healthmedicinet

We first investigated the compatibility of the DZP injection with infusion fluids.
For sample preparation, diluted DZP injection was passed through a 0.2 ?m inline filter
to remove precipitate, and then the DZP concentration was determined by HPLC (Figure 1). As a control, we also diluted the Horizon injection with acetonitrile/water (50:50 v/v),
in which the DZP was completely dissolved. As shown in Figure 1, the profile of the sample was slightly different to that of the control. In most
samples, the DZP concentrations of the sample were obviously lower than those of the
control at the dilution ratios ranging from 2× to 20×, indicating that significant
precipitation of DZP occurred under the experimental conditions. The precipitation
behavior was similar to that from visible inspection 2].

Figure 1. Precipitation behaviors as a function of (a) marketed products of DZP and (b) infusion
fluids. (a) Horizon injection or Cercine injection was diluted with purified water. (b) Horizon injection was diluted with saline, purified water, 5% glucose injection,
and Soldem 3A. Horizon injection was completely dissolved in acetonitrile/water (50:50 v/v),
and it was used as a control. Each value represents the mean?±?S.D., N?=?3.

This study also investigated the effects of formulations and infusion fluids on the
compatibility. The profiles of Horizon injection and Cercine injection were identical
(Figure 1a). According to their packaging inserts, their formulations are almost the same;
thus, this result seems to be reasonable. The profiles, as a function of infusion
fluids, were similar to each other (Figure 1b), suggesting that the infusion fluid also appears to have little influence on the
compatibility of DZP. This result is supported by previous studies 2],6]. For example, Morris compared the compatibility of DZP injection after dilution with
various infusion fluids (5% dextrose, saline, Ringer’s injection and lactated Ringer’s
injection), and then reported that there was no perceptible difference caused by the
difference in the infusion fluids 2]. Mason et al. determined the solubility of DZP in 5% dextrose, lactated Ringer’s
injection and saline, and found that there was little change in the solubility of
DZP even though the diluents were changed 6]. DZP is a weakly dissociated base with a pKa of 3.3, attributed to deprotonation
of its conjugate acid at the 4-position nitrogen atom 7]. When the pH of the solution is below the pKa, the solubility of DZP is changed substantially
6]. However, the pH of commonly used infusion fluids is much higher than the pKa 6], therefore it is to be expected that changing the infusion fluids will have no significant
effect on the compatibility of DZP injection.

In the above experiments, we confirmed the dilution conditions under which significant
precipitation occurred. We next investigated the solubility of DZP in the diluted
injections. The formation of precipitate is thought to be caused by the DZP concentration
of the sample exceeding the solubility. Thus, we expected that samples with precipitate
have reached saturation, while samples without precipitate have a capacity to dissolve
more DZP.

For preparation of the saturated solutions, excessive amounts of DZP powder were mixed
with DZP injections diluted with purified water. Afterwards, the precipitate and DZP
powder were removed by filtration and then their DZP concentrations were measured
by HPLC. The observed solubility curve is shown in Figure 2. We note that the solubility of DZP was quite different from our expectation. Except
for the original DZP injection (1×), all saturated solutions showed lower DZP concentrations
than those of the diluted DZP injections. We also observed similar results from ¹H-NMR spectra from the samples; compared with the sample without DZP powder, the sample
with the powder (saturated solution) showed lower concentration of DZP (data not shown).
This result indicates that the infusion fluid admixtures were supersaturated solutions
of DZP. We further observed that the DZP concentration was substantially decreased
by the addition of metal chip to the diluted DZP injection (data not shown). This
is a common feature of supersaturation. Newton et al. also suggested the possibility
of supersaturation of DZP 8]. They investigated the solubility of DZP in infusion fluid admixtures. The solubility
of DZP in injection at a 1:17.7 dilution ranged from 0.04 to 0.05 mg/mL. On the other
hand, when they conducted this study, it had already been reported that the lowest
volume ratio that did not result in a precipitate was a 1:40 dilution 2]. They recognized that the DZP concentration at 1:40 dilution was 0.12 mg/mL, and
the concentration was higher than the solubility they clarified. They subsequently
suggested the supersaturation of DZP in the diluted injection.

Figure 2. Solubility curve of DZP in the injection diluted with purified water. DZP injection was diluted with purified water, in various dilution ratios, at 25°C,
after which excessive amounts of DZP powder were added to the DZP injections diluted
with water to prepare the saturated solutions. Each value represents the mean?±?S.D.,
N?=?3.

For further information, as a preliminary experiment, we evaluated the stability of
the supersaturated state of DZP in infusion fluid admixtures. After dilution of DZP
injection, we left the sample as it was at room temperature for a few days, and then
measured the concentration once again. No change in the DZP concentration was observed
(data not shown).

With regard to the stability of DZP in infusion fluid admixtures, there are a number
of articles in which this is addressed. For one thing, it is known that DZP is substantially
absorbed by polyvinyl chloride (PVC); a substantial reduction in DZP concentration
was found after storage of the infusion fluid admixture in a container made of PVC
6],9],10]. In contrast, DZP in the infusion fluid admixtures seems to be stable as long as
it is stored in PVC-free containers. For example, there is a report that the DZP concentration
in the infusion fluid admixture remained unchanged over a 168 h period when stored
in a glass or polyethylene bottle 9]. In our study, all experiments were performed using PVC-free containers (we used
polypropylene tubes or glass bottles); thus, it was assumed that the supersaturated
state of DZP would be maintained over a long period if the sample was left as is.

In the next phase of the study, we analyzed the precipitate. Jusko et al. have reported
that the precipitate comprises almost solely DZP 1]. However, the details of the precipitate had yet to be investigated. We then performed
a wide variety of analyses.

Figure 3 shows microscopic images and FTIR spectra of the precipitate and analytical-grade
DZP. The IR spectra are very similar, indicating that the precipitate was mostly composed
of DZP. A little difference was shown in carbonyl stretching region (1800-1700 cm^?1). The difference may represent the interaction between DZP and some other substances.
That is because there is a good possibility that ingredients of DZP injection were
coexisted in the precipitate. PXRD patterns of the precipitate appear to be almost
the same as for analytical-grade DZP (Figure 4). Thus, in terms of the crystal form, there was no significant difference between
the DZP of the precipitate and the analytical-grade DZP. In addition, small peak around
2 theta?=?4° were observed. Although the details are still unclear, it probably be
derived from ingredients of the DZP injection. We dissolved the precipitate in CD₃OD, and then acquired the ¹H-NMR spectrum (Figure 5). Besides the signals for DZP, signals for benzoic acid and benzoic alcohol were
also observed, which clarified that some ingredients of DZP injection were present
in the precipitate. We also recorded the DSC curve of the precipitate (Figure 6). The endothermic peak corresponding to DZP was obviously lower than in the case
of the analytical grade. The depression of the melting point is caused by the coexistence
of ingredients in the DZP crystal.

Figure 3. Microscopic images (a) and FTIR spectra (b) of the precipitate and analytical-grade
DZP. The region of interest for the measurement of the IR spectra is indicated by a square.
The aperture sizes used for the precipitate and analytical-grade DZP were 100 ?m?×?100 ?m
and 30 ?m?×?30 ?m, respectively. The y-axis of each spectrum has arbitrary units.

Figure 4. Powder X-ray diffraction patterns of the precipitate and analytical-grade DZP.

Figure 5. ¹H-NMR spectra of the precipitate, analytical-grade DZP, benzoic acid and benzyl alcohol
dissolved in CD₃OD.

Figure 6. DSC thermograms of the precipitate and analytical-grade DZP. The y-axis of each spectrum has arbitrary units.

DZP has a wide range of indications; thus, there is a good possibility that the injection
is administered to patients simultaneously with other injectable drugs. Much research
has been carried out on the compatibility of various injectable products with DZP
injection. As far as we know, the injectable drug products are the following: Hextend
11], Precedex 12], Depacon 13], acetaminophen 14], doripenem 15], linezolid 16], and tirofiban hydrochloride 17]. In all the studies, significant precipitation was reported; a white turbid precipitate
was observed immediately after the mixing of two injections. All experiments were
performed by mixing two injectable products at a 1:1 dilution. Furthermore, all injectable
products mixed with DZP injection were water-based injections and free from organic
solvent.

Our study revealed that a 1:1 dilution resulted in significant precipitation. The
precipitation observed is probably caused by incompatibility of DZP injection with
other injectable products. The precipitate comprises mostly DZP.