provides a good balance between effective sweating and skin protection during outdoor exercise in the heat.

When humans exercise, the increase in metabolic heat production should be matched by a greater heat transfer to the environment to maintain core temperature within safe limits and prevent deterioration of performance or the extreme of exertional heat stroke.1,2  However, when protective clothing and equipment are worn during sport training and competition, convective, conductive, and evaporative heat losses are limited.3 

During outdoor physical activity, particularly in hot and humid environments, humans need protection from solar radiation without compromising sweat production and evaporation, a major path for heat dissipation. A few studies4,5  have shown that clothing does indeed limit convection and evaporation. With this in mind, authors69  have investigated sweat-production patterns in males and females, and several manufacturers have claimed to design clothing that facilitates thermoregulation by improving sweat evaporation from the skin. The use of sunscreen lotions, however, has not received much attention in association with a potential limitation of sweat production. These products are widely used and aggressively promoted to protect unclothed skin from ultraviolet rays, but they may interfere with effective sweat production and, if used over large skin areas, they may impair thermoregulation.

The objective of our study was to measure the possible effects of 2 water-resistant sunscreen products on local sweat production in men and women exercising in the heat and to compare these effects with the expected inhibition resulting from the use of an antiperspirant (AP). We hypothesized that the latter would significantly reduce sweat rates when compared with a no-lotion condition, whereas the 2 sunscreens would have either significant but smaller effects or nonsignificant, undetectable effects.

METHODS

After undergoing basic screening (Physical Activity Readiness Questionnaire10  and questioning about known skin allergies), 20 physically active, apparently healthy college students volunteered for the study: 10 men (age = 22.5 ± 2.8 years, height = 1.771 ± 0.069 m, mass = 70.2 ± 11.0 kg) and 10 women (age = 22.2 ± 3.2 years, height = 1.625 ± 0.075 m, mass = 57.7 ± 7.9 kg). We obtained written informed consent from all participants before the study, and the University of Costa Rica Science and Ethics Committee approved all procedures in advance.

Procedures

Using a repeated-measures design (randomized crossover design), we tested each participant under all 4 skin treatments: sunscreen A (SA; [organic chemical sun filter] Banana Boat Sport Performance Sun Filter 50; Edgewell Personal Care Company, St Louis, MO), sunscreen B (SB; [inorganic physical sun block] Nivea Sun Solar Protection 50; Beiersdorf, Hamburg, Germany), AP (Rexona V8 Tuning Roll-On; Unilever, London, UK), or control (C, no treatment; Table 1). We used both the left and right subscapular regions, allowing 2 treatments to be tested on each participant in each session. We assigned treatments in random order to the left or right subscapular region and the first or second day of testing.

Table 1. 

Main Skin Lotion Characteristics (as Reported by the Manufacturers)

Main Skin Lotion Characteristics (as Reported by the Manufacturers)
Main Skin Lotion Characteristics (as Reported by the Manufacturers)

Testing procedures have been previously described11  but are detailed here in English. We instructed all participants to avoid diuretics 24 hours before the tests and to drink at least 1 L of water the night before testing. To allow us free access to the scapular region, women dressed in a comfortable top covering the upper body and men were shirtless.

Each participant reported to the laboratory on 2 consecutive days, at the same time of day. Upon arrival, participants voided their urine, and we measured urine specific gravity (USG) using a refractometer (model URC/Ne; ATAGO Co Ltd, Tokyo, Japan). We obtained their nude and dry body mass to the nearest 20 g on a scale (model e-Accüra SB710-120; Bright Advance Corporation, Shanghai, China). Both subscapular regions were wiped clean with gauze and distilled water and marked with a waterproof eyeliner (model ColorStyle; Revlon, Inc, New York, NY) according to the International Society for the Advancement of Kinanthropometry international standards for anthropometric assessment.12 

We applied the assigned lotion to the marked area (0.162 g or 10 mg/cm2  of skin) with the help of a stencil and left it to dry for 60 minutes while the participant rested in a climate-controlled chamber. Then we measured body temperature in triplicate using a tympanic thermometer (model ThermoScan Pro 4000; Braun GmbH, Kronberg, Germany). Excess sunscreen or AP was removed with dry gauze before a sweat patch (model MSX-6446; 3M, Brookings, SD) was placed over each marked region.

Participants exercised on stationary bicycles for 20 minutes in the chamber at a dry-bulb temperature of 30.2°C ± 0.4°C and relative humidity of 58% ± 4.3%. Exercise intensity was maintained between 78% and 80% of maximum heart rate, which we estimated according to the formula of Gellis et al13 : maximum heart rate = (207 − [0.7 × age]). On completion of the exercise period, we removed the sweat patches, weighed them, and recorded sweat-collection time and obtained nude and dry body mass to the nearest 20 g for each participant.

We calculated local sweat rate (LSR) according to the method used by Maughan et al,14  Smith and Havenith,15  and Morris et al16  and refined in our laboratory.11  We placed each new patch in a labeled, hermetically sealed plastic bag and weighed it on a precision scale (model GX-200; A&D Company, Limited, Tokyo, Japan) to the nearest 1 mg. Then we removed the patch from the bag, peeled the backing off, and applied it to the skin. Once the test was completed, we placed the patch in the same bag with all its parts and sealed it until it was weighed. Weight gain corresponded to sweat absorbed during testing; 1 g was considered to be equivalent to 1 mL. We divided sweat volume by testing time (recorded to the nearest second) and by the area covered by the patch (0.1617 dm2), to obtain μL/min·dm2 .

We rescheduled testing for any participant who arrived hypohydrated (as assessed by USG with a cutoff point of ≥1.020)17  and any female participant in transition between 2 menstrual-cycle phases; the latter was to avoid extraneous body temperature differences.18,19 

Statistical Analysis

To characterize the sample, we calculated descriptive statistics for age, height, baseline body mass, and total body sweat rate. We performed a mixed, 2-way sex-by-treatment (2 × 4) analysis of covariance with LSR as the dependent variable and working heart rate, USG, pre-exercise body temperature, baseline body mass, change in body mass, room temperature, and relative humidity as covariates. In addition, we performed two 1-way, repeated-measures analyses of variance (left side versus right side and session 1 versus session 2) with LSR as the dependent variable. For a power of .90, we calculated that a sample of 10 men could identify differences as small as 16.4 μL/min·dm2  and a sample of 10 women could identify differences as small as 1.73 μL/min·dm2 : that is, one-half and one-tenth, respectively, of the difference in average sweat rates for men and women (32.8 μL/min·dm2  and 17.3 μL/min·dm2 , respectively) found when participants used AP in a previous study.11 

RESULTS

Descriptive data are shown as mean ± SD. Men and women were the same age (22.5 ± 2.8 and 22.2 ± 3.2 years, respectively), but men were taller (height = 1.771 ± 0.069 m versus 1.625 ± 0.075 m, repectively; P = .001) and heavier (mass = 70.2 ± 11.0 kg versus 57.7 ± 7.9 kg, respectively; P = .001) and had a greater total body sweat rate estimated from body mass loss (1454 ± 759 mL/h versus 876 ± 438 mL/h, respectively; P = .001).

As shown in Table 2, no baseline, environmental, or exercise condition was different among skin treatments (P > .05).

Table 2. 

Reference Values

Reference Values
Reference Values

Local sweat rates are shown in Figure 1. The mixed 2-way analysis of covariance demonstrated no significant interaction between sex and treatment (P = .439). A main effect was noted for sex, with a greater mean sweat rate in men (115.6 μL/min·dm2 ; 95% confidence interval [CI] = 109.6, 121.5 μL/min·dm2) than in women (97.0 μL/min·dm2 ; 95% CI = 91.0, 102.9 μL/min·dm2; P = .0007).

Figure 1. 

Scapular localized sweat rate. Columns represent mean values; error bars, 95% confidence intervals. a The sunscreen A and control treatments were not different (P = .48). b The antiperspirant and sunscreen B treatments were not different (P = .10).

Figure 1. 

Scapular localized sweat rate. Columns represent mean values; error bars, 95% confidence intervals. a The sunscreen A and control treatments were not different (P = .48). b The antiperspirant and sunscreen B treatments were not different (P = .10).

There was also a main effect of skin treatment (P = .001). Post hoc analysis indicated that LSR was lower for the SB (99.3 μL/min·dm2 ; 95% CI = 93.1, 105.5 μL/min·dm2) and AP (88.3 μL/min·dm2 ; 95% CI = 82.0, 94.7 μL/min·dm2) treatments than for the SA (114.8 μL/min·dm2 ; 95% CI = 108.8, 120.6 μL/min·dm2) or C (122.6 μL/min·dm2 ; 95% CI = 116.2, 129.0 μL/min·dm2; P < .01) treatments. No differences were found between the SB and AP (P = .10) or between the SA and C (P = .48) treatments.

The LSR did not differ between the left and right sides (P = .654; Figure 2A). Mean sweat rates were 103.4 μL/min·dm2  (95% CI = 85.4, 121.4 μL/min·dm2) for the left and 109.1 μL/min·dm2  (95% CI = 91.1, 127.1 μL/min·dm2) for the right. Finally, LSR was not different between sessions (P = .397), with 104.5 μL/min·dm2  (95% CI = 91.5, 117.5 μL/min·dm2) for session 1 and 108.2 μL/min·dm2  (95% CI = 94.6, 121.8 μL/min·dm2) for session 2 (Figure 2B).

Figure 2. 

Scapular localized sweat rate by A, side, and B, session. Columns represent mean values; error bars, 95% confidence intervals. Mean differences were not significant between sides (P = .654) or sessions (P = .397).

Figure 2. 

Scapular localized sweat rate by A, side, and B, session. Columns represent mean values; error bars, 95% confidence intervals. Mean differences were not significant between sides (P = .654) or sessions (P = .397).

DISCUSSION

Our main result was that the application of SB to the skin hindered sweat production to the same extent as an AP. This was not the case for SA, which had no measurable effect on LSR when compared with the C treatment. The effect of the lotions was not differentiated by sex.

The SB reduced LSR by 17.2% compared with C and 12.5% compared with SA. At this point, it is not possible to explain the major differences between the sunscreens. We tested 2 commercially available products with the same SPF but many different specific ingredients. The major active ingredients were different: oxybenzone for SA and titanium dioxide for SB. Also, we can speculate that phenylbenzimidazole sodium sulfonate, a secondary ingredient consisting of very small particles and used in combination with titanium dioxide in SB, may penetrate the skin and physically interfere with sweating; there may be a chemical inhibition of sweat gland activity as well. Future researchers should identify those ingredients that interfere with effective sweating.

In previous studies20,21  of sunscreen, no differences in sweat rate, core temperature, or exercise performance have been observed. Connolly and Wilcox20  evaluated the physiological responses of 22 men who exercised at 50% to 60% Vॱo2max in the heat (temperature = 32°C, relative humidity = 54%) for 45 minutes with or without sunscreen (30 mL sunscreen/m2 of bare skin). No differences were found for rectal temperature, blood lactate, plasma volume changes, oxygen consumption, heart rate, perceived exertion, or total body sweat loss during exercise. However, because they provided no details about the brand or type of sunscreen tested, it is possible that they used a product similar to our SA, which had no effect on LSR. House and Breed21  compared thermoregulation during 60 minutes of stepping exercise in dry heat (temperature = 40°C, relative humidity = 20%) for 8 males and 4 females under 3 conditions: after the application of 20 g/m2 of alcohol-based sunscreen, oil-based sunscreen, or no sunscreen (C). No differences were noted in sweat output (P = .939), heart rate, core temperature, or skin temperature (all P values >.05). However, the initial conditions (ie, body temperature, hydration status) and exercise intensity were not carefully controlled in the study, limiting the strength of the conclusions. No ingredients in the sunscreens used were reported, so it is also possible that they were similar to SA in our study. The physiological relevance of the LSR reductions we observed with SB remains to be tested; thermoregulatory effects will likely depend on the ratio of the clothed skin area to the bare skin area treated with sunscreen.

The AP had the expected effect on LSR, which was 25.5% less than for the C condition, consistent with a previously reported 24% reduction (90.3 versus 118.7 μL/min·dm2  for AP and C, respectively; P = .001).11  This reduction in sweat output is far from the 50% claimed by some manufacturers. Nevertheless, we consider it high enough to advise physically active people and athletes against the use of AP over wide skin areas when effective sweating is important for thermoregulation.

In summary, this study showed a potential conflict between thermoregulation (the need for effective sweat production to promote evaporation) and the need for skin protection during outdoor exercise in the heat.22  Future authors should aim to identify the specific ingredients in sunscreens that interfere with effective sweating and their physiological mechanisms, as well as the thermoregulatory implications. Meanwhile, we advise athletes and physically active people to limit the use of AP and SB if they are concerned about reducing sweat rates during exercise in the heat.

ACKNOWLEDGMENTS

This study was supported by the University of Costa Rica, project VI-838-B0-315; the authors' work was independent of the funders. Our sincere thanks to Dr Gilbert Brenes for his support with statistical design and analysis.

REFERENCES

REFERENCES
1
Hadad
E,
Rav-Acha
M,
Heled
Y,
Epstein
Y,
Moran
DS.
Heat stroke: a review of cooling methods
.
Sports Med
.
2004
;
34
(
8
):
501
511
.
2
Hosokawa
Y,
Adams
WM,
Stearns
RL,
Casa
DJ.
Heat stroke in physical activity and sports
.
Pensar Mov
.
2014
;
12
(
2
):
1
21
.
3
McCullough
EA,
Kenney
WL.
Thermal insulation and evaporative resistance of football uniforms
.
Med Sci Sports Exerc
.
2003
;
35
(
5
):
832
837
.
4
Zhang
XH,
Li
J,
Wang
YY.
Effects of clothing ventilation openings on thermoregulatory responses during exercise
.
Indian J Fibre Text Res
.
2012
;
37
(
2
):
162
171
.
5
Davis
JK,
Bishop
PA.
Impact of clothing on exercise in the heat
.
Sports Med
.
2013
;
43
(
8
):
695
706
.
6
Havenith
G.
Interaction of clothing and thermoregulation
.
Exog Dermatol
.
2002
;
1
(
5
):
221
230
.
7
Gavin
TP.
Clothing and thermoregulation during exercise
.
Sports Med
.
2003
;
33
(
13
):
941
947
.
8
Ueda
H,
Inoue
Y,
Matsudaira
M,
Araki
T,
Havenith
G.
Regional microclimate humidity of clothing during light work as a result of the interaction between local sweat production and ventilation
.
Int J Cloth Sci Technol
.
2006
;
18
(
4
):
225
234
.
9
Kwon
J,
Choi
J.
Clothing insulation and temperature, layer and mass of clothing under comfortable environmental conditions
.
J Physiol Anthropol
.
2013
;
32
(
11
):
1
8
.
10
Adams
R.
Revised Physical Activity Readiness Questionnaire
.
Can Fam Physician
.
1999
;
45(992.995):1004–1005.
11
Aburto-Corona
JA,
Aragón-Vargas
LF.
Capacidad del método de parches de sudoración para detectar el efecto de un antitranspirante [Capacity of a sweat patch method to detect the effect of an antiperspirant]
.
Pensar Mov
.
2015
;
13
(
1
):
1
10
.
12
Marfell-Jones
M,
Olds
T,
Stewart
A,
Carter
L.
International standards for anthropometric assessment
.
The International Society for the Advancement of Kinanthropometry Web site
. ,
2015
.
13
Gellis
RL,
Goslin
BR,
Olson
RE,
McDonald
A,
Russi
GD,
Moudgil
VK.
Longitudinal modeling on the relationship between age and maximal heart rate
.
Med Sci Sports Exerc
.
2007
;
39
(
5
):
822
829
.
14
Maughan
RJ,
Merson
SJ,
Broad
NP,
Shirreffs
SM.
Fluid and electrolyte intake and loss in elite soccer players during training
.
Int J Sport Nutr Exerc Metab
.
2004
;
14
(
3
):
333
346
.
15
Smith
CJ,
Havenith
G.
Body mapping of sweating patterns in male athletes in mild exercise-induced hyperthermia
.
Eur J Appl Physiol
.
2011
;
111
(
7
):
1391
1404
.
16
Morris
NB,
Cramer
MN,
Hodder
SG,
Havenith
G,
Jay
O.
A comparison between the technical absorbent and ventilated capsule methods for measuring local sweat rate
.
J Appl Physiol
.
2013
;
114
(
6
):
816
823
.
17
Armstrong
LE,
Pumerantz
AC,
Fiala
KA,
et al.
Human hydration indices: acute and longitudinal reference values
.
Int J Sport Nutr Exerc Metab
.
2010
;
20
(
2
):
145
153
.
18
Moran
VH,
Leathard
HL,
Coley
J.
Cardiovascular functioning during the menstrual cycle
.
Clin Physiol
.
2000
;
20
(
6
):
496
504
.
19
Lacerda
MG,
García
AM,
Cruz
C,
Calcagno
JC,
Reis
FM,
Silami-Garcia
E.
Effects of menstrual cycle on sweating during exercise performed in hot and dry environment
.
Biochem Physiol
.
2013
;
2
:
1
5
.
20
Connolly
DA,
Wilcox
AR.
The effects of an application of suncream on selected physiological variables during exercise in the heat
.
J Sports Med Phys Fitness
.
2000
;
40
(
1
):
35
40
.
21
House
JR,
Breed
M.
Sunscreen use reduces sweat evaporation but not production
.
Presented at: 15th International Conference on Environmental Ergonomics; February 11–15
,
2013
;
Queenstown, New Zealand.
22
Adams
BB.
Sports Dermatology
.
New York, NY
:
Springer;
2006
:
106
123
.