1 Institute of Anaesthesia and Intensive Care and 2 Institute of Hygiene, Catholic University of the Sacred Heart, Largo Francesco Vito 1, 00168 Rome, Italy
* Corresponding author: Institute of Anaesthesia and Intensive Care, Largo Francesco Vito 1, 00168 Rome, Italy. E-mail: f.cavaliere{at}rm.unicatt.it
Accepted for publication November 26, 2004.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Methods. Ten critically ill hypoalbuminaemic patients (<24 g litre1) and 10 critically ill normoalbuminaemic patients (>32 g litre1) were included in this study. They underwent sedation with propofol, aimed at a Ramsey sedation score of 45. The Diprifusor was used to achieve target propofol plasma concentrations that ranged between 0.6 and 1.5 mg litre1. Propofol concentration was measured by high-performance liquid chromatography 5 min, 15 min, 30 min, 1 h, 2 h, 4 h, 6 h and 8 h after starting TCI. The accuracy of TCI was evaluated by calculating performance errors [PE=100x(measured concentrationpredicted concentration)/predicted concentration], absolute and relative individual median performance errors (MDAPE and MDPE) and divergence (the slope of individual regression lines between PEs and time).
Results. PEs [median (range)] were 7 (65, 79) in hypoalbuminaemic patients and 2 (53, 188) in normoalbuminaemic patients; absolute PEs were 21 (1, 79) and 22 (0, 188). No significant difference was observed between the two groups. MDPE, MDAPE and divergence values were also similar. In most patients the accuracy of TCI increased with time because higher PE values were observed during the first 30 min.
Conclusions. Hypoalbuminaemia does not affect the accuracy of Diprifusor during sedation with propofol in critically ill patients.
Keywords: anaesthetics i.v., propofol ; complications, hypoalbuminaemia ; infusion, target controlled ; sedation
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Hypoalbuminaemia, one of the frequently observed physiological derangements in critically ill patients, can affect the pharmacokinetics of drugs that strongly bind to albumin.710 Usually 98% of plasma propofol is bound to this protein,11 12 but the unbound fraction may increase in the presence of hypoalbuminaemia, as suggested by studies performed in vitro on plasma from critically ill patients.12 Consequently, it seemed useful to verify the degree of accuracy of the Diprifusor in the presence of low plasma albumin concentrations. The aim of this study was to compare Diprifusor performance during sedation in hypoalbuminaemic and normoalbuminaemic patients.
![]() |
Methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Each patient was sedated with propofol; this was given through the central venous catheter using Diprifusor TCI (Zeneca, UK). The target concentration was set at 1 mg litre1 initially; subsequently, it was adjusted to achieve a Ramsey sedation score between 4 (a brisk response to a light glabellar tap) and 5 (a sluggish response to a light glabellar tap). Each patient was weighed and the measured weight was used for TCI calculations. Before starting the sedation, patients were questioned for pain and, if required, morphine was given in small bolus doses of 12 mg i.v. until satisfactory analgesia was obtained [the patient scored 1 or 2 on a scale between 1 (no pain) and 5 (maximal pain)]. In addition, colloids 810 ml kg1 were given to prevent arterial hypotension. During sedation, if systolic arterial pressure decreased to <90% of the presedation values, normal saline 0.51 litre was given. Persistent hypotension was treated with small doses of a vasoconstrictor (etilefrine hydrochloride, 12 mg i.v.).
The first 8 h of sedation was taken for this study. Monitoring consisted of heart rate, arterial pressure, respiratory rate and pulse oximetry. Nine heparinized arterial blood samples were collected: one before TCI and the others 5 min, 15 min, 30 min, 1 h, 2 h, 4 h, 6 h and 8 h after starting TCI. Blood was rapidly centrifuged and plasma was stored at 60°C. Total propofol concentration was determined by high-performance liquid chromatography using Plummer's method13 after solidliquid extraction with Oasis cartridges (Waters, USA). Plasma albumin was measured in the presedation blood sample by the central laboratory of the hospital using a colorimetric assay (ALB plus; Roche Diagnostics, Germany) using a Roche/Hitachi 917 analyser. Coefficients of variation of propofol and albumin estimations were 4.9 and 3.0% respectively.
The performance error (PE) corresponding to each sample was calculated as PE=100x(measured concentrationtarget concentration)/(target concentration).
In each subject, median absolute performance error (MDAPE) was calculated as the median of the absolute PEs, i.e. ignoring positive or negative signs, and median performance error (MDPE) was calculated as the median of PEs calculated with positive and negative signs maintained. MDPE was taken to indicate bias14 (i.e. the systematic tendency to under- or overestimate the measured concentration); MDAPE was taken to indicate inaccuracy, providing information on the typical size of the differences between measured and targeted concentrations. The divergence, i.e. the slope of the regression line between PE and the time elapsed from starting TCI, was taken to indicate time-related changes in predicting TCI. In comparison with crude PE and absolute PE (APE) values, MDPE, MDAPE and divergence are not influenced by within-subject variance.
Statistical analysis was performed using the software Statistica for Windows (StatSoft, USA). Data were tested for normality by examination of histograms and by the ShapiroWilks test and were accordingly reported as mean (SD) or median (range). Comparisons were performed using the MannWhitney U test.
A priori power analysis for planning sample size was performed with the software G*Power obtained from the University of Düsseldorf's website.15 We assumed that hypoalbuminaemia would increase the Diprifusor's inaccuracy by at least 50% of the standard deviation of the population and would consequently have a significant impact on clinical practice, i.e. we arbitrarily decided to search for a medium size effect according to Cohen's convention.16 In power analysis this corresponds to choosing a value of 0.5 for parameter d, defined as the ratio between the difference of the two means and the standard deviation of the population. The analysis was first applied to an independent t-test with equal group sizes and equal , and this showed that, to achieve a power (1ß) of 0.851 with
=0.05, it was necessary to compare 146 samples (73 vs 73). Then the sample size was corrected for the MannWhitney U-test by dividing 146 by the asymptotic relative efficiency (or Pitman efficiency) of the U-test, which is 0.955. The final sample size was 152.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Hypoalbuminaemia is frequently observed in critical illness and results from transcapillary leak, decreased synthesis, large losses of body fluids and dilution caused by fluid resuscitation. Although albumin has several functions, such as maintaining colloid osmotic pressure in the vascular system, transporting fatty acids and bilirubin, and scavenging free radicals,18 correction of hypoalbuminaemia by albumin administration has not decreased the risk of death; in fact it has been associated with increased mortality.18 We did not correct hypoalbuminaemia in patients included in this study because to do so is considered unnecessary and potentially dangerous.19 20 Nevertheless, some workers suggest that severe hypoalbuminaemia (<20 g litre1) should be corrected21 and that albumin infusion may decrease fluid requirements and tissue oedema.22 Septic patients, who were excluded from this study, may also represent an exception and benefit from albumin administration.23 24
Hypoalbuminaemia may affect both the pharmacodynamics and the pharmacokinetics of drugs that are highly bound to albumin.710 By increasing the free drug concentration in plasma, hypoalbuminaemia may increase the drug concentration at effect sites and enhance the pharmacological actions.25 In this study, however, no increase in the sedative effect of propofol was apparent because targeted concentrations were similar in the two groups.
The pharmacokinetic effects of low plasma albumin are mainly related to the changes in clearance and distribution of the drug.26 Renal clearance usually parallels free drug concentration because only the unbound drug is filtered by the glomerulus and high free drug concentrations may increase tubular secretion. Also, hepatic clearance may be affected by changes of free drug concentrations, but usually hepatic blood flow and intrinsic clearance of unbound drug have larger influence. Theoretically, TCI inaccuracy resulting from abnormal drug clearance may not be immediately evident, but may become apparent later, when the influence of drug distribution is less pronounced. By contrast, in this study we observed a downward trend in TCI inaccuracy, which suggested that propofol clearance was not appreciably affected, although 8 h was probably not long enough to get conclusive data in this regard. Other studies have also reported larger Diprifusor inaccuracy initially after starting TCI, and have hypothesized that this might be caused by incomplete mixing.5
Hypoalbuminaemia usually increases the volume of distribution of drugs highly bound to albumin26 and this may potentially affect TCI accuracy. For TCI systems, the target concentration is kept constant by keeping the content of the central compartment constant. Such content is calculated by multiplying the targeted concentration by the estimated volume of the central compartment. If the propofol central compartment had been increased in hypoalbuminaemic patients, a bias would have been apparent towards measured drug concentrations being lower than targeted concentrations, i.e. towards negative predictive errors. By contrast, in this study no significant bias of Diprifusor accuracy was seen in either group.
A possible explanation for these results is that the effect of hypoalbuminaemia may be too small in relation to the overall degree of accuracy achieved by TCI devices. The power of this study was adequate to show a medium-sized effect, i.e. an effect that is at least equal to half the standard deviation of the population. Swinhoe and colleagues reported a standard deviation of APE of about 42% in a study on Diprifusor accuracy.5 Assuming that the standard deviation of the population is near this value, a medium-sized effect of hypoalbuminaemia would correspond to a difference of 21 in APE between hypoalbuminaemic and normoalbuminaemic patients. Although this difference may appear large, we thought that a lower effect would not be of clinical importance, given the large variability of PE.
A second possible mechanism could be that propofol binds to albumin linearly in the range of concentrations tested in this study. Hypoalbuminaemia does not affect the pharmacokinetics of drugs that are highly, but linearly bound to albumin27 because the percentage of free drug is unaffected. The relationship between bound and free propofol in plasma has not been investigated in relation to albumin concentrations, but only to changes in total propofol concentration.28 This relationship was not linear and free propofol percentage increased by 30% (from 1.2 to 1.7%) from the lowest (0.5 mg litre1) to the highest propofol concentrations tested (16 mg litre1). Nevertheless, the relationship can be regarded as linear in the lower part of the curve, suggesting non-saturable binding sites at concentrations up to two- or three-fold higher than those used for sedation. A low-capacity, high-affinity site primarily involved at total concentrations lower than 1020 mg litre1 may compensate for a loss of global capacity for propofol binding.28
In conclusion, this study shows that, during sedation with propofol, the accuracy of the Diprifusor is unaffected in the presence of hypoalbuminaemia, although our data cannot entirely rule out the risk of a loss of precision in long-term sedation. Further studies are needed to evaluate Diprifusor accuracy in hypoalbuminaemic patients at higher propofol concentrations, such as those used during anaesthesia.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
2 Gray JM, Kenny GNC. Development of the technology for Diprifusor TCI system. Anaesthesia 1998; 53: 227[ISI][Medline]
3 McMurray TJ, Johnston JR, Milligan KR, et al. Propofol sedation using Diprifusor target-controlled infusion in adult intensive care unit patients. Anaesthesia 2004; 59: 63641[CrossRef][ISI][Medline]
4 Van den Nieuwenhuyzen MCO, Engbers FHM, Vuyk EJ, Burm AGL. Target-controlled infusion systems. Role in anaesthesia and analgesia. Clin Pharmacokinet 2000; 38: 18190[ISI][Medline]
5 Swinhoe CF, Peacock JE, Glen JB, Reilly CS. Evaluation of the predictive performance of a Diprifusor TCI system. Anaesthesia 1998; 53: 617[ISI][Medline]
6 Servin FS. TCI compared with manually controlled infusion of propofol: a multicentre study. Anaesthesia 1998; 53: 826[ISI][Medline]
7 Pichette V, Geadah D, du Souich P. Role of plasma protein binding on renal metabolism and dynamics of furosemide in the rabbit. Drug Metab Dispos 1999; 27: 815
8 Romano S, Fdez de Gatta MM, Calvo MV, Caballero D, Dominguez-Gil A, Lanao JM. Population pharmacokinetics of amikacin in patients with haematological malignancies. J Antimicrob Chemother 1999; 44: 23542
9 Mimoz O, Soreda S, Padoin C, Tod M, Petitjean O, Benhamou D. Ceftriaxone pharmacokinetics during iatrogenic hydroxyethyl starch-induced hypoalbuminemia: a model to explore the effects of decreased protein binding capacity on highly bound drugs. Anesthesiology 2000; 93: 73543[CrossRef][ISI][Medline]
10 Lagneau F, Perbet S, Delefosse D, Wernet A, Stocco J, Marty J. Drugs pharmacokinetics in ICU patients: consequences of hypoalbuminemia upon drugs monitoring and dosing scheme. Intensive Care Med 2004; 30: 1247[CrossRef][ISI][Medline]
11 Costela JL, Jiménez R, Calvo R, Suárez E, Carlos R. Serum protein binding of propofol in patients with renal failure or hepatic cirrhosis. Acta Anaesthesiol Scand 1996; 40: 7415[ISI][Medline]
12 Zamacona MK, Suárez E, Aguilera L, Rodriguez-Saslain JM, Aguirre C, Calvo R. Serum protein binding of propofol in critically ill patients. Acta Anaesthesiol Scand 1997; 41: 126772[ISI][Medline]
13 Plummer GF. Improved method for the determination of propofol in blood by high performance liquid chromatography with fluorescence detection. J Chromatogr 1978; 421: 1716
14 Varvel JR, Donoho DL, Shafer SL. Measuring the performance of computer-controlled infusion pumps. J Pharmacokinet Biopharm 1992; 20: 6394[ISI][Medline]
15 University of Düsseldorf. www.psycho.uni-duesseldorf.de/aap/projects/gpower/. Accessed September 20, 2004
16 Cohen J. Statistical Power Analysis for the Behavioural Sciences, 2nd edn. New York: Academic Press, 1977
17 Schüttler J, Kloos S, Schwilden H, Stoeckel H. Total intravenous anaesthesia with propofol and alfentanil by computer assisted infusion. Anaesthesia 1988; 43: 27[ISI]
18 Quinlan GJ, Mumby S, Martin GS, Bernard GR, Gutteridge JM, Evans TW. Albumin influences total plasma antioxidant capacity favorably in patients with acute lung injury. Crit Care Med 2004; 32: 7559[CrossRef][ISI][Medline]
19 Alderson P, Bunn F, Lefebvre C, Li WP, Li L, Roberts I, Schierhout G. Human albumin solution for resuscitation and volume expansion in critically ill patients. Cochrane Database Syst Rev 2002; 1: CD001208[Medline]
20 Polimood TB, Park GR. Debate: albumin administration should be avoided in the critically ill. Crit Care 2000; 4: 1515[CrossRef][ISI][Medline]
21 Consensus conference. Use of albumin solutions in surgical anesthesia and surgical intensive care of adults. Ann Fr Anesth Reanim 1996; 15: 407568[ISI][Medline]
22 Haynes GR, Navickis RJ, Wilkes MM. Albumin administration. What is the evidence of clinical benefit? A systematic review of randomized controlled trials. Eur J Anaesthesiol 2003; 20: 77193[ISI][Medline]
23 Finfer S, Bellomo R, Boyce N, French J, Myburgh J, Norton R, SAFE study investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350: 224756
24 Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med 1999; 341: 4039
25 Wood M. Plasma binding and limitation of drug access to site of action. Anesthesiology 1991; 75: 7213[ISI][Medline]
26 Grandison MK, Boudinot FD. Age-related changes in protein binding of drugs. Implication for therapy. Clin Pharmacokinet 2000; 38: 27190[ISI][Medline]
27 Bertucci C, Domenici E. Reversibile and covalent binding of drugs to human serum albumin: methodological approaches and physiological relevance. Curr Med Chem 2002; 9: 146381[ISI][Medline]
28 Mazoit JX, Samii K. Binding of propofol to blood components: implications for pharmacokinetics and for pharmacodynamics. Br J Clin Pharmacol 1999; 47: 3542[CrossRef][ISI][Medline]
29 Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993; 270: 295763[Abstract]