56
where X = dose, V = volume of distribution of the central
0 c
compartment.
Validity of the terminal rate constant. Proper estimation (Appendix
II) of the terminal rate constant (and half-life) depends upon a)
analytical sensitivity; b) number of terminal plasma concentration
values, the time interval between these values, and the number of
terminal half-lives over which the samples were collected; c) selection
of the compartment model; and d) proper weighting of the data. Upon
acute IV bolus administration of buprenorphine in dogs at the 0.7-2.6
mg/kg doses used, the plasma concentrations of buprenorphine were below
20 ng/ml at 1000 min (See Figs. 4-9). Thus the available analytical
sensitivity of 5 ng/ml did not permit accurate estimation of the
terminal half-life. For example, at the 2.5632 mg/kg (Study #3) IV bolus
dose of buprenorphine in dog C, the estimated terminal rate constant
obtained from a semilogarithmic plot of the terminal phase plasma data
against time (n=12) was 1.7 X 10 4 (half-life = 4040 min) +_ 0.70 X
-4 -1
10 (SE) min Thus the range that would include the 95% confidence
limits for this rate constant would be 1.48 X 10 ^ (half-life = 47000
min) to 3.3 X 10 4 (half-life = 2111 min) min 1 (See Table 3).
The terminal half-life significantly depends upon the number of
compartments assumed. Consider dog A (Study #1). Fitting of the data
weighted by the inverse of the concentration to a 3-compartment model
-4 -1
gave a terminal rate constant of 7.6 X 10 min (half-life = 916
min). When the same data were fitted to a 4-compartment model, the
terminal rate constant estimated by using the computer program (Appendix
I) was 3.78 X 10-4 +0.984 X 104 (SE) min1 (half-life = 1840 min,
n=3; The 95% confidence limits; + smt, where t=12.71; 433 min to time