Medicine & Science in Sports & Exercise:
BASIC SCIENCES: Original Investigations
Salicylate-Based Analgesic Balm Attenuates Pressor Responses from Skeletal Muscle
RAGAN, BRIAN G.1,5; NELSON, AMANDA J.1,2,5; BELL, GERALD W.1,5; IWAMOTO, GARY A.2,3,4
1Athletic Training Education Program, 2Departments of Veterinary Biosciences, 3Molecular and Integrative Physiology, the 4Neuroscience Program, and 5Kinesiology and Community Health, University of Illinois at Urbana-Champaign, Urbana, IL
Address for correspondence: Gary A. Iwamoto, ATC, Ph.D., Department of Veterinary Biosciences, University of Illinois at Urbana-Champaign, 2001 S. Lincoln Ave., Urbana, IL 61802; E-mail: firstname.lastname@example.org.
Submitted for publication April 2007.
Accepted for publication July 2007.
Purpose: Analgesic balms (AB) are widely used in sports medicine. We previously have examined effects of various counterirritant-based AB on pressor responses evoked by muscular contraction (MC), mediated by group III and IV muscle afferents known to produce exercise and nociceptive responses. Our purpose was to examine trolamine salicylate-based analgesic balm (TS) effects.
Methods: Ten healthy, adult male and female cats were used. Decerebration under halothane allowed elimination of anesthesia. Electrical stimulation of L7 and S1 ventral roots evoked static MC (30 s). After control runs, commercial TS (10% concentration) was applied to the skin over the contracting muscles of one hind limb (N = 5). MC was evoked every 10 min, alternating between sides. Ipsilateral (T = 0, T + 20, T + 40, T + 60 min) and contralateral (T − 10, T + 10, T + 30, T + 50 min) responses were analyzed. Five additional cats received AB minus TS.
Results: There were significant attenuations in both peak mean arterial pressure (MAP), in the last 12 s and the last 6 s of the 30 s of MC for both contra- and ipsilateral limbs occurring at T + 50 and T + 60 min after TS application, respectively. No significant changes in heart rate (HR) responses were seen for either the ipsi- or contralateral stimulation. There were no changes in MAP or HR in control cats.
Conclusions: These results indicate that TS affects the end of the 30 s of MC, which is thought to be mainly chemically mediated through group IV afferents. TS represents the salicylate class of AB and has no counterirritant properties. TS works as an inhibitor of cyclooxygenase (prostaglandin formation) and is, at least in part, blood borne.
Athletes and the physically active have a long history of using analgesic balm (AB) for relief of minor aches and pains associated with exercise. Topical analgesic use has grown into a $150 million annual industry, with almost 60 million adults in America using AB at some time (2). Despite their commercial popularity, there are limited data addressing the efficacy and mechanism of action of AB. There are two different general types of AB products: counterirritants and salicylates. By definition, counterirritants irritate the skin, increase cutaneous blood flow, stimulate or depress pain receptors, and stimulate thermoreceptors (2). Salicylates are one form of a number of nonsteroidal antiinflammatory drugs (NSAID). Salicylates are thought to work via the alteration of prostaglandin metabolite formation through the cyclooxygenase pathways (24). There are two major pathways: cyclooxygenase-1 (COX-1), which is found in blood vessels; and cyclooxygenase-2 (COX-2), which is responsible for the production of prostaglandins in settings of inflammation (8).
Previous work from our laboratory has focused on AB and the various active ingredients. We have demonstrated that a combination product containing a strong counterirritant (capsaicin) and a salicylate (methyl salicylate) attenuated the pressor response to muscular contraction (MC) after 20 min of application in decerebrated cats (7). The pressor response evoked by MC is known to be mediated by group III and IV afferents in the contracting muscle, which are associated with both MC and nocioceptive responses. Similar attenuation effects were seen with the individual active ingredients capsaicin (17) and menthol (19), which are both strong counterirritants. One unique finding from our previous work was the difference in latency of effects seen on the pressor response. Both capsaicin and menthol AB alone produced attenuation at 20 min after application, with responses returning to normal at 40 min. The combination product (counterirritant and salicylate) (7) showed attenuation at both 20 and 40 min after application. Although not tested separately, the longer latency effects (40 min after application) were thought to be attributable to the salicylate portion of the AB.
The results of these experiments, using the combination of capsaicin and methyl salicylate in decerebrate cats, were later verified in the exercising human (5). These investigators also showed attenuation effects after 50 min post application to the skin overlying the muscles involved in hand grip exercise. The effects could not be attributable to changes in muscle blood flow or central command.
Salicylates are popular active ingredients in AB, despite the paucity of evidence of their effectiveness. Although we have seen effects on the pressor responses with methyl salicylate in the combination product, the effects of the salicylate could not directly be attributed to the salicylate portion. Methyl salicylate is a common active ingredient in AB that is usually combined with a counterirritant. However, methyl salicylate has counterirritant properties (3,8).Trolamine salicylate (TS), on the other hand, has no counterirritant properties and is routinely the sole active ingredient in commercially available AB. The FDA has classified TS as safe but not proven effective, because of the lack of effectiveness data (4). Thus, the purpose of this study was to examine the effects of TS-based AB on pressor responses evoked by MC. We hypothesized that the application of TS would attenuate the pressor response evoked by MC.
The methods have been described in detail previously (7). All procedures were conducted in conformance with the American College of Sports Medicine policy statement on research with experimental animals and was approved by the University of Illinois institutional animal care and use committee under protocol S9B 250/8349. The University of Illinois at Urbana-Champaign is accredited by the American Association for Accreditation of Laboratory Animal Care (AAALAC) and has an assurance statement on file with the National Institutes of Health (A3118-01).
General animal preparation.
Briefly, 10 adult cats of either sex were used for this experiment. The cats were anesthetized with 1-5% halothane combined with a gaseous mixture of 67% nitrous oxide and 33% oxygen. End-tidal CO2 was kept between 3.5 and 4.5% after a tracheal tube was inserted and connected to a ventilator, which ensured proper airflow. An external jugular vein was cannulated to allow for infusion of fluids. A common carotid artery was cannulated to monitor and measure blood pressure. Heart rate (HR) was established from the blood pressure signal and was calculated beat to beat with a Gould Biotach. Body temperature was maintained at 37°C, aided by a heating pad under the animal and intermittent activation of a standard light bulb shining on the dorsal surface. The hair was shaved over the head, back, and hind limbs, using electrical clippers. The hair was not shaved more closely, to avoid damaging the skin. An L4-L7 laminectomy was performed, to allow eventual access to lumbar and sacral ventral roots. Triceps surae (gastrocnemius and soleus) tension was monitored by attaching a piece of calcaneous bone and calcaneal tendon on both sides to two separate force transducers. Care was taken to minimize the skin incision to the area just over the hock (ankle) region. The animal‘s body was then mounted in a Kopf stereotaxic frame and Goteborg spinal-fixation device. A midcollicular decerebration was carried out, to eliminate the need for anesthetic use and the influence of higher brain centers. After the decerebration, the animals were weaned from the halothane within 15 min and from the ventilator within 2 h. Using a Reichert optical dissection microscope, the L7 and S1 ventral roots were isolated, placed on bilateral platinum hook electrodes, and immersed in a pool of mineral oil. A catheter was inserted, and urine output was monitored throughout the experiment. Figure 1 provides a general diagram of the acute animal preparation.
FIGURE 1-The general...Image Tools
Muscular contraction and trolamine salicylate application.
A stabilization period of 120 min was allowed after all animal preparations and surgery to allow vital signs to return to normal conditions, including HR, blood pressure, and unassisted respiration. After stabilization, L7 and S1 ventral roots were electrically stimulated for 30 s using 0.1-ms pulses at 30 Hz and 3× motor threshold from an AstroMed Grass S48 stimulator. Submaximal stimulus frequencies were used as a precautionary measure against the onset of fatigue. The tension produced by the static contraction of the triceps surae was monitored during each stimulation period. The resting tension of the triceps surae was initially placed at 200-400 g or less than the optimum for tension generation, so that this mechanism might be engaged at a later time to increase evoked tension as possible fatigue was encountered. The general scheme of stimulus presentation for the main experimental intervention is shown in Figure 2. The ventral roots were then stimulated to evoke the pressor response from the static MC. The stimulation periods alternated from the ipsi- to the contralateral side at 10-min intervals, which provided a 20-min recovery period for the muscles on each side. This was done to minimize peripheral fatigue, which would result in reduced tension. When evoked tension is substantially reduced, it attenuates the pressor response (9). A baseline was established from at least three consecutive trials in which consistent muscle tension, mean arterial pressure (MAP), and HR measurements were obtained bilaterally (i.e., the three right-side measurements were consistent with each other, and the three left-side measurements were consistent with each other). A common TS AB (10% trolamine salicylate) (N = 5), which is the common dose available over the counter, or AB minus the active ingredient (TS) (N = 5), was applied approximately 2 mm thick to a cotton cast pad and placed on the caudal thigh and leg on one side within 10 s of the response obtained at T = 0. This area corresponds to the dermatomes L7 and S1, which are well known in at least the dog (13). Light massaging for the initial 10-15 s and maintenance of normal body temperature ensured that the AB came in contact uniformly with the skin. Care was also taken to not apply the AB to open wounds. The alternating stimulation continued for 50-60 min after application, at 10-min intervals. The timing in this experiment was longer than the methods previously used on counterirritants (7,17,19). The monitoring time was extended to 60 min because of the suspected mechanism of action of TS through skin penetration (1). Furthermore, the attenuation effects of 40-min latency, which were seen in our previous work with the combination of counterirritant and salicylate (7), were not present with counterirritants alone (17,19).
FIGURE 2-Diagram of ...Image Tools
Both the ipsi- and contralateral responses were analyzed using repeated-measures ANOVA with a within-subject factor of time (ipsilateral: T = 0, T + 20, T + 40, T + 60; contralateral: T − 10, T + 10, T + 30, T + 50) for two dependent variables (MAP and HR). Tukey‘s post hoc tests were performed on significant findings. Significance was set at P < 0.05. The effects of TS application on the ipsilateral side were assessed with respect to the overall peak response, regardless of where it occurred in the time course of the contraction. Because the type of afferent input is dependent on the time during the stimulation bout, in which the effects were measured, the response seen in the first 12 s, the last 12 s, and the last 6 s of the 30-s stimulation period were analyzed. The last 6 s of the stimulus period were examined to ensure that the MAP and HR responses were largely mediated by chemical events.
Figure 3 illustrates a single example of ipsilateral MAP and HR responses to static muscular contraction (MC) before and after TS application. In this case, changes in peak MAP induced by static ipsilateral MC were profoundly attenuated at the end of the response obtained at T + 60. Figure 4 provides a graphical representation of the TS effects on the peak responses of five cats, for the overall peak, first 12 s, last 12 s, and last 6 s of the responses during the 30-s stimulus period. There was a significant difference in overall peak MAP (F3,12 = 5.33; P < 0.05) at T + 60 min on the ipsilateral side. The mean (M ± SEM) of the peak MAP for the ipsilateral side just before application T = 0, T + 20, T + 40, and T + 60 min were 35 mm Hg ± 7, 31 ± 9, 28 ± 5, and 22 ± 4,respectively.
Consideration of afferent input timing.
The type of afferent fiber activity corresponds to specific time frames within a stimulation period, with mechanically sensitive afferents responding at the onset of MC, and chemically sensitive afferents responding later (11,12,14). There was no significant difference in peak MAP during the first 12 s of the response (F3,12 = 2.71; P > 0.05; power = 0.51). Significant differences were found in the last 12 s and the last 6 s responses for MAP at T + 60 min (last 12 s: F3,12 = 4.33; P < 0.05; last 6 s: F3,12 = 4.03; P < 0.05). The mean (M ± SEM) of the peak MAP for the ipsilateral side in the last 12 s were T = 0 = 33 mm Hg ± 8, T + 20 = 28 ± 4, T +40 = 21 ± 4, and T + 60 = 15 ± 5. The peak MAP responses at the very end (last 6 s) were T = 0 = 34 mm Hg ± 8, T + 20 = 27 ± 3, T + 40 = 20 ± 2, and T + 60 = 12 ± 5. There were no significant differences in HR observed with any condition.
No differences were found in overall peak MAP (F3,12 = 2.56, P > 0.05). The afferent timing results also showed no differences in the peak MAP during the first 12 s (F3,12 = 1.57, P > 0.05). There were significant effects observed on the peak MAP responses during the last 12 s and the last 6 s of the muscle stimulation period at T + 50 min after application (last 12 s: F3,12 = 3.48, P < 0.05; last 6 s: F3,12 = 4.29, P < 0.05) on the side on which no TS was applied (Fig. 5). The mean (M ± SEM) of the peak MAP for the contralateral side in the last 12 s were T − 10 = 26 mm Hg ± 6, T + 10 = 27 ± 7, T + 30 = 19 ± 4, and T + 50 min = 10 ± 3. The mean values of the peak MAP for the end response (last 6 s) for T − 10, T + 10, T + 30, and T + 50 min were 23 ± 6, 23 ± 7, 16 ± 5, and 6 ± 2, respectively.
FIGURE 5-Results fro...Image Tools
In five control animals, no significant differences in MAP responses were identified during the 60 min of testing. This demonstrates that the AB base was not the cause of the observed attenuation.
The results demonstrate that salicylate-based AB without counterirritant properties can attenuate pressor responses evoked by MC similar to those effects seen with the counterirritants, capsaicin (17), and menthol (19). There was a decrease in the overall peak MAP response at 60 min after application of TS on the ipsilateral side. The time course of the effects from the TS was longer than that seen with counterirritants. Past AB work with counterirritants (17,19) has shown significant overall peak attenuation at 20 min after application, but the response returned to baseline at 40 min after application. This difference in time course may be attributed to the differences in the putative mechanism of action of the active ingredients. Counterirritants produce strong sensory stimulation quickly, with the effects not long lasting. Counterirritants are thought to hyperstimulate and elicit an opiate system response (15,16). Conversely, salicylates must penetrate the skin rather than stimulating its surface, and they are thought to limit the production of prostaglandins through the cyclooxygenase pathways (24).
These differences suggest a delay in the latency of the effects of salicylates versus counterirritants. Furthermore, these differences help explain our previous observations in which a combined product containing counterirritant and salicylate was tested (7). Whereas a combined product had a short latency effect, this was likely attributable to the counterirritant portion of this AB. However, it is likely that the response at 40 min was a combination of a waning counterirritant effect (17) and, perhaps, the beginning of the salicylate effect. In the present study, which examines a pure salicyate effect, it is likely that the effects become statistically significant only at 50-60 min. We and others have theorized that counterirritants act by closure of spinal- and/or brainstem-mediated gate control mechanisms (3,7,17,19). The counterirritant component, which consists of capsaicin, may also deplete afferents of peptides or neurotransmitters (6,10). Salicylates have been shown to penetrate the dermis and to work subcutaneously over time (23,24). It seems likely that the TS penetrated the skin to have some effects on sensory nerve endings through both local penetration and blood-borne routes, presumably on prostaglandin metabolite formation (1). The presence of a contralateral effect certainly could not be attributed to a direct local penetration. Further, the CNS was likely not affected directly by penetration by crossing the blood-brain barrier (18). The most likely remaining route for contralateral effects would, then, be blood-borne effects on peripheral receptors.
The effects were also apparent during the last 12 s of the MC stimulus, indicating that the effects of TS may target group IV afferents, which are stimulated by chemical events (11). This selectivity to chemical processes and events was also evident, with the significant reduction of mean peak MAP during the last 6 s of the response. This time period was chosen to further ensure that the responses would be mediated largely by group IV afferents. It is known that the cyclooxygenase pathway and prostaglandins contribute to pressor responses evoked by MC (25). Furthermore, cyclooxygenase blockade has been shown to attenuate the responses of group IV muscle afferents to static MC (20). This selectivity for chemical events is further supported by the lack of attenuation seen during the initial 12 s of the MC stimulus, which, in turn, is thought to be primarily activating group III afferents, which signal largely mechanical events. It is known that group III sensitivity can be affected by prostaglandin levels, because their responses increase when exposed to arachidonic acid (prostaglandin precursor) (21). Thus, it is somewhat surprising that no changes were observed during the first 12 s of the response to MC. However, taken together, these data clearly show that the TS-based AB can have attenuation effects, which are likely related to an effect of cyclooxygenase inhibition.
The significant contralateral effects on the last 12 s of the response on the contralateral side were unlike the results from the counterirritant studies in which no significant contralateral effects were identified. There were only attenuation effects seen in the last 12 s (chemical) of the contralateral response. This also supports the proposed mechanism of action of salicylate-based AB, acting on chemical events such as prostaglandin metabolite formation. There were no significant effects identified on portions of the response thought to be mediated by nonchemical events. As previously mentioned, these results suggest that, at least in part, the TS affected the contralateral side by being blood borne to the contralateral limb.
On the basis of the work of Dawson etal. (5), in humans it is unlikely that the attenuations could be attributable to changes in overall limb blood flow. However, another factor in these observations could be localized skin blood flow. It has been reported to increase three to four times with AB application (22), which would seem to be an effect of topically applied AB products. Skin blood flow changes could alter or compound the behavior of sensory receptors. It is well known that changes in skin temperature (determined in part by skin blood flow) can lead to changes in sensory circuits and may even lead to analgesic effects (3). However, AB does not lead to temperature changes (26).
Although there were significant contralateral effects on the TS cats, there were no changes in the responses of the control cats that had the AB minus the active ingredient TS for the 60-min time course. Thus, it is likely that the effects were from the TS and not simply the condition of the cats through the longer experiment.
Type of pain and analgesic balms.
The time course and presence of effects in the TS cats in the present study are also important for practical reasons. It seems that the different active ingredients used in AB may target different portions of the response and at different latencies. Both capsaicin and menthol effects are quickly demonstrated by the attenuation of the overall peak responses at 20 min but are not long lasting, with responses returning to baseline levels at 40 min after application. These data, with the longer effects seen with the combination product (1% CAP, 12.5% methyl salicylate) at 40 min (7), 50 min (5), and the results from the present study with TS effects at 50 min and 60 min, seem to indicate that formulations could be developed to produce both immediate and long-lasting effects from AB. Also, products may be potentially developed to target specific afferents and pain, using the knowledge of the portion of the response the active ingredient effects. The last 12 s of the response to MC thought to be mediated by group IV afferents has been shown to be responsive to AB along with overall responses. Although the first 12 s of the response (group III) has not been significantly attenuated with any AB, capsaicin decreased the first 12-s responses from 20.7 ± 3.6 mm Hg initially to 12.7 ± 2.6 mm Hg at 20 min after application (17). The future may hold the tailoring of AB to specific types of pain conditions.
We conclude that the pressor responses caused by activation of group III and IV muscle afferents can be significantly attenuated by the application of AB containing the active ingredient TS. Changes in peak MAP induced by static ipsilateral MC were significantly attenuated at 60 min. This time course of action differs considerably in comparison with our previous work, which investigated specific counterirritants (17,19). However, the data seem consistent with those observed in an experiment in which a combination of counterirritant and salicylate was tested (7). It is likely that TS exerts its effects through an inhibition of cyclooxygenase-specifically, COX-2. This further suggests that the effects from TS are slower to develop and, thus, take effect when the counterirritant effects are waning.
We gratefully acknowledge Chattem Inc. for providing the inactive control substance used in this study. This research was supported in part by National Heart, Lung and Blood Institute grants HL 37400 and the University of Illinois Athletic Training Research Fund.
1. Baldwin, J. R., R. A. Carrano, and A. R. Imondi. Penetration of trolamine salicylate into the skeletal muscle of pig. J. Pharm. Sci.
2. Barone, J. N. Topical analgesics: how effective are they? Phys. Sportsmed. (Minneapolis)
3. Bell, G. W., and W. E. Prentice. Infrared modalities. In: Therapeutic Modalities in Rehabilitation
, W. E. Prentice (Ed.). New York, NY: McGraw Hill, 2005, pp. 290-292.
4. Center For Drug Evaluation And Research. OTC Ingredient List P-Z Updated August 2006 Ingredient Review Panel Report Drug Category
, Rockville, MD: U.S. Food and Drug Administration, 2006.
5. Dawson, A. N., B. Walser, and M. Jafarzadeh. Topical analgesics and blood pressure during static contraction in humans. Med. Sci. Sports Exerc.
6. Dray, A. Mechanism of action of capsaicin-like molecules on sensory neurons. Life Sci.
7. Ichiyama, R. M., B. G. Ragan, G. W. Bell, and G. A. Iwamoto. Effects of topical analgesics on the pressor response evoked by high threshold muscle afferents. Med. Sci. Sports Exerc.
34: 1440-1445, 2002.
8. Insel, P. A. Analgesic, antipyretic, and anti-inflammatory agents and drugs employed in the treatment of gout. In: Goodman and Gilman‘s The pharmacological basis of therapeutics
, J. G. Hardman and L. E. Limbird (Eds.). New York, NY: McGraw-Hill, 1996.
9. Iwamoto, G. A., and B. R. Botterman. Peripheral factors influencing expression of pressor reflex evoked by muscular contration. J. Appl. Physiol.
10. Jessel, T. M., L. L. Iversen, and A. C. Cuello. Capsaicin-induced depletion of substance P from primary sensory neurones. Brain Res.
11. Kaufman, M. P., and H. V. Forster. Reflexes controlling circulatory, ventilatory and airway response to exercise. In: Handbook of Physiology, Section 12: Exercise: Regulation and Integration of Multiple Systems
, L. B. Rowell and J. T. Shepherd (Eds.). New York, NY: Oxford University Press, 1996, pp. 381-446.
12. Kaufman, M. P., J. C. Longhurst, K. J. Rybicki, J. H. Wallach, and J. H. Mitchell. Effects of static muscular contraction on impulse activity of groups III and IV afferents in cats. J. Appl. Physiol.
13. Kitchell, R. L., and H. E. Evans. The spinal nerves. In: Anatomy of the Dog
, H. E. Evans (Eds.). Philadelphia, PA: W.B. Saunders, 1993, pp. 829-893.
14. Kniffki, K. D., S. Mense, and R. F. Schmidt. Responses of group IV afferent units from skeletal muscle to stretch, contraction, and chemical stimulation. Exp. Brain Res.
15. Lebars, D., A. H. Dickinson, and J. M. Besson. Diffuse noxious inhibitory controls (DNIC). I. Effects on dorsal horn convergent neurons in the rat. Pain
16. Lebars, D., A. H. Dickinson, and J. M. Besson. Diffuse noxious inhibitory controls (DNIC). II. Lack of effect on non-convergent neurones, supraspinal involvement and theoretical implications. Pain
17. Nelson, A. J., B. G. Ragan, G. W. Bell, R. M. Ichiyama, and G.A. Iwamoto. Capsaicin based analgesic balm decreases pressor responses evoked by muscle afferents. Med. Sci. Sports Exerc.
18. Parepally, J. M. R., H. Mandula, and Q. R. Smith. Brain uptake of nonsteroidal anti-inflammatory drugs: ibuprofen, flubiprofen, and indomethacin. Pharm. Res.
19. Ragan, B. G., A. J. Nelson, G. W. Bell, J. H. Foreman, and G. A. Iwamoto. Menthol-based analgesic balm attenuates pressor responses evoked by muscle afferents. Am. J. Vet. Res.
20. Rotto, D. M., J. M. Hill, H. D. Schultz, and M. P. Kaufman. Cyclooxygenase blockade attenuates the response group IV muscle afferents to static contraction. Am. J. Physiol.
21. Rotto, D. M., H. D. Schultz, J. C. Longhurst, and M. P. Kaufman. Sensitization of group III muscle afferents to static contraction by products of arachidonic acid metabolism. J. Appl. Physiol.
22. Shellock, F. G. Effect of a topically applied counterirritant/analgesic on skin blood flow. Med. Sci. Sport Exerc.
23. Singh, P., and M. S. Roberts. Skin permeability and local tissue concentrations of non-steroidal anti-inflammatory drugs after topical application. J. Pharm. Exp. Ther.
24. Singh, P., and M. S. Roberts. Dermal and underlying tissue pharmacokinetics of salicylic acid after topical application. J. Pharmacokinet. Biopharm.
25. Stebbins, C. L., Y. Maruoka, and J. C. Longhurst. Protaglandins contribute to cardiovascular reflexes evoked by static muscular contraction. Circ. Res.
26. Trowbridge, C. A., D. O. Draper, J. B. Feland, L. S. Jutte, and D. L. Eggett. Paraspinal musculature and skin temperature changes: comparing the Thermacare HeatWrap, the Johnson & Johnson Back Plaster, and the ABC Warme-Pflaster. J. Orthop. Sports Phys. Ther.
SALICYLATES; THERAPEUTIC MODALITIES; PAIN; NSAID
©2007The American College of Sports Medicine
What does "Remember me" mean?
By checking this box, you'll stay logged in until you logout. You'll get easier access to your articles, collections,
media, and all your other content, even if you close your browser or shut down your
To protect your most sensitive data and activities (like changing your password),
we'll ask you to re-enter your password when you access these services.
What if I'm on a computer that I share with others?
If you're using a public computer or you share this computer with others, we recommend
that you uncheck the "Remember me" box.
Highlight selected keywords in the article text.
Data is temporarily unavailable. Please try again soon.