Stability of ranitidine and thiamine in parenteral nutrition solutions

BASIC NUTRITIONAL

INVESTIGATION

Nutrition Vol. 13, No. 6, 1997

Stability of Ranitidine and Thiamine in Parenteral Nutrition Solutions THOMAS

From

G. BAUMGARTNER, PHARMD, MED,” GEORGE N. HENDERSON, JANET FOX, PHARMD,“F AND UMA GONDI, PHD§

PHD,**

“Shands Hospital and Colleges of *,**Medicine and *$Pharmacy, University of Florida, Florida, and tGlaxo Research Institute, Research Triangle, North Carolina

Gainesville,

Date accepted: 21 October 1996

ABSTRACT

Our objectives were to ascertain the stability of thiamine HCl (3 mg/L) and ranitidine HCl ( 150 mg/L) at room and refrigeration temperatures in a central vein formula of parenteral nutrition (PN) solution (containing 6% amino acid, 25% carbohydrate, macro- and microminerals, and multivitamins) and to determine the effect of ranitidine on the stability of thiamine. Stability of thiamine and ranitidine in PN solutions was also compared with PN salt solutions, which contained no amino acids or carbohydrates, to indirectly ascertain the impact of these macronutrients on the stability of these moieties. High-pressure liquid chromatography (HPLC) methods were developed to measure thiamine and ranitidine in the PN mixture. Stability studies were conducted in triplicate and each sample was assayed in duplicate using newly developed HPLC methods. Refrigeration provided stability for both ranitidine and thiamine for extended periods of time. At room temperature, ranitidine was also shown to be stable for about 188 h; there was, however, significant degradation of thiamine at 24 h with, and without, addition of ranitidine. The time required for 10% of thiamine to degrade was calculated to be 12.9 h for the PN mixture containing multivitamins and ranitidine; 11.1 h for the PN mixture containing multivitamins alone: and 33.4 h for the PN mixture containing only thiamine HCl. This work suggests that the concentration of thiamine in this central vein PN formula, with or without ranitidine, falls below the 90% acceptable stability within 24 h. Nutrition 1997: 13547-553. OElsevier Science Inc. 1997 Key words: thiamine stability, ranitidine stability, parenteral nutrition solutions, kinetics

INTRODUCTION

Total parenteral nutrition (PN) has evolved into a morbidityminimizing, lifesaving modality.’ To reduce intravenous catheter manipulations, thereby minimizing the risk of (or potential for) contamination, the addition of medications and other additives to rather complex PN solutions has become common practice. Cost assessment of the potential and real-cost savings by the addition of drugs to PN solutions has also promoted the addition of medication2 It is clear that hospital, ambulatory, industrial, and community pharmacists share a common interest in learning more about the stability of these complex solutions. Unfortunately, nearly all stability studies dealing with medication( s) in nutrient solutions focus on medication and not nutrient stability. In particular, information about micronutrient stability as a consequence of adding medication to PN is virtually nonexistent in the literature. Such knowledge is vital for the

welfare of patients, especially those on chronic PN treatment in the hospital or home, since vitamin deficiencies from longterm PN have been reported’ and may be related to their physical and/or chemical instability in this complex array of nutrients and medications. As it turns out, vitamin micronutrients may be the most labile components of PN mixtures that include amino acids, carbohydrates, trace elements, electrolytes, and sometimes fat emu1sions.4 DeRitter’ has suggested that the least stable vitamins in solution are vitamin A, vitamin K, ascorbic acid, cyanocobolamin, folic acid. pantothenic acid/panthenol, and thiamine. Of the variety of medications that are now routinely admixed to PN solutions, one of the most common is the histamine-2 (H2) antagonist. This admixing has been shown to be cost effective, both from a labor and morbidity/mortality perspective, as it minimizes multiple breaks in the intravenous system.

This study was supported by a grant from Glaxo, Inc., Research Triangle Park, NC. Correspondence to: Thomas G. Baumgartner, PharmD, MEd, Shands Hospital, Box 1003 16, University of Florida, Gainesville, FL 32610-0316, USA.

Nutrition 13:547-553, 1997 OElsevier Science Inc. 1997 Printed in the USA. All rights reserved.

ELSEVIER

0899.9007/97/$17.00 PI1 SO899-9007(97)00034-8

548

STABILITY

OF RANITIDINE

AND THIAMINE

In addition, the elimination of wide swings in medication peaks and troughs associated with intermittent infusion is of benefit to the patient.627An audit of the administration of ranitidine in PN solutions indicated that the drug is generally used appropriately when administered in this manner and that most patients have positive clinical outcomes.’ There has been, to our knowledge, no investigation of the stability of vitamins in the presence of ranitidine HCl. We have chosen to investigate the stability of ranitidine, a commonly used H2 antagonist, and thiamine in a central formula PN solution, used at Shands Hospital at the University of Florida, as a first step in addressing this information gap. Thiamine was the first of the vitamins chosen because of the severity of adverse effects during deficiency states and because thiamine stores in the body are marginal and last only weeks. Permanent cerebellar damage (i.e., Wemicke-Korsakoff syndrome) and death as result of refractory lactic acidemia have been reported in the setting of severe thiamine deficiency? Deaths were associated with a recent national shortage of thiamine-containing multivitamins and their omission from PN solutions for <5 wk.” Thiamine drives pyruvate from the glycolytic lactose cycle toward acetyl coenzyme A, followed by the high-energy-producing Krebs citric acid cycle. Unlike other B vitamins, it activates the guanylate cyclase/cyclic guanosine monophosphate system but not the adenylate cyclase system. The active ester is thiamine pyrophosphate (TPP), which serves as a coenzyme in the decarboxylation of a-keto acids and keto sugars. TPP also functions in the transketolase reaction of the hexose monophosphate shunt involving two sugar phosphates. Moderate deficiency impairs carbohydrate metabolism in peripheral neurons (dry beriberi) while severe deficiency impairs heart and blood vessel carbohydrate metabolism and results in high-output congestive heart failure (wet beriberi). The objective of our study was to determine the stability of thiamine in a commonly used, centrally administered PN solution in the presence and absence of other vitamins, and with and without ranitidine at room temperature and under refrigeration. The stability of ranitidine was also determined, in the presence and absence of multivitamins including thiamine, under each of the study conditions. MATERIALS

IN PARENTAL

NUTRITIONAL

SOLUTIONS

ment. Each formula was compounded using the Clintec (Deerfield, IL) nutrition software that sends data to the AutoMix and the MicroMix mixing devices (Clintec, Deerfield, IL, USA). One liter of PN solution contained 60 g amino acids, 250 g carbohydrate, 35 mEq sodium chloride, 40 r&q potassium acetate, 5 mEq calcium gluconate, 7.5 mM potassium phosphate, 16 mEq of magnesium sulfate, combination trace elements, multivitamins, and sterile water. The PN formula simulated a commonly used, centrally administered (large vein) solution at Shands Hospital. The PN solutions were compounded on the study day, stored under refrigeration, and protected from light until used. To start the stability study, intravenous solution of ranitidine HCl (6 mL of 25 mg/mL solution, 150 mg/L final concentration) and/or thiamine HCl (30 ,QL of 100 mg/mL solution, 3 mg/L final concentration) or multivitamin (one-unit vial MVI-12 [lo mL]) was injected into respective bags (1-L Viaflex bags) containing PN using sterile hypodermic syringes. Study solutions and their medication/nutrient components are listed in Table I. Each of the study solutions were prepared in triplicate and sampled at 0, 3, 6, 12, 24, 36, 72, and 168 h and a few at 240 and 408 h. Two samples (10 mL each) were withdrawn using sterile needles and syringes at each time point and immediately frozen. For the refrigerated (4°C) stability study, the bags were placed on ice while sampling. All procedures were carried out under sterile conditions. Known standards were stored frozen for the same length of time, and fresh standards were used to verify the stability of frozen standards. Samples were assayed for thiamine and ranitidine using validated HPLC methods. pH Determinations The pH of the samples was determined at 0 and $I h and at the end of the study using a pH meter model ( Accumet 915, Fisher Scientific). HPLC A Perkin-Elmer (Norwalk, CT) Series 410 Bio Liquid Chromatograph equipped with a Perkin-Elmer model LC-95 UVVIS detector was used for the HPLC analyses. The signals were integrated using PE-Nelson (Cupertino, CA) Analytical

AND METHODS

Materials Chemicals were obtained as follows: thiamine HCl standard (USP, Washington, DC, USA), thiamine HCl 100 mg/mL intravenous solution (Elkins-Sinn, Philadelphia, PA, USA); disodium hydrogen phosphate (ACS grade, Fisher Scientific, Pittsburgh, PA, USA); acetonitrile and triethylamine (high-pressure liquid chromatography [HPLC] grade, Fisher Scientific); sodium octanesulfonate (Eastman Kodak, Rochester, NY, USA); amino acids (Travasol lo%, Travenol Laboratories, Inc., Deerlield, IL, USA) ; carbohydrate (70% dextrose injection, 2-L ViaIIexTu bag, Travenol Laboratories) ; sodium chloride, potassium acetate, and calcium gluconate (IMS, Ltd., South El Monte, CA, USA) ; potassium phosphate (American Regent, Shirley, NY, USA) ; magnesium sulfate (Abbott Laboratories, Abbott Park IL, USA) ; combination trace elements (MTE-6, Fujisawa/Lyphomed, Deerfield, IL, USA); multivitamins (MVI-12, single injection, Astra, Westborough, MA, USA) ; sterile water for injection (Travenol Laboratories); and ranitidine HCl 25 mg/mL injection (Zantac, Glaxo Pharmaceuticals, Research Triangle Park, NC, USA). Ranitidine reference standard was supplied by Glaxo Research Institute (Research Triangle Park, NC, USA). Methods Preparation and Sampling Procedure for PN Solutions PN solutions were prepared aseptically in the Shands Hospital pharmacy PN-compounding area in the laminar flow environ-

TABLE I. THIAMINE AND RANlTIDINE STABILITY

#

Temp. F-1

Solution

1

2.5

PN

2

25

PN

3 4 5 6 7 8

25 25 25 25 25 4

PN PN PN PN saltt PN salt7 PN

9

4

PN

10

4

PN

Added ingredients Multivitamins, ranitidine HCl Multivitamins, ranitidine HCl Multivitamins Thiamine HCl Ranitidine HCI Thiamine HCl Ranitidine HCl Multivitamins, ranitidine HCl Multivitamins, ranitidine HCl Multivitamins

I-L Viaflex polyvinyl chloride bags used for study. t Contains no amino acids and carbohydrates, PN, parenteral nutrition.

STUDIES Compound studied

Thiamine HCl Ranitidine

HCl

Thiamine HCl Thiamine HCI Ranitidine HCl Thiamine HCl Ranitidine HCl Thiamine HCl Ranitidine Thiamine

HCI HCl

STABILITY Integrator mL/min.

OF RANITIDINE

software

AND THIAMINE

IN PARENTAL

model 2600 or 2700. The flow rate was 1

Assay of thiamine in PN solutions. Since a literature review of methods for the analysis of thiamine in PN solutions containing multivitamins proved to be unsatisfactory, an HPLC method was developed in our laboratory. A wavelength of 280 nm and a Zorbax (DuPont; Wilmington, DE) phenyl column (4.6 mm X 25 cm) with a phenyl guard column was employed for the analysis. The mobile phase was 70% 0.1 M monobasic sodium phosphate containing 0.01% Et3N, 15% CHCN, and 15% 50 mM sodium octanesulfonate. After uniform agitation, 50 PL of PN solution was analyzed by HPLC. Assay of ranitidine in PN solutions containing multivitamins. An HPLC method was developed for the analysis of ranitidine in PN. A Zorbax C8 column (4.6 mm X 25 cm) with a C8 guard column was used for the analysis. Detection was at a wavelength of 322 nm. Elution was performed using 85% 0.1 M ammonium acetate, 13% CHCN, and 2% I-PrOH. After uniform agitation, 20 PL of PN solution was analyzed by HPLC. Kinetics Thiamine degradation data did not exhibit zero-order or simple (monoexponential decay) first-order kinetics. Hence, the data were subjected to curve stripping and fitting using the RSTRIP program (MicroMath, Salt Lake City, UT, USA), using mono-, bi-, and triexponential decay modes. From the goodness-of-fit statistics and using the Akai Information Criteria, “.” the exponential decay best fitting the data was chosen. All of the thiamine data best fitted the biexponential decay, which is expressed by the following equation (Equation 1) :

C = Ae-“’ + Bemo’ where C is the concentration at time t, A and B are concentration terms, and (Y and p are the rate constants. Stripping and fitting of the data provided values for A, B, cy, and p. These were then substituted in the above equation and, using the solver option (iteration using Newton’s method of convergence) of the Microsoft Excel program (Microsoft Inc, Redmont, WA),13 time for 10% and 50% degradations were calculated. Ranitidine degradation data exhibited apparent zero-order kinetics. Half-life and time to 10% degradation were calculated from the equation (Equation 2) : t = (C - Co)/k where Co is the initial concentration, C is the concentration at time t, and k is the rate constant. For zero-order reactions, k is the slope of the time-concentration plot. RESULTS

NUTRITIONAL

SOLUTIONS

549

solution had no peaks after 5.43 min under these conditions. Figure 1A shows the HPLC of thiamine HCl in a PN solution containing multivitamins and ranitidine. The linearity and accuracy were tested by constructing standard curves of PN solutions spiked with varying amounts of thiamine HCI (1-5 pgl mL). In the initial testing, standards were prepared fresh and refrigerated until use. Reproducibility was tested by determining the concentration of six samples. Initial testing showed that there was no difference in the standards prepared in the PN solution or phosphate buffer (0.2 M at pH 3.5) in quantifying thiamine in PN mixtures; hence, phosphate buffer standards were used. These could be stored frozen for 4 months without any degradation. The developed method was stability-indicating because none of the decomposition products of thiamine, ranitidine, or PN solution interfered with the analysis of thiamine. To check the possibility of interference from degradation of the PN solutions, the PN solution without vitamins and ranitidine and the PN solution with only ranitidine added were left at room temperature, unprotected from light, for 2 wk (maximum time stability was followed) and then analyzed using the new HPLC method. No interfering peak was found, indicating that none of the decomposition products of ranitidine or PN solution interfered with the analysis. Thiamine (0.3 mg/mL) was heated at 100°C for 3, 6, 12, and 24 h in 0.2 M NaOH, 0.2 M HCl, or pure water. The acid or base solution was neutralized. All three solutions were diluted with PN and analyzed by the new HPLC method. Peak purity was further verified using a diode array detector (Perkin-Elmer LC-235C), which measures the UV spectrum at three sections of the HPLC peak. The correlation coefficient was 0.989. The intraday and interday coefficients of variation were 1.6% and 2.4%, respectively. Ranitidine An HPLC method was developed to analyze ranitidine directly from PN mixtures without any sample preparation. A C8 column and UV detection at 322 nm were employed. Ranitidine had a retention time of 7.18 min. Figure 1B shows the analysis of ranitidine in PN containing ranitidine and multivitamins. The method developed was stability-indicating because none of the decomposition products of multivitamins, ranitidine, or the PN solution interfered with the analysis of ranitidine. The PN solution without vitamins and ranitidine and the PN solution with multivitamins were left at room temperature, unprotected from light, for 2 wk and then analyzed using the new HPLC method. No interfering peak was found. To force decomposition, ranitidine HCl was heated in various solutions as reported by Bullock et al.18 The decomposed solutions were diluted with PN solution and analyzed by the new HPLC method. Peak purity was further verified using a diode array detector (PerkinElmer LC-235C). The correlation coefficient was 0.997. The intraday and interday coefficients of variation were 1.8% and 2.3%, respectively.

HPLC Method Development Thiamine Several methods have been reported for the analysis of thiamine in multivitamin mixtures and biological tluids.14 Many of the methods require oxidation to thiochrome derivatives before analysis. Microbiologic, I5 oxidation to thiochrome, I6 and reverse phase HPLC I4 methods have been reported for the assay of thiamine in PN solutions. The HPLC method reported by van der Horst et al.” used a complex procedure involving two solvent delivery systems, purge and trap, and column switching. A simpler method for the analysis of thiamine in PN solutions was developed in our laboratory. As stated in the METHODS section, a phenyl column was employed for the HPLC analysis. Thiamine HCl had a retention time of 14.24 min and the PN

Thiamine Stability The results of the thiamine stability study at room temperature (25°C) are plotted in Figure 2, and the calculated halflives and times for 10% degradation are presented in Table II. In Figure 2, the points represent experimental values and the lines represent calculated values based on Equation 1. The stability of the vitamin up to 24 h is clinically relevant since the national standard of care is to administer a PN solution for a maximum of 24 h. The studies were continued to 168 h to determine the half-life of thiamine degradation. Thiamine degraded in the PN solution by an apparent first-order biexponential decay. Thiamine had a calculated half-life of 912 h (38 d) in the PN containing multivitamins and ranitidine, 495 h (21

550

STABILITY

OF RANITIDINE

AND THIAMINE

IN PARENTAL

NUTRITIONAL

SOLUTIONS

FIG. 1. (A) HPLC analysis of thiamine in PN solutions containing multivitamins on a phenyl column (thiamine 14.2 min). (B) HPLC analysis of ranitidine in PN solutions containing multivitamins on a C8 column. HPLC, high-pressure liquid chromatography; PN, parenteral nutrition. d) in the PN solution containing multivitamins, 1116 h (47 d) in the PN solution containing thiamine HCl, and no significant degradation of thiamine HCI in the PN salt solution (without carbohydrates and amino acids). The calculated values extend beyond the range of observed values and, hence, have to be treated as approximate. The clinically relevant 24-h data are 85.4% of thiamine remaining in the PN solution containing multivitamins and ranitidine, 82.9% in the PN solution containing multivitamins, 93.5% in the PN solution containing thiamine HCl, and no significant degradation observed for thiamine HCl in PN salt solution. From the data, it is clear that there is significant degradation of thiamine in the PN solution with multivitamins within 24 h both in the presence and absence of added ranitidine. Based on the apparent first-order biexponential degradation, the time required for 10% of thiamine to degrade in each of the solutions was calculated and they are as follows: 12.9 h in the PN solution containing multivitamins and ranitidine, Il. 1 h in the PN solution containing multivitamins, and 33.4 h in the PN solution containing thiamine HCl. No significant degradation was observed for thiamine HCl in the PN salt solution. Ranitidine had no deleterious effect on

the stability of thiamine in the PN solution and, if anything, had a protective effect in that thiamine was more stable in the PN solution containing ranitidine. The thiamine stability study at refrigeration temperature (4°C) showed negligible degradation of thiamine in PN solution up to 24 h in the presence and absence of ranitidine. After 168 h (7 d), 93.5% of thiamine remained in the presence of ranitidine and 97.4% of thiamine was left in the absence of ranitidine. Ranitidine

Stability

The results of the ranitidine stability study at room temperature (25°C) are shown in Figure 3. Although the stability of ranitidine up to 24 h is clinically relevant, the studies were continued to 408 h to determine the half-life of ranitidine degradation. Ranitidine degraded in the PN solution by a pseudo zero-order reaction. Ranitidine had a calculated half-life of 39 d in the PN solution containing multivitamins and ranitidine, and 46 d in the PN solution containing only ranitidine (no multivitamins). No significant degradation of ranitidine occurred in the PN salt solution (without carbohydrates and amino acids). At 24 h, 98.7% of ranitidine remained in the PN solution

STABILITY

OF RANITIDINE

AND

THIAMINE

IN PARENTAL

NUTRITIONAL

- I-

5.51

SOLUTIONS

PN +Multivitatnins

+ Ranitidine

- -A- - PN + Ranitidine -O-

I1

60 0

II 40

II

II 80

PN salt soln. +Ranitidine

I1

II 120

II

II 160

II 200

Time (hrs> FIG. 2. Thiamine stabilitv in PN solutions at room temnerature (25°C). Points are experimental values and the lines are calculated using biexponential decay Equation 1. PN, parentera nutrition. containing multivitamins and ranitidine, and 97.9% of ranitidine in the PN without multivitamins. No significant degradation was observed for ranitidine in the PN salt solution. The time required for 10% of ranitidine to degrade in each of the solutions was calculated and they are as follows: about 188 h in the PN containing multivitamins, 220 h in the PN without

TABLE

multivitamins, and no significant degradation was observed for ranitidine in the PN salt solution. There was essentially no degradation of ranitidine in any of the solutions during storage at 4°C. Color For solutions containing multivitamins, there was a darkening of the yellow color after 48 h at room temperature; however, little color change was observed for solutions refrigerated 7 d.

II.

TIME FOR 10% (T,,) AND 50% (T,,) DEGRADATION OF THIAMINE AND RANITIDINE IN PN SOLUTIONS AT ROOM TEMPERATURE

PH The results of pH measurements are presented in Table III. There were no significant changes in pH in any of the solutions measured.

Ranitidine

Thiamine

DISCUSSION

T,o

T 50

T,o

T50

PN + multivitamins

12.9 h

+ ranitidine PN + multivitamins

188 h (7.8 d)

941 h (39.2 d)

11.1 h

912 h (38 d) 495 h (21 d)

220 h (9.2 d)

1107 h (46.1 d)

Stable

Stable

Study’condition

PN + ranitidine PN + thiamine Ranitidine + PN salt solutiont Thiamine + PN salt solution

33.4 h

1116 h (47 d) -

Stable

Stable

PN, parenteral nutrition. t Contains no amino acids and carbohydrates.

-

Our stability studies at 25°C showed that thiamine levels drop below 90% of the initial concentration in 12.3 h in the PN solutions containing multivitamins and ranitidine, 11.1 h in the PN solutions containing multivitamins, 33.4 h in the PN solutions containing thiamine HCl alone, and no significant degradation was observed for thiamine HCl in the PN salt solutions. Thiamine degraded in PN solution by a biexponential apparent first-order reaction, as evidenced by the excellent fitting of the data to the biexponential decay curves (Fig. 3). This suggests that thiamine has two modes of degradation in PN solutions, with each of the modes exhibiting first-order kinetics. The same biexponential decay of thiamine was also observed in subsequent studies in PN solutions. Thiamine had a half-life of 38 d in PN solutions containing multivitamins and ranitidine, 21 d in PN solutions containing multivitamins, and 47 d in PN solutions containing thiamine HCI. Thiamine was remarkably stable in PN salt solutions without amino acids and carbohydrates, but containing comparable trace element and elec-

STABILITY

552

OF RANITIDINE

AND THIAMINE

-M-A-O-

100

50

0

150

IN PARENTAL

NUTRITIONAL

SOLUTIONS

PN + Multivitamins + Ranitidine PN + Ranitidine PN salt soln. + Ranitidine

200

250

300

350

400

Time (hrs) FIG. 3. Ranitidine stability in PN solutions at room temperature (25°C). PN, parenteral nutrition. trolyte concentrations. This strongly suggests that neither the trace elements nor the electrolytes are directly involved in the degradation of thiamine and that either the amino acids or carbohydrates (or both) and/or one or more of the other vitamins are responsible for the degradation of thiamine through their direct or indirect interactions. Thiamine, as a component of multivitamin mixtures, degrades faster, regardless of the presence or absence of ranitidine, than thiamine alone. There is also a possibility that one of the pathways for thiamine degradation may be accelerated by one or mom of the vitamins. Bowman and NguyenI reported that thiamine is degraded by sulfite that can be found as a preservative in a number of PN additive solutions. Using microbiologic assay, Chen et al.” found no significant degradation of thiamine in PN solutions for 8 h under artificial or indirect light, but found 26% degradation in sunlight. Our findings are that there was significant

TABLE III. MEAN PH OF PN SOLUTIONS (FROM THREE DETERMINATIONS) Study # (from Table I)

Time (h)

pH

Time (h)

pH

Time (h)

pH

1 and 2 3 4 5 6 7 8and9 10

0 0 0 0 0 0 0 0

5.65 5.60 5.60 5.68 6.20 6.30 5.70 5.65

24 24 24 24 24 24 24 24

5.60 5.65 5.62 5.62 6.24 6.31 5.65 5.66

168 168 168 168 300 300 168 168

5.62 5.64 5.62 5.65 6.24 6.29 5.67 5.66

PN, parenteral nutrition.

degradation of thiamine at 24 h under ambient temperature (25’C) and light (artificial room light). Vitamins that are sensitive to oxidative conditions and/or pH include vitamins A, K, C, B,, B2, and Bi2. It has been reported that thiamine degrades under the inlhtence of heat, higher pH, light, and oxidative conditions? It was found to be most stable from pH 2.5-5.0,19 and its stability decreases rapidly as pH values increase from 5.0. The PN solutions used in the current study have a pH of approximately 5.8, and this can be a contributing factor in the degradation of thiamine. Riboflavin can oxidize thiamine to thiochrome, and thiamine is known to degrade folic acid and cyanocobalamine. The extent of these vitamin interactions in the complex PN solutions is unknown. Ranitidine appears to have a protective effect on the stability of thiamine in PN solutions. It is possible that it competes for the same oxidizing agent ( s ) that is (are) responsible for the oxidation of thiamine, thus blocking one of the pathways of its degradation. Under refrigeration, thiamine is stable for at least 7 d under all of the conditions investigated in the present study. Ranitidine, a frequently prescribed” Hz-receptor antagonist, is well established as a potent inhibitor of gastric acid secretion. It is effective in the maintenance and treatment of gastroesophageal reflux disease (including erosive esophagitis), benign gastric and duodenal ulcer, and hypersecretory conditions such as Zollinger-Ellison syndrome.” Unapproved uses include the prevention of upper gastrointestinal bleeding, the prevention of aspiration pneumonitis, the prophylaxis of stress ulcers, and the prevention of gastric nonsteroidal-induced ulceration. The stability of ranitidine in PN solutions has been reported by several investigators.‘8.22.23Walker and Bayliff 23studied the Canadian formulation of ranitidine in PN and found it to be stable for 48 h. Bullock et al.‘* studied the stability of both ranitidine and amino acids in PN solutions; ranitidine retained at least 90% potency for 24 h and did not appreciably affect amino acid concentrations. In PN solutions containing lipids (total nutrient admixtures or 3-in- 1 solutions), ranitidine stability for 24 h or more has been shown.24,25

STABILITY

OF RANITIDINE

AND

THIAMINE

IN PARENTAL

Although sucralfate, antacids, and Hz-receptor antagonists are equivalent in reducing clinically important bleeding, sucralfate has been shown to decrease the incidence of nosocomial pneumonia and is associated with lower mortality compared with antacids and/or Hz-receptor antagonistsz6; this has been supported with meta-analytical investigation.*‘,** Ranitidine was found to be stable under the clinically relevant conditions tested in the central vein PN formula. It took approximately 7.8 d for 10% to degrade at room temperature, and negligible degradation was noted under refrigeration or simulated home-use conditions. This time is considerably longer than the 24 h*’ or 48 h Ix reported for other PN formulations but shorter than that reported by Grimble et al.,25 who found 0.3% (without lipids) and 2.2% (with lipids) ranitidine degradation in 42 h at room temperature for PN mixtures previously kept at 4°C for 14 d. The major differences between the two studies that reported shorter stability for ranitidine and the present study are the PN amino acid composition and concentrations. The present study used a PN formula containing 6% amino acids, whereas the other two studies used either 2% or 4% amino acids. All three studies used about 25% dextrose solutions. It is possible that the higher buffering capacity of the 6% amino acids enhanced the stability of ranitidine. It is evident from this study and those reported in the literature that long-term stability of ranitidine is dependent on the PN formula

NUTRITIONAL

SOLUTIONS

553

used. Although Allwood and Martin’” have suggested that ranitidine is stable for 24 h, they highlight that ranitidine stability is contingent on a variety of factors. Of most importance is the oxidative potential of the solution that is dependent on the amino acid source, the presence of micronutrients, and the type of container. Our study indicates that at least one of the micronutrients, thiamine, does not appear to impact on ranitidine stability; however, in order to more fully understand the longterm stability of ranitidine in PN mixtures. its stability will need to be further studied in a variety of PN formulations. We conclude that the addition of ranitidine to a standard central PN formulation has no adverse effect on thiamine stability at room or refrigeration temperature. Thiamine remained stable for at least 7 d under refrigeration, but it undergoes significant degradation in <24 h at room temperature. This degradation is accelerated by the presence of other vitamins. Ranitidine was also found to be stable for at least 48 h at room temperature and 7 d under refrigeration in all of the solutions studied, regardless of the presence or absence other vitamins. There were only slight changes

of thiamine

or

in pH over time

in any of the PN solutions. Finally, and most importantly, this work suggests that the concentration of thiamine decreases by > 10% in this central vein PN formula, with or without ranititline, in <24 h at room temperature.

REFERENCES TG. Total parenteral nutrition. In: Baumgartner TG, 1. Baumgartner ed. Clinical guide toparenteral micronutrition, Fujisawa, Deerfield, IL. USA. Inc., 1991:607 2. Pearson VE, King LE. An analysis of potential and real cost savings by the addition of ranitidine to total parenteral nutrition solutions. Hosp Pharm 1992;27:610 3. Howard L, Chu R, Feman S, et al. Vitamin A deficiency from long term parenteral nutrition. Ann Intern Med 1980;93:576 4. Nordfjeld K, Pederson JL, Rasmussen M, et al. Storage of mixtures for total parenteral nutrition III. Stability of vitamins in PN mixture. Clin Hosp Pharm 1984;9:293 formulations. J Pharm Sci 5. DeRitter E. Vitamins in pharmaceutical 1982;71:1073 6. Sanders SW, Fuchi KN, Moore JG, Bishop AL. Pharmacodynamics of intravenous ranitidine after bolus and continuous infusion in patients with healed duodenal ulcers. Clin Pharmacol Ther 1989;46:545 7. Ballesteros MA, Hogan DL, Koss MA, et al. Bolus or intravenous infusion of ranitidine: effects on gastric pH and acid secretion. A comparison of efficacy and cost. Ann Intern Med 1990; 112:334 8. Hershey AG, Rosen GH, Foster MD, et al. Audit of ranitidine administration in parenteral nutrient solutions. Am J Hosp Pharm 1991;48:104 enzyme 9. Butterworth RF, Kril JJ, Harper CG. Thiamine-dependent changes in the brains of alcoholics: relationship to the WemickeKorsakoff syndrome. Alcohol Clin Exp Res 1993; 17 (5 ) : 1084 total paren10. Anonymous. Deaths associated with thiamine-deficient teral nutrition [published erratum appears in MMWR 1989 Feb 10;38(5):79]. MMWR 1989 Jan 27;38(3):43 11. Akaike H. A Bayesian analysis of the minimum AIC procedure. Ann Inst Statist Math Part A 1978;30:914 F, Peris-Ribera JE, Garcia-Carbonnell MC, Aristor12. Torres-Molina ena JC, Granero L, Pla-Delfina JM. Nonlinearities in amoxycillin pharmacokinetics I. Disposition studies in the rat. Biopharm Drug Disp 1992; 13:23 Excel 5.0 User’s Guide, Redmont, WA: Microsoft Cor13. Micros+ poration. 1994 14. Kawasaki T. Vitamin B 1:thiamine. In: De Leenheer AP. Lambert

(For additional

perspective

15. 16. 17.

18.

19.

20. 21. 22.

23. 24.

25. 26.

27.

28.

29.

WE, Nelis HJ, eds. Modern chromatogruphic analysis of vitamins. New York, NY: Marcel Dekker, 1992:319 Chen MF, Boyce HW, Jr, Triplett L. Stability of the B vitamins in mixed parenteral nutrition solution. J Paren Ent Nutr 1983;7:462 Bowman BB, Nguyen P. Stability of thiamine in parenteral nutrition solution. J Paren Ent Nutr 1983;7:567 van der Horst A, Martens HJM, de Goede PNFC. Analysis of water soluble vitamins in total parenteral nutrition solution by high pressure liquid chromatography. Pharm Weekbl Sci 1989; 11: 169 Bullock L, Parks RB, Lampasona V, et al. Stability of ranitidine hydrochloride and amino acids in parenteral nutrient solutions. Am J Hosp Phann 1985;42:2683 Henderson GN, Whalen PO, Darr RA. et al. Development of an oral drug formulation for dichloroacetate and thiamine. Drug Develop In&s Pharm 1994; 20:2425 Grant SM. Langtrv HD. Broaden. Ranitidine. Drugs 1989;37:801 Package insert-on ranitidine (product informati& #404 1275). Glaxo Pharmaceuticals, Research Triangle Park, NC Williams MF, Hak LJ, Dukes G. In vitro evaluation of the stability of ranitidine hydrochloride in total parenteral nutrient mixtures (abstr P-424E). Am J Hosp Pharm 1990;47:1574 Walker SE, Bayliff CD. Stability of ranitidine hydrochloride in total parenteral nutrient solution. Am J Hosp Pha&n 1985;42:590 Sandmann B. Hancock B, Jacobi J. et al. Stability of ondansetron in total nutrient admixture (TNA) with and without ranitidine (abstract). American Society of Hospital Pharmacists Abstracts, Orlando, FL, December, 1992 Grimble GK, Hunjan MK, Paynes-James JJ, et al. Zantac and TPN. Br J Intens Care 1991;32 Cook DJ. Stress ulcer prophylaxis: gastrointestinal bleeding and nosocomial pneumonia. Best evidence synthesis. Stand J Gastroenterol. 1995;21O(Suppl):48 Cook DJ, Reeve BK, Scholes LC. Histamine-2-receptor antagonists and antacids in the critically ill population: stress ulceration versus nosoco mial pneumonia. Infect Control Hosp Epidemiol 1994; 15:437 Cook DJ, Reeve BK, Guyatt GH, et al. Stress ulcer prophylaxis in critically ill patients. Resolving discordant meta-analyses. JAMA 1996;275:308 Allwood MC, Martin H. Factors influencing the stability of ranitidine in TPN mixtures. Clin Nutr 1995; 14: 171

see Editorial

Comment

on page 571. )