The Effect Of Atropine On Respiratory Sinus Arrhythmia Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

RSA is the change in heart rate seen throughout the breathing cycle; heart rate increases upon inspiration and decreases on expiration. It is measured by an E: I ratio and this is calculated by dividing the average R-R interval on expiration by the average R-R on inspiration. Twenty volunteers were recruited and given oral atropine at a dose of 20mcg/kg (n=10) or placebo (n=10), whilst attached to an ECG and given a spirometer to measure volume and flow changes in the subject. 6 breathing cycles were performed over a minute, with each cycle of inspiration and expiration lasting ten seconds. Readings were taken every ten minutes pre-dose and every 15 minutes post dose for a total of three hours. The results showed a significant increase in heart rate and a significant decrease in E: I ratio in subjects receiving atropine relative to those receiving placebo. The E: I ratio at the peak of atropine activity was 1.23 (SD=0.05, n=10) in the control and 1.11 (SD=0.05, n=10) in the atropine group. A negative correlation is seen between the change in heart rate and E: I ratio, as the heart rate increased the E: I ratio decreased, showing that a decrease in RSA is seen when oral atropine is administered. A secondary hypothesis was tested, suggesting that RSA changed throughout the day. This hypothesis was inconclusive as data proved insignificant, indicating that further investigation is required.


Respiratory Sinus Arrhythmia (RSA) is the natural occurrence of an increase in heart rate during inspiration and decrease in heart rate during expiration. Atropine is a drug known to increase heart rate by inhibiting muscarinic receptors. This study attempts to discover whether human respiratory sinus arrhythmia (RSA) is affected by oral atropine, whether they are correlated, and also in what manner it is altered.

Heart rate control

Figure 1 Overview of heart rate innervations.The heart is the organ responsible for pumping blood around the body. It collects deoxygenated blood from the body in the right atrium, pumps it into the right ventricle and on to the lungs so that carbon dioxide can be exchanged for oxygen. The oxygenated blood then returns to the left atrium, where it is pumped to the left ventricle and is pumped round the body. Although the heart is myogenic, a region called the sinoatrial (SA) node sets the rate that all the cells contract at. Every heartbeat is started when the SA node sends an impulse through the heart to another area of specialised tissue called the atrioventricular (AV) node, where the signal is delayed, before the ventricles contract. The delay allows the atria to empty and the ventricles to fill. The heart is innervated by sympathetic and parasympathetic (vagal) nerves that are regulated by the medulla of the brain. In the medulla, the nucleus tractus solitaris (NTS) receives input from baroreceptors and chemoreceptors around the body, so that the heart rate can be altered in accordance to the body's needs. The heart rate is controlled by stimulating sympathetic or vagal nerve fibres. Sympathetic stimulation increases the heart rate, whereas vagal stimulation decreases the heart rate. Vagal stimulation also acts on the sympathetic nerves to inhibit its activity, and this is reciprocated [1] (Figure 1).

The electrical impulses that the SA node and AV node produced can be measured, non-invasively, using an electrocardiogram (ECG), in which electrodes are placed around the body at specific points. The ECG measures the voltage between two sets of electrodes. In a healthy individual an ECG produces a trace with 5 waves; P,Q,R,S and T waves. These are usually grouped into; a P-wave, a QRS complex and a T-wave. The P-wave indicates SA node activity, the QRS complex indicates depolarisation of the left and right ventricles, the measurement of two R-peaks is the length of one heartbeat, and the T-wave indicates ventricular repolarisation. An ECG is used for diagnosing abnormal heart conditions or ensuring that the heart is beating correctly. Abnormal ECGs can indicate electrolyte dysfunctions and other cardiac problems such as myocardial infarction and atrial fibrillation1.

Respiratory sinus arrhythmia (RSA)

Respiratory sinus arrhythmia (RSA) is a natural phenomenon where the changes in heart rate coincide with changes in the breathing cycle. On inspiration the heart rate increases and upon expiration heart rate decreases. This is shown on an ECG by measuring R-R intervals and monitoring the changes during breathing (Figure 2). The longer the R-R interval is the slower the heart rate. Many factors affect RSA such as age, posture, disease state and frequency and depth of breathing. There is a negative correlation between age and variation of heart rate [2] , such that the magnitude of RSA decreases with age. The magnitude of RSA is considered to be controlled by feedback from baroreceptors [3] and volume receptors, due to blood flow and pressure changes on stretch receptors, combined with interactions between respiratory and cardiac centers in the medulla [4] . Posture plays a large part as when standing RSA is less pronounced than when laying horizontally [5] , [6] . A poor amount of sleep causes an erratic RSA [7] , possibly due to an alteration in circadian rhythm. Disease state is a key factor because a condition such as asthma can cause the magnitude of RSA to increase due to the increased parasympathetic drive [8] . However in a cardiac disease state such as myocardial infarction, a reduced RSA is seen due to poor vagal tone, which is clinically connected to a high mortality rate [9] . Frequency and depth of breathing also plays an important role in RSA magnitude, as rapid, shallow breathing produces a less evident variation than deep breathing [10] . Heart period variability and heart rate variability are defined as the changes in R-R intervals, and is a broad term that encompasses RSA.

Figure 2- A normal RSA trace, showing volume of the breath (blue line), the ecg trace (green line) and the R-R (pink line). As the subject breathes in (the blue line rising) the r-r intervals decrease and increase on expiration.

RSA is measured using an expiration: inspiration (E: I) ratio; this is the ratio of the average R-R interval during expiration compared to the average R-R interval during inspiration [11] . The closer the E: I ratio is to one the more the R-R intervals during expiration and inspiration are becoming equal. Conversley; the larger the ratio, the greater the amplitude of RSA. Other ways of measuring RSA are the mean heart rate range, which is the average heart rate for inspiration taken away from the expiration, and the ratio of the shortest R-R interval to the longest R-R interval.

Theories on Why RSA occurs

Heart rate changes are controlled by the medulla oblongata, which contains the nucleus ambiguus. The nucleus ambiguus has parasympathetic neurons for the heart, which are cardioinhibitory, and allow quick changes in blood pressure. This increases parasympathetic input to the vagus nerve, which in turn slows down the rate of SA node firing. During expiration heart rate decreases due to activation of cells in the nucleus ambiguus activating the parasympathetic drive. During inspiration heart rate increases as inhibitory signals are received from baroreceptors, and so the parasympathetic drive is suppressed. Reasons for this change in heart rate are also thought to be due to inspiration causing an increase in the volume of the lungs, which also constricts blood vessels, so the heart rate must increase to keep the pressure steady. There are many theories given for why the body needs to alter its heart rate throughout the breathing cycle. One theory is that RSA improves the energy efficiency of gas exchange, [12] as during inspiration blood flow into the vena cava (of the heart) is increased due to decreased intrathoracic pressure, caused by movement in the diaphragm. This increase in volume would require a faster heartbeat to pump the extra blood around the body, and during inspiration the lung volume is also larger so the surface area of the alveoli are also larger which would improve gas exchange if there was a larger volume of blood [13] . Another theory is that RSA saves energy by "saving heartbeats" [14] , as the body expends less energy by suppressing the heart rate on expiration, this links in with the previous theory that it is more efficient. Both these theories may be true and combine to make respiration and pulmonary gas exchange more efficient.

Control Of RSA

Literature on how RSA is controlled is varied as there appears to be more than one system that contributes to RSA control. It was initially thought that RSA is vagally controlled with muscarinic receptors, and there appears to be evidence that this is the case. There is a linear correlation between heart rate variation and parasympathetic control [15] , however there is increasing support that nicotinic receptors and acetyl choline (ACh) are involved in RSA [16] . Studies have shown that ACh inhibits cardioinhibitory parasympathetic neurons [17] , which decrease cardiac activity and help alter RSA. Neff et al, shown that γ-aminobutyric acid (GABA), an inhibitory neurotransmitter, increased during inspiration in rats, and so is linked to RSA. It was decreased when a nicotinic antagonist was added16. However this experiment was conducted on rats, and it has not been proven that rats have a similar RSA to mammals; for this reason the credibility of the paper is questionable. There have been other papers that suggest that the magnitude of RSA is affected by adrenergic receptors. Thus in a study by Pitzalis, et al [18] , two different beta-blockers were trialled and the effect on the RSA was measured. The results from this showed that beta blockade enhanced RSA. This is also the conclusion drawn from other studies [19] , [20] . However, other works have concluded that an adrenergic blocker has little or no effect on RSA and heart rate variablilty [21] . If RSA is under adrenergic control, then the atropine administered in this trial will have no effect on RSA, indirectly this trial may add to the evidence, proving or disproving vagal control of RSA.

Polyvagal theory

Another theory that has been put forward to explain the conditions when RSA is not directly correlated to vagal inhibition is the polyvagal theory. It attempts to solve the Vagal Paradox (Table 1), which is an inconsistency based on one vagus verve. This concept was introduced in 1995 by Porges [22] as an evolutionary step for mammals, from reptiles who still have the single vagus. Porges suggests that there are 2 types of nerve within the vagus nerve itself, one vegatative vagus, that originates in the dorsal motor nucleus (DMN), and the smart vagus that originates in the

Table 1. The Vagal Paradox

1. Increased vagal tone produces neurogenic bradycardia

2. Decreased vagal tone produces suppression of RSA

3. Bradycardia occurs during periods of suppressed RSA

nucleus ambiguus (NA). In terms of RSA, the theory is that both sets of vagii terminate in the SA node, however one set of nerves mediate neurogenic bradycardia, and another set mediates RSA. It is hypothesized that the DMN vagus is responsible for neurogenic bradycardia, and the NA vagus mediates the amplitude of RSA. It also speculated that the two types of vagii compete with each other, thus ensuring that the DMN vagus does not cause the heart rate to go too low, and so the NA vagus protects the heart and body. This has been proven in some studies as there is evidence of a primitive vagal system in reptiles, shown by little or no respiratory variation in lizards [23] . However this hypothesis is under scrutiny as other research articles have argued that there are inaccuracies within its logic. The polyvagal theory suggests that reptiles have no RSA because they have not evolved the second vagus nerve, however there is literature to support the opposite, as studies have found cardio-respiratory coupling in reptiles [24] , [25] . Grossman and Taylor [26] argue that the DMN nerve is only mediated under experimental irritation, and there is no evidence of DMN affecting vagal heart rate under normal circumstances. They state that RSA can still dissociate from vagal tone such that the polyvagal theory cannot explain. It concludes that RSA is a measure of final vagal effects, after any other processes have had their effect. This is a fair conclusion, however it does not provide an insight into the mechanisms that underlie RSA.


Blocking the vagal nerve increases heart rate, and a naturally occurring agent that does this is atropine. Atropine is an antagonist which competes for muscarinic receptors and increases SA node firing by blocking the vagal nerve, thereby increasing heart rate. It is an anticholinergic agent that has two isomers, the levo form being the active form of the drug. The therapeutic use for it is ophthalmic preparations used in dilating the pupil for surgery. Atropine is licensed for use as a treatment for salivation, lacrimation, urination, diaphoresis, gastrointestinal motility, emesis (SLUDGE) symptoms caused by organophosphate poisoning. The dose is very high at 2mg for adults and 20 mcg/kg for children [27] i.v. or i.m. An overdose of atropine can result in tachycardia, decreased sweating, blurry vision, decreased lacrimation and vasodilation. These symptoms can be reversed by physostigmine or pilocarpine.

Literature states that atropine does not produce an effect on respiratory parameters; one such article suggests that changes in vagal tone do not cause a change in respiratory rate or tidal volume [28] . This is important as these factors have well-defined influences on RSA. Therefore changes in these respiratory parameters do not need to be taken into account after atropine administration. The trial does not use a fixed breathing rate or depth; it was dependant on how comfortable the subject felt. This is variable between individuals, and this has an effect on the magnitude of RSA. This study also used propranolol as an adrenergic block, which could have affected tidal volume and respiratory rate. The sample size was ten, and there were no calculations done to confirm that this would produce significant results.

Primary Hypothesis

This study is based upon the hypothesis that oral atropine will have a significant effect on RSA and override the phenomena, leveling out the heart rate but at an increased level. I have arrived at this theory as there is literature that backs up a decrease in RSA once atropine has been introduced. In a paper investigating the effect of atropine on R-R intervals [29] , it was found that the maximum response was a 97% decrease in the ratio of R-R intervals during inspiration and expiration, and a dose response curve, which shows increasing the atropine dose causes a greater decrease in RSA. It used many measures of RSA including the ratio of the longest to the shortest R-R interval, ratio of the 15 and 30th intervals and mean R-R intervals, these showed significant decreases in RSA. However this study was conducted on ten volunteers and a power calculation was not performed, so a larger sample size may be needed to support the reliability of the results. Wheeler, et al [30] have also noted that heart rate variability was abolished when atropine was administrated; however this paper has limited credibility as it focused on cardiac denervation in diabetics. A significant outcome of this paper was that heart rate variation was reduced in diabetics that suffered from autonomic neuropathy, implying that vagal innervations are essential for RSA. Unfortunately the controls used were two healthy volunteers, which suggest the results are not significant as a small sample size was used, and the volunteers were on different drugs. This paper does not add much to the argument that atropine reduces RSA, however it does argue that heart rate variability is vagally controlled. A study done on rhesus macaque monkeys confirmed atropine significantly reduced RSA three hours after administration4. The heart rate for this study was increased at all doses, along with heart period variance. Although the study was done on monkeys, they are mammals and are a good substitute to human mammalian RSA. The baseline for this study was taken over four weeks ensuring it was a reliable, and the dosing was done per kilogram of body weight. However it found the amplitude of some P-waves fluctuated and even reversed, which could be explained by potassium fluctuations and cardiac arrhythmias indicating unhealthy subjects.

Secondary Hypothesis

A secondary hypothesis is investigated; RSA is different in the morning when compared to the afternoon. This hypothesis is based on circadian changes throughout the day, which could have an effect on RSA. For example blood pressure fluctuates throughout the day and is found to be higher in the morning than in the evening [31] - [32] . As RSA is thought to be due to blood pressure changes, diurnal variation could be an important factor. It has also been found that FVC is at its peak during the afternoon and declines slightly throughout the day [33] . This may also have an effect on RSA as deep, slow breathing produces a more pronounced RSA curve with greater amplitude than rapid shallow breathing [34] . It is even thought that heart rate itself varies throughout the day, being higher during the night and lower during the day [35] , which can directly affect the R-R intervals and alter results. Current articles suggest that RSA is at its lowest during the day and gradually increases to its peak during the night. One such paper looked into circadian rhythm of patients with coronary artery disease (CAD), and concluded that in both sets of data heart rate variability increased during the night and declined during the day, but the effect was significantly pronounced in the healthy volunteers [36] , [37] . The study was conducted on 40 people, of similar age. Beta-blocking therapy had been withdrawn in these patients so it did not interfere with results. The healthy volunteers were used as a control but provided useful information to generate a hypothesis. Furlan, et al [38] attempts to explain the changes in circadian rhythms. It suggests that there is an increase in vagal activity throughout the night, coupled with a decrease in sympathetic activity that corresponds to sleep. Upon waking the vagal system decreased with an increase in sympathetic activity. It is stated that physical activity is a major factor in the morning increase in atrial pressure, a marker of sympathetic tone, implying that physical activity could mediate sympathetic drive and decrease heart rate variability.



Before the trial was conducted ethical approval was sought for and given by the University of Nottingham Medical School Ethics Committee. Twenty subjects, ages 18-25, were initially chosen to participate; they had to fulfill criteria that were specified in the questionnaire (Appendix 1) and go through screening to ensure there were no unusual parameters present. At least 8 hours sleep was recommended and caffeine, alcohol, strenuous physical activity and any food were prohibited for 3 hours before testing. Smokers were not included in this trial, which meant that subjects smoking more than 5 cigarettes per week were excluded. Any potential subjects on long-term medication, were excluded from the trial. Signed consent was obtained before study participation was carried out (Appendix 2). In our initial twenty subjects we found that two were unsuitable for testing; one produced an irregular ECG, and the other struggled to maintain inspiration and expiration for six full cycles so data was difficult to extract. We recruited two more volunteers, who were more suitable and discarded the data that was not usable from the two unsuitable subjects.

Screening Procedure

The subjects were given an information sheet (Appendix 3), outlining what the trial would involve, and then were screened for regular peak flow values, a regular blood pressure, a suitable BMI, and vital lung capacity. Peak flow was assessed using a Peak Flow Mini Wright Standard Peak Flow Meter EU scale (Clement Clarke International). This was done three times and the maximum value checked against a nomogram (Appendix 4). The nomogram gives an indication of whether the airways are constricted, and suggest severity. Blood pressure was measured using an inflatable cuff and a stethoscope, and an electronic blood pressure monitor (Seinex SE-9400 Full Auto Arm Blood Pressure Monitor). The exclusion values were <90/60mmHg or >140/90mmHg. These values were chosen as blood pressure over 140/90 is considered hypertension by NICE [39] . BMI was obtained using the subjects height and weight (digital scales used: Salter, Model Number 915) subjects were excluded if their BMI was outside the range of 18.5-27.5. Subjects were told to expel as much air as they could into a Vitalograph Spirometer Gold Standard, so their forced expired volume in one second (FEV1) and forced vital capacity (FVC) could be obtained this was done three times and the maximum value used. This was done so that lung function could be assessed for any obstructive or restrictive lung diseases. The screening criterion was summarized in a document (Appendix 5), these criteria were chosen to characterize a healthy individual, and exclude factors that may influence RSA.

Experimental Design

Experimental Procedure

Figure 3. The set up of the spirometer, the subject places their mouth over the mouthpiece and inhales and exhales. While an ECG is attached to them. An MLA1026 re-useable mouthpiece (ADInstruments) was attached to a MLA304 disposable filter. This was then attached via clean bore tubing, to a MLT1000L Respiratory Flow Head 1000L ADInstruments. The flowhead was connected to ML818 spirometer PowerLab 15T-Data acquisition system (ADInstruments) and this was linked to a computer with LabChart 7.03 Software (ADInstruments) to pick up the spirometry readings and the ECG trace (Figure 3).

The equipment was calibrated before each subject began; the flow head was standardized by using the Vitalograph 1 litre precise syringe to force one litre of air through the flow head and the trace adjusted in the computer. The subjects were then brought in and seated for ten minutes to allow heart rate to level out.

The subject was attached to an electrocardiogram (ECG) and given the spirometer to gradually breathe through when readings were being taken. The ECG was a three lead ECG, with one electrode on each shoulder, and one on the right ankle, taking care to avoid any large areas of muscle. The subject was told to lie on a bed which was propped up at a 45 degree angle. They were asked to relax and not to move whilst the readings were being taken, as this affected the ECG readings. Before readings were taken, the spirometer pod was calibrated, and a short test run conducted to ensure accuracy. A PowerPoint presentation instructing the subject to 'breathe in' gradually for five seconds, and 'breathe out' gradually for five seconds at regular intervals was shown to the subject. The readings lasted one minute, which encompassed six full cycles of inspiration and expiration.

Readings were taken at -30, -20, -10, 15, 30, 45, 60, 75, 90, 105, 120, 135 and 150 minutes (Figure 4) using a stopwatch. Subjects were given nose clips to ensure air inspired or expired was only through the mouth. At time 0 the atropine or the placebo was given orally in 200ml of apple juice to disguise any taste which would distinguish either agent. The dose of atropine given was 20mcg/kg, which is a clinically safe dose for an adult, as atropine is given orally at 0.6-1.2mg as an antispasmodic [40] . This is outlined in a procedure document given to people conducting the trial (Appendix 6). The subject was allowed to have a maximum of 500mls water to drink.

Figure 4- Timeline of readings taken (blue arrows) and atropine or placebo given (time 0)

The trial was double blind placebo controlled, so neither the subjects nor the people running the trial knew which subjects were to have atropine or placebo until the trial had ended. The trial lasted three hours in total, subjects were trialled in the morning (10am-1pm) or the afternoon (2pm-5pm), two subjects per time slot.

Data Analysis

The intervals between each R-wave, the R-R interval, for each heartbeat were measured in miliseconds. These were recorded and split into expiration (E) and inspiration (I). E:I ratios for the six inspirations and expirations over the minute were calculated by dividing the average expiration by the average inspiration. The heart rate for each minutes reading was calculated by measuring R-R intervals. With the data collected, standard deviations were performed to allow the F-test to be performed. The F-test indicated whether there was equal variance or not. If there was equal variance then a T-test was performed. If one of the values displayed unequal variance, a modified T-test was performed.


Group Demographic

The groups were split, to try and obtain an equal number of each population subtype:



Age (18-25)



Male Gender



Female Gender



Caucasian Ethnicity



Asian Ethnicity



Heart Rate

The data collected based on the heart rate, displayed an overall increase in the atropine group when compared to the control. The data indicates a peak 90 minutes after administration, an increase from 74.7±8.4 beats per minute (bpm) (SD= 8.4, n=10) one reading before dosing, to 84.7 bpm (90 minutes, SD=14.8, n=10), an increase of 13.5%, after which a decrease in heart rate is seen. When comparing the values at time 90 between the control (65.8, SD=5.5, n=10) and atropine the difference is 18.9 bpm, an increase of 28.7%.

Figure 5- The changes in heart rate over the course of the experiment. The atropine appears to take effect after 30 minutes and peaks at 90 minutes before declining toward the control. The statistically significant points of amended results are marked with an asterisk (P<0.05). Individual graphs with full error bars can be found in Appendix 7






*The control data remains roughly constant throughout; the control group had an average heart rate that ranged between 63.9 and 69.5 bpm during the time course of the experiment, whereas heart rate for the atropine group ranged from 70.4 to 84.2 bpm (Figure 5). This indicated that there was a significant inhibition of the muscarinic receptors caused by atropine to increase the heart rate. The baseline heart rate recorded (-30 to -10 minutes) for the control group was lower than that of the atropine group by an average of 7.8 bpm. To overcome this, the set of data was subtracted from its middle baseline value (-20 minutes) for both the placebo and atropine groups. This gave the change in heart rate, which allowed statistical testing to be done and compare the placebo against the atropine. An F-test was performed to check whether variance was equal then a modified T-test was performed. Results from this showed that there were statistically significant differences between the two sets of data (P<0.05).

Respiratory Sinus Arrhythmia (RSA)

Figure 6- E: I ratio changes throughout the experiment in placebo and atropine group. The control overlaps the atropine until time 30 where the atropine appears to take effect for E: I ratio and heart rate. The E: I ratio declines and plateaus from 60-75 The statistically significant points are marked with an asterisk (P<0.05). Individual graphs with full error bars can be found in Appendix 7




*The measure for RSA is the E: I ratio, a decrease in E I ratio indicates a decline in RSA. In the control group the E: I ratio fluctuated between 1.18 and 1.25 for the duration of the experiments. The atropine group had an E: I ratio consistent with the placebo group from -30 to 30 minutes; after this a sharp decrease in RSA is seen. From time -10, the baseline value before dosing, to time 75 the values are; 1.20 (time -10, SD=0.1, n=10) and 1.11(time 75, SD=0.05, n=10), a decrease of 0.9. When compared to the control at time 75 the values are 1.23 (SD=0.05) for the control to 1.11 in the atropine group, this is a difference of 0.12. The values then begin to increase towards the control (Figure 6).

A correlation coefficient was determined by producing a graph plotting "change in E: I ratio" against the corresponding "change in heart rate" for each time interval. The "changes in" values were calculated by subtracting each E: I ratio from its time -20 value, and each heart rate from its time -20 value. This was done for placebo and atropine groups to obtain two separate data groups for each measure. The resulting data was then plotted on a graph to illustrate any correlation.

The correlation coefficient quantifies the association between E: I ratio and heart rate and provides a number between -1 and 1. The closer the value is to -1 or 1 the stronger the correlation. The correlation coefficient obtained for the placebo group is -0.44, and for the atropine group it is -0.86, indicating that in the atropine group there is a strong correlation between heart rate and E: I ratio over time (Figure 7). A T-test was conducted on the correlation coefficient, and the atropine was more significant (P<0.05) than the placebo.

Figure 7-Correlation between heart rate and E: I ratio, the atropine group shows a negative correlation, whereas the placebo group has a weaker correlation.

The Influence of Time On RSA

To test this hypothesis the subject data was grouped into 4 groups; atropine groups AM and PM, and placebo groups AM and PM. The values for 90 minutes for the heart rate and E: I ratio were chosen as there was a significant difference at that time. The values for the groups were averaged and standard deviations performed on them, this was done for both heart rate and E: I ratio.

For heart rate the placebo groups do not differ from the morning to the afternoon, however in the atropine group there was a change. In the morning the heart rate was 88.8 (SD=13.8, n=7), and the afternoon it decreased to75.2 (SD=15.1, n=3) bpm. This is a difference of 13.6 bpm in the afternoon, a 15.3% decrease (Figure 8).

3 7 3 7

3 7 3 7Figure 8- Heart rate in subjects examined in the morning and afternoon, the number in the bar denotes the number of subjects tested.

The results show that the E: I ratio is higher in the afternoon than the morning. In the placebo group the E: I ratio increased by 0.08; from 1.17 (SD=0.06, n=7) in the morning to 1.25 (SD=0.07, n=3) in the afternoon. In the atropine group the morning value was 1.1 (SD=0.09, n=7) compared to 1.16 (SD=0.05, n=3) in the afternoon, a change of 0.05 (Figure 9). No piece of data showed significant value.

3 7 3 7figure 9- RSA of subjects examined in the morning and the afternoon at time 90 minutes, the numbers in the bars represent the number of subjects tested.


The original hypothesis was that atropine would increase the overall heart rate and decrease the amplitude of RSA. The results from this study show that atropine exerted a significant effect on both heart rate and E: I ratio. RSA magnitude decreases after oral atropine is administered, as the heart rate increases. This agrees with literature on atropine effects on RSA29-30, however this is different from previous trials as either intra-venous or intra-muscular atropine was used instead of oral. A dose of 20mcg/kg, which would equate to 1.4mg for a 70kg individual; was enough to significantly increase the heart rate and reduce the magnitude of RSA. This considerable change indicates oral atropine could be used clinically to measure the activity of the vagal nerve in patients. For example it could be used to assess the degree and progression of vagal neuropathy in diabetic patients30.

Atropine could be blocking the muscarinic receptor, preventing the vagus nerve from inhibiting heart rate, indicating RSA is vagally controlled. This conclusion can be reached from our results because atropine is a known muscarinic antagonist and has no effect on nicotinic receptors. In atropine subjects, the drug took an effect as the heart rate increased, and there is a significant decrease in E: I ratio without the use of nicotinic blockade. This agrees with much of the literature citing RSA as a vagally controlled phenomenon4, 15. The peak of the atropine appeared to be at different times for each parameter measured; however the onset of action for atropine was after 30 minutes in both sets of data. According to a study by Miller at al, approximately 25 minutes is the time before effects of atropine are seen [41] . This would suggest that both the heart rate and RSA have muscarinic influence, i.e. are simultaneously controlled by the vagal nerve. The heart rate peaked at 90 minutes whereas the E: I ratio peaked 75 minutes after dosing. This difference in time could be explained by other mechanisms overriding the RSA, faster than the atropine affecting the heart rate. A correlation was found between the change in heart rate and the change in E: I ratio; in the atropine group as the change in heart rate increased, the E: I ratio decreased. The correlation coefficient value obtained was -0.86, which is a strong negative correlation as it is close to -1. The value for the placebo group was -0.44, indicating little or no correlation which is to be expected within a placebo. This implies that the heart rate is a key factor in RSA production and magnitude. The atropine value proved to be significant, whereas the placebo group did not. The significant relationship suggests a correlation between the change in heart rate and the change in RSA, however the peak effect on each parameter are at different times. Another factor that is not taken into account is the diversity between individuals.

The data may also support the polyvagal theory, as it is possible that a second set of neurons within a set of nerves mediates RSA. The data shows an increase in heart rate, which could be caused by the neurogenic blockade of the proposed DMN vagus, allowing the NA vagus to alter RSA. The polyvagal theory states there is competition between the two sets of vagii22 which could be an explanation for the slight change in heart rate on inspiration and expiration. However the same could be said of a univagal system. According to our findings it appears there is no nicotinic aspect to RSA influence as vagal blockade is near to abolishing RSA. This is contrasting with other papers that suggest that there is an influence16; a possible explaination for this could be that nicotinic receptors can alter RSA but not control it, and so cannot alter the level of RSA as dramatically. A way of testing whether there is a nicotinic aspect to RSA would be to repeat this experiment with a nicotinic antagonist, and see if there is any recorded effect. The results of that trial could be compared with this to establish which has a greater impact on RSA. Another way to test this would be to have subjects' trial atropine and a nicotinic antagonist, and then atropine alone, and compare differences.

There could be other explanations to the changes seen in RSA, as atropine may not just be having an effect on the vagus nerve. A study shows that atropine has an influence on baroreceptor sensitvity [42] , which caused bradycardia, after low dose atropine was used.

The baseline heart rate of the placebo group is lower than that of the atropine group but the E: I ratios are roughly equal; there appears to be no obvious explanation for this as the trial was double-blind placebo controlled and the subjects receiving atropine was not revealed until all subjects had been trialled. This could have occurred due to an increased room temperature during the trials of simultaneous atropine subjects. In further studies the room temperature would have to be monitored and keep constant where possible.

A power calculation was conducted after testing using a web-based package [43] , with the default settings our results achieved a statistical power of 84.9%, and minimum sample size of three subjects in each group were needed to accept the outcome. This is a positive result as our study consisted of ten subjects and allows us to accept the results and conclusions with greater confidence.

Secondary Hypothesis

The outcomes for this hypothesis were that in the control there were no difference in the morning and afternoon for heart rate but for the atropine group, a decrease was seen in the afternoon. For RSA both placebo and atropine groups showed an increase in the afternoon compared to the morning.

There is shown to be diurnal variations in factors affecting RSA, such as blood pressure31-32. It is suggested that blood pressure is higher in the morning, and so would adjust the RSA. The heart rate may not change as vasoconstriction and vasodilatation can alter to keep the heart rate steady. The RSA magnitude may increase in the afternoon due to a change in forced vital capacity (FVC)33, which is shown to be highest in the afternoon. The results from this trial are a contrast as previous research done indicates a decline in RSA throughout the day.

Although this is a good starting point, further testing needs to be done as the data was statistically insignificant. This could be due to the unequal numbers of subjects as there was not an even split between the AM and PM parameter as there were eight more volunteers in the morning than the afternoon. In any further trials this ratio would need to be evened out. The timing of each session could have been too close together; from 10am-1pm and 2 pm-5pm. If the trial was conducted again the two sessions would have to be further apart, such as 10am-1pm and 5pm-8pm to distinguish morning from afternoon more clearly. To come to a reliable and significant conclusion, another trial would have to be run as the data here is statistically insignificant and varied.

Further Considerations

The experiment gave good results but some details may need to be monitored. A three lead ECG was used, which was enough for the experiment; however it is more common to use a five or twelve lead ECG. This may have yielded more accurate results as we found some R peaks to be indistinguishable from other peaks, with more leads we could be more confident with which peaks showed the R-R interval.

Whilst performing the experiment it was noted that ECG readings varied when there was noise, such as talking, in the background. When this was noted, noise was kept to a minimum when readings were being taken. However this was noted halfway through the experiment and for future testing noise levels would have to be kept at a minimum throughout.

Breathing technique was different among subjects, despite instructions to gradually breathe in and out; some subjects had to hold their breath to finish the five seconds in some cycles. It was decided that if the volume of air plateaus or if the R-R intervals were at the change between inspiration and expiration, they would not be measured for that time. This was at the analyzers' discretion, which would introduce a subjective element and reduces consistency.

If the subject was disinterested or was not fully compliant, they could have been fidgeting or not be in a relaxed state which could alter the resting heart rate, thereby affecting the ECG readings. It is difficult to gauge whether a subject is relaxed. To eliminate the problem of disinterest, an incentive or reward could be given at the end of the trial.

Our subjects were asked to rate their own fitness in the questionnaire. This also introduced a subjective element, which would need to be made as objective as possible, should this trial be repeated. It is known that fitness affects RSA magnitude [44] , so it is important to establish a minimum level of fitness for each subject.

Ideally, given more time, it would be useful to have people as their own control; testing them with atropine and placebo. This would eliminate problems such as subjective fitness, breathing technique, and metabolism rate of atropine. However it would be difficult to bring people for two three hour sessions on different days, and to maintain consistency this would have to be at the same time slot, in the same room.


After analyzing the data, there is a significant change in RSA after oral atropine administration. A change in both heart rate and E: I ratio (RSA) was changed. The heart rate began to increase at 30 minutes and peaked at 90 minutes after dosing. The E: I ratio also increased at 30 minutes, but appeared to peak 75 minutes after dosing. The difference in E: I ratio between the control and atropine group was 0.12. This change proved to be statistically significant, as T-tests, F-tests and a standard deviation analysis were performed. This agrees with current literature, which also shows a decreased RSA after atropine introduction. The correlation chart shows there is a correlation between HR and RSA, implying heart rate is an important factor in RSA. From this set of results it appears that RSA would be vagally controlled due to the impact a muscarinic antagonist had. The data is further enhanced by the power calculation, revealing a statistical power of 84.9% with a minimum of three subjects to accept this hypothesis.

In the secondary hypothesis it initially appears that RSA increases in the afternoon compared to the morning. The results indicate at time 90 in the placebo group, E: I ratio increased by 0.7 and by 0.6 in the atropine group. However the data is unreliable as statistical testing proved insignificant, and the number in each group varied. These results differ from other studies which have suggested a decline in RSA throughout the day. More testing, specifically for this hypothesis must be carried out to achieve an accurate conclusion.