The production of salivary secretory immunoglobulin A (sIgA) is the major function of the mucosal immune system, and sIgA is the predominant immunoglobulin in saliva and other mucosal secretions. It has been described as "the first line of defense" against pathogens and antigens presented at the mucosa, such as cold-causing viruses (4,41). Specifically, decreased levels of sIgA have been associated with subsequent episodes of upper respiratory tract infection (13,26), and sIgA has therefore been suggested to be the most useful clinical biomarker to predict the incidence of this infection (13).
Saliva composition and secretion of sIgA can be modified by both parasympathetic and sympathetic nerve stimulation (29). Because intensive exercise is associated with enhanced sympathetic nervous system activity, it seems logical to assume that physical activity could modify secretion of saliva and its constituent proteins (4). Indeed, decreased sIgA secretion rates have been reported after bouts of strenuous exercise (22,23,28,39). In contrast, after moderate-intensity exercise (typically below 70% maximal oxygen uptake), secretion rates of sIgA are generally unaffected (1,20,27). However, it should be noted that data in the literature are conflicting, and several studies have shown no decrease or even increases in sIgA secretion rate after strenuous exercise (1,4,5,40).
To our knowledge, the effect of controlled bouts of exercise on mucosal immune function has yet to be analyzed in spinal cord-injured (SCI) individuals. An understanding of this is of interest for many reasons: First, certain parameters of innate immunity are depressed in the SCI population in resting conditions (9). This may potentially lead to differences when comparing mucosal immune function after exercise between paraplegic (PARA) and non-spinal cord-injured (NON-SCI) individuals. Second, it has been suggested that differences in exercise protocols may have an effect on sIgA concentration and secretion rate (39); thus, it is of interest to examine wheelchair propulsion and how this may influence exercise-induced immune responses. Third, given the aforementioned effects of sympathetic activation on sIgA in the able-bodied population, the decreased sympathetic outflow in tetraplegic (TETRA) individuals (35,36) may affect their immune response after exercise.
Analyzing the mucosal immune function in SCI individuals is of practical relevance. The most common causes of death in SCI individuals are related to respiratory illnesses (8), and individuals with a higher lesion level seem to be at a higher risk for pulmonary complications (21). Underlying mechanisms include lesion-dependent losses of respiratory muscle innervation, which lead to impairments in respiratory muscle function (3). Whereas low-level PARA individuals are comparable with able-bodied individuals with respect to respiratory and upper body strength, respiratory muscle strength in tetraplegia is markedly reduced (16). This results in a decreased ability to cough and clear secretions, and as a consequence, various types of respiratory diseases such as dyspnea, pneumonia, or respiratory failure occur in SCI individuals (8). Furthermore, uncoordinated autonomic control in TETRA individuals may be responsible for abnormal bronchial secretion, airway hypersensitivity, and other respiratory issues (19). On the other hand, it should also be noted that despite depressed resting levels, SCI persons are still capable of fortifying their systemic immune system, as shown after an electrically stimulated exercise in TETRA individuals (24). However, data about effects of exercise on mucosal immune function in this population are scarce.
Knowledge about the adaptations of the mucosal immune function after exercise in wheelchair athletes could serve as a base of health promotion and monitoring in this specific population, using sIgA as a biomarker, which also has a practical advantage because saliva is easy to collect. Salivary α-amylase may support the protective function of sIgA because it can bind to some oral bacteria (33). Given that α-amylase has been proposed as a noninvasive marker of the autonomic nervous system (25), the effect of exercise on this enzyme is of particular interest, especially in individuals with a disrupted sympathetic nervous system, i.e., TETRA individuals. Therefore, the purpose of this investigation was to explore the effects of a 60-min constant load (CL, moderate) and intermittent (IM, strenuous) bout of laboratory-controlled exercise on the sIgA and α-amylase response in a group of wheelchair athletes. Because of the reduced sympathetic outflow in TETRA individuals, we hypothesize a less pronounced sIgA decrease and α-amylase increase induced by strenuous exercise in this population.
Twenty-three male wheelchair athletes volunteered to participate in this study, which was approved by the university's ethics committee. Participants consisted of eight motor complete TETRA, seven motor complete PARA, and eight NON-SCI individuals, competing in wheelchair basketball, rugby, and tennis. All participants performed their sport on a national level at least; a summary of their physical, physiological, and sport characteristics is presented in Table 1.
Participants visited the laboratory on two occasions, separated by a minimum of 4 d. After reporting to the laboratory between 09:30 and 11:00, participants provided written informed consent and completed separate health, training, and disability questionnaires. A food diary was provided, and participants were asked to eat and drink the same types and amounts of food before both visits, to abstain from caffeine, and not to perform any exercise 24 h before the tests. To minimize the risk of autonomic dysreflexia, participants emptied their bladder immediately before each exercise session. In a preliminary testing phase, which was performed in both visits, exercise intensities for the main trial were obtained. The same standardized lunch was then provided for both visits; participants were not allowed any other food intake apart from lunch during the laboratory tests, and drink intake was limited to water ad libitum. The main trial was conducted 1 h after lunch (13:00-15:30 for all participants). For one visit, this consisted of a CL exercise block; for the other visit, of an IM exercise block. The order of visits was randomized and counterbalanced for all participant groups. All exercise tests were performed in the participants' competition wheelchair on a motorized treadmill (h/p/cosmos, Traunstein, Germany). Tire pressure was controlled and kept identical for both visits.
Participants performed six to eight submaximal CL 4-min exercise blocks at 1.0% gradient and ascending speeds to elicit physiological responses covering a range from 40% to 80% peak oxygen uptake (V˙O2peak). This was followed by a 15-min passive recovery. A graded exercise test to exhaustion (GXT) was then performed at a constant speed, which was chosen according to the responses elicited during the submaximal exercise blocks. The gradient at the start of the GXT was 1.0% for all groups; the gradient was then increased by 0.3% every minute for PARA and NON-SCI and by 0.1% every 40 s for TETRA to achieve total GXT test times between 8 and 14 min. After the GXT, participants sat quietly for 2 min and recovered actively at a low intensity (1.0 m·s−1 at 1.0% gradient) for 5 min. To confirm the V˙O2peak attained in the GXT, they then performed a verification stage, designed as a test to exhaustion at the same constant speed but at a gradient that was higher than the maximal gradient achieved during the GXT (+0.6% for PARA and NON-SCI +0.3% for TETRA). The GXT and the verification stage were terminated when participants were unable to maintain the speed of the treadmill, and verbal encouragement was given throughout the test. Spirometric data were recorded continuously with an online gas analysis system (METALYZER 3B; CORTEX Biophysik GmbH, Leipzig, Germany), and oxygen uptake (V˙O2) of each submaximal exercise block was averaged during the final minute, whereas V˙O2peak was defined as the highest average value over 30 s for both the GXT and the verification stage. From the linear workload-V˙O2 relationship and from the higher of the two V˙O2peak readings attained in the GXT and the verification stage, speeds at 1.0% gradient for 40%, 60%, and 80% V˙O2peak were then calculated for each individual from the preliminary testing of the first visit and used for both main trials. The results of the preliminary testing of the second visit were ignored, but all procedures were carried out to ensure an identical preload before the main trials.
Participants were weighed to the nearest 0.1 kg. They then performed 60 min of exercise at 1.0% gradient, which was divided by a 5-min break after 30 min, allowing data collection for midexercise responses. In the CL trial, the speed was set constantly at 60% V˙O2peak for the whole duration of exercise. In the IM trial, participants completed twenty 2-min periods at 80% V˙O2peak, each separated by a 1-min recovery at 40% V˙O2peak (Fig. 1).
Data collection for the main trial.
Timed unstimulated saliva samples were collected into sterile plastic containers before, mid, after, and 30 min after exercise. For this, participants rinsed their mouth with water and sat still with their head slightly tilted forward with minimal orofacial movement. Participants were allowed to consume water ad libitum apart from 6 min before each collection. Before, mid, and after exercise, small capillary blood samples were obtained from the earlobe to measure blood lactate concentrations using a lactate analyzer (YSI 1500 SPORT; YSI Incorporated, Yellow Springs, OH). For participants who terminated the main trial before 60 min because of exhaustion, a saliva and capillary blood sample was obtained at the time of exhaustion and treated as postexercise data. Further, participants were asked to indicate their rating of perceived exertion (RPE) using a scale ranging from 6 to 20 (7) every 5 min during the CL trial and after each period at 80% V˙O2peak in the IM trial. Heart rate (HR) was continuously recorded using an HR monitor (Polar PE4000; Polar, Kempele, Finland), whereas respiratory data were recorded during four 4-min intervals (Fig. 1) using a calibrated online gas analysis system (METALYZER 3B; CORTEX Biophysik GmbH).
Saliva samples were frozen and stored at −20°C immediately after collection. For analysis, samples were defrosted and weighed to the nearest 10 mg. Saliva volume was estimated assuming saliva density to be 1.00 g·mL−1 (11), and saliva flow rate was calculated from saliva volume and collection time. Samples were then spun for 2 min at 13,400 rpm. A sandwich ELISA was conducted using flat-bottomed microtitration plates (Nunc-Immuno Plate; Thermo Fisher Scientific, Roskilde, Denmark). After coating the plates with a rabbit antihuman capture antibody (Dako UK, Ely, United Kingdom), they were washed and blocked with a blocking protein solution (2.0% bovine serum albumin in phosphate-buffered saline). Depending on saliva flow rate, saliva samples were diluted in phosphate-buffered saline by 1:500-1:2000 (samples with a lower flow rate were diluted by a higher factor). Purified IgA from colostrum was used as a standard (Sigma-Aldrich, St. Louis, MO). Duplicate samples of 50 μL were applied to the plates and incubated overnight at 4°C. After washing the plates, a detection antibody (Dako UK) was applied, and plates were incubated at 25°C for 90 min. After a final wash, a coloring substrate (OPD substrate; Dako UK) was added, and the absorbance of the individual samples was determined spectrophotometrically at 490 nm (Opsys MR; Dynex Technologies, Inc., Chantilly, VA). SIgA secretion rate was calculated by multiplying sIgA concentration with saliva flow rate.
Salivary α-amylase activity was measured using the same spectrophotometer and microtitration plates mentioned above. Briefly, 20-μL saliva diluted 1:100 in 1.0 mmol·L−1 CaCl2 was mixed with 180 μL of amylase reagent (Infinity Amylase; Thermo Electron, Melbourne, Australia). The plate was incubated at 25°C, and the increase in absorbance at 405 nm was recorded for minutes 1 and 3. The difference in absorbance per minute was multiplied by 1994, which is a reagent and temperature-specific factor provided by the manufacturer of the amylase reagent.
All samples from the same participant were analyzed in duplicate on one microplate. The coefficient of variation of the methods on the basis of analyses of these duplicate samples was 2.1% ± 2.0% for sIgA and 1.7% ± 1.5% for α-amylase.
Data processing and statistical analyses.
The SPSS 16.0 statistical package (SPSS, Inc., Chicago, IL) was used for all statistical analyses. We based our sample size calculation on previous unpublished pilot data of relative increases in sIgA secretion rates (22% ± 16%) comparing before with immediately after an intensive training session in elite PARA athletes from the same population as the present study. Using G*Power 3.1.2 (Heinrich Heine University, Duesseldorf, Germany), we calculated we would need seven participants in each group to detect a similar change in sIgA secretion rate, with an effect size of 1.38, 90% power, and an α of 5%.
Means and SD were computed for all variables. A logarithmic transformation was applied to sIgA and amylase data to achieve normality and homogeneity of data. Normality was checked with the Shapiro-Wilk test, and homogeneity with the Levene statistic. A three-way (group × time × exercise type) ANOVA was then applied to these data. To compare the observed interaction effects of the analysis of sIgA data, PARA and NON-SCI data were collapsed into one group, and multiple two-way (group × time) ANOVAs were conducted to compare different time points, applying a Bonferroni correction to take into account these multiple comparisons. For all comparisons where the assumption of sphericity was violated, a Greenhouse-Geisser correction was applied.
V˙O2 data were averaged during 1 min at minutes 3, 18, 44, and 53. RPE and HR data were averaged during each 30-min exercise block separately. V˙O2 data were compared between trials using Bonferroni-corrected t-tests. Saliva flow rate, RPE, HR, and blood lactate concentration data violated the normality and/or homogeneity of variances' assumption required for parametric testing. Therefore, multiple Wilcoxon signed rank tests were used to analyze RPE, HR, and blood lactate concentration data between trials, and multiple Kruskal-Wallis tests were used for between-groups analyses. Saliva flow rate data were analyzed with multiple Wilcoxon signed rank tests, comparing preexercise data to data mid, after, and 30 min after exercise. For all multiple comparisons, Bonferroni corrections were applied. Statistical significance for all analyses was accepted at P < 0.05.
All participants completed the CL trial, although three participants terminated the IM trial early (one TETRA at 44 min, two NON-SCI at 30 and 38 min). The IM trial resulted in significant elevations of V˙O2, blood lactate concentration, and RPE when compared with the CL trial in all three participant groups. However, no differences between groups were found in V˙O2 and blood lactate concentration at any time (P > 0.05, Table 2). HR was consistently lower in the TETRA group when compared with the other groups (P < 0.05), whereas RPE did not differ between groups (P > 0.05, Table 2). Finally, water consumption during exercise did not differ between groups (P > 0.05).
The effect of exercise resulted in an elevated sIgA concentration (P < 0.001) and secretion rate (P < 0.05, Fig. 2), which remained elevated 30 min after exercise, whereas saliva flow rate was unaffected by exercise (P > 0.05, Table 3). Both sIgA concentration and secretion rate did not differ between participant groups at any time point, and the responses evoked by exercise did not differ between the CL and the IM trials (P > 0.05). A group × time interaction (P < 0.05) indicated a different development of sIgA secretion rate (but not IgA concentration), and after visual inspection of data, group × time interactions between TETRA and a collapsed group of PARA and NON-SCI showed a further increase of sIgA secretion rate for TETRA after midexercise, whereas sIgA secretion rates for PARA and NON-SCI decreased (P < 0.05, Fig. 3). As a consequence, 60 min of exercise resulted in a more pronounced overall increase of sIgA secretion rate in TETRA when compared with the other groups (P < 0.05).
The effect of exercise resulted in an elevated α-amylase activity (P < 0.001), which returned to resting levels 30 min after exercise. In analogy to the sIgA concentration data, α-amylase activity did not differ between participant groups at any time point, and the responses evoked by exercise did not differ between the CL and the IM trials (Table 3).
The main finding of this study is that resting levels and the main trends in sIgA secretion rate after exercise do not differ when comparing TETRA, PARA, and NON-SCI participants and result in an increase of this parameter during 60 min of both CL and IM exercise. However, TETRA participants exhibit a different pattern in the evolution of sIgA secretion rate, resulting in a greater magnitude of increase from before to after levels. Furthermore, we showed that α-amylase activity increases during exercise and returns to resting levels 30 min after exercise, irrespective of the type of exercise and the participant group. It should be noted that the participants of this study comprised a highly trained group, with V˙O2peak values similar to or exceeding existing literature of international-level wheelchair athletes (14,15). This means, with respect to training status, they were comparable to participants of many previous studies investigating sIgA and/or α-amylase responses to exercise (5,20,22,31,32,40).
The sIgA responses evoked by exercise in wheelchair athletes are in line with findings in able-bodied athletes using a similar (though soccer-specific) protocol to our study, where no differences in sIgA concentration and secretion rate have been found between CL and IM exercise, despite higher RPE scores in the IM trial (31). The same research group also found increased sIgA secretion rates after IM exercise (32), which again is in line with our findings but is challenged by Walsh et al. (40), who found no changes in sIgA secretion rate after IM exercise of a similar protocol. Likewise, similar α-amylase responses have been observed after controlled bouts of moderate (1,20) and intense (1,40) exercise.
It is commonly accepted that the sympathetic nervous system may at least be partly responsible for the changes in salivary markers, such as sIgA and α-amylase (1,4,5,10,32). Hence, it may seem surprising that no distinct differences in the mucosal immune response between TETRA and the PARA and NON-SCI group were found. TETRA individuals represent a model with no centrally mediated sympathetic nervous control (19) because centrally mediated sympathetic stimuli do not activate the decentralized part below the level of lesion (12). The disrupted autonomic innervation of the heart therefore leads to the observed decreased HR, which is a common observation in TETRA individuals (14,16,35). Most importantly, in our case, the innervation of the salivary glands in TETRA individuals is disrupted as well because it originates from the upper thoracic segments, although it remains unclear precisely where in this region (29). In rats, it has been shown that sympathectomy results in a decreased sIgA secretion (30). Furthermore, both parasympathetic and sympathetic stimulations of rat salivary glands evoke changes in both saliva flow rate and sIgA secretion (29). Because we are not aware of any scientific study investigating sIgA in human TETRA individuals during exercise, we conclude that the differences in the responses observed in denervated animal models may be species-related or may stem from other exercise-related factors, such as circulating metabolites (37), which may alter the autonomic neural output.
During exercise, TETRA individuals may compensate the lack of a centrally mediated neural drive with a spinal reflex. The observed increase in epinephrine and norepinephrine in TETRA after bladder stimulation (18) or electrically stimulated cycle exercise (6) both support the theory of a remaining but qualitatively altered sympathetic function due to reflex activity. Moreover, a hyperresponsiveness of α-adrenoreceptors in TETRA individuals (2) may further compensate for some of the lack of the centrally mediated neural drive. With respect to physical exercise, reflex activity may also be driven by afferent signals from mechanoreceptors (38), and it is possible that a reflex increase in sympathetic outflow due to muscle acidosis (37) is responsible for the observed increase in sIgA secretion rate and α-amylase activity during exercise. We did not find any difference in blood lactate concentration between our participant groups, suggesting that any reflex activity stemming from muscle acidosis may result in similar responses in all participant groups. With respect to mucosal immune function, this would underline the value of exercise for a population that is disadvantaged with regard to immune function (9,42).
Although the main trends in sIgA secretion rate are similar between groups and show an exercise-induced increase, the lack of centrally mediated sympathetic control may manifest itself in the fine-tuning of the mucosal immune response. It seems that sIgA secretion rate is downregulated after 30 min of exercise in PARA and NON-SCI, whereas this is not the case in TETRA individuals. Therefore, it is possible that centrally mediated sympathetic signals may have a greater effect in the down-regulation, whereas the aforementioned peripheral signals contribute to an up-regulation of sIgA secretion rate. A further interesting observation is that sIgA secretion rate before exercise in TETRA individuals tends to be slightly reduced. However, after 60 min of exercise, this is increased to average values found in the other groups.
Salivary α-amylase is increasingly being used as a biomarker for autonomic nervous system activation, although it has been pointed out that the correlation of α-amylase and sympathetic markers, such as norepinephrine or epinephrine, is relatively small (25). In the present study, no differences in α-amylase between groups were found. However, it should again be noted that compensatory mechanisms, such as sympathetic reflex activity, may enable individuals who lack central sympathetic drive to access parts of their sympathetic system. This suggestion is supported by the reduced but still existent increase in epinephrine and norepinephrine after exercise in TETRA individuals (34). Therefore, the observed increase in α-amylase in TETRA individuals is not necessarily proof that this biomarker should not be used to deduce sympathetic overall drive. However, it should clearly not be used as a biomarker for sympathetic central drive.
Autonomic dysreflexia is a condition sometimes found in high-level SCI individuals and includes a high outflow of epinephrines (17). It must be noted that no self-reported episodes of autonomic dysreflexia and no episodes of acute bradycardia, indicating autonomic dysreflexia, were noted during any of the exercise tests. Furthermore, only two of the TETRA participants had a history of exercise-related autonomic dysreflexia. However, to control autonomic dysreflexia more closely, we encourage data collection relating to objective symptoms of autonomic dysreflexia (such as blood pressure, skin color, or sweating) in future exercise studies with TETRA individuals. This is of particular importance because sympathetic nervous system hyperactivity and the concomitant increased outflow of epinephrines may affect sIgA and α-amylase secretion.
It is concluded that TETRA, PARA, and NON-SCI wheelchair athletes likewise experience the positive acute effects of exercise on markers of the mucosal immune function. However, the impaired autonomic nervous system in TETRA seems to influence the fine-tuning of their sIgA response when compared with PARA and NON-SCI. This results in a greater increase of sIgA secretion rate in TETRA, which may potentially result in more pronounced immunoprotective effects in this group. Furthermore, the results of this study question the use of α-amylase as a marker of centrally mediated autonomic nervous system activity.
The authors thank Louise Croft and John Lenton for their expertise and contribution during laboratory testing and the support from the Great Britain Wheelchair Basketball Association and the Great Britain Wheelchair Rugby Association. Appreciation is also extended to all sportsmen who volunteered to participate in this study. No direct funding was received for this work, other than the support from the corresponding institution.
The authors are not aware of any conflict of interest.
The results of the present study do not constitute endorsement by the American College of Sports Medicine.
1. Allgrove JE, Gomes E, Hough J, Gleeson M. Effects of exercise intensity on salivary antimicrobial proteins and markers of stress in active men. J Sports Sci
2. Arnold JM, Feng QP, Delaney GA, Teasell RW. Autonomic dysreflexia in tetraplegic patients: evidence for α-adrenoceptor hyper-responsiveness. Clin Auton Res
3. Baydur A, Adkins RH, Milic-Emili J. Lung mechanics in individuals with spinal cord injury: effects of injury level and posture. J Appl Physiol
4. Bishop NC, Gleeson M. Acute and chronic effects of exercise on markers of mucosal immunity. Front Biosci
5. Bishop NC, Walker GJ, Scanlon GA, Richards S, Rogers E. Salivary IgA responses to prolonged intensive exercise following caffeine ingestion. Med Sci Sports Exerc
6. Bloomfield SA, Jackson RD, Mysiw WJ. Catecholamine response to exercise and training
in individuals with spinal cord injury. Med Sci Sports Exerc
7. Borg GA. Psychophysical bases of perceived exertion. Med Sci Sports Exerc
8. Brown R, DiMarco AF, Hoit JD, Garshick E. Respiratory dysfunction and management in spinal cord injury. Respir Care
9. Campagnolo DI, Dixon D, Schwartz J, Bartlett JA, Keller SE. Altered innate immunity following spinal cord injury. Spinal Cord
10. Chicharro JL, Lucía A, Pérez M, Vaquero AF, Ureña R. Saliva composition and exercise. Sports Med
11. Cole AS, Eastoe JE. Biochemistry and Oral Biology
. 2nd ed. London (UK): Wright; 1988. p. 476.
12. Corbett JL, Frankel HL, Harris PJ. Cardiovascular reflex responses to cutaneous and visceral stimuli in spinal man. J Physiol
13. Fahlman MM, Engels HJ. Mucosal IgA and URTI in American college football players: a year longitudinal study. Med Sci Sports Exerc
14. Goosey-Tolfrey V, Castle P, Webborn N, Abel T. Aerobic capacity and peak power output of elite quadriplegic games players. Br J Sports Med
15. Goosey-Tolfrey VL, Batterham AM, Tolfrey K. Scaling behavior of V˙O2peak
in trained wheelchair athletes. Med Sci Sports Exerc
16. Haisma JA, van der Woude LH, Stam HJ, Bergen MP, Sluis TA, Bussmann JB. Physical capacity in wheelchair-dependent persons with a spinal cord injury: a critical review of the literature. Spinal Cord
17. Jacobs PL, Nash MS. Exercise recommendations for individuals with spinal cord injury. Sports Med
18. Karlsson AK, Friberg P, Lönnroth P, Sullivan L, Elam M. Regional sympathetic function in high spinal cord injury during mental stress and autonomic dysreflexia. Brain
. 1998;121(Pt 9):1711-9.
19. Krassioukov A. Autonomic function following cervical spinal cord injury. Respir Physiol Neurobiol
20. Li TL, Gleeson M. The effect of single and repeated bouts of prolonged cycling and circadian variation on saliva flow rate, immunoglobulin A and α-amylase responses. J Sports Sci
21. Lucke KT. Pulmonary management following acute SCI. J Neurosci Nurs
22. MacKinnon LT, Ginn E, Seymour GJ. Decreased salivary immunoglobulin A secretion rate after intense interval exercise in elite kayakers. Eur J Appl Physiol Occup Physiol
23. MacKinnon LT, Jenkins DG. Decreased salivary immunoglobulins after intense interval exercise before and after training
. Med Sci Sports Exerc
24. Nash MS. Immune responses to nervous system decentralization and exercise in quadriplegia. Med Sci Sports Exerc
25. Nater UM, Rohleder N. Salivary α-amylase as a non-invasive biomarker for the sympathetic nervous system: current state of research. Psychoneuroendocrinology
26. Neville V, Gleeson M, Folland JP. Salivary IgA as a risk factor for upper respiratory infections in elite professional athletes. Med Sci Sports Exerc
27. Nieman DC, Henson DA, Austin MD, Brown VA. Immune response to a 30-minute walk. Med Sci Sports Exerc
28. Nieman DC, Henson DA, Fagoaga OR, et al. Change in salivary IgA following a competitive marathon race. Int J Sports Med
29. Proctor GB, Carpenter GH. Regulation of salivary gland function by autonomic nerves. Auton Neurosci
30. Proctor GB, Carpenter GH, Garrett JR. Sympathetic decentralization abolishes increased secretion of immunoglobulin A evoked by parasympathetic stimulation of rat submandibular glands. J Neuroimmunol
31. Sari-Sarraf V, Reilly T, Doran DA. Salivary IgA response to intermittent and continuous exercise. Int J Sports Med
32. Sari-Sarraf V, Reilly T, Doran DA, Atkinson G. The effects of single and repeated bouts of soccer-specific exercise on salivary IgA. Arch Oral Biol
33. Scannapieco FA, Solomon L, Wadenya RO. Emergence in human dental plaque and host distribution of amylase-binding streptococci. J Dent Res
34. Schmid A, Huonker M, Barturen JM, et al. Catecholamines, heart rate, and oxygen uptake during exercise in persons with spinal cord injury. J Appl Physiol
35. Schmid A, Schmidt-Trucksäss A, Huonker M, et al. Catecholamines response of high performance wheelchair athletes at rest and during exercise with autonomic dysreflexia. Int J Sports Med
36. Stjernberg L, Blumberg H, Wallin BG. Sympathetic activity in man after spinal cord injury. Outflow to muscle below the lesion. Brain
. 1986;109(Pt 4):695-715.
37. Victor RG, Bertocci LA, Pryor SL, Nunnally RL. Sympathetic nerve discharge is coupled to muscle cell pH during exercise in humans. J Clin Invest
38. Vissing J, Wilson LB, Mitchell JH, Victor RG. Static muscle contraction reflexly increases adrenal sympathetic nerve activity in rats. Am J Physiol
. 1991;261(5 Pt 2):R1307-12.
39. Walsh NP, Bishop NC, Blackwell J, Wierzbicki SG, Montague JC. Salivary IgA response to prolonged exercise in a cold environment in trained cyclists. Med Sci Sports Exerc
40. Walsh NP, Blannin AK, Clark AM, Cook L, Robson PJ, Gleeson M. The effects of high-intensity intermittent exercise on saliva IgA, total protein and α-amylase. J Sports Sci
41. Woof JM, Kerr MA. The function of immunoglobulin A in immunity. J Pathol
42. Yamanaka M, Furusawa K, Sugiyama H, et al. Impaired immune response to voluntary arm-crank ergometer exercise in patients with cervical spinal cord injury. Spinal Cord
Keywords:©2011The American College of Sports Medicine
TRAINING; ORAL IMMUNE FUNCTION; UPPER RESPIRATORY TRACT INFECTION; WHEELCHAIR BASKETBALL; WHEELCHAIR RUGBY