Flocculation of Intravenous Fat Emulsions

State of the art review: Intravenous fat emulsions: Current applications, safety profile, and clinical implications

OBJECTIVE

To review the current state of the science regarding intravenous fat emulsions (IVFEs), with an emphasis on their safety profile.

DATA SOURCES

Articles were identified via a search of the MEDLINE database, including publications from 1979 to December 2009, using a search string that included the terms parenteral nutrition, lipid emulsion, fat emulsion, IVFE, safety, adverse effect, neonate intralipid, and terms describing a range of specific adverse events (AEs) such as pancreatitis.

STUDY SELECTION AND DATA EXTRACTION

We selected articles that allowed us to compare the results of clinical trials involving delivery of medications via IVFEs with the historical use and effects of IVFEs in parenteral nutrition, with an emphasis on AEs. We focused on 2 drugs in current use that are administered intravenously in lipid emulsions: propofol and clevidipine.

DATA SYNTHESIS

Clearance of the fat particles in IVFEs is mediated by the enzyme lipoprotein lipase. AEs are more likely if the rate or duration of IVFE administration exceeds the enzyme's clearance capacity. AEs are also more likely after administration of a 10% IVFE formulation than a 20% formulation, because the higher concentration of free phospholipid in the 10% formulation interferes with lipoprotein lipase activity. AEs can be reduced by administering IVFEs at a dosage < or = 2.5 g/kg/day and at a rate < or = 0.11 g/kg/h. The anesthetic agent propofol, which is formulated in a 10% IVFE, has been used clinically for 25 years. Typical AEs associated with propofol use include infection, high plasma triglyceride concentrations, and pancreatitis. Recent clinical trials involving clevidipine, which is formulated in a 20% IVFE, have demonstrated a low rate of lipid-related AEs.

CONCLUSIONS

The results of this review demonstrate that IVFEs are well tolerated when administered in accordance with guideline recommendations.

The pathogenesis of fat embolism

Fat embolism is a common autopsy finding in patients with or without a history of trauma. There are two basic mechanisms causing fat to embolize. Depot-derived fat embolism arises by disruption of depot fat, usually as a result of trauma, allowing direct entry into the bloodstream. Plasma-derived fat embolism is caused by agglutination of endogenous or infused exogenous fat such as Intralipid, with consequent embolism. Chylomicrons and Intralipid liposomes are known to undergo calcium-dependent agglutination by C-reactive protein (CRP), and this may play a role in vivo in this type of fat embolism.

The all-in-one nutrient solution in parenteral nutrition

Dextrose, lipid emulsion, amino-acids, electrolytes, vitamins and trace elements were mixed in a well defined order to prepare 10 different total nutrient admixture solutions. The pH and osmolality of these solutions did not change significantly during 14 days of storage at 4 degrees C. Mean diameters of lipid particles on different occasions for the different solutions varied from 0.31 +/- 0.10 mum to 0.48 +/- 19 mum with 95% of particles less than 0.6 mum. Particles greater than 6 mum were not observed. Cultures were all negative. 243 patients received 5101 infusions of the admixture fluids. Patients on treatment for more than 1 week showed improved nitrogen balance. Serum transferrin and albumin were increased when treatment was ongoing for more than 2 weeks. No adverse reactions or abnormal laboratory findings were observed. One patient, with only duodenum and a part the colon left, has been maintained in good health for more than 55 months using this kind of nutritional support.

Three-in-one parenteral nutrition: a safe and economical method of nutritional support for infants

The medical records of 20 infants under 1 year of age who received parenteral nutrition (PN) for a minimum of 1 week were reviewed. Patients were divided into two groups based on the method of PN administration. One group received PN by the traditional two-plus-one method where lipid emulsion is given separately from the dextrose-amino acid mixture. The second group received PN by the three-in-one method where lipid emulsion is admixed with the dextrose and amino acids. There were no statistically significant differences between the two groups in the amount of calories, lipid, dextrose, protein, calcium, and phosphorus received per kg of body weight. Average daily weight gain was not statistically different, and no obvious trends were noted in serum biochemical parameters between the two groups. Average total PN days for the groups (13 +/- 7 days for the two-plus-one group vs 39 +/- 35 days for the three-in-one group) were significantly different (p less than 0.05). A cost comparison of the two methods of PN administration indicated that the two-plus-one method costs our hospital $11.78 more per day than the three-in-one solution. We conclude that the three-in-one method of PN administration is safe, efficacious, and cost effective for infants under 1 year of age.

Effect of amino acid solutions on total nutrient admixture stability

The stability of total nutrient admixture (TNA) systems containing a soybean oil emulsion (Soyacal) has been reported with only one amino acid (AA) solution. This study was conducted to compare the physical stability of 10 TNA systems varied only by the AA solution used. All systems contained electrolytes, vitamins, trace elements, and heparin. The volume ratios of AA, dextrose 70% and lipid emulsion 20% were 1:1:1. Solutions were stored at 4 degrees C for 14 days and then held at ambient temperature for an additional 4 days. Each TNA was serially analyzed on days 0, 1, 3, 5, 7, 14, 15, 16, 17, and 18 for gross visual appearance, pH, osmolality, mean particle diameter, and particle size distribution. The AA solutions evaluated include: Travasol 8.5% and 10%; Aminosyn 8.5% and 10%, Aminosyn RF 5.2%; FreAmine III 8.5% and 10%; FreAmine HBC 6.9%; HepatAmine 8.0%; and NephrAmine 5.4%. The pH values of the Aminosyn solutions were lower than those of other products; however, the pH values of all TNA systems were greater than 5.7, which supported particle stability. The osmolality was not affected by the AA solutions. Based upon particle size distribution, 95% of all particles were less than 0.608 micron in diameter, with means ranging from 0.286 micron to 0.309 micron. The largest particle observed by light microscopy was 6.9 micron. These data indicate that TNA systems prepared with the 10 AA solutions and Soyacal 20% are physically stable.

Total nutrient admixture: a review

The TNA system of nutritional support has become very popular and offers some unique advantages over the traditional method of administering TPN to hospitalized and home patients. However, these advantages as outlined in this review, must be carefully weighed against potential disadvantages before the TNA system is employed as a nutritional support modality. It should also be noted that the stability of TNA systems is not well established since many stability studies do not provide specific information regarding formulations tested. In addition, many studies do not utilize methods to determine the entire spectrum of particle size and distribution. Droplet size in TNA systems attain a diameter several times larger than the 0.2 to 0.4 micron of manufacturer's lipid emulsions and naturally occurring chylomicrons. Although the administration of the TNA system has not been associated with any acute toxicity, the long-term consequences of infusing droplets larger than 0.4 micron is not definitely known. In addition, the biological implications of using the TNA system need to be elucidated. Subtle differences in the properties of the lipid emulsion can affect the way it is metabolized by the body. Wretlind has mentioned that two apparently similar soybean oil emulsions, Intralipid, and Lipofundin are handled differently by the body. Minor differences in the phospholipid layer of the droplets were postulated as a cause. Certainly the nature of the emulsifying layer of phospholipid on TNA system droplets is modified and therefore may be metabolized differently. The recent report of enhanced growth of microorganisms in TNA systems is also worrisome.(ABSTRACT TRUNCATED AT 250 WORDS)

Comparison of total nutrient admixture stability using two intravenous fat emulsions, Soyacal and Intralipid 20%

The safe clinical use and physical stability of a total nutrient admixture (TNA) system containing a soybean oil emulsion (Intralipid) has been reported. A study was conducted to compare the physical stability of four admixtures which were divided into two groups based upon the ratios (1:1:1 and 2:1:1) of amino acids (8.5%), dextrose (70%), and fat emulsion (20%). The fat component in each group contained either a new soy bean fat emulsion, Soyacal, or Intralipid. The quantities of all electrolytes, trace elements and vitamin additives were the same. All solutions were stored at 4 degrees C for 28 days and then held at ambient temperature for 5 days for a total 33-day study period. Each admixture was serially analyzed on days 0, 1, 2, 3, 5, 7, 14, 21, 28, 29, 30, 31, 32, and 33. Examination of gross visual appearance and determinations of pH were performed. Osmolality was measured by means of freezing point depression (Advanced Digimatic Osmometer, Advanced Instruments, Inc., Needham Heights, MA). A Brookhaven particle analyzer was used to measure lipid particle size and particle size distribution. Electron and light microscopy were used to verify maximum particle size and distribution on days 0, 7, 14, 21, 28, and 29. The type of lipid emulsion used did not affect the pH or osmolality of the admixtures. Admixtures prepared with the 2:1:1 ratio had slightly higher pH (0.07) and lower osmolality (350 mOsm/kg). The range of mean diameters for the admixtures prepared with Soyacal and Intralipid were 0.283 to 0.310 micron and 0.314 to 0.351 micron, respectively.(ABSTRACT TRUNCATED AT 250 WORDS)

Effects of pH, temperature, concentration, and time on particle counts in lipid-containing total parenteral nutrition admixtures

It has been standard practice in the United States to separate lipid emulsion from the other components of total parenteral nutrition (TPN) due to the reported instability of admixed intravenous lipid emulsions. Some clinicians, however, have combined all TPN components into one container and administered these admixtures to patients without apparent difficulties. Infusion of all nutrients from one container has many advantages. In this study standard and concentrated admixtures were aseptically prepared using generally accepted guidelines of the nutritional requirements for a 70-kg patient. Treatments of standard and concentrated admixtures consisted of: storing at 4 degrees C without adjusting the pH; increasing the pH to 6.6 and storing at 4 degrees C; increasing the pH to 6.6 and storing at room temperature. Samples were monitored for 3 weeks by means of Coulter Counter analysis, pH determinations, and visual observations. The pH of the admixtures did not change over 3 weeks. Mean counts of particles with sizes between 1.6 and 25.4 mu increased over time for each treatment group. Within treatments, concentrated admixtures had significantly greater particle counts than the corresponding standard admixtures. Within the standard and within the concentrated admixtures, the particle counts were significantly greater for group one than for group three. Particle counts in group two tended to lie between the values of group one and three. Visual signs of emulsion deterioration were greatest in those admixtures in which the pH was not adjusted and occurred earlier in concentrated admixtures.

Administration of fat emulsions with nutritional mixtures from the 3-liter delivery system in total parenteral nutrition

A series of studies was performed to test the efficacy and safety of a parenteral lipid emulsion, Lipofundin S, when given as part of a complete nutritive mixture from the three-liter bag total parenteral nutrition (TPN) delivery system. In vitro stability studies with mixtures corresponding to high and low nutritional intakes showed the fat emulsion to be stable during refrigerated storage for at least 6 days. The clinical use of Lipofundin S in 3-liter TPN bags was studied in 39 consecutive patients requiring TPN, and there were no untoward side-effects. Nitrogen balance was maintained in patients with pancreatitis, those recovering postoperatively, and those with miscellaneous conditions. However, patients with multiple trauma remained in negative balance. The ability of sera, from patients on TPN to agglutinate Lipofundin S was compared to that from healthy controls, and acutely ill patients not on TPN. Patients on TPN showed a higher degree of in vitro creaming than acutely ill controls, and this may have been related to the severity of the underlying illness. These studies suggest that this parenteral lipid emulsion can be safely administered to patients requiring TPN when given from the 3-liter bag delivery system.

Creaming and plasma clearance rate of intravenous fat emulsion in critically ill patients

Intravenous fat emulsion incubated with serum or plasma in vitro may result in the aggregation of fat (creaming). Twenty critically ill patients were tested for in vitro creaming of the fat emulsion Intralipid. An intravenous fat tolerance test was used to determine the plasma clearance rate of Intralipid in each patient. Eleven patients (55%) were found to be creamers. These patients had a higher mean plasma clearance rate of Intralipid than non-creamers (5.73 +/- 0.56 vs. 2.77 +/- 0.37% per min; p < 0.001); however, the rates of both groups were within the range reported in normal healthy subjects. Mean C-reactive protein concentration was significantly higher (p < 0.01), and albumin levels were lower (p < 0.01) in creamers compared to non-creamers. Ionized calcium levels did not differ between the two groups. The results of this study indicate that in vitro creaming is common in acutely ill patients. The clinical significance of creaming is probably minimal since creamers tolerated 50 to 100 g/day of intravenous fat emulsion while receiving total parenteral nutrition. Creaming was uncommon when the fat was mixed with blood in vitro.

Effect of electrolytes in the presence of some trace elements on the stability of all-in-one emulsion mixtures for total parenteral nutrition

In a previous study we showed that the 'all-in-one' emulsion mixtures investigated for total parenteral nutrition are stable for 1 month when kept in ethylene vinyl acetate polymer (EVA) bags and refrigerated at 4 degrees C. Analogous experiments were performed with the addition of two different solutions of trace-elements: one was prepared in the pharmacy while the other was commercially obtained (Addamel, Kabi Vitrum). A Coulter counter TA II with a 30 microns orifice was used for globule size measurements. Particles with diameters ranging from 0.6 up to 16 microns were measured. No significant changes in particle size were noticed during storage and the routine addition of trace elements to the 'all-in-one' mixtures was thus possible. These emulsions can be stored without stability problems for at least 8 days.

Study of intravenous hyperalimentation: effect of selected amino acids on the stability of intravenous fat emulsions

In an attempt to resolve the conflicts relevant to intravenous fat emulsion compatibility two amino acids, L-valine and L-histidine monohydrochloride monohydrate, were chosen as intravenous additives, and these substances were investigated as to whether they could induce significant changes in the physicochemical properties of an intravenous soybean oil emulsion. The stability of the fat emulsion was evaluated in terms of gross visual observation, particle size distribution, surface tension, pH, and the zeta potential. It was found that (a) no significant difference was produced on the particle size distribution up to 7 d after admixing, (b) the mean diameter of oil globules was approximately 0.21-0.24 micron, and (c) the fat emulsion was sufficiently stable, with a zeta potential ranging between -30 and -60 mV. The relationship of the isoelectric point (pI) of the amino acids and the pH of the buffer solutions was determined to be one of the most important factors in the stability of the fat emulsion.

Agglutination of intralipid by sera of acutely ill patients

The serum of over half of a group of acutely ill patients agglutinated "Intralipid' a fat emulsion based on soya bean oil designed for intravenous infusion. This reaction is probably precipitated by C-reactive protein in the presence of calcium ions. Post-mortem examinations of patients who had received intralipid showed evidence of microembolism which could have been caused by agglutination of intralipid. If this is the case then intralipid should perhaps not be given to these acutely ill patients.

A study of intravenous emulsion compatibility: effects of dextrose, amino acids, and selected electrolytes

Microscopic and electronic counting procedures as well as visual observations for creaming and flocculation were employed to quantitatively and qualitatively measure the effects of dextrose, amino acids and various mono- and di-valent cations on the globule size distribution of the soybean oil emulsion 10%, Intralipid. A linear regression analysis was demonstrated to successfully profile much of the stability data. Results indicated that divalent cations caused flocculation in the emulsion's internal phase immediately upon or shortly after the addition of their salts. The rate and extent of flocculation intensified with increasing ionic concentration. Amino acids, apparently acting at the oil/water interface, delayed divalent cation-induced flocculation; however, they did not prevent emulsion stability loss. The addition of dextrose 5% or 12.5% brought about a reduction of emulsion pH and significant globule coalescence 72 hours after admixture. Monovalent cations (i.e., Na+, K+) induced a progressive loss of emulsion stability over the 72-hour course of the experiments, the effect of function of ionic concentration. From the data, a model has been generated to predict significant changes (p less than 0.05) in Intralipid's globule size distribution upon addition of solute and exposure to room temperature. Further recommendations of solute admixture with the intravenous emulsion are also included.

Splenic lipidosis after administration of intravenous fat emulsions

Spleens showing fatty infiltration and necrosis of the pulp were found at necropsy on several patients who had received intravenous fat emulsions during their terminal illnesses. The postmortem findings are described and the clinicopathological correlation is discussed with special reference to the phenomenon of creaming of the emulsion.

The effect of Intralipid on the estimation of serum bilirubin in the newborn infant

After noting rises in measured serum billirubin in premature infants undergoing parenteral nutrition, a series of experiments was designed to ascertain the effect of the presence of Intralipid on the actual laboratory analysis. It has been shown that the most commonly used spectrophotometric analysis of bilirubin and the estimation of direct bilirubin by micromethods are affected by the presence of Intralipid in the specimens. Centrifugation of the blood samples at 15,000 rpm for 10 minutes and analysis of the aqueous layer overcomes, in both methods, the interference and error introduced by the presence of Intralipid chylomicrons in patient sera. Alternatively spurious bilirubin measurements due to hyperlipidemia may be avoided by using a bilirubinometer equipped with a filter which eliminates the effect of Intralipid interference.