VOLUME 5 • NUMBER 2 □ OCTOBER 1998 The South African Journal of Sports Medicine The Official Publication of the South African Sport Medicine Association R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) very one f e e l s pain at some time . . . Fobemoo • S i g n i f i c a n t a n a l g e s i c , a n t i - i n f l a m m a t o r y and a n t i p y r e t i c action. • R a p i d t h e r a p e u t i c r e s p o n s e e n s u r e d 1. • A c t i o n P a c k c o n t a i n i n g 15 tab lets. • P a c k o f 12 s u p p o s i t o r i e s . HE Flurbiprofen 100 mg. knollTablets 100 mg. suppositories 100 mg. BASF P h a r m a R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) EDITORIAL The content of the second issue of the Journal in 1998 is representat ive of the diversity of scientific interests among members of the South African Sports Medicine Association (SASMA). I believe that the strength of the Association lies in this diversity, and 1 remain convinced that Sports Medicine is best practiced and researched in a multi-disci­ plinary setting. I'll is issue has a mmiber of unique contributions. Firstly, Dr Myburgh contributed the invited Review Art icle for this Edition. The importance of achieving an optimum peak bone mass must be emphasized. In particular, the positive role of physical activity to achieve peak bone mass must be eucoiu-aged in children and adolescents. As pointed out by Dr Myburgh, it is important to remember that the voimg female athlete with menstrual dysfunction can develop a low bone density, and clinicians must be aware of this potential danger. The Editorial Board of this Journal will con­ tinue to publish high quality Invited Review Articles of similar caliber to the one written by Dr Myburgh. Secondly, the original research articles in this edition are once again of a high standard, and represent different areas of Sport Mcdienc. Dr Du Toil and his colleagues designed a unique study to examine the force absorption and rebound characteristics of cricket batting pads. Their results are likely to influence the design and choice of pads used by play­ ers. Ms Fuller and her colleague investigated the potential use of an infrared auditory canal thermometer to measure the core temperature in athletes - a clinical measurement that is of extreme importance in diagnosing heat illness. They conclude that infrared auditory canal thermometry is too variable and does not provide the clinician with a “ quick and easy” measure of core tempcratiu e. Thirdly, we are happy to introduce a new form of presenting research data in die form of a brief report. This format allows researchers to publish data that is of interest, but where the content, depth or limitations preclude it from being published as a full research article. Dr Marino reports on the urinary catecholamine excretion during outdoor sports rock climbing. He concludes that urinary catecholamine excretion does not increase after a strenuous climb of 7-8 minutes. Finally, we are proud to publish the Position Statement 011 “ Ethics in Sports Medicinc” , which has been reproduced, with permission, from the International Sports Medicinc Federation (FIMS). We will, in future, continue to publish Position Statements, as these are a very useful yardstick by which we can evaluate our approach to problems (clini­ cal and other) in Sports Medicine. 1 would like to lake this opportunity to thank the International Sports Medicine Federation (FIMS) for allowing us to publish their material. Now it just remains for me to wish you all a happy festive season, a restful holiday, and great sporting and personal success in 1999. Prof MP Schwellnus, MBBCh, MSc (Med), MD, FACSM EDITOR South African Journal of Sports Medicine SPORTS MEDICINE OCTOBER 1998 1 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) THE SOUTH AFRICAN JOURNAL OF SPORTS MEDICINE VOLUME 5 NUMBER 2 OCTOBER 1998 Editor - in - Chief Prof MP Schwellnus University of Cape Town Senior Associate Editors Prof M Mars University of Natal Dr M Lambert University of Cape Town News Editor Dr P Mac Farlane General Practitioner, Cape Town Editorial Board ProfY Coopoo University of Durban Westville Dr K Myburgh University of Stellenbosch Prof TD Noakes University of Cape Town Prof G Rogers University of the Witwatersrand Prof K Vaughan University of Cape Town The Editor The South African Journal of Sports Medicine PO Box 115, Newlands 7725 Tel: (021) 686-7330 Fax: (021) 686-7530 CONTENTS Editorial 1 MP Schwellnus Review Article: 3 Exercise and peak bone mass: An update K H Myburgh Research Article: 9 The force absorption and rebound characteristics of cricket batting pads at four impact velocities R Stretch, E du Toit, T Edwards, B Pretorius Research Article: 14 Comparison of oral and infrared auditory canal thermometry in sports participants A Fuller, D Mitchell Brief Report: 19 Urinary catecholamine excretion during outdoor sport rockclimbing F Marino, J Booth Position Statement: 22 Code of Ethics in Sports Medicine International Sports Medicine Federation (FIMS) SASMA News 24 PRODUCTION ADVERTISING Andrew Thomas Andrew Thomas REPRODUCTION Output Reproduction PUBLISHER Glenbarr Publishers cc PRINTING Tel: (O il) 442-9759 INCE The views expressed in individual articles are the personal views of the Authors and are not necessarily shared by the Editors, the Advertisers or die Publishers. No articles may be reproduced without the written consent of die Publishers. 2 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) REVIEW a r t ic l e Exercise and peak bone mass: An update KH Myburgh (PhD) Department o f Human and Animal Physiology, University o f Stellenbosch, South Africa INTRODUCTION Exercise has been widely recommended as a means of preventing osteoporosis. The rationale for this recom­ mendation was based largely on the inference that immobilisation and weightlessness lead to bone loss and therefore exercise should lead to bone gain. Since peak bone mass is a major determinant of bone mass and fracture risk later in life, any factors that could enhance peak bone mass can be considered to be ben­ eficial in preventing osteoporosis in the long-term. Exercise is a lifestyle factor that can be modified and if it does indeed increase peak bone mass, it may offer protection against bone fragility in old age. New evi­ dence suggests that exercise (luring growth can lead to large increases in bone mass.2,1” If these increases are maintained into early adulthood, they wall result in increased peak bone mass. In contrast, during adult­ hood exercise results in relatively small gains in bone mass and appears to play a more important role in maintenance of bone density and possibly prevention of early bone loss. Exercise may not always be beneficial however, since athletic primary and secondary' amen­ orrhea may lead to either a failure to gain bone or bone loss. Current nomenclature: In vitro measurement of bone breaking strength has determined that bone den­ sity can account for up to 80 percent of the variance/" Therefore bone density is used as a surrogate measure of the breaking strength of bone. In vivo measurement of the mineral content of the skeleton can be done using dual energy X-ray absorptiometry (DXA) or quantitative computed tomography (QCT). Both deter­ mine an estimate o f bone mineral content, which can be expressed as either a mass or a density. Bone mass is the amount of mineral in grams (g) unadjusted for size. Bone mass divided by the area of the region analysed (g /cn r) is termed areal bone density. Although areal bone density only adjusts for the width and length of a bone (not depth), it is the unit most commonly reported in the literature. Volumetric bone density is the bone mass divided by the volume of the CORRESPONDENCE: Dr KH Myburgh Dept Human & Animal Physiology, University of Stellenbosch, Private Bag X I Matieland 7602 Tel: 27 - 21 - 808 3149 Fax: 2 7 - 2 1 - 808 3145 E-mail: khm@maties.stm.ac.za region measured (g/cnr1) and can be calculated from DXA scans by using equations that estimate the size of the vertebra and femoral neck.5 This estimate is termed bone mineral apparent density (BMAD). The term ‘density’ without a clarifying prefix can be misleading because it does not distinguish between areal, volumetric or apparent bone density. Analysis of samples of bone has shown that true bone density does not increase with age or size. In other words, the chemical composition of the skeleton remains con­ stant during growth and adulthood.49 “ It is important to acknowledge that an increase or decrease in areal or volumetric bone density is not a change in the chemi­ cal composition o f bone but rather a change in total bone length, bone size, cortical thickness, medullary width and/or the biomechanical organisation of the trabecular structure.40 When is peak bone mass achieved? Peak bone mass is the term used to describe the max­ imal lifetime amount o f bone tissue aquired in individ­ ual bones and the whole skeleton. It is the conse­ quence o f the nett accrual of bone mass due to growth during the childhood years and the balance between accrual and resorption rates in the adult pre­ menopausal period. Approximately 80 to 85 percent of peak bone mass has been accrued by the age of menar- che. About half of this is achieved during pre-pubertal growth (~10 to 12 years) and the other half is achieved very rapidly in the 2 to 4 years of pubertal growth.14 Bone mass accrual continues slowly after puberty and contributes a further 15 to 20 percent to the peak bone mass. While it is well accepted that bone mass con­ tinues to be accrued after linear growt h has ceased (at about 16 years in females) there is still debate about when peak bone mass is achieved. There have been reports that peak bone mass may be achieved from as early as 17 to 18 years to as late as 35 years.17 *’-54 Theintz et al. (1992) showed no increase in bone density at the femoral neck or the lumbar spine after 17 to 19 years of age.54 In contrast Mazess and Barden (1991) report that peak bone mass may be achieved later in adulthood. They reported small differences in bone density between women aged 20-24 and women aged 30-34: lumbar spine was 3% higher in the older cohort, but no differences were noted in the femur.33 There are several reasons for these conflicting results: a) measurement of bone mass at one site may not be representative of the entire skeleton or other sites within the skeleton; b) expression o f results as bone mass may differ from expression of results as bone density due to changes in bone shape and c) the results of cross-sectional studies can be influenced by SPORTS MEDICINE OCTOBER 1998 3 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) mailto:khm@maties.stm.ac.za the different cohorts.*’ The results of prospective studies support the notion that peak bone mass may be achieved later rather than earlier in adulthood. Bennell et al. (1997) studied subjects aged 17 to 26 years for one year.3 This study showed increases in t he range of 2.2% in total body bone mass in the female subjects. In a longer study 156 women were followed for up to 5 years and peak bone mass was reported to be achieved between 28 and 30 years of age.4'1 There is also debate about how long peak bone mass is maintained before bone loss begins. There may be a plateau in bone mass or bone loss may occur short­ ly after peak bone density has been achieved. Two studies support the concept of early bone loss: 1) Matkovic et a], (1994) reported that the bone density o f the proximal femur was about 15 percent lower in older premenopausal women compared to younger pre­ menopausal women,32 and 2) Bonjour et al. (1991) reported that the spines of adolescents (aged 14 to 19 years; n =24) had 10% higher trabecular bone density than young adults (25 to 35 years; n = 24).4 Not all studies report a statistically significant decline in bone density with age during the pre-menopausal period.133'’ The maintenance of bone mass in the lumbar spine and femoral neck has been reported in longitudinal studies of women in their thirties.33 35 Individual variation may explain these conflicting results. This individual vari­ ation in the rate of change and direction of change in bone mass was highlighted by Mazess and Barden (1991) who reported that there was no detectable rela­ tionship between age and bone mass in a large group of subjects aged 20-39 years. However, 63 subjects decreased spine bone density by7 more than 2%, where­ as 100 subjects showed change of less than 2% in either the positive or negative direction and a further 68 subjects gained more than 2%.33 In summary, there is clear evidence that the major­ ity of bone mass is accrued during pre- and peri-puber- tal years and that bone mass continues to be accrued at a slower rate in the post-pubertal years. There is still controversy7 about when peak bone mass is achieved and how long it is maintained, but the timing o f the attainment of peak bone mass and bone loss is most likely7 a) site-specific and b) variable between individuals. This heterogeneity between individuals in ay be the result of the interaction between genetic and environmental determinants that influence bone mass. The genetic determinants o f peak bone mass The variance in peak lumbar spine bone density varies between -20% and +20% of the mean. Family studies have shown that 60 to 90% of this variation between individuals is genetically7 determined/10 It has also been found that the genetic effect is greater at the lum­ bar spine (up to 90%) than the femoral neck (up to 70%).S1 A finding that is clinically7 important is that pre­ menopausal women with a maternal family history7 of osteoporosis have low bone density,1 and peri- menopausal women with a family history7 of hip frac­ ture also have low bone density.™ Interestingly, daugh­ ters of mothers with hip fractures had low bone densi­ ty at the hip, while daughters of mothers with spine fractures had low bone density at the spine50 51 indicat­ ing that this familial association may also be site spe­ cific. While a large proportion of the variance in bone density may be genetically7 determined, environmental factors also account for a clinically important propor­ tion of the variance.4" These influences may not be independent as it has been suggested that there is a common genetic control of both muscle mass and bone mass, both of which can also be affected by exercise. "1 The osteotrophic effect o f exercise on bone density in adult premenopausal women Early7 studies of exercising subjects seemed to indicate that exercise could substantially improve bone mass. For example, the dominant arm in tennis players has higher bone mass than their non-dominant arm.-1 The reported differences between the two arms in compet­ itive club players were between 8% and 13%, and even larger differences were reported in the professional playrers. However, cross-sectional data of mature ath­ letes does not give an indication of whether the adap­ tation occurred in childhood or adulthood. Therefore, to investigate the effects of exercise on adult bone mass, a longitudinal approach is preferable. The effect o f aerobic exercise on bone density in the adult has been reviewed, and the following consensus has been reached: early cross-sectional studies were confounded by sampling bias and led to the belief that exercise in the adult could lead to large increases in bone density (up to 20%). Intervention studies tended to show mixed resul ts ranging from small increases (2 to 3%), to no change or even bone loss.15 Weight training was hypothesised to have a greater osteotrophic effect than jogging or walking since weight training places a greater load on the skeleton to which it must adapt. Intervention studies in pre­ menopausal women ranging from 4 to 24 months have shown inconsistent findings.16 1H 29 47 53 Some o f these studies reported small but statistically significant changes of 1 to 2 percent; others showed no change in bone mass or bone loss. Friedlander et al. (1995) reported that the change in lumbar spine bone density alter 2 years of aerobics and weight training in adult women (aged 28 years),10 was similar to that found by Snow-Harter et al. (1992) after 8 months (1.3% for both studies),83 indicating the possibility of a plateau hi the adaptation to exercise. In the study the femoral neck bone density increased even less (0.5 ± 0.5%) with weight training but this was in contrast to a group of women who only7 did stretching exercises and who lost 1.9 ± 1.0% (p<0.05). Although weight training places a high load on the skeleton, it is possible that the impact is not high since there is no momentum at the initial contact. The osteotrophic effect on the skeleton of high and low impact loading has also been investigated, albeit in postmenopausal women. Ken- et al. (1996) randomised 56 subjects to a high or low impact exercise group.27 4 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Both groups exercised three times per week for 12 months. Increases were found in the high impact group at the trochanter, Wards triangle and ultra dis­ tal radius o f between 1.7 ± 4.1% and 2.4 ± 4.3%). There was no increase at the femoral neck or other radial sites. There were no changes at any site (relative to baseline) in the low impact high repetition group and a control group lost bone density at most sites (changes between 0.8 ± 5.2 % and -1.4 ± 2.3 %). Sports where impact loading is high, may also lead to increased bone density. Higher bone density has been reported in athletes who play volleyball and bas­ ketball,45 and in college gymnasts.1'’ The extent of the differences can be substantial (see Table 1). It should be remembered that firstly the results could be affect­ ed by selection bias, and secondly it can not be deter­ mined from these studies what proportion of these benefits were gained from exercise during childhood and adolescence, and what proportion was gained from exercise in adulthood. The clinical relevance of exercise in the prevention o f osteoporosis lies in the reduction of fracture risk late in life. It is generally accepted that a 10% increase in bone density' is associated with halving the risk of femoral neck fracture." Moderate exercise of any type in adulthood is unlikely to result in such large increas­ es in bone mass. But a long term commitment to an active lifestyle, including exercise that places a mechanical load on the skeleton, through the pre-, peri-, and postmenopausal years may help reduce frac­ ture risk by reducing bone loss and ind irectlv by reduc­ ing the risk of falling by improving muscular strength, coordination and balance. The effect o f menstrual irregularity on hone density Research in the mid-eighties and subsequent research has confirmed the long term risk of osteopoenia asso­ ciated with athletic amenorrhea. " :,1M The decrements in bone density are seen relative to both sedentary' peers, and eumenorrheic athletic controls. The skele­ tal site most affected by menstrual dysfunction is the lumbar spine, a site containing a large proportion of trabecular bone."131 The deficit in bone density relative to eumenorrheic athletes has been reported to be as high as twenty percent.31 Up to fotu percent of trabecu­ lar bone can be lost in the first year of secondary' amen­ orrhea and bone loss continues for at least two years.38 More recent studies have reported that oligomenor­ rhea is also associated with low' bone density,28 34 despite no episodes of amenorrhea.34 There appears to be a relationship between bone density and the sever­ ity of current and previous menstrual history expressed either by category','1 or the number of cycles per year since age thirteen years.114 It has also been sug­ gested that more subtle hormonal disturbances may' affect bone mass, even w'hen menses is regular. Women with more than one short luteal phase per y'ear or anovulatory cycles, may lose bone mass due to decreased progesterone secretion.4' But De Souza et al. (1997) follow'ed exercising w'omen for three months and reported that luteal phase insufficiency w'as asso­ ciated with decreased progesterone, but not decreased bone density, wliereas reduced estrogen production in the follicular phase, despite regular menses, w'as asso­ ciated with low'er bone density.8 Can exercise offset hone loss associated with menstrual dysfunction ? Despite low'er bone density at Lite spine, amenorrheic athletes have either high or normal bone density at w'eight bearing sites. One explanation for this trend is that exercise may' offset the negative effects of amen­ orrhea at the w'eight bearing sites. Tw'o factors argue against this explanation: 1) bone density o f the w'eight bearing sites may be higher than the non-w'eight bear­ ing sites because of previous loading before exposure to estrogen deficiency' and 2) trabecular sites are more at risk of early' bone loss and cortical sites later bone loss, so that only' trabecular bone loss is evident at the stud}' time. Pearce et al. (1997) have reported low'er bone density at the w'eight bearing sites only' in ballet dancers with longer time periods of oligomenorrhea.41 Exercise with higher impact loading than ballet may offer protection at the w'eight bearing sites. In a group of figure skaters with a substantial history' of menstrual irregularity bone density w'as enhanced in the low'er limbs, but not in the lumbar spine.52 Robinson et al. (1995) reported that gymnasts had higher bone density at all sites compared with runners w'ho had a similar incidence of menstrual dy'sfunc- TABLE 1: Lumbar spine and lower limb bone density in gy mnasts, volley ball and basketball play ers compared to non-athletes. Group LS BMD Diff vs control FN BMD Diff vs Control Reference g/cnr % g/cm 2 % Controls 0.11 ± 0.11 0.97 ± 0.10 no 46 Gymnasts 0.17 ± 0.13 + 5.4 % 1.09 ± 0.12 + 12.4 % no 46 Calcaneus BMD Controls 1.15 ± 0.03 0.424 ± 0.019 no 45 Volley'ball 1.32 ± 0.04 + 14.8 % 0.536 ± 0.017 + 26.4 % no 45 Basketball 1.29 ± 0.03 + 12.1 % 0.575 ± 0.022 + 35.6 % no 45 Abbreviations: LS BM D = Lumbar spine bone mineral density; F N B M D = femoral neck hone mineral densityj; D iff = difference SPORTS MEDICINE OCTOBER 1998 5 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) tion.4 son test for the force absorption and rebound distance of the cricket batting pads Force Absorption Rebound Distance PI P2 P3 PI P2 P3 Speed 1 P3 -10.81* -4.60 P3 -18.95* -11.25* 25.49* 53-84 24.49* -15.18 -12.74 - 5.04 52.84 13.18 P4 -19.66* -11.96 -3.812 54.16* 14.49* 15.49* -13.45 - 5.75 2.40 82.51 42.84 43.84 Speed 2 P3 -8.89* -3.00 P3 -21.22* -15.27* 101.82 130.45 49.15 * t -66.98* -15.32 -9.38 7 7 .7 8 -38.35 P4 -21.53* -15.58 - 3.26 112.15* -3.98 48.68* -15.63 - 9.69 2.64 140.78 24.65 77.32 Speed 3 P2 -3.70 4.30 P3 -16.30* -16.60* 77.14* 122.19 34.81* -64.86* -8.29 - 8.60 79.86 -19.81 P4 -19.54* -19.85* -7.25 76.81* -22.86 19.47* -11.54 -11.84 0.76 121.86 22.19 64.53 Speed 4 P2 -2.22 5.95 P3 -16.93* -18.79* 142.94* 181.72 78.61* -83.72* - 8.76 -10.62 117.39 -44.94 P4 -21.17* -23.03* -8.32* 118.94* -43.39* 20.94* -13.00 -14.86 -0.15 157.72 -4.61 5 9 .7 2 *Siynificant difference (P<0.01) 12 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) At all four velocities PI recorded a greater post- impact rebound distance than the other pads. P2 showed similar rebound distances to P3 at SI and to P4 at S2 and S3. P2 showed rebound distances that were significantly greater than P4 at SI and signifi­ cantly smaller than P3 at S2, S3 and S4, and P4 at S4. P3 was significantly greater than P4 at all the impact velocities, while P4 was significantly smaller than all the pads at all four the impact velocities, with the exception o f recording similar rebound distances to P2 at S2 and S3. At the slower velocities P4 showed a sig­ nificantly smaller rebound distances than PI, P2 and P3, with the exception of similar scores to P2 at S2. P3 recorded significantly smaller values than PI at SI and S2 and P2 at SI. P2 and P3 showed similar rebound characteristics at SI. At the faster velocities P4 showed significant!}7 smaller rebound characteristics than PI and P3, with similar scores to P2 at S3. P2 showed significantly smaller rebound distance to P4 at S4. Further investigation yielded that the 99% Scheffe simultaneous confidence intervals for the contrast that involved a comparison o f P2 with the other three pads, were significant at S2, S3 and S4, indicating that the force absorption and rebound distance characteristics of this pad rated the best. P2 did not differ sigffifi- cant.ly from the other pads at SI. DISCUSSION The principal finding of this study was that there is a significant difference between the impact kinetics of die cricket batting pads at various impact velocities. These differences would be as a result of the differ­ ences in die structure and composition of die protec­ tive part of the pads, widi some makes of pads better able to absorb the impact forces of the ball providing the batsman greater protection, while others were bet­ ter able to reduce die rebound distance of the ball after impact. From die findings it would appear diat die batting pads best able to absorb the impact forces of a ball bowled at fast to express velocities would be diose manufactured from a polyurethane material (PI). A second advantage, aldiough not tested experimentally, is that these pads are very7 light and comfortable to wear and as a result they should not have as great a negative effect on the running speed of the batsman as die heavier pads currendy used. In limited-overs crick­ et where an off-field umpire uses slow motion TV replays to decide whether a batsman is run-out or not, the player who runs fast between the wickets has a dis­ tinct advantage and the success or failure often depends 011 die speed with which he is able to run between the wickets. The pads that are currendy in use have a greater ability7 to reduce the rebound distance of the ball after impact with the pads. This reduced ball-pad coeffi­ cient of restitution would be as a result of die structure and composition of the protective parts of the pads which are less rigid.However, the rebound characteris­ tic of the pads could eidier be an advantage or a dis­ advantage when batting, depending on the type of match and the match situation. Li Test cricket, where it is more common for fielders to field in close proxim­ ity to the batsman, pads widi a large post impact rebound distance of the ball could result in the bats­ man being caught “bat-pad” (die ball deflected from die bat onto the pads and then being caught by a field­ er). Wearing the more traditional pads, which showed smaller rebound distance, would reduce this rebound distance thus reducing the risk o f dismissal in this manner. Hie converse could, however, apply in limited- overs cricket matches where the fielders are normally a distance from the batsman. The batsman wearing pads with a greater post-impact rebound distance may have an advantage by being able to score runs or leg-byes from balls deflecting off the pads. A disadvantage of PI, aldiough not tested experi­ mentally, was that the ball striking the pad made a sim­ ilar sound to that of a ball striking the bat and could result in a batsman being incorrecdy adjudged caught by7 the wicketkeeper or fielder. The manufacturers are, however, hi the difficult position o f having to balance the impact absorption and rebound characteristics of the pads widi the comfort required to wear them for long periods at a time. Cricket pad manufacturers, aware that the batting pads behave differentiy7 under various impact conditions due to the structure and composition of the protective lay7- ers o f die pads, need to further investigate the impact properties of the components or combinations of com­ ponents they use in their batting pads. Only through continual research in the design and composition of the materials used in the manufacture of batting pads, will die players gain maximum protec­ tion and comfort. REFERENCES 1. Temple R. Cricket Injuries: F ast Pitches change the Gentleman’s Sport. Pliys Sportsmed 1 9 8 2 :1 0 ( 6 ) :1 86-19 2. 2. Blonstein JL. Medical aspects o f amateur boxing. Proc R Soc Med 19 66 ;59 :64 99. 3. Stretch R A . Injuries to South African cricketers playing at first-class level. Sports Med 1 9 8 9 :4 (I) :3-20. 4. Stretch R A . The incidence and nature o f injuries in first- league and provincial cricketers. S A fr M ed J 1 9 9 3 ;8 3 (5 ) -.339-342. 5. Stretch R A . The seasonal incidence and nature o f injuries in schoolboy cricketers. S A fr Med J 1 9 9 5 ;8 5 (1 1 ):1 1 8 2 -1 1 8 4 . 6. Jones NP, Tidlo A B . Severe eye injuries in cricket. B r J Sports Med 19 8 6 ;2 0 ( 4 ) : 1 78-179. 7. Stretch R A , Tyler J. The force absobtion characteristics o f cricket batting gloves at four impact velocities. S A fr J Sports Med 1 9 9 5 ;2 ( 3 ) :2 2 © 2 9 . 8. Hyrosomallis C. Shock absorption o f cricket leg guards and batting gloves. Paper presented at Australian Conference o f Science and Medicine in Sport, National Convention Centre,Canberra, Australia. 1996. 9. Payne \VR, H o y G, Laussen SP, Carlson JS. What research tells the Cricket Coach. Sports Coach 1 9 8 7 ;1 0 (4 ) -.17-22. 10. Abernethij B. Mechanism o f skill in Cricket Batting. A u s t J Sport 1 9 8 1 ;1 3 ( 1 ) :3 -W . 11. Penrose T, Foster D , Blanksbg B. Release velocities o f Fast Bowlers during a Cricket Test Match. Supplement to A u st J Health Phys Ed Rec. 1976:71:2-5. 12. Nigg BM. The validity and relevance o f tests used for the assessment o f sports surfaces. Med Sc Sport Exercise 1 9 9 0 :2 2 (1 ) :1 31-13 9. □ SPORTS MEDICINE OCTOBER 1998 13 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) RESEARCH ARTICLE Comparison of oral and infrared auditory canal thermometry in sports participants A Fuller (BSc Hons) D Mitchell (PhD) Department o f Physiology, University o f the Witwatersrand, Medical School, Parktown 2193 . ABSTRACT Objective: To investigate whether recently developed infrared auditory canal thermometers, which offer numerous advantages over conventional oral glass- mercury thermometers, provide measurements equiv­ alent to those of the oral glass-niercurv thermometers before and after exercise. Design: Open trial Setting: Sporting arenas and gymnasiums. Interventions: Oral (T,„.,|) and auditory canal tempera­ tures (Tuc) were recorded in 45 adult persons partic­ ipating in one of four sports; field hockey, squash, high-impact aerobics or swimming. Results: T.1(. correlated linearly with Toral before and after exercise, but there was considerable variability between the two measures o f body temperature, with upper and lower 95 % prediction intervals determined over the range o f Tac differing by at least 2 "C. Although we did not measure core body temperature, post-exer­ cise measurements of Tac and Tonl] did not indicate tlie expected rise in body temperature, and both sites appear to underestimate blood temperature. In addi­ tion, the ear canal thermometer recorded anomalously low readings of post-exercise Tolal at the site o f tlie swimming pool. Conclusions: Our data indicate that the relationship between auditory canal temperature and oral tempera- true is variable and unpredictable in sports partici­ pants. Audi ton' canal thermometry is unlikely to offer an improvement on oral thermometry for tlie screening of body temperatiue in tlie sports arena. Key words: body temperature; exercise; tympanic membrane temperatiue; hyperthermia INTRODUCTION Hyperthermia and hypothermia constitute serious health risks, both for the elite athlete and the recre­ ational sportsperson. Management of the risk requires the measurement of core body temperature, which is a particularly formidable logistic problem when events can attract tens o f thousands o f participants. Numerous techniques have been used, clinically and in the laboratory, to monitor body temperature. The gen­ eral thermal status o f tlie body is best represented by CORRESPONDENCE: Andrea Fuller Department of Physiology University of the Witwatersrand, Medical School 7 York Road, Parktown, 2193 Johannesburg, South Africa Tel: 0 1 1 6 4 7 2363 Fax: O il 643 2765 E-mail: 127and y@cli iron. wi t.s .ac. za the temperatiue of mixed venous blood in the right ventricle or pulmonary7 artery.316 However, temperatiue measurement at this site necessarily is invasive and usually is carried out only in critically ill patients. Other estimates of body core temperature trade off accuracy against practical considerations of conve­ nience and compliance, hi tlie spoils arena, especially when there are multiple participants, one is con­ strained to use measures, at least for the primary screening of body core temperatiue, which forfeit accu­ racy for convenience. Even though it is documented that sublingual oral temperatiue (T<)rill) is not a good index of blood temperature,"1 it is die measiue of body temperature most commonly used in sports partici­ pants; it is used because it is convenient and socially acceptable. However, measurement of Toral requires the voluntary co-operation of die subject and careful probe placement.7 Glass-mercury thermometers are associat­ ed with risks of mercury spillage and cross-infection, and are slow, requiring at least three minutes of closed-moudi measurement and providing maximum accuracy after about eight minutes.12 Furthermore, the accuracy of the reading, which is not good at best, can be influenced adversely by the ingestion of liquids, gum chewing, and open moutii breadiing.7222'' Recendy, “ tympanic membrane” thermometers have become available commercially, at relatively low cost, and are being used increasingly in hospitals. These instruments record infrared radiation emitted from the tympanic membrane or, more usually, die ter­ minal auditory canal, and offer several practical advan­ tages over other devices used for body temperature measurement. Tlie tiiemiometers are non-invasive, easy to use, and provide a digital display of tempera­ ture in a few seconds.22 Unlike oral thermometers, auditory canal dierniometers are not influenced by recent liquid ingestion or inadvertent mouth opening,23 and there is no discomfort or risk of infection for sub­ jects. H ie usefulness of diese thermometers has been demonstrated in pediatric patients, intensive care unit patients, nursing home residents, outpatients, and research laboratories.022 Tlie potential o f the infrared auditory canal diennometer to simplify and shorten the process of temperature measurement makes it an appealing device for use in sports participants. If it is to be a viable sports thermometer, it should be no less accurate than the conventional oral thermometer. As with the oral thermometer, the auditory canal ther­ mometer cannot be used to monitor the temperature of a person widi heat illness, where a better measiue of body core temperature is required; its role might be to screen for those sports participants who require more intensive monitoring. The purpose of our study was to determine whether die infrared auditory canal thermometer provides measurements equivalent to those of the oral glass- mercury diennometer before and after exercise. We compared auditor}' canal and oral temperature mea­ surements in adults participating in a variety of sports. 14 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) METHODS Toiai and auditory canal temperatures (Tac) were recorded in subjects participating in four sports, at dif­ ferent times of day; field hockey, squash, high-inipact aerobics, and swimming. In total, 45 health}' subjects volunteered, and their ages are show'll in Table I. The protocol for our study was approved by the Committee for Research on Human Subjects, University of the Witwatersrand. TABLE: I Subject ages Age (yr.) Sport Mean ± SD Range Hockey (n = 11) 20.5 ± 1.4 18 - 23 Squash (n = 10) 21.0 ± 1.2 19 - 23 Aerobics (n = 12) 22.8 ± 2.8 20 - 30 Swimming (n = 12) 22.9 ± 8.2 18 - 48 We measured Tol.a] using a glass-mercury ther­ mometer placed in the subject’s posterior sublingual pocket for three minutes. During this period, Tac was recorded using an infrared thermometer, the Genius FirstTemp Model 3000A (Intelligent Medical Systems, Carlsbad, CA). Auditory canal temperatures were recorded in both ears, first in the right ear and imme­ diately thereafter in the left ear. Temperatures were measured according to the manufacturer's instruc­ tions; the subject’s head was gently restrained while the probe tip was inserted far enough into the ear canal to seal the tympanum from ambient air. To min­ imise operator variability, Tac was measured by a single investigator. The subjects rested at ambient tempera­ ture, at die site of subsequent activity, for at least five minutes before we recorded rest temperatures. Post­ exercise temperatures were recorded within two min­ utes after completion of exercise. Swimmers did not towel-dry themselves until both Tuc and Toral had been measured; their ear canals were, however, dried before measurements of Tac were taken. The duration of exer­ cise for each sport (excluding warm-up time) was 70 minutes for hockey, 50 minutes for squash, 60 minutes for aerobics and 25 minutes for swimming. Post-exer­ cise temperatures were not obtained from three sub­ jects for technical reasons. The glass-mercury thermometers and the auditory canal thermometer were calibrated after completion of the study, against a high accuracy quartz thermometer (Quat 100/200, Heraeus, Hanau, Germany). The oral thermometers were immersed in an insulated water bath set at a variety of controlled temperatures. The auditor}7 canal thermometer was pointed at an infrared black body, constructed as a matt black re-entrant cone set up in the water bath. When calibrated, both the auditory canal thermometer and the oral thermome­ ters had an accuracy of better than 0.1 ’C over their measurement ranges. Dr}7 and wet-bulb ambient temperatures were mea­ sured at 10-minute intervals at each sports site using a sling psyclirometer. A psychrometric chart was used to calculate vapour pressure and relative humidity. Environmental conditions at each site are shown in Table II. The heated swimming pool was located indoors in a health club complex, and pool tempera­ ture was about 25 'C. TABLE II: Environmental conditions (mean ± SD, n = 4 to 8) Sport Tdh Twb RH VP CC) CC) (%) (kPa) Hockey 21.1 ± 1.1 12.1 ± 1.0 36 ± 4 0.9 ± 0.1 Squash 21.0 ± 0.7 13.8 ± 0.4 46 ± 6 1.2 ± 0.1 Aerobics 20.1 ± 0.5 13.7 ± 0.3 51 ± 2 1.2 ± 0.0 Swimming 18.0 ± 0.1 15.0 ± 0.1 74 ± 1 1.5 ± 0.1 T Jb = dry bulb temperature; Tw|, = wet bulb temperature; R H = relative humidity7; VP = vapour pressure. The relationship between Tai. and Toral was evaluat­ ed using Pearson product-moment correlation analyses and linear regression. Values of p < 0.05 were consid­ ered significant. RESULTS Left T.u: was significantly correlated with right Tac at rest before exercise (r = 0.64, P < 0.0001, n = 45; left Tac = 12.7 + 0.64 right Tllc; right Tuc = 13.2 + 0.63 left Tac), and mean left Tac (35.9 ± 0.5 °C) and mean right Tac (35.9 ± 0.5 "O were the same. Furthermore, the slope of a regression line forced tlirough the origin was not significantly different from one (Figure 1), so there was no particular bias for one side of the head. However, only about 40 % of the variability7 in the audi­ tor}7 canal temperature of one ear was associated with variability7 in the other ear. Measures of contralateral T.lc were not equivalent; the mean absolute difference between left Tac and right Tac was 0.3 ± 0.3 ‘C, and exceeded 1 *C for individual subjects. Tac , in all further analyses, therefore was expressed as the mean of left Tac and right T.lc. Figure 1: Left and right auditor}7 canal temperatures in the same subjects, at rest, before participation in sports events. The linear regression line (solid line) and 95 % prediction intervals (dashed lines) of an equation constrained to go through the origin are shown. (Right Tac = (1.00 ± 0.01) x left Tac, r = 0.52, P < 0.0001, n = 45). Mean oral and auditor}7 canal temperatures record­ ed for participants in each sport, both at rest and post- exercise, are shown in Table III. Mean oral tempera­ tures were typical of those usually recorded in young adults, in Johannesburg. At rest, Tuc and Tonll were sig­ nificantly correlated (Figure 2), with Tolal somewhat SPORTS MEDICINE OCTOBER 1998 15 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) higher than Tac at tlie lower temperatures, and some­ what lower at the high temperatures. However, the upper and lower 95% prediction intervals differed by 2 ’C (Figure 2), implying that at any Tac , Tornl could be substantially higher or lower than Tac in individual subjects. TABLE III: Oral (Toral) and auditory canal (Tac) temperatures o f subjects recorded before and after exercise (mean ± SD) Toni, CC) Sport Rest Post -exercise Tac CC) Rest Post -exercise Hockev 36.2 ± 0.4 35.5 + 0.5 36.0 ± 0.3 35.6 ± 0 5 (n = 10) Squash 36.1 ± 0.5 35.8 ± 0.7 35.8 ± 0.6 35.3 ± 0.9 (n = 10) Aerobics 35.8 ± 0.6 36.2 ± 0.6 35.7 ± 0.6 36.0 ± 0.5 (n = 11) Swimming 35.9 ± 0.7 36.2 ± 0.9 36.0 ± 0.5 33.6 ± 0.7 (n = 11) Figure 2: Oral temperature (Tnrai) as a function of audi­ tory' canal temperature (Tac, mean of both canals) in subjects at rest before exercise. The linear regression line and 95 % prediction intervals are shown. (Toral = 0.63 Tac + 13.3, r = 0.56, P < 0.001, n = 45). Post-exercise temperatures for hockey, squash and aerobics participants are shown in Figure 3. Both Tac and ToraI tended to drop (luring the sports events. Tac and Torati for all three sports considered together, were still significantly correlated after the events. As with rest temperatures, variability between Tac and T()ral of individuals was considerable, with the upper and lower 95 % prediction intervals differing by between 2.2 and 2.4 "C over the range of measured temperatures (Figure 3). The mean Tac measured after swimming was very low (33.6 + 0.7 so the temperatures after swim­ ming were analysed separately. Measurements o f Tac and Toral were significantly correlated (Figure 4). However, tlie mean absolute difference between Ta(. and Toral was 2.6 + 0.6 "C, and the 95 % prediction intervals differed by between 2.9 and 3.3 ‘C over the range of measured Tac. The greatest discrepancy between Tac and Tora| was therefore evident in swimmers. We con- Figure 3: Oral temperature (Toral) as a function of audi­ tory canal temperature (Tac, mean o f both canals), after exercise in hockey players (open circles, n = 10), squash players (open triangles, n = 10), and aerobics participants (closed circles, n = 11). The linear regres­ sion line and 95 % prediction intervals are shown. (Toral = 0.48 Tac + 18.8, r = 0.54, P < 0.005, n = 31). sidered the discrepancy too great to be the result sim­ ply of an inherent difference between the measure­ ment sites, and so measured Tac in two observers, who had been at tlie poolside but not in the water; in both cases, Tac was recorded as less than 34 ’C. Figure 4: Oral temperature (TonJ) as a function of audi­ tory canal temperature (Tuc, mean of both canals), after swimming. The linear regression line and 95 % predic­ tion intervals are shown. (Toral = 1.01 Tac + 2.4, r = 0.76, P < 0.01, n = 11). DISCUSSION The main aim o f our study was to determine whether temperatures indicated by an auditory canal ther­ mometer bore a defined relationship to those indicat­ ed by an oral thermometer, when the measurements were made simultaneously in sports participants. If that had been the case, the speed and convenience of the auditory canal thermometer would make it an attractive instrument for screening participants in events which impose a risk o f hyperthermia or hypothermia, replacing tlie oral thermometer, but not replacing the rectal, oesophageal or blood thermome­ 16 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) ter necessary for moiiitoring the body core temperature of those with signs or symptoms of heat or cold illness. Our results indicate that the relationship between auditory canal temperature and oral temperature is variable and unpredictable in sports participants, before and after their events. Indeed, the results lead us to doubt whether either thermometer can be used safely to screen spoils participants. With neither thermometer does the problem He with an intrinsic inaccuracy of the measuring instru­ ment. When calibrated against a quartz thermometer, by water immersion in the case of the oral thermome­ ter and using a re-entrant cone blackbody in the case of the infrared detector in the auditory7 canal ther­ mometer, both thermometers achieved an intrinsic accuracy of 0.1 "C over the range of measured temper­ atures, better than necessary7 for physiological screen­ ing o f body temperature. Li the case o f one sport, namely swimming, the main problem with the audito­ ry canal thermometer resulted from its failure to record temperatures properly in the field. For all sports, a second problem arose, namely physiological variability and inconsistency between the tempera­ tures indicated by the oral and auditory7 canal ther­ mometers. We detected the poolside problem with the audito­ ry canal thermometer through the anomalously low readings of post-exercise Tac recorded at the site of the swimming pool (Table III). We believe that the infrared thermometer was adversely affected by water, either in the air or condensing, as a result of the high relative humidity (74 %), or because o f direct wetting (despite swimmers drying their ear canals). If water indeed caused the malfunction, sweat accumulation in the auditory7 canals of subjects may also affect the func­ tioning of the infrared thermometer. We recorded audi­ tory7 canal temperatures of less than 34.5 ‘C in three squash players who had sweated profusely (Figure 3). We do not-know whether the problem is peculiar to the make of the thermometer we used, or whether it is a general problem o f all infrared auditory7 canal ther­ mometers. If it is a general problem, it is a serious deficiency; it not only would exclude the use of the thermometers for water sports (where hypothermia is the thermal risk), but also exclude them from use in environments where humidity contributes to the risk o f hyperthermia. That does not mean only hot humid environments; several football deaths have been reported when Ta was below 24 'C, but the relative humidity exceeded 95 %.14 One of the physiological problems we encountered with the auditory7 canal thermometer was lack of bilat­ eral symmetry between the temperatures of the two ear canals. Several papers describing the clinical use of auditory7 canal thermometers have claimed that the canals do have the same temperature, even in the pres­ ence of unilateral ear disease.11’ " 24 It may be that, in the clinical setting, the mean absolute difference of 0.3 "C between the two ear canals, which we observed, is considered tolerable. The criterion sometimes used to conclude that there is bilateral symmetry, namely that in a large group of subjects the mean difference between the two canal temperatures is not significant­ ly different from zero, clearly is an inappropriate use of statistics, and reflects only variability between the sub­ jects. We decided to use the average temperat ure of the two auditor}7 canals. The difference between that aver­ age temperature and oral temperature measured simultaneously was much greater than could be accounted for by the intrinsic inaccuracy of the ther­ mometers. Despite similarities in mean temperatures (Table III) and statistical correlation between Tac and Torat (Figure 2, 3 and 4), the oral glass-mercury and infrared auditor}7 canal thermometers yielded temper­ atures which were not interchangeable. In resting sub­ jects, only 31 % of the variance in Tac was associated with the variance in T„ra|. Furthermore, for any partic­ ular value of T.1C, Tora, of an individual could be more than 1 ‘C above or below the mean expected value for the group (Figure 2). During exercise there are changes in heat balance at the head, for example those resulting from open-mouth breathing and scalp sweat­ ing, and one would expect the relationship between Tac and T(lra| to be even more variable than at rest. Indeed, we found greater inconsistency in the relationship between Tac and T(lral after exercise. For example, at Tac just below 36 "C, oral temperature was below 35 "C in one hockey player and almost 37 "C in one aerobics participant. We therefore have to conclude that there is no fixed relationship between auditor}7 canal tem­ perature and oral temperature in sports participants, at least for the sports and conditions we investigated. The absence of a relationship between auditor}7 canal temperature and oral temperature could arise because either one, or both, of the sites at which the thermometers measure temperature have a tempera­ ture not consistently related to blood temperature. Though we did not measure core body temperature, it is likely that neither site reflected blood temperature properly, since neither site demonstrated the expected rise in temperature associated with increased meta­ bolic heat production during exercise (Table III). Although it continues to be used in the sports arena, oral thermometry has been known, for at least thirty years, to provide an unreliable index o f blood temperature during exercise.21 Measurements of T(mll are lower than temperatures at deeper sites, in an inconsistent way, and are affected by both breathing patterns and ambient air temperature.12111 Head skin temperature also has a strong influence on oral tem­ peratures,15 and low skin temperature could have con­ tributed to low oral temperature in the swimmers and when there was free evaporative cooling from the head. Infrared auditory7 canal thermometers are much newer devices, and experience with them outside the hospi­ tal setting is limited.20-22 They are often referred to as infrared tympanic thermometers; the temperature at one particular site of the tympanum is thought by some to reflect the temperature of blood perfusing the brain.124 However, the Genius thermometer which we used does not have the resolution necessary to mea­ sure temperature of the tympanic membrane only, and detects infrared radiation emanating from both the ear drum and the surrounding tissue, and the same is true for other makes o f infrared thermometer.0H Auditory canal temperature is lower than tympanic membrane temperature (T,y).20 True tympanic temperature can be measured by placing either a thermistor or a thermo­ couple in direct contact with the ear drum. Positioning of such a thermometer is a thoroughly unpleasant experience for subjects,2"* and contact tympanic ther­ mometry cannot be used in the sports arena. The fact that Tac may underestimate body core tem­ perature would not prevent its use, as a field ther­ mometer, if the relationship between Tac and blood temperature was fixed, in sports participants, and uncontaminated by other variables. However, that is not the case. Li particular, auditory canal temperatures are influenced by head skin temperature/'13ls 17 Heat SPORTS MEDICINE OCTOBER 1998 17 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) loss from die human head constitutes a major compo­ nent of total heat loss during exercise, particularly when there is additional convective cooling in sweating subjects;3'19 fanning the face of hyperthermic subjects reduces TIV.S IM The effect of head skin temperature on Tac arises not primarily from the pinna, but from tlie scalp directly above and behind the ear.13 Hockey play­ ers, unlike tlie other sports participants in our study, exercised outdoors in windy conditions. Tlie air move­ ment, arising from ambient wind and tlie players’ motion, would have induced convective and evapora­ tive heat loss from scalps which may be the reason that post-exercise Tac was lower than Tac at rest. In our view, therefore, auditory canal th e r m o m e t r y is unlikely to offer an improvement on oral thermome­ try for the screening of sports participants for hyper­ thermia or hypothermia, and we reaffirm that oral thermometry provides an inconsistent underestimate o f blood temperature after sport. It remains remotely possible that Tac does have some recognizable rela­ tionship with blood temperature in sports participants; the public nature of our study prevented us measuring rectal temperature (Trc). Correlation of Tlu. with Trc in a variety of sports participants still needs to be done. In one recent study,24 there was poor agreement between measurements of Tac and T n: in athletes with suspect­ ed exertion-induced heat exhaustion. Tac on admission to the field medical centre was on average 1.2 "C lower than Trc. In addition, statistical variability between the two measures o f body temperature was such that mea­ surements of' T .li: could not be used confidently to pre­ dict a rectal temperature greater than 38 ’C. The inves­ tigators concluded that the use of infrared tympanic membrane thermometry can result in misdiagnosis of heat exhaustion.9 Therefore, it seems unlikely that there will be a recognizable relationship between Tac and Trc in sports participants. ACKNOWLEDGEMENTS We thank Candice Downing, T anya Bohm and Carey Eddy for assistance with the data collection. This work was supported by the Foundation for Research Development. REFERENCES 1. Benzinger TII. On physical heat regulation and the sense o f temperature in man. Proc Nat A ca d Sci USA 1959: 45: 645-659. 2. Benzinger TH, Taylor GW. Cranial measurements o f inter­ nal temperature in man. In: H ardy JD ed. Temperature: Its Measurement and Control in Science and Industry. New York: Reinhokl Publishing Corp., 1963: 111-120. 3. Brengelmann GL. Dilemma o f body temperature measure­ ment. In: Shiraki K , Yousef M K eds. Man in Stressful Environments. Springfield: Thomas, 19 87 : 5-22. 4. Brinnel H, Cabanac M. Hyperthermia and human brain cooling. In: Shiraki K . Yousef M K eds. Man in Stressful Environments. Springfield: Thomas, 1987: 87-97. 5. Brinnel H , Cabanac M. Tympanic temperature is a core temperature in humans. J Therm Biol 1989: 14: 47-53. 6. Chamberlain JM, Terndrap TE, Alexander DT, Silverstone F A , Wolf Klein G, O D onnell R, Grandner J. Determination of normal ear temperature with an infrared emission detec­ tion thermometer. A nn Emerg Med 19 95 : 2 5 :15 -2 0. 7. Erickson R. Oral temperature differences in relation to ther­ mometer and technique. iVurs R es 1 9 8 0 :2 9 : 157-164. 8. Fraden J, Lackey RP. Estimation o f body sites tempera­ tures from tympanic measurements. Clin Pediatr 1 9 91 : 3 0 (Suppl): 65-70. 9. H ansen R D , Olds TS, Richards D A , Richards CR, Leelarthaepin B. Infrared thermometry in the diagnosis and treatment of heat exhaustion. Int J Sports Med 1996: 17: 66-70. 10. Kelly B, Alexander D. Effect o f otitis media on infrared tympanic thermometry. Clin Pediatr 1 9 91 : 3 0 (Suppl.) :46- 48. 11. Kenney RD, Fortenberry JD, Surratt SS, Ribbeck BM, Thomas WJ. Evaluation o f an infrared tympanic membrane thermometer in pediatric patients. Pediatrics 1990: 85 : 854-858. 12. Mairaux P, Sagot JC, Candas V. Oral temperature as an index o f core temperature during heat transients. Eur J A p p l Physiol 1983: 50 : 33 1-34 1. 13. Marcus P. Some effects o f cooling and heating areas o f the head and neck on body temperature measurement at the ear. Aerospace Med 1973: 44 : 39 7-40 2. 14. McArdle W D, Katch FI, Katch, VL. Exercise Physiology: Energy, Nutrition and Human Performance. 3rd Edition, Philadelphia: Lea and Febiger, 1991: 569. 15. McCaffrey TV, McCook R D . Wurster R D . Effect o f head skin temperature on tympanic and oral temperature in man. J A pp l Physiol 1975: 3 9 : 114-118. 16. Mitchell D, Laburn HP. Pathophysiology o f temperature regulation. Physiologist 19 85 : 28 : 507-517. 1 1 ■ Nadel ER, Horvath SM. Comparison o f tympanic mem­ brane and deep body temperatures in man. Life Sci 1970: 9: 869-875. 18. Nielsen B. Natural cooling o f the brain during outdoor bicy­ cling? Pflugers Archives 1 9 88 : 4 1 1 : 45 6-46 1. 19. Rasch W, Samson P, Cote J, Cabanac M. Heal loss from the human head during exercise. J Appl Physiol 1 9 91 : 71: 590- 595. 20. Shenep JL, Adair JR, Hughes WT, Roberson PK, Flynn PM, Brodkey TO. Fullen GH, Kennedy WT, Oakes LL, Marina NM. Infrared, thermistor, and glass-mercury ther­ mometry for measurement o f body temperature in children with cancer. Clin Pediatr 1991: 3 0 (Suppl.): 36-41. 21. Strydom NB, Wyndham CH, Willianis CG, Morrison JF, Bredell G A G , Joffe A . Oral/rectal temperature differences during work and heat stress. J Appl Physiol 1965: 2 0 : 283- 287. 22. Terndrup TE. A n appraisal o f temperature assessment by infrared emission detection tympanic thermometry. A nn Emerg Med 1992: 2 1 : 1483-1 492 . 23. Terndrup TE, Allegra JR, K ealy J A A . Comparison o f oral, rectal, and tympanic membrane-derived temperature changes after ingestion o f liquids and smoking. A m J Emerg Med 1 9 89 : 7: 150-154. 24. Terndrup TE, Wong A . Influence o f otitis media on the cor­ relation between rectal and auditory canal temperatures. A J D C 19 91 : 145: 75-78. ' □ 18 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) Brief report: Urinary catecholamine excretion during outdoor sport rockclimbing F Marino PhD, Med BPE J Booth MSp Sc, Bed Human Movement Studies Unit, Human Performance Laboratory Charles Sturt University, Bathurst, NSW 2 7 9 5 Australia ABSTRACT Objectives: The purpose of this study was to examine the response of urinary excretion of catecholamines after a bout of outdoor sport rockclimbing. Design, setting & subjects: This study was undertaken outdoors in the Blue Mountains in NSW, Australia. Seven elite rockclimbers were recruited for the study. Each had previous experience with the climb route. The subjects were required to climb the route as fast as possible. Heart rate (HR) and urinary excretion of catecholamines (epinephrine, EPI; norepinephrine, NE) were measured pre-climb, immediately post-climb and 30 min post-climb. Results & conclusions: The duration of the climb was ~7 min, 36 s. FIR increased from 74 bpm (pre-climb) to 157 bpm during the climb. The rise in cate­ cholamine excretion was not significant compared to pre-climb samples at either of the sample times. It is concluded that training induced adaptations to this type of exercise coupled with the methodological con­ straints of the study are responsible for the non-signif­ icant increases in catecholamine excretion. INTRODUCTION Rockclimbing has increased in popularity in recent years. The follow-on has seen improvement in climb­ ing standards and competitions.1 However, unlike other popular sports t here are not much scientific data on rockclimbers and rockclimbing with most ol the available data related to typical climbing injuries.13 More recently, however, anthropometric data has become available.45 Data on the physiological respons­ es and requirements of rockclimbing is somewhat scarce with inferences mainly being drawn from other related activities such as mountaineering and gymnas­ tics. However, sport rockclimbing is unlike the more studied sports with the exercise itself characterised by isometric muscle contractions and gymnastic type movements. Furthermore, the sport may be considered ‘high risk’ which adds a stressful component to the physiological demands. CORRESPONDENCE: Frank Marino Human Movement Studies Unit Human Performance Laboratory Charles Sturt University, Bathurst, NSW 2795 Australia Tel: 61 + 2 + 63 384268 Fax: 61 + 2 + 63 384065 Email: frnarino@csu.edu.au Blood and urine catecholamine concentrations have been typically used as an index of sympathetic activity and have been shown to differ between trained and untrained subjects.0 Because rockclimbing activity relies heavily on static muscular contractions, increased sympathetic activity would be expected.7 Therefore, it would be of significant interest to evalu­ ate the sympathetic drive elicited by a bout of outdoor rockclimbing. Furthermore, the common cate­ cholamine response exhibited after similar or ‘high risk’ type activity such as rockclimbing is an increased epinephrine (EPI) and norepinephrine (NE) response.8 For example, the urinary levels of EPI and NE in para­ chutist trainees were found to increase for initial jumps and then subsequently decline over the training period for both pre-jump and post-jmnp values. This decrease hi EPI and NE is indicative of a reduced adrenomedulary discharge as a consequence of famil­ iarisation to a previously stressful event. This type of physiological response has yet to be studied in trained sport rockclimbers. Therefore, the aim of t li i s study was to examine the response of uri­ nary7 excretion of catecholamines after a bout of out­ door sport rockclimbing. m e t h o d s Subjects: Seven competitive rockclimbers (6 male, 1 female) were recruited for this study. The mean climb­ ing experience of the group was 8.9 ± 1.2 years with the most difficult outdoor ascent made without pre­ view or fall ranging from 6b - 7a (UK grading system). All subjects were in good health as determined by a medical and health questionnaire. The study was approved by the University Ethics in Human Research Committee. The mean (± SE) age, height, mass and sum of nine skinfolds were 25 ± 1 years, 175.7 ± 2.7 cm, 62.6 ± 3.3 kg, and 61.3 + 4.2 mm, respectively. Climbing protocol: The climb was graded 5c and was conducted on a rockface in the Blue Mountains of New South Wales, Australia. All subjects had previous expe­ rience witli the climb on at least three other occasions. The length of the climb was 24.4 m and overhanging by = 4m throughout at an elevation o f 890 m. On one occa­ sion approximately 5 - 7 d prior to the climb a resting sample of urine was collected after subjects abstained from alcohol, caffeine ingestion and vigorous exercise for a minimum period of 24 h. The reasons for this were that the route to the climb was lengthy with sub­ jects required to walk for approximately 20 mill and in some instances the terrain physically demanding. Flence, a pre-climb sample at die climb site would have influenced the outcome of the urine concentration o f catecholamines. Once arrived at the climb site sub­ jects rested for at least 30 min before micturition. At SPORTS MEDICINE OCTOBER 1998 19 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) mailto:frnarino@csu.edu.au the climb site the subjects were briefed as to the con­ ditions of the climbing protocol and prepared for climbing with a waist harness secured to a 10 mm dynamic rope fed through a ring bolt at tlie top of the rockface and to a belayer at the start of the climb. The briefing and preparation was approximately 20 - 30 min duration. Subjects were required to climb the route as quickly as possible. On completion of the climb each subject was lowered to tlie ground and given a specimen bottle for collection of urine. The specimen bottle and urine sample were returned with­ in ( 5 min of completing the climb. On returning the urine sample the subjects were given a bottle contain­ ing 350 nil of water to ingest over a period of 15-20 min before giving another urine sample 30 min post - climb. During the climb heart rate (HR) was monitored con­ tinuously with a Sports Tester (Polar-Electro, Oy., Finland). Catecholamine determination: The urine samples were collected in specimen bottles and acidified to pH 2-3 with 15 ml of hydrochloric acid. Immediately after collection the samples were stored on ice and subse­ quently frozen (-20‘C) until analysis. All samples were analysed in the same run to avoid inter-assay variation. The method of determination was by high performance liquid chromatography according to Pillai.9 This method requires less urine (0.5 ml vs. 4 ml) and is quicker with adjustment of pH facilitated by a visual indicator. Statistics: The data were analysed by AiMOVA with repeated measures on time for HR and for urinary' cat­ echolamine concentrations (pre-climb ae post-climb se 30 min post-climb). When significant main effects were found Tukey’s HSD post-hoc procedure was employed to locate the source o f significant difference. Significance was accepted when P < 0.05. All values are expressed as means ± SE. RESULTS Climb duration and HR: Tlie time taken for the climb was 7 min, 36 s (range 6 min 28 s - 9 min 54 s). Pre­ climb HR was 74 ± 5 bpm and increased to 145 ± 10 bpm (P < 0.05) and 157 ± 8 bpm after 1 and 5 min of climbing, respectively. IiR peaked at 83 bpm above resting values. Urinary catecholamine concentrations: Tlie results of the catecholamine analysis are shown in Figure 1. Resting urinary concentrations of EPI, NE and total catecholamines (CAX0X) were 89.1 ± 20.1, 315.5 ± 78.8, 404.6 ± 98.3 nmoles.L', respectively. The rise in uri­ nary excretion of EPI was not significant (P = 0.26) with levels reaching 118.5 ± 21.0 nmoles.L 1 and 180.5 ± 73.3 nmoles.L-1 immediately post-climb and 30 min post-climb, respectively. A similar non significant (P = 0.35) finding for tlie urinary excretion of NE was observed with levels reaching 304.6 ± 42.6 nmoles.L'1 immediately post-climb and rising to 731.4 ± 430.3 nmoles.L1 30 min post-climb. CATOt were not signifi­ cantly changed (P = 0.34) with values rising to 423.2 ± 60.1 nmoles.L"1 immediately post-climb and 911.8 ± 501.8 nmoles.L1 (range 195 - 4407 nmoles.L1) 30 min post-climb. cholamine excretion in trained sport rockclimbers during outdoor climbing. Sample times are at pre-climb (1), immediately post-climb (2 ) and, 30 min post-climb (3). DISCUSSION This is the first study to examine tlie sympathetic ner­ vous system response in elite rockclimbers while climbing in tlie field. Tlie levels of NE and EPI excre­ tion observed in tlie pre-climb urine sample were with­ in the normal range (EPI < 101 nmoles.L'1 ; NE < 701 nmoles.L1) as reported by tlie pathology laboratory' (Barratt & Smith, Pathology, Orange, NSW, Australia). Furthermore, NE concentration was ( 3.5 times that of EPI and is considered normal.1" Therefore, it is rea­ sonable to assume that tlie sampling and determina­ tion procedures were satisfactory for this type of field experiment. Although urinary excretion of cate­ cholamines is not ordinarily regarded as an accurate measure of sympathoadrenal activity," tlie geographi­ cal setting and isolation of the investigation prevented tlie use of serial venous blood sampling. Even though there was a tendency' for CAT0T con­ centration to increase after the climb, tlie post-climb sample was not statistically' different compared to the pre-climb sample. Tlie large SE associated with tlie 30 min post-climb CATtot is most likely' related to the large range of values recorded for both EPI and NE and the fact that one subject in particular responded with very' high values for both amines. However, a very dis­ tinct pattern did emerge from the immediate post­ climb sample to the 30 min post-climb sample for both amines. EPI excretion showed a more linear rise at all three sample times, whereas NE only showed an increase for the 30 min post-climb sample. This increase in post-climb NE can be partly explained by the delayed sympathetic spill from neuronal sites into the circulation and eventually into the tuine.8 However, given that there was no statistical signifi­ cance between samples at the various measurement times, it is difficult to speculate as to the role that the sympathetic nervous system may have played during tlie climb. Nonetheless, several points can be raised with respect to the present findings. For instance, pre­ vious work has shown increases in plasma cate­ cholamine concentration when isometric handgrip exercise is performed at levels above 20% of maximum voluntary contraction (MVC).7 Although MVC was not measured in this study, it would be reasonable to assume that subjects performed isometric exercise 20 SPORTS MEDICINE OCTOBER 1998 R ep ro du ce d by S ab in et G at ew ay u nd er li ce nc e gr an te d by th e P ub lis he r (d at ed 2 01 2. ) above 20% MVC given that they were required to lift their own body weight and that the climb was over­ hanging by = 4 m thereby increasing the negative effects of gravity. Furthermore, recent evidence indi­ cates that trained climbers are able to generate signif­ icantly higher forces on grip strength tests compared to either recreational or non-climbers.1' Although, rockclimbing is characterised by inter­ mittent isometric muscle contractions, some dynamic muscle work is also performed. Ail elevated HR of at least 30 bpm during short-term dynamic exercise is required to stimulate a rise in plasma catecholamine concentration.13 In this study, HR increased by = 83 bpm above the pre-climb value. Hence, a rise in cate­ cholamine concentration would be expected. However, even at high work rates EPI levels are not always ele­ vated compared to NE.14 This might partly account for the non-significant increase in urinary concentrations of catecholamines observed hi this study. Exercise intensity is also a determinant of sympa­ thetic st imulation. Previous work suggests that at least a 30% increase in V02 must be elicited for an increase in catecholamine concentration.11 A recent investiga­ tion has shown that rockclimbers use less than 50% of V02lllax during a standard indoor climb.15 However, work in our laboratory indicates that rockclimbers use up to 75% o fV 0 2ma, ( = 44 ml.kg.miir1) if the V02max is expressed as a function of graded climbing (unpub­ lished observations). These data indicate that rock- climbing can elicit an exercise intensity above that which is normally required to stimulate sympathetic activity. Moreover, a given VOg elicited by dynamic arm work increases catecholamine concentrations to a higher level than if the same V02 is elicited by leg work.11517 Although the present findings are not statisti­ cally significant, the pattern of urinary catecholamine excretion seem to support previous findings. There are at least three possible explanations why the rise in urinary catecholamine concentrations failed to reach statistical significance. First, plasma concen­ trations of catecholamines have been shown to be attenuated during acute exercise following train- jHg_n.i8.i9 Thjs sympathetic adaptation has also been observed in subjects with physically trained arms whilst performing handgrip exercise.2021 Tliis might partly explain the blunt catecholamine response in this study. Second, the results support the notion that trained individuals do not necessarily perceive the activity as ‘high risk’ as less trained individuals who exhibit signs of subjective fear.8 Also, the methodolog­ ical features of the study may have contributed to a reduced sympathetic response than would have other­ wise been expected. For instance, the climbers were secured to a top rope rather than lead climb and all climbers had previous experience on the route, all of which may have attenuated the subjective fear. In summary, elite sport rockclimbers do not exhib­ it significant]}7 elevated catecholamine concentrations as measured in urine after a bout ol moderatlcv diffi­ cult outdoor climbing. This may be due to training induced adaptations to the exercise and the method­ ological constraints of the investigation. It is suggested that further field work be undertaken to confirm these findings so that physiological responses to such a high­ ly skilled activity be quantified and used in order to provide a practical approach to training and prepara­ tion for competition and that the stresses related to this activity be understood. REFERENCES 1. Bollen SR. Soft, tissue injury in extreme rockclimbers. Br J Sp Med 1988; 22 : 145-147. 2. Bollen SR, Gunson CK. Hand injuries in competition climbers. B r J Sp Med 1990; 24 : 16-18. 3. Iloltzhausen LM, & Noakes TD. Elbow, forearm, wrist, and hand injuries among sport rockclimbers. Clin J Sport Med 1996; 6: 196-203. 4. Watts PB, Martin DT, & Durtschi S. Anthropometric pro­ files oj elite male and female competitive sport rock climbers. J Sports Sc 1993; 11:113-117. 5. Iloltzhausen LM, Schwellnus MP, & Noakes TD. Anthropometric and muscle strength measurements o f com­ petitive South African sport rockclimbers. Sth A j J Sports Med 1996; 4:13-18. 6. Mazzeo RS. Catecholamine responses to acute and chronic exercise. Med Sci Sports Exerc 1991: 7: 839-845. 7. Seals DR, Chase PB, Taylor JA . Autonomic mediation o f the pressor response to isometric exercise in humans. J Appl Physiol 1988; 64: 2 1 90 -2 196 . 8. Hansen JR, Stoa I