1094 
MONITORING 
Fhzcjuaficij Locking wHh Vagal Stimulation 
1 Imputse/stimulus : 
Figure 4. — Stimulus period versus mean heart period. 
creases. Figure 3 also shows lines drawn to 
indicate the ratios 2:1, 1:1, and 1:2. Note that 
there is both harmonic and sub-harm.onic cou- 
pling to the stimulation period. 
Figure 4 shows a section of the records when 
the number of pulses in each burst increased 
but the period was held constant. Note that 
while there is greatly increased stimulation, the 
resulting heart period is unchanged. 
These observations indicated to us that our 
conception of the vagal heart rate control sub- 
system was too simplistic. We then continued to 
examine in more detail this vagal-heart pace- 
maker sub-system. 
The next experiment that we carried out was 
to stimulate the vagus, again using the digital 
computer as a control system to place the stim- 
uli in specific phases of the cardiac cycle. The 
trigger in this instance was the "P" wave re- 
corded from the atrial cavity. The computer de- 
tected the P onset (program lines 59 and 60). 
The 1800 then transmitted pulses to the labora- 
tory and drove the Grass stimulator. The pa- 
rameters, such as the number of pulses and the 
delay between the "P" and the stimulator, were 
all recorded. The subsequent cardiac periods 
were detected and printed out on-line in the 
laboratory (program lines 61 to 68), and the 
data were stored sequentially in the disc file 
(program line 70 to 73). An incremental plotter 
presented a hard copy of raw and mathe- 
matically transformed data on-line. 
These features were particularly useful. The 
first advantage was to judge the quality of the 
data (program line 69), whether or not the 
computer had determined the P wave onsets 
correctly, in which case the data could be given 
permanently to the disc file. Second, the graphic 
presentation of the data permitted us to judge 
the confidence with which the data were being 
acquired. The human interaction with the 
graphic output permitted a reduction in the 
programming time, eliminating the develop- 
ment of sophisticated error detection schemes. 
The digital computer was useful because we 
could change the stimulation parameters 
through a wide variety of alternatives simply 
by typing in the appropriate values in the array 
and initiating the transfer to the 1800. It was 
possible to get a completely flexible system of 
experimental control and data reduction and 
storage with a minimum of program develop- 
ment. Approximately 1,000 stimulations were 
carried out in 18 dogs.^ The time from the 
conception of the experiment to the first com- 
plete experiment was one work week. Actually, 
it took only four tries to develop the entire 
system and get it operational and on the air. 
It took us another three months, however, to 
collect all the experimental data. 
The influence of the timing of vagal stimula- 
tion for heart rate is depicted in Figure 5. A 
burst placed about 200 milliseconds after the P 
wave onset results in a prolonged P-P interval. 
A burst placed later in the cycle results in no 
prolongation of that cycle, and a diminished 
prolongation of the subsequent beat. 
Figure 6 summarizes the effect of varying the 
number of pulses and the timing of those pulses 
on the heart period. P represents time zero of 
the cycle in which the stimulus was carried out. 
The vertical axis represents the period in which 
the stimulation was carried out. Thus, in this 
example, a stimulus with one pulse placed at 
time zero, coincident with a P wave results in 
a cycle length of 550 msec. If the stimulus is 
placed 300 msec after the P wave, then the 
cycle is not inhibited. However, the following 
cycle is. We have elected to plot that effect as 
being placed before the reference P wave. As 
the stimulation is placed closer and closer to 
the reference P wave, the inhibition of that 
