| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Hydroxyethyl starches (HES) are almost exclusively excreted glomerularly, in part after hydrolysis by amylase. HES 130/0.4 (Voluven®; Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany) was developed to improve pharmacokinetics whereas preserving the efficacy of volume effect. We studied the dependency of pharmacokinetics of HES 130/0.4 on renal function. Nineteen volunteers with stable, non-anuric renal dysfunction, ranging from almost normal creatinine clearance (CLcr) to severe renal impairment (mean CLcr: 50.6 mL · min-1 · 1.73 m-2), were given a single infusion of 500 mL 6% HES 130/0.4 over 30 min. HES plasma concentrations were determined until 72 h, urinary excretion until 7296 h. CLcr had been obtained at least twice before and twice after dosing. Standard pharmacokinetic calculations and regression analysis were performed. Area under the time concentration curve (AUC0inf) clearly depended on renal function comparing subjects with CLcr <50 with those with CLcr 50 (ratio 1.73). Peak concentration (Cmax, 4.34 mg/mL) as well as terminal half-life (16.1 h, model independent) were not affected by renal impairment. At CLcr 30, 59% of the drug could be retrieved in urine, versus 51% at CLcr 15<30. The mean molecular weight of HES in plasma was 62,704 d at 30 min, showing lower values with increased renal impairment (P = 0.04). Pre-dose amylase concentrations inversely correlated with baseline CLcr. Residual HES plasma concentrations after 24 h were small in all subjects ( 0.6 mg/mL). We conclude that HES 130/0.4 (500 mL 6%) can be safely administered to patients even with severe renal impairment, as long as urine flow is preserved, without plasma accumulation. IMPLICATIONS: Dependency of the pharmacokinetics of hydroxyethyl starch 130/0.4 on renal function was studied. The area under the time concentration curve increased moderately with more severe renal dysfunction; however, small plasma concentrations were observed after 24 h. Terminal half-life and peak concentration remained unaffected by renal impairment.
Hydroxyethyl starch (HES) solutions are artificial colloids for intravascular volume replacement derived from corn starch amylopectin. HES solutions are polydisperse, comprising a distribution of molecular sizes. The polyglucose chains closely resemble glycogen with predominant 14 bindings. The main characteristics of a HES type are mean molecular weight (Mw, weight average), molar substitution (MS, i.e., mol hydroxyethyl residues per mol glucose subunits), and C2/C6 ratio (the substitution pattern at the glucose subunit carbon atoms). There are large regional differences in the use of different HES specifications. For decades in the United States, only HES 450/0.7 has been available for volume therapy. The predominant starch in Europe is HES 200/0.5, whereas in Japan only HES 70/0.5 is available. Safety issues of a colloid such as HES are related to molecular properties, resulting in different in vivo Mw, plasma concentrations over time, and tissue storage. Pharmacology, especially regarding half-lives, duration of volume effect, posology, and possible side effects, significantly differs among the different specifications.
However, all HES preparations are excreted renally. Other ways of excretion are negligible (1). Molecules above the renal threshold are filtered glomerularly only after hydrolysis by serum According to the European label for HES 130/0.4 and the United States hetastarch label, these can be used in patients with non-oligoanuric renal impairment. Therefore, information on the dependency of pharmacokinetics of HES on renal function is a crucial contribution to increase therapeutic safety, and was assessed in this study for HES 130/0.4 (6%).
After approval of the study by the local ethics committee, subjects gave written informed consent for their participation. All procedures were performed according to Good Clinical Practice standards. Nineteen volunteers (6 women, 13 men) with different degrees of stable, non-anuric renal dysfunction were included. Phase I facilities and volunteer database (Pharm PlanNet Clinical Pharmacology Institute, Moenchengladbach, Germany) were used. Subject recruitment followed a predefined stratification according to the Food and Drug Administration guidance document (15) to cover a wide range of renal impairment, and to allow for regression analysis rather than comparing arbitrary renal function groups only. Six subjects belonged to the group with "severe" renal impairment (creatinine clearance [CLcr] 15<30 mL · min-1 · 1.73 m-2), 4 to "moderate" (30<50), 5 to "mild" (50<80), and 4 almost normal subjects showed a CLcr in the interval 80<120. Stable renal function was verified by performing at least 2 separate measurements of CLcr using 24-h urine collection and a serum sample. Creatinine was measured by the Jaffé method. The standard formula was applied for CLcr determination: urine concentration · urine volume · body surface area · serum concentration-1 · 1.73 m-2 · 1440 min-1. The mean of both values was taken before the subjects were assigned to their respective stratum. Volunteers with unstable renal function, severe hepatic or cardiac disease, or severe anemia were not included in the study. Under supervision of an anesthesiologist, 500 mL of 6% HES 130/0.4 solution (Voluven®; Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany) was infused over a 30-min period by using an infusion pump. Venous blood samples were drawn into lithium heparinized tubes before the start of infusion, 5, 10, and 30 min, 1, 2, 4, 6, 8, 24, 48, and 72 h after the start of infusion. Urine samples were collected quantitatively up to 96 h. The HES either in plasma or urine samples was precipitated by acetone. After the removal of the supernatant, HES precipitate was dissolved in sodium azide solution and hydrolyzed into glucose with trifluoroacetic acid. After drying and dissolving in buffer solution, the glucose was determined enzymatically with a glucose analyzer. The determination of in vivo mean Mw was performed by using low-angle laser lights scattering and high-performance liquid chromatography validated for physiologic samples.
The primary objective of this study was to investigate whether renal impairment had an effect on the pharmacokinetics of HES. Significance tests were exploratively applied each at nominal This was statistically addressed by exploring the relationship between CLcr and each of the pharmacokinetic variables applying correlation and regression analyses. A subject was judged valid for analysis if CLcr was found stable during the screening phase showing values between 15 and <120 mL · min-1 · 1.73 m-2, and if primary pharmacokinetic variables (area under the time concentration curve [AUC(0inf)] and peak concentration [Cmax]) could be derived from the analytic blood sample profile. Correlation analyses were also performed to investigate the relationship between CLcr, age, body mass index, height, weight, and the pharmacokinetic variables (both model independent and model based) calculating Pearson correlation coefficients. This also included consideration of partial correlations that describe the association between two variables controlling for other potentially confounding variables. In addition, HES mean Mw, baseline CLcr, and serum amylase were also the subject of partial correlation analyses. Descriptive statistics were calculated for each of the pharmacokinetic variables, including geometric means and corresponding coefficients of variation.
In addition, two predefined subgroups with CLcr Data were analyzed by using SAS version 6.12 (SAS Institute Inc., Cary, NC). Pharmacokinetic variables were derived by using WinNonLinTM version 3.0 (Pharsight Corp., West El Camino Real, CA) for both the model independent approach and the two-compartmental modeling. Safety was investigated by documentation of adverse events (AE), routine clinical laboratory, physical examination, blood pressure (BP), pulse rate, and electrocardiogram. Specifically, determination of CLcr from 24-h urine (day 3, 4) was performed twice after drug exposition. Cumulative HES excretion and residual HES plasma concentrations were regarded as especially relevant for product safety, in addition to the primary pharmacokinetic variables AUC(0inf) and Cmax.
All subjects were valid for pharmacokinetic evaluation. Means (±SD) of demographic/baseline values were the following: age 54 (±14, range 3077) yr, weight 77.3 (±15.2) kg, height 172 (±10) cm, systolic BP 144 (±19) mm Hg, diastolic BP 83 (±10) mm Hg, heart rate 68 (±10) min-1. CLcr at baseline was on average 50.6 mL · min-1 · 1.73 m-2 ranging between 15.4 and 100.8 mL · min-1 · 1.73 m-2, with a mean intraindividual CV of 9% for the baseline measurements done in each subject. Maximal CV was observed in 1 subject in the severe renal group based on CLcr values of 12.2 and 18.6 mL · min-1 · 1.73 m-2; all other CVs were less than 17%. Increased restriction of renal function was associated with older age and higher baseline systolic BP. For 8 of the 15 subjects with CLcr values <80, the etiology of the impaired renal function was not known. For two subjects, diabetes mellitus was reported as the underlying disease. Each of the diagnoses, immunoglobulin A (IgA)-neuropathy, nephrolithiasis, sarcoidosis, suspected hypertensive nephropathy, and polycystic kidney disease, was documented once. HES plasma concentrations were larger in the 10 subjects with CLcr <50 than in the 9 subjects with CLcr >50. Figure 1 shows the mean courses of HES concentration up to 72 h after infusion. Table 1 shows the most relevant pharmacokinetic variables. No differences were apparent when comparing the two renal groups (15<30) and (30<50). The same is true for a comparison between the renal groups (50<80) and (80<120).
The study showed that AUC(0inf), and consequently HES total plasma clearance, clearly depended on renal function (i.e., on CLcr). Figure 2 displays the correlation between baseline Clcr and AUC(0inf). The coefficient of correlation was calculated as -0.76 being significantly different from zero (P < 0.001). This relationship seemed even stronger when controlling for the potentially influencing factors of age, body mass index, height, and weight: the corresponding partial correlation coefficient resulted to -0.89.
Subjects with a CLcr <50 had, on average, a 73% larger AUC(0inf) than subjects with a CLcr >50. The 95% confidence interval for the mean AUC ratio of 1.73 was 1.44, 2.07, which indicated that a more than doubled AUC in moderate-to-severe renally impaired subjects can be excluded. Inversely, HES total plasma clearance decreased with decreasing baseline CLcr (Fig. 3); however, the proportional decrease was less than expected from the reduction in CLcr for a compound such as HES, which is glomerularly filtered only.
The plasma Cmax was not affected by renal impairment (r = -0.07, P = 0.77). Also, renal impairment did not influence terminal half-life, which was on average 16.1 h and ranged between 14.1 and 18.9 h (r = 0.31, P = 0.19). The cumulative amount of drug excreted into urine within 72 to 96 h was 59% of the total dose infused in subjects with a CLcr >30, whereas it was 51% in the severely impaired group (15<30 mL · min-1 · 1.73 m-2; Fig. 4).
HES pharmacokinetics are often described using a two-compartmental model (14). The two-compartmental model also provided a valid fit to the individual HES plasma concentration-versus-time curves in this study.
The The mean Mw of the original 6% solution was analyzed as 130,652 d. In the first sample taken (after 0.5 h), mean Mw was 62,704 d, and in the second sample (after 1 h), the mean value was 59,117 d. However, the Mw tended to decrease with decreasing baseline CLcr (r = 0.63, P = 0.04 after 0.5 h). Baseline serum amylase was inversely correlated with CLcr (r = -0.72, P < 0.001), and Mw in plasma after 1 h was also inversely correlated with baseline amylase (r = -0.76, P < 0.001) (Fig. 5). In subjects with higher baseline amylase, lower HES Mw after 1 h can be regarded as an indicator of accelerated intravascular breakdown, because there was no indication that the lower plasma Mw was caused by a selective retention of smaller HES molecules in these subjects. In fact, partial correlation analysis provided evidence for an association between baseline amylase activity in the metabolism of HES independent of CLcr.
There was no deterioration in renal function after dosing, as determined by repetitive CLcr measurements in the individual subjects. On average, CLcr even increased by 7.8 mL · min-1 · 1.73 m-2 (SD = 16.6, P = 0.054) from prestudy to poststudy. All AE had been recorded throughout the study, whether or not causally related. There was no drop-out. Eight of 19 subjects experienced 17 AE, none of them serious or otherwise important, all lasting <24 h, except for urinary infection. Headache, nausea, transient vision disturbances, and precordial pressure (all rated as "mild") were classified as possibly related to the study medication, but could alternatively be explained by the volunteer setting. The other AE had an unlikely relationship to the study drug. Interestingly, there was no AE at all in the worst renal function group (CLcr 15<30). Amylase increased postdose, which however was judged as clinically irrelevant (predose mean: 62.5; 8 h: 96.9; 24 h: 75.4; 48 h: 65.4 U/L), because this is a well known effect for HES based on the formation of macro-amylase (16), which is excreted more slowly and not related to pancreatic dysfunction. The transient, induced amylase increase did not correlate with in vivo Mw (data not shown), as opposed to the correlation of baseline amylase and Mw described above. Overall, no clinically relevant laboratory changes occurred. Also, the coagulation variables studied were unremarkable: partial thromboplastin time 35 ± 6 s at 8 h (baseline 33 ± 6 s), thrombin time 16 ± 1 s at 8 h (16 ± 1 s), and platelets 237 ± 49/nL (238 ± 54/nL).
The pharmacokinetics of HES 130/0.4 were evaluated in subjects with stable renal insufficiency. HES 130/0.4 was developed to improve pharmacokinetics of current medium Mw HES whereas maintaining efficacy. Studies in healthy volunteers after single (14) and multiple administration of 500 mL for 10 days (2) showed rapid renal excretion and no plasma accumulation. In contrast, older starches, such as HES 450/0.7 accumulate in plasma after multiple infusions (17,18). This has been known since the 1970s, but is often not acknowledged. Large residual concentrations of colloid after 24 hours should be avoided, because they are not expected to exert a relevant therapeutic effect, but could be related to potential side effects. In this study, we found residual concentrations 24 hours after 500 mL HES 130/0.4 to be small, even in subjects with severe renal dysfunction. In the worst renal function group (CLcr 1530), the mean HES plasma concentration at 24 hours was 0.5 mg/mL. Persisting large plasma concentrations beyond the duration of volume effect (which lasts 46 hours in the case of HES 130/0.4) (19) should be avoided, because potential harm will not be balanced by the benefit. As a comparison, 24 hours after the same amount of hetastarch (450/0.7, 6%) was given to volunteers with normal renal function, the concentration was 4 mg/dL (17). In a recent study with pentastarch (200/0.5, 500 mL, 10%), approximately 1.5 mg/dL HES was found in plasma after 24 hours, again in normal volunteers (20). Hence, residual plasma concentrations after 24 hours in our subjects with mild-to-severe renal impairment were smaller than those in normal volunteers after hetastarch or pentastarch in the cited studies. In a pilot study, Gassmayr et al. (21) revealed no clinically relevant differences in half-time and plasma HES concentrations using a single HES 200/0.5 (10%) infusion in patients with renal impairment compared with healthy subjects; therefore, no undue risks with an improved HES specification were to be expected. The current study with HES 130/0.4 for the first time gives extensive pharmacokinetic data of a HES in dependency of well defined, stable, non-anuric renal dysfunction. Cmax and terminal half-life were not dependent on renal function, whereas AUC increased moderately. HES total plasma clearance in renal dysfunction decreased less than expected from the reduction in CLcr. This might be attributed to the higher baseline amylase levels in subjects with renal dysfunction, because higher baseline amylase levels were also associated with lower in vivo Mw, indicating increased HES metabolism. Data analysis showed that the effect of amylase can be postulated as independent of baseline renal function although both amylase and Mw were closely related with baseline renal function expressed as CLcr. Thus, the results suggest that the increased baseline amylase levels (caused by the renal insufficiency itself) in part compensated for the reduced glomerular filtration rate, explaining the higher than expected HES clearance. The infusion volume was fixed to 500 mL in all subjects to allow comparability to the previous data generated in normal volunteers (14). Because the four groups were comparable with regard to body weight, the individual differences concerning the dose per kilogram of body weight did not affect the study results and conclusions presented in this report. A limitation of all pharmacokinetic studies with colloidal infusion solutions is that they influence their own volume of distribution (which was not a primary variable of this study) because of their volume-expanding effect. Overall, HES 130/0.4 in subjects with mild-to-severe renal impairment showed excellent urinary excretion compared with HES 130/0.4 in healthy volunteers (14), and small residual plasma concentrations after 24 h when compared with other HES preparations in healthy volunteers. After our single-dose application, we found no deterioration in renal function, as measured by CLcr determination in duplicate after exposition. Dehne et al. (22) did not find an effect of HES on renal function or sensitive tubular markers in surgical patients without prior renal impairment. Kumle et al. (23) studied the influence of 3 different intravascular volume replacement regimens (HES 70/0.5, HES 200/0.5, and gelatin) on renal function in elderly patients without prior renal dysfunction undergoing major abdominal surgery. As assessed by sensitive markers of renal function, all three regimens could be used safely for volume replacement without renal dysfunction. Schortgen et al. (24) compared HES 200/0.62 (6%) to a 3% gelatin solution in patients with severe sepsis, and described HES 200/0.62 to be a risk factor for deterioration of renal function. However, as noted in several letters (25), the conclusions of that study should not be generalized. Baseline renal function is a known major risk factor for the development of postoperative renal dysfunction, as determined in many other studies (2628). Among the HES solutions, it is important to differentiate between those with a longer-lasting volume effect accompanied by plasma accumulation after repetitive use (high Mw and/or highly substituted, HES 450/0.7 or 200/0.62), and medium Mw starches (HES 200/0.5, HES 130/0.4), which are more easily excreted renally, as shown for HES 130/0.4 in this study. Residual HES plasma concentrations after 24 hours were small. Considering these and the pharmacokinetic results of this study, we conclude that HES 130/0.4 (500 mL 6%) can be administered to patients even with severe renal impairment, as long as urine flow is preserved, without the risk of plasma accumulation. Further studies in renally impaired patients in the perioperative setting and in intensive care are justified.
Supported by a grant from Fresenius Kabi Deutschland GmbH, Bad Homburg, Germany.
Presented in part at the 21st International Symposium on Intensive Care and Emergency Medicine, Brussels, Belgium, March 2023, 2001.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|