[Previous Months][Date Index][Thread Index][Join - Register][Login]
[Message Prev][Message Next][Thread Prev][Thread Next]

[IP] Novolog and Humalog

Do you really want to know? The data in the figure two curve of the Lilly 
Humalog sheet can be converted to an equation that produces that curve when 
graphed. With that equation the data can be calculated for any time or time 
increments. A table of such values at 3 minute intervals was generated. Then, 
because the Animus delivers a dollop each 3 minutes and because the response 
to each such is simply additive to whatever is in the blood already, many 
columns of repeated numbers were placed in the spread sheet. Each column was 
time offset from the one to its left by 3 minutes. That gave a table of what 
each time displaced dollop was doing. The leftmost column of the spread sheet 
page with that stuff on it is the simple sum of what is in that row, row by 
row. That is, the net insulin is the sum of the individual mini doses each 
playing through at 3 minute separations. That is how it was done and that is 
what you will see in the spread sheet. However, you do not have an Animus and 
I need to know how the Minimed puts in the basal. How much and how often? 
That and the drop test will allow me to generate a similar but customized 
basal transient curve for you. The drop test first derivative (the slope) 
produces the counterpart to fig two. From there the steps are similar but 
differ in the magnitudes and probably the time steps because of minimed vs 
animus basal delivery timing. Your friend is overlooking the fact that the 
diffusion data is implicit in the Humalog data sheet curve. That is a real 
response in typical of a group of people. Add to that the knowledge that 
double the dose produces double the result when insulin is injected (that is, 
the response to dose is  linear) and the application of linear superposition 
takes care of the calculation of how mini doses staged 3 minutes apart 
combine. Is it good to three decimal places? No. Does it need to be? No. We 
are only attempting to relate the basal responses to the basal rate time 
block that is responsible and for that super precision is not needed. As the 
basal response is monitored, that same dose dynamic gives at least a notion 
of what insulin intensity that was passed during the transient accounted for 
the momentarily flat baseline. That facilitates the estimate of what the next 
attempt at basal rate should be as one works through the proportional 
scaling. There is no fundamental dependence on highly precise numbers in the 
above basal rate adjustment  process so the fact that some reasonable 
simplifying assumptions about the mechanism of combination of mini doses was 
made should not bother anyone.  Regarding potency, the assumption is that the 
fig 2 novolog curve and the fig 2 humalog curve are taken under similar 
conditions of injection site. They are the average responses of a group of 
people so individual variation is accounted for. The respective AUC of the 
two gives the single bolus potency comparison. When we did that with R and H 
I calculated .9075 and we later measured .91. I have not done that 
calculation for Novolog vs H because I do not see us using both. I did do it 
for the basal transient curves (which relate through a different ratio than a 
single bolus ratio) because I suspected you would attempt to do a pump switch 
to Novolog and I wanted to forestall a hypo created in ignorance of the 
for HELP or to subscribe/unsubscribe, contact: HELP@insulin-pumpers.org
send a DONATION http://www.Insulin-Pumpers.org/donate.shtml