Colloids Versuss Crystalloids - Article Example

www.allwriting.net Sumanta Sanyal Dated: 5/05/2006 Introduction In recent times there has been considerable and continuous controversy over the risk factors associated with application of colloids to critically ill patients. Clinical studies have shown that there is considerable difference in the effects of colloids and crystalloids on a range of physiological parameters. This difference in effects has been clinically studied in the course of meta-analyses of mortality in randomized trials of critically ill patients applied with either colloids or crystalloids as comparative fluids applied for volume expansion. Nevertheless, at the very beginning of this analysis, it is noted that review of recent literature has revealed that, though such meta-analyses and other studies have revealed that colloid usage in the critically ill indicates higher than normal risk of mortality, affected physiological parameters are not universally defined nor has there been any progress towards emergence of any definite applications guidelines. (Schierhout and Roberts, 1998) Since the particular colloid being investigated here is albumin some little is being said here about the colloid. Albumin is a synthetic colloid derived from pooled plasma collection and hence there is always the risk of infection. Aside from this there are other reasons why the controversy over its and other colloids’ usage continues. These are as follows. They have a hitherto undefined potential for adverse effects. They are relatively costlier than crystalloids. There is a lack of consensus regarding their application. (American Thoracic Society Consensus Statement, 2004) The American Society Consensus Statement is a very comprehensive document on colloids and it has listed a number of adverse effects of colloid application with particular physiological explanations. This analysis believes that such information should be more forthcoming if the controversy is to be cleared once and for all. In the light of that it is also noted that the analysis report includes in a synopsis some of the salient characteristics of albumin. This is because only by studying the chemistry of the colloid and its subsequent interaction with human physiology, both normal and diseased, can the exact mechanism or mechanisms of the adverse effects be unraveled. In this context not only the chemistry but also the genetics, which primarily initiates the chemistry, should be studied. Only the physiological chemistry is touched upon here in a rudimentary fashion. No genetics is provided. Objectives of the Analysis: To quantitatively determine the effect on mortality of administering albumin or plasma protein fraction on critically ill patients. Albumin: The Colloid Albumin was the first natural colloid to be used clinically as a volume expander and it remains the standard colloid in comparison to other colloids being used clinically on critically ill patients. Colloids: Colloids are usually large molecules that do not pass through the thin capillary walls and they exert an oncotic pressure that restores fluid balance. They are usually applied to restore intravascular volume and improve tissue perfusion. (Albumin, US Pharmacist, 2000) Clinical usage of albumin: Albumin is applied to increase intravascular oncotic pressure and expand intravascular volume in patients with hypovolemic shock, severe burn injury, adult respiratory distress syndrome (ARDS), ascites, liver failure and pancreatitis. It is also clinically applied to patients undergoing cardiopulmonary bypass. It is also sometimes used to treat neonatal hyperbilirubinemia, hypoproteinemia and nephritic syndrome. (Albumin, US Pharmacist, 2000). It is also noted at the very beginning that, at present, the usage of albumin in patients for the above purposes is under controversial analysis with the US Food and Drug Administration (FDA) strongly advising physicians to be discrete in the use of albumin (FDA, CBER, 1998). It bases its cautious notice on the Cochrane Injuries Group report that shall be more comprehensively dealt with later in this analysis report. The FDA further cautions, as on that date, that physicians should wait till more research has been done on the adverse effects of albumin therapy and some positive decision has been reached upon its continued usage. Albumin is a protein colloid. Its costs are high compared to non-protein colloids like hetastarch and dextran. It is also higher on costs to other alternatives like crystalloids (Ringer’s lactate) and normal saline. This high cost has created a controversy among physicians. (Albumin, US Pharmacist, 2000) It is also necessary to understand how colloids work in comparison to crystalloids. In patients with intact intravascular linings colloids remain within the intravascular space and cause less edema than crystalloids. Also, fewer amounts of colloids are needed compared to crystalloids to initiate a particular degree of volume expansion. (Albumin, US Pharmacist, 2000) Colloidal action: Colloids are defined according to their molecular weight and size. Monodisperse colloidal solutions are comprised wholly of particles with uniform weight and size. Polydisperse solutions may be made up of many types of particles that differ in molecular weight and size. (Albumin, US Pharmacist, 2000) Nevertheless, in solutions, the osmotic pressure is determined by the solute’s kinetic energy and the concentration of the solute and not on its molecular weight or size. (Albumin, US Pharmacist, 2000) However, molecular weight and particularly size does determine how fast a solute can be cleared off the intravascular space of a patient. In this size is inversely proportional to osmotic pressure and directly proportional to the time of residence within the intravascular space. (Albumin, US Pharmacist, 2000) The ideal colloid: Clinically an ideal colloid would produce a sustained increase in plasma volume without posing a risk for anaphylaxis, edema and infection. The true aim of this paper is to investigate to what extent albumin can be called an ideal colloid in the clinical sense, if at all. (Albumin, US Pharmacist, 2000) Physical Characteristics: Albumin is one of the three principal proteins in plasma. Under normal physiological conditions its concentrations in plasma is 4.5 g/dL. It is a monodisperse solution with an average molecular weight of 66,300 D to 69,000 D. The colloid is highly water-soluble and carries a strong negative charge of –17 at pH 6.4 to 7.4. This negative charge explains why albumin is removed so rapidly when plasmapheresis is employed to remove harmful substances attached to the colloid. (Albumin, US Pharmacist, 2000) Albumin comprises 75-80 % of the normal colloid oncotic pressure and 50-60 % protein in plasma. 1 gram of intravascular albumin binds to 18mL of water at normal oncotic pressure. Albumin enters the intravascular space by two routes – passing directly through the hepatocytes into the sinusoids, and traversing the space of Disse (the gap between the hepatocytes and the sinusoids) and entering the hepatic lymphatic system and the thoracic ducts. Tests show that, after 2 hours, 90 % of the albumin remains in the intravascular space. The half-life of albumin in plasma is 16 hours and it is removed from the body by the reticuloendothelial system. Under normal pathological conditions daily loss of albumin from the intravascular space is 10 %. (Albumin, US Pharmacist, 2000) Albumin is produced in the liver at the rate of 150-200 mg/kg/day. Under normal liver conditions thyroid hormone and cortisol stimulate RNA production to induce albumin synthesis. The main synthetic mechanism is the colloidal oncotic pressure (COP) within hepatocytes. The serum albumin concentration does not provide feedback for albumin synthesis. Fall or rise in concentration does not affect synthesis as the COP mostly determines it. (Albumin, US Pharmacist, 2000) Binding Sites: Albumin is unique among colloids and crystalloids because it has the ability to bind reversibly to both cations and anions. It has three binding sites – an acidic, a basic and a neutral site. This enables it to play important roles in transport of lipids and lipid-soluble materials. It is a carrier molecule of fatty acids, hormones, enzymes, dyes, trace metals and drugs and this property enables it to regulate the extracellular concentrations of these substances. Also, since protein-bound substances are rendered inactive by it, albumin indirectly controls the biological activities of these substances. These binding properties complement its volume expanding capabilities. (Albumin, US Pharmacist, 2000) Required Studies: Some of the studies that are required to correctly ascertain how albumin adversely affects the physiologies of critically ill patients are as below: Acute toxicity studies Multiple dose toxicity studies Mutagenicity studies Pharmacokinetic studies Antigenicity studies (Zander, 2006) Though most of these studies are probably conducted for synthesised albumin yet, since there is no consensus on the issue of its application in critically ill patients, more obviously has to be done. The Cochrane Injuries Group Albumin Reviewers’ Report The ‘Cochrane Injuries Group Albumin Reviewers’ Report’ is one of the primary meta-analyses conducted on randomized trails of critically ill patients infused with either colloids or crystalloids or both in varying concentrations and varying infusion times. It is one of the primary documents on which this analysis itself bases its procedure. Though the exact methodology applied by the reviewers of the Cochrane Injuries Group is different from that applied by this analysis nevertheless the basic logic coincides sufficiently to enable this analysis to declare that it has adopted its methodology from that report. Findings: The report, prepared by Dr. Ian Roberts, declares that in patients with acute and chronic illness serum albumin concentration is inversely related to risk of death. It is noted that, for the purpose of the meta-analysis, mortality was taken as endpoint. The principal features of the report are inserted hereafter. For each patient category the risk of death in the albumin treated group was higher than of the comparison group. For hypovolaemia, the relative risk of death after albumin administration was 1.46 (95 % confidence interval 0.97-2.22). For burns the relative risk was 2.4 (1.11-5.19). For hypoalbunaemia the relative risk was 1.69 (1.07-2.67). Pooled risk of death with albumin administration was 1.68 (1.26-2.23). Pooled difference in the risk of death with albumin was 6 % (95 % confidence interval 3 % to 9 %) with a fixed effect model. The overall findings suggest that for every 17 critically ill patients treated with albumin there is an additional death. (Ian Roberts, 1998) Against this distinguished backdrop the analysis shall conduct its won study and it shall be seen how that fares against this primary report. Methodology The analysis was conducted on the patient care data comprised for 72 critically ill patients in intensive care units. It is noted that the original data sheet contained data on the patient’s ailment but these are too diverse for such a small sample to allow separate mortality analysis on albumin administration by ailment type, as in the Cochrane report which utilized 30 randomized controlled trials on 1419 randomized patients. Thus, pooled risk of death with albumin administration is derived. The dosage rate, frequency and duration are also too diverse for such a small collection to be analyzed upon with any logical outcomes. Table 1 of the appendix contains the patient mortality data. 0 implies no death while 1 implies death. Since these are the two primary variables the analysis bases itself on these are taken into consideration. The indicator column is there as a control against which regression analysis can be done. It contains all zeroes, implying no death, against which the principal and furthest regression is death at 1. Variables: The two main variables are patients with only crystalloid dosage (C ) and patients dosed with both crystalloids and colloids, albumin this case (CA). . Criteria for inclusion: Of the 72 patients whose data were initially collected only 64 had been included for the purpose of the analysis. This simply because there are 32 patients who had been dosed with both albumin and crystalloid while there are 40 patients who had been dosed with only crystalloids. To make even pairs for variables analysis it is noticeable that only 32 pairs are possible. . Statistical methods employed: Descriptive statistics: Since the range of the variables is small at 0 (no death) to 1 (death) and there are only two possible variable values 0 and 1, descriptive statistics with calculation of mean and standard deviation is expected to reveal large trends within the two variables CA (with albumin) and C (only crystalloids). The descriptive statistics are enhanced with computation of Pearson’s correlation coefficient and p-values at 5 % significance levels. Multiple correspondence analysis: A multiple correspondence analysis has been conducted at 5 % significance levels to analyze how the values are ranged across the two sets of variables. Chi-square test: A Chi-square test has also been run to test the independence of the two sets of data. It has a significance level of 5. Results and Discussion Descriptive statistics: Table 2 depicts the summary statistics. It is observable that the means for the two sets of variables are 0.25 for the CA group and 0.094 for the C group. This is significant in the sense that the CA group tends towards morality much more than the C group. This is the first indication within the analysis that albumin administration has more risk of death than crystalloid administration alone. The standard deviations too, at 0.44 for the CA group and 0.296 fro the C group, points in the same direction as the means. Table 3 plots the Pearson’s correlation matrix at 0.062, much less than 1 the perfect correlation, for the two sets of variables. This signifies that the two sets of variables are almost insignificantly correlated. Table 4 gives the p-values at 0.736. This signifies that though correlation is insignificant the probability of occurrence of values within the two variables is above 0.5. This does not signify much as the only two possible values of the variables are 0 and 1. Nevertheless, where the C group has observable less mortality figures this signifies that the CA group has a higher mortality rate. Multiple correspondence analyses: The total inertia is 1. Table 5 gives the eigenvalues and percentages of inertia. Table 6 gives the variables contribution. The rest of the analyses data have been discarded as they are not deemed essential for this small level of analysis. It is found from Table 6 that the relative weight for the CA-0 (no death) is less at 0.375 compared to the C-0 group at 0.453. The same is true for the CA-1 (death) at 0.125 against C-1 at .0.047. This signifies a relatively much higher risk of death with albumin administration against only administration of crystalloids. From Table 5 it is found that, for the CA group, out of 32 patients administered with albumin there were 8 deaths. For the C group with the same number of patients the death figures are low at 3. Value of alpha is at 0.05 implying almost insignificant association between the two sets of variables. Chi-square test: Table 7 gives the p-value at Read More