Browsing by Subject "Head Injuries"

Now showing 1 - 1 of 1
  • No Thumbnail Available
    Item
    Open Access
    An investigation of the characteristics of, and variation in the non-steady state pharmacokinetics of diphenylhydantoin after head trauma
    (1995) Jooste, Jacqueline Fraser; Folb, Peter
    Diphenylhydantoin (DPH) is a drug, introduced by Merritt and Putnam in 1938 (Porter and Rogawski 1992), which to date is still widely used and is effective for all types of seizures except absence seizures. This study was undertaken in an attempt to identify and explain the possible mechanisms responsible for the apparent low serum DPH concentrations measured in neurosurgical, head trauma patients, receiving a loading dose of DPH as prophylaxis against post-traumatic epilepsy (PTE). The pharmacokinetics of a single loading dose infusion of DPH, at non-steady state, were established. The impact of head trauma on the behaviour of the parent drug DPH as a protein-bound complex in serum (total DPH) and in the free unbound form in plasma (free DPH), together with the major metabolite 5-(p-hydroxyphenyl)-5-phenylhydantoin (p-HPPH), was investigated. A simple High Pressure Liquid Chromatography (HPLC) analytical method was developed to allow for the simultaneous measurement of the concentrations of DPH (total or free) and the metabolite, p-HPPH. The range of acceptable linearity, for this particular method, was established as 0.31 - 40.0 mg/L for DPH and 0.25 - 4.0 mg/L for p-HPPH. The reproducibility for the new method was depicted by a correlation coefficient (r) of 0.999 for both DPH and p-HPPH respectively. The good between-run precision of this particular method was reflected by the low coefficient of variation for DPH (4.60% - 7.98%) and p-HPPH (6.10%). In order to evaluate the impact of varying degrees of trauma on a group of neurosurgical patients a new scoring system was developed for the purposes of the study. Four scoring systems were chosen, namely the Glasgow Coma Scale (GCS), the Acute Physiologic and Chronic Health Evaluation (APACHE II) score, the Injury Severity Score (ISS) and a recently developed scoring system which is based on the evaluation of the CT Scan. These scoring systems were chosen to ensure all aspects of injury, i.e. physical, neurological, physiological and patient prognosis, could be investigated. A resultant quartile score was assigned to each of the individual scores that were rated, using StatGraphics™. Each quartile score was then summed to derive a "Total Injury Score" (TIS) describing all aspects of the injury. The TIS was well correlated with each individual score and was a good predictor of survival (p = 0.04) even though the CT scan score tended to bias the TIS by its weaker performance. A total of 25 patients who had experienced some form of neuro-trauma were admitted to the study. The median age of the group was 32 years. The loading dose given had a median value of 12.82 mg/kg (8.82 - 18.18 mg/kg). This loading dose was much lower than the recommended loading dose of 15 - 18 mg/kg. This low loading dosage was thought to be partly responsible for the low serum DPH levels noted in the neurosurgical wards. The above-mentioned analytical procedure was used to measure the drug levels thereby establishing the non-steady state pharmacokinetics (Vmax, Km and the apparent volume of distribution Vd) of a single loading dose of DPH. The median total DPH levels measured at 2, 12 and 24 hours were seen to be in the defined therapeutic range, but decreased with time as follows: 17.29 mg/L, .14.2lmg/L, and 13.75 mg/L respectively. In a similar fashion the free DPH levels decreased with time, i.e. 1.95 mg/L, 1.53 mg/Land 1.11 mg/Lat 2, 12 and 24 hours respectively after the end of the loading dose. The metabolite levels behaved differently in that the median levels increased from 0.045 mg/L at 2 hours to 0.075 mg/Lat 12 hours then decreased to 0.033 mg/Lat 24 hours. The free fraction of DPH, calculated as the percentage of free DPH relative to the total DPH level, decreased over the first periods investigated, i.e. at 2 and 12 hours (11.96 and 8.82 respectively), and then at 24 hours increased slightly (10.0%). Various investigators have described situations of increased DPH free fraction in head-trauma patients (Lund et al., 1971; Elfstrom 1977; Crowley et al., 1987; Toler et al., 1990). Others have gone further to propose that changes in protein binding, as a result of trauma, together with increases in the free fraction, result in increased plasma clearance (Boucher et al., 1987; Griebel et al., 1990; Anderson et al., 1991). No specific relationship was discernible over the 24-hour period investigated between the extent of a patient's injury (TIS) and the total serum DPH, the free plasma DPH or the serum metabolite concentration that was measured. Similarly, no relationship between the pharmacokinetic parameters of DPH and the extent of trauma was established (p > 0.05). In our group of head trauma patients the free fraction of DPH was found to be inversely correlated with serum albumin concentrations only after the initial distribution phase, i.e. 2 hours after the end of the loading dose infusion (r = -0.49: p < 0.03). Pharmacokinetic parameters were established using a clinical pharmacokinetic software program, Abbottbase™ PKS. Abbottbase™ PKS initially uses a population nomogram, based on population-typical data from a group of idiopathic epileptics, to predict serum concentrations or kinetic parameters and then to generate individualised data based on Bayesian regression methodology in a critically ill head trauma group. The accuracy of the initial non-specific population-typical nomogram is examined against the Bayesian methodology. The initial estimate, or population-typical values of the various pharmacokinetic parameters, were found to be significantly different (p < 0.0001) from those established for the trauma group by Bayesian methodology. Thus the null hypothesis was rejected. The basic Michaelis-Meneten kinetic parameters, Vmax and Km, together with the apparent volume of distribution (V d) of DPH determined over a period of 24 hours, after a loading dose infusion, are presented. The Michaelis-Menten constant, Km, and the V d for the DPH were within the normal range quoted in the literature: 4.95 µg/ml and 0.62 L/kg [42.8 L] respectively. The only remarkable finding was that the head trauma group had a much lower Vmax value (median of 4.59 mg/kg/day or 323 mg/day) than the Vmax values quoted in the literature. Head injured patients who had experienced additional body trauma appeared to have lower Vmax values than those patients with just a head injury (p < 0.05). However, it must be noted that the majority of pharmacokinetic investigations of DPH are carried out at steady state and in fairly "healthy" patients with idiopathic epilepsy. Hence there are no strictly comparable data for comparison with the present study. The lower Vmax implies that a slower rate of metabolism occurs during the first 24-hour period in the head trauma patient. The reason for this altered metabolism was difficult to assess but believed to be due to a number of factors such as altered disposition of the drug and a cumulative effect of the often complex manifestation of acute traumatic stress. These include an altered basal metabolic rate, changes in plasma protein binding, altered distribution patterns and low serum albumin (such as was found in our group: median albumin of34 g/L). A distinct relationship (p < 0.001) between the loading dose given (in mg/k$) and the serum total DPH levels measured reflected the significance of the V d when designing an appropriate loading dose for the head trauma patient. The above relationship signals the need to dose accurately in terms of patient weight to achieve adequate DPH levels during prophylaxis rather than using an empirical loading dose of either 750mg or 1000mg of DPH which is standard practice in a trauma setting. For optimal management of the critically ill head trauma patient, an individualised dosing plan is recommended which includes a weight-based loading dose followed by a blood level taken 12 - 24 hours after the end of the loading dose infusion. The level is then subjected to Bayesian analysis in order to predict a maintenance dose that will achieve and maintain adequate serum levels. Clearly, these results are preliminary and reflect the need for more in-depth investigation of the disposition of DPH in a critically ill head trauma group. Specific nomograms for various sub-populations of patients must be established to obtain accurate predictions of both the doses given and the levels achieved. This is reinforced by the suggestion of the present study that patients with head trauma (and possibly other critically ill sub-populations) should be treated with the awareness that drugs that are hepatically metabolised may have altered pharmacokinetic and pharmacodynamic properties.