-
- Posts: 11
- Joined: Thu Dec 29, 2011 1:43 pm
TL;DR: I developed method of rimantadine assay which gives good result with fresh samples, but questionable results during stability studies. I need help to understand underlying processes.
Here is the problem (prepare for long read):
I study drug formulation, which has following composition:
Paracetamol - about 360 mg
Cetirizine HCl - about 2,5 mg
Rimantadine HCl - about 75 mg
Ascorbic acid - about 125 mg
Mixture CaCO3/sorbitol (70:30)- about 90 mg
Citric acid - about 200 mg
Aspartam - about 50 mg
Flavor ‘Orange' / ‘Lemon’- about 50 mg
Dextrose or fructose - up to 10 g (about 9.05 g)
(Don’t ask me why, I’m not the developer of the formulation).
I have some issues with assay of rimantadine.
I chose the most logical method of analysis - GC/FID.
I use naphtene as internal standard, which I add to organic part to control its loss due to evaporation.
At first, I tried to analyze rimantadine as described in USP for rimantadine tablets - dissolving sample in water, adding NaOH solution and extracting with hexane or heptane (tried both with similar results). Recoveries were 90-98 % (with lower values at lower concentrations). I tried double and even triple extraction, they significantly increased sample preparation time with no real improvement of recovery.
Then I came up with idea of using chloroform: it dissolves both Rimantadine and Rimantadine HCl. Moreover, most of other components practically do not dissolve in it, so I can easily get rid of them by filtering. After that I shake filtrate with concentrated NaOH solution to remove HCl, and I get a solution of Rimantadine in chloroform, which I analyze on HP-5 column at 170 C isotherm with run time 6 min. During validation, method showed excellent linearity at range 2 – 4 mg/ml (50 – 150 %) and recovery values were 99 – 101 %.
Later I read an article (in Russian printed journal) about rimantadine HCl determination in tablets of somewhat similar composition by direct injection of ethanol solution of tablets, which seemed questionable to me, since Rimantadine HCl, unlike Rimantadine, is not exactly a volatile substance. Still I tried it and got a feeling, that the principle of the method is to achieve more or less reproducible decomposition of Rimantadine HCl in injection port and during run and to find appropriate temperature gradient, which would compensate peak tailing and peak deformation. I was able to achieve acceptable peak shape in standard, but not in test solution (most likely due to interference from citric acid).
Anyway, that was preamble.
During stability study in accelerated conditions, and to lesser extent in normal conditions, I found decrease of rimantadine HCl content during storage (from 75 mg down to 62-67 mg). And that is quite strange in itself, since rimantadine HCl is a very robust substance. You can check its structure. I might be ignorant of some obvious things, but I can't see any probable way it can decompose in the given mixture of solids. Also of notice, that decrease was higher in fructose samples comparing to dextrose.
I began to suspect my method and went back to roots and used modified USP method with hexane extraction. I got low results again. I tried method described in article mentioned above - results tended to be low, though it was hard to say because of peak deformation. However, I also found out, that if I mix ethanol solution with hexane and NaOH solution (kind of hybrid of USP and article methods) I get my rimantadine back to 72-76 mg. The method had some issues with linearity, but the results were clearly higher. Moreover, addition of 5 ml of ethanol to 25 ml of chloroform (my method) also significantly increased amount of rimantadine found (up to normal). I checked accuracy of my method with freshly prepared model mixture and got excellent results. I made different model mixtures (without every single component with or without fructose), placed them at 40 C / 75 % RH in barely closed flasks for 3 days, and analyzed using chloroform. I got lower content of rimantadine in every mixture, with the bigger decrease in the mixtures without CaCO3 and the lesser in mixtures without citric acid. However all the mixtures turned into cake (especially without fructose), which did not disintegrate readily in chloroform, so I’m not sure whether the results are due to decomposition of rimantadine or can be attributed to extraction / sample preparation issues.
Also I found that:
1. addition of 1 ml of water to 25 ml chlorofom (before filtering and addition of NaOH solution) significantly lowers Rimantadine content, most likely due to distribution of Rimantadine HCl between water and chloroform.
2. I can use butanol-1 instead of chloroform and Rimantadine content in that case is even more than 75 mg (method is raw and requires some work on it).
3. Content of other active components (paracetamol, cetirizine, ascorbic acid, CaCO3) and average mass of content doesn't change significantly.
So, in summary, I assume, that loss of rimantadine during storage is apparent in nature (meaning perceived rather than real). I think water/humidity and maybe temperature play major part in the process of making part of rimantadine ‘invisible’. I suspect, that fructose samples exhibit bigger decrease than dextrose, because fructose is more hygroscopic than dextrose (?). I believe that alcohols can somehow reverse ‘the invisibility’, but I have no idea how. And in the end, I’m at a loss. I need to understand what’s going on to find appropriate solution.
That is why I ask for any help, any information, from personal experience or literature or any other source, any assumptions and ideas on the matter, any advice or suggestion on what to do or what to look at.
Here is the problem (prepare for long read):
I study drug formulation, which has following composition:
Paracetamol - about 360 mg
Cetirizine HCl - about 2,5 mg
Rimantadine HCl - about 75 mg
Ascorbic acid - about 125 mg
Mixture CaCO3/sorbitol (70:30)- about 90 mg
Citric acid - about 200 mg
Aspartam - about 50 mg
Flavor ‘Orange' / ‘Lemon’- about 50 mg
Dextrose or fructose - up to 10 g (about 9.05 g)
(Don’t ask me why, I’m not the developer of the formulation).
I have some issues with assay of rimantadine.
I chose the most logical method of analysis - GC/FID.
I use naphtene as internal standard, which I add to organic part to control its loss due to evaporation.
At first, I tried to analyze rimantadine as described in USP for rimantadine tablets - dissolving sample in water, adding NaOH solution and extracting with hexane or heptane (tried both with similar results). Recoveries were 90-98 % (with lower values at lower concentrations). I tried double and even triple extraction, they significantly increased sample preparation time with no real improvement of recovery.
Then I came up with idea of using chloroform: it dissolves both Rimantadine and Rimantadine HCl. Moreover, most of other components practically do not dissolve in it, so I can easily get rid of them by filtering. After that I shake filtrate with concentrated NaOH solution to remove HCl, and I get a solution of Rimantadine in chloroform, which I analyze on HP-5 column at 170 C isotherm with run time 6 min. During validation, method showed excellent linearity at range 2 – 4 mg/ml (50 – 150 %) and recovery values were 99 – 101 %.
Later I read an article (in Russian printed journal) about rimantadine HCl determination in tablets of somewhat similar composition by direct injection of ethanol solution of tablets, which seemed questionable to me, since Rimantadine HCl, unlike Rimantadine, is not exactly a volatile substance. Still I tried it and got a feeling, that the principle of the method is to achieve more or less reproducible decomposition of Rimantadine HCl in injection port and during run and to find appropriate temperature gradient, which would compensate peak tailing and peak deformation. I was able to achieve acceptable peak shape in standard, but not in test solution (most likely due to interference from citric acid).
Anyway, that was preamble.
During stability study in accelerated conditions, and to lesser extent in normal conditions, I found decrease of rimantadine HCl content during storage (from 75 mg down to 62-67 mg). And that is quite strange in itself, since rimantadine HCl is a very robust substance. You can check its structure. I might be ignorant of some obvious things, but I can't see any probable way it can decompose in the given mixture of solids. Also of notice, that decrease was higher in fructose samples comparing to dextrose.
I began to suspect my method and went back to roots and used modified USP method with hexane extraction. I got low results again. I tried method described in article mentioned above - results tended to be low, though it was hard to say because of peak deformation. However, I also found out, that if I mix ethanol solution with hexane and NaOH solution (kind of hybrid of USP and article methods) I get my rimantadine back to 72-76 mg. The method had some issues with linearity, but the results were clearly higher. Moreover, addition of 5 ml of ethanol to 25 ml of chloroform (my method) also significantly increased amount of rimantadine found (up to normal). I checked accuracy of my method with freshly prepared model mixture and got excellent results. I made different model mixtures (without every single component with or without fructose), placed them at 40 C / 75 % RH in barely closed flasks for 3 days, and analyzed using chloroform. I got lower content of rimantadine in every mixture, with the bigger decrease in the mixtures without CaCO3 and the lesser in mixtures without citric acid. However all the mixtures turned into cake (especially without fructose), which did not disintegrate readily in chloroform, so I’m not sure whether the results are due to decomposition of rimantadine or can be attributed to extraction / sample preparation issues.
Also I found that:
1. addition of 1 ml of water to 25 ml chlorofom (before filtering and addition of NaOH solution) significantly lowers Rimantadine content, most likely due to distribution of Rimantadine HCl between water and chloroform.
2. I can use butanol-1 instead of chloroform and Rimantadine content in that case is even more than 75 mg (method is raw and requires some work on it).
3. Content of other active components (paracetamol, cetirizine, ascorbic acid, CaCO3) and average mass of content doesn't change significantly.
So, in summary, I assume, that loss of rimantadine during storage is apparent in nature (meaning perceived rather than real). I think water/humidity and maybe temperature play major part in the process of making part of rimantadine ‘invisible’. I suspect, that fructose samples exhibit bigger decrease than dextrose, because fructose is more hygroscopic than dextrose (?). I believe that alcohols can somehow reverse ‘the invisibility’, but I have no idea how. And in the end, I’m at a loss. I need to understand what’s going on to find appropriate solution.
That is why I ask for any help, any information, from personal experience or literature or any other source, any assumptions and ideas on the matter, any advice or suggestion on what to do or what to look at.
