- Email: [email protected]
The Determination of Phosphorus in Haemophilus influenzae Type b Conjugate Vaccines by Inductively Coupled Plasma-Atomic Emission Spectrometry
Biologicals (2000) 28, 227–231 doi:10.1006/biol.2000.0261, available online at http://www.idealibrary.com on
The Determination of Phosphorus in Haemophilus influenzae Type b Conjugate Vaccines by Inductively Coupled Plasma-Atomic Emission Spectrometry* Laura A. Swartz, Joseph J. Progar and Joan C. May Center for Biologics Evaluation and Research, United States Food and Drug Administration, Rockville, MD 20852, U.S.A.
Abstract. This study describes a method for the determination of phosphorus in lyophilized Haemophilus influenzae type b conjugate vaccines by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). The concentration of polysaccharide is directly related to the concentration of phosphorus as measured in the laboratory. Phosphorus is present in the polyribosyl-ribitol phosphate (PRP) group of the Haemophilus influenzae type b conjugate vaccine. The repeating unit of PRP is 3-B-D ribose[1-1]ribitol-5phosphate. Phosphorus in the final container is measured in µg per dose. The amount of PRP is calculated from this and reported in µg per dose. The Haemophilus influenzae type b conjugate vaccine was analyzed for phosphorus content within the range of 1·34 to 2·02 µg phosphorus per ml. The relative difference of phosphorus concentrations determined by the ICP-AES method from the phosphorus concentrations determined by the traditional colorimetric molybdate method ranged from 2·2 to 10·6%. Phosphorus spike recovery for the vaccine ranged from 93 to 99% (1·93±0·13 µg P/ml). The phosphorus determination of NIST SRM 3139 phosphorus spectrometric solution differed by 3·0% from the certified phosphorus value (10·00 mg P/ml).
Introduction Haemophilus influenzae type b conjugate vaccines are used to prevent bacterial meningitis in humans.1–6 Haemophilus influenzae type b conjugate vaccines have been approved for routine use in children 2 months of age and older for prevention of bacterial meningitis caused by Haemophilus influenzae type b (Hib). Prior to the routine administration of Haemophilus influenzae type b conjugate vaccines, Hib was the leading cause of infant meningitis in the United States.1–5 Recent studies have shown that the use of conjugate Haemophilus influenzae type b vaccines has resulted in a dramatic decline in meningitis among children.7,8 In the United States, from 1987 to 1995, there has been a *Presented in part at NRC-CNRC meeting, The Future of Spectroscopy: From Astronomy to Biology, SainteAdele, Quebec, Canada, September 1994. To whom correspondence should be addressed: Laura A. Swartz, Center for Biologics Evaluation and Research, 1401 Rockville Pike, Rockville, MD 20852-1448, U.S.A. 1045–1056/00/040227+05 $35.00/0
96% decrease in the incidence of invasive Hib disease among children <5 years old.9 The polysaccharide, polyribosyl-ribitol phosphate (PRP), which surrounds Hib is the major virulence factor.1 The concentration of polysaccharide in these vaccines can be calculated from the concentration of phosphorus measured in the laboratory. Phosphorus is present in the PRP group of Haemophilus influenzae type b conjugate vaccine. The repeating unit of PRP is 3-B-D ribose[1-1]ribitol-5-phosphate.10 The measured phosphorus (g P per 0·5 ml dose) is converted to PRP (g PRP per 0·5 ml dose) based on the stoichiometric ratios. The method currently used to determine phosphorus content in Hib conjugate vaccines is the colorimetric molybdate method.11 In this method, phosphate present in the vaccine reacts with ammonium molybdate to form a phosphomolybdate complex. As this complex is reduced, a blue colour develops and the absorbance is measured with a visible spectrometer set at 820 nm. The amount of phosphorus contained in the vaccine
228
L. A. Swartz et al.
is determined by comparison of the sample absorbance with the absorbances obtained from a series of phosphorus standard solutions.11,12 With the colorimetric molybdate method, as with any colorimetric procedure, there is the possibility that chemical components of the matrix will interfere with the assay. Also, the colorimetric molybdate method is a time-consuming method that involves the preparation of a series of reagents and several additional steps including sample digestion. In this study inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was investigated as an alternative analytical technique. The sample is aspirated into an argon plasma, which is a partially ionized gas having a temperature ranging from 6000 to 10 000K. As the sample traverses the plasma, desolvation, vaporization, atomization and ionization processes take place in turn and at di#erent locations within the plasma.13 As a result of the high temperatures achieved by the plasma, e#icient excitation of the analyte atoms occurs, as well as ionization of the atoms and subsequent excitation of the resulting ions. The wavelength of the light emitted when the atom or ion returns to a more stable configuration is specific for each element. The emission intensity of the element of interest is measured at one of its characteristic wavelengths and then this result is compared with the emission intensities obtained from a series of standard solutions.
Materials and methods Reagents
Ammonium phosphate, dibasic, crystal ((NH4)2HPO4) and ultrapure nitric acid (HNO3) were obtained from the J.T. Baker Chemical Co. (Philipsburg, NJ, U.S.A.). Reagent Grade water, obtained from a Milli-RO4/ MilliQ system (Millipore, Bedford, MA, U.S.A.) was used in all solutions. Reference materials and samples
Phosphorus Spectrometric Solution, National Institute of Standards and Technology (NIST), Standard Reference Material (SRM) 3139 (NIST, Department of Commerce, Gaithersburg, MD, U.S.A.) was used as a reference standard for phosphorus. This SRM has a certified content of 10·00 mg/ml P in 0·8% (v/v) hydrochloric acid (HCl).
Table 1. ICP-AES Instrumental Parameters Spectrometer Wavelength RF power Observation Height Aspiration Rate Plasma gas flow Auxiliary gas flow Nebulizer gas flow Nebulizer type Torch attachment
Perkin-Elmer Model 6500 213·618 nm 1·11 kW 14 mm 1 ml/min 18 l/min 0·50 l/min 1·0 l/min Cross-Flow Torch Extension
The samples used in this study were commercially prepared single dose lyophilized Haemophilus b polysaccharide conjugate vaccines. Instrumentation
Emission measurements were made with a Perkin-Elmer Inductively Coupled Plasma-Atomic Emission Spectrometer (ICP-AES), Model 6500 (Perkin-Elmer, Norwalk, CT, U.S.A.). The instrumental parameters are listed in Table 1. All measurements were made using the phosphorus line of 213·618 nm. Calibration
A stock standard solution of 5·00 mg/ml phosphorus in 5% (v/v) HNO3 was prepared by dissolving 2·132 g of ammonium phosphate in deionized water. Then 5 ml of concentrated HNO3 was added and the solution was brought to volume with deionized water in a 100 ml volumetric flask. A calibration curve was constructed by diluting the stock solution (1:100) to create a working standard solution. This solution was then used to make standards for the calibration curve by adding various volumes to separate 10 ml volumetric flasks. A five point standard curve, ranging from 0·50 to 5·00 g/ml phosphorus was used for analysis. Procedure
Four or more vials containing lyophilized Haemophilus b polysaccharide conjugate vaccine samples were reconstituted to the appropriate volume with deionized water and pooled. The samples tested were all from one manufacturer. A working phosphorus control solution containing 100 g/ml of phosphorus was prepared by diluting the NIST SRM (1:100) with deionized water. To obtain a 2·00 g/ml phosphorus control solution, the working phosphorus control solution was diluted
Determination of Phosphorus in Haemophilus influenzae
(1:50) by pipetting 0·2 ml of this solution into duplicate 10 ml volumetric flasks. To evaluate analyte recovery, duplicate samples of the phosphorus control and the vaccine were spiked with 2·00 g/ml phosphorus. The spike was added to evaluate the possibility of suppression or enhancement of the intensity signal caused by differences in sample matrices.14 The ‘‘spiked’’ sample was read by ICP-AES in the same manner as the other samples and standard solutions. The sample was corrected for enhancement or suppression in proportion to the percent recovery of the ‘‘spike’’ added to the sample. The concentration of phosphorus found in the sample was multiplied by the following correction factor:
229
Table 2. Determination of phosphorus in the NIST control* No. of tests
Av. g P/mlSD
CV (%)
% Error†
22
1·940·13
6·7
3·0
*Spiked results, average of NIST phosphorus spectrometric standard, SRM 3139, measurements taken on 11 di#erent test dates. †% relative error from the 10·00 mg P/ml certified NIST spectrometric standard (the 10·00 mg P/ml standard was diluted to 2·00 g P/ml for the analysis).
Known amount P standard added to ‘‘spiked’’ sample
Table 3. Mean spike recovery, ICP-AES Method for 2·00 g P/ml spike added to haemophilus b polysaccharide conjugate vaccine
(Total assayed P in ‘‘spiked’’ sample)(assayed P in unspiked’’ sample)
Sample
Samples, controls, standards, and blanks were aspirated directly into the ICP-AES. Intensity readings were recorded in triplicate after a 20 s read delay in order to allow for plasma stabilization. A calibration curve was constructed as a plot of intensity versus phosphorus concentration in the standards. Linear regression yielded an equation from which amounts of phosphorus in the samples were calculated.
Results The ICP-AES instrumental parameters are listed in Table 1. The limit of detection for phosphorus, using ICP-AES at wavelength 213·62 nm, was determined to be 0·08 g P/ml. This was accomplished by calculating the concentration of phosphorus equal to three times the standard deviation of the background signal of ten matrix blanks.15 The vaccine samples were found to contain phosphorus within the manufacturer’s specification of 0·67 to 1·01 g P/0·5 ml dose (1·34 to 2·02 g P/ml) which is approximately sixteen times higher than the calculated limit of detection for phosphorus. The emission intensity readings were linear over the range of the calibration standards employed (0·50 to 5·00 g/ml P). The average r2 value was calculated and found to be greater than 0·99. The accuracy of the ICP-AES method was established by the analysis of the phosphorus content of the NIST control. The results are listed in Table 2. The phosphorus concentration results for the NIST
A B C D E F G H I
No. of determinations
% recovery
4 4 8 3 4 5 2 3 4
95 99 93 94 95 97 97 95 95
control (Table 2) di#ered from the certified value of 10·00 mg/ml by 3·0%. In order to further establish accuracy, the Hib conjugate vaccine was spiked with a known amount of phosphorus. The vaccine samples that were tested consisted of di#erent lots from one vaccine manufacturer. The tests were run on many di#erent test dates. The percent recovery of the phosphorus spike ranged from 93 to 99%. This is summarized in Table 3. The concentration of phosphorus in the vaccines was measured by both the molybdate method and the ICP-AES method. The results are compared in Table 4. The vaccine samples that were tested for this study consisted of di#erent lots from one vaccine manufacturer. The tests conducted using the molybdate method were performed by the vaccine manufacturer. This table shows that comparable results were obtained by the ICP-AES method. Percent variation between the two methods, relative to the molybdate method, ranged from 2·2 to 10·6% (Table 4).
230
L. A. Swartz et al.
Table 4. Comparison of phosphorus results for haemophilus b polysaccharide conjugate vaccine (from one manufacturer) by the ICP-AES method and the molybdate method Av. g P/ml Sample A B C D E F G H I
Molybdate method* 1·82 1·70 1·80 1·64 1·50 1·56 1·78 1·70 1·84
ICP-AES method† 1·680·26 1·620·18 1·840·28 1·600·10 1·540·22 1·460·44 1·840·54 1·520·28 1·680·20
(n=5)§ (n=4) (n=8) (n=3) (n=4) (n=5) (n=5) (n=3) (n=4)
Percent‡ di#erence 7·7 4·7 2·2 2·4 2·7 6·4 3·4 10·6 8·7
*Results determined by vaccine manufacturer. †Results calculated with spike correction; testing was conducted on many di#erent test dates. ‡Percent di#erence between ICP-AES and molybdate results relative to the molybdate results. §n is the number of determinations.
To establish precision for the ICP-AES method, the phosphorus concentration in the NIST control was determined in duplicate on eleven di#erent test dates. The %CV was calculated to be 3·0. This is summarized in Table 2. Discussion By using a NIST control and also by spiking the samples, the ICP-AES method was shown to demonstrate both analytical accuracy and precision when determining phosphorous concentrations at the low levels typically found with Hib conjugate vaccine preparations (Tables 2 and 3). The percent di#erence between the molybdate and ICP-AES methods (Table 4) shows that the ICP-AES provides comparable results for phosphorus determination in the vaccine matrix. Traditional colorimetric procedures often have the potential for being complicated as well as timeconsuming. These methods generally involve the preparation of a series of reagents. Not only is it time-consuming to prepare many reagents, but a time factor is also required for complex formation and full colour development. Reagent stability is another concern. To have an accurate analysis, it is critical that reagents remain stable not only during colour complex formation but also long enough to allow for sample absorbance readings. Selection of a suitable method of analysis requires that an analyst consider not only potential timeconsuming processes, but also whether or not
sample pretreatment is necessary. In the current molybdate method, sample must be digested with perchloric acid prior to complex formation and analysis. Complete digestion of a polysaccharide sample with perchloric acid can take many hours. Also, it is during sample pretreatment stages, such as digestion, that there is an increased potential for analyte loss. Analyte loss may occur by thermal vaporization, adsorption of analyte to the digestion vessel or more often during transfer of sample from the digestion vessel to the dilution flask or sampling container. Loss of analyte inevitably translates into a loss of analytical accuracy and precision. The ICP-AES method o#ers a viable alternative to the molybdate method for phosphorous determination in Hib conjugate vaccines. This method avoids most of the pitfalls commonly encountered with colorimetric procedures. Several advantages are apparent with ICP methodology, namely, ease of analysis, greater sample throughput and the relative freedom from analytical interferences. Sample pretreatment simply involves dilution of sample with deionized water prior to actual analysis. Sample digestion is not necessary, thus avoiding the use of hazardous chemicals such as perchloric acid.16 In addition, there is a lower potential for analyte loss since there is less sample manipulation. Since the sample is aspirated directly into the argon plasma, there is no requirement to have an incubation period in order to allow for complete complex formation and colour development. Other than a short 20 s time interval, in order to allow for
Determination of Phosphorus in Haemophilus influenzae
sample stabilization within the plasma, signal intensity readings are almost instantaneous. Additionally, due to the high temperature of the plasma (6000 to 10 000K) chemical interferences are significantly reduced. Organic compounds are destroyed in the plasma. Thus, other chemical compounds normally found in vaccines, such as phenol, thimerosal, formaldehyde, and 2-phenoxyethanol, are not expected to interfere. The recovery studies performed for the development of the ICP-AES method show that the protein carrier in the conjugate portion of the vaccine does not interfere with the phosphorus determination. The recovery studies show (Table 3) a range from 93 to 99% for the phosphorus recoveries and, therefore, no significant interference from the conjugate protein. Interferences from inorganic elements are avoided by selecting a wavelength that is specific for the element of interest. The ICP-AES method, thus, is a viable alternative to the molybdate method for phosphorous determination in lyophilized Hib conjugate vaccines. Further studies will be necessary to evaluate the method for determining phosphorus in Hib conjugate vaccines adsorbed to aluminium adjuvants. 2000 United States Government References 1. Centers for Disease Control. Haemophilus b conjugate vaccines for the prevention of Haemophilus influenzae type b disease among infants and children two months of age and older: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR 1991; 40(RR-1): 1–7. 2. West DJ et al. Safety and immunogenicity of a bivalent Haemophilus influenzae type b/hepatitis B vaccine in healthy infants. Pediatr Infect Dis J 1997; 16: 593–599. 3. Plumb JE, Yost SE. Molecular size characterization of Haemophilus influenzae type b polysaccharideprotein conjugate vaccines. Vaccine 1996; 14: 399–404.
231
4. Adams WG et al. Decline of childhood Haemophilus influenzae type b (Hib) disease in the Hib vaccine era. JAMA 1993; 269: 221–226. 5. Anderson EL, Decker MD, Englund JA et al. Interchangeability of Conjugated Haemophilus influenzae Type b Vaccines in Infants. JAMA 1995; 273(11): 849–853. 6. Booy R, Hodgson S, Carpenter L et al. E#icacy of Haemophilus influenzae type b conjugate vaccine PRP-T. The Lancet 1994; 344: 362–366. 7. Rosenstein NE, Perkins BE. Update on Haemophilus influenzae serotype type b and meningococcal vaccines. Pediatr Clin N Am 2000; 47(2): 337–352. 8. Schuchat A, Robinson K et al. Bacterial meningitis in the United Sates in 1995. Active Surveillance Team. N Engl J Med 1997; 337(14): 970–976. 9. Lee AK, Crutcher JM. Oklahoma notes decline in Haemophilus influenzae: invasive Haemophilus influenzae disease among children aged <5 years— Oklahoma, 1990–1997. J Okla State Med Assoc 1999; 92(6): 276–277. 10. Anderson PW, Pichichero ME et al. Vaccines consisting of periodate-cleaved oligosaccharides from the capsule of Haemophilus influenzae type b coupled to a protein carrier: structural and temporal requirements for priming in the human infant. J Immunol 1986; 137(4): 1181–1116. 11. Chen PS, Toribara TY, Warner H. Microdetermination of Phosphorus. Anal Chem 1956; 28: 1756–1758. 12. Skoog D. Principles of Instrumental Analysis, 3rd Edn. Saunders College Publishing, 1985: p. 210. 13. Boss CB, Fredeen KJ. Concepts, Instrumentation, and Techniques in Inductively Coupled Plasma Atomic Emission Spectrometry. Norwalk, CT, U.S.A., The Perkin Elmer Corporation, 1989: pp. 2.5–2.7. 14. Rains TC. Determination of Aluminum, Barium, Calcium, Lead, Magnesium, and Silver in Ferrous Alloys by Atomic Emission and Absorption Spectrometry. In: Javier-Son A (ed.) New Analytical Techniques for Trace Constituents of Metallic and Metal-Bearing Ores, ASTM STP 747. Philadelphia, Pennsylvania, U.S.A., American Society of Testing and Materials, 1981: pp. 43–54. 15. U.S. Pharmacopeia National Formulary. The United Sates Pharmacopeial Convention, Inc. 1995: p. 1983. 16. The Merck Index, 12th Edn, 1996: pp. 7299.
Received for publication 6 December 1999; accepted 24 November 2000
The Determination of Phosphorus in Haemophilus influenzae Type b Conjugate Vaccines by Inductively Coupled Plasma-Atomic Emission Spectrometry* Laura A. Swartz, Joseph J. Progar and Joan C. May Center for Biologics Evaluation and Research, United States Food and Drug Administration, Rockville, MD 20852, U.S.A.
Abstract. This study describes a method for the determination of phosphorus in lyophilized Haemophilus influenzae type b conjugate vaccines by inductively coupled plasma-atomic emission spectroscopy (ICP-AES). The concentration of polysaccharide is directly related to the concentration of phosphorus as measured in the laboratory. Phosphorus is present in the polyribosyl-ribitol phosphate (PRP) group of the Haemophilus influenzae type b conjugate vaccine. The repeating unit of PRP is 3-B-D ribose[1-1]ribitol-5phosphate. Phosphorus in the final container is measured in µg per dose. The amount of PRP is calculated from this and reported in µg per dose. The Haemophilus influenzae type b conjugate vaccine was analyzed for phosphorus content within the range of 1·34 to 2·02 µg phosphorus per ml. The relative difference of phosphorus concentrations determined by the ICP-AES method from the phosphorus concentrations determined by the traditional colorimetric molybdate method ranged from 2·2 to 10·6%. Phosphorus spike recovery for the vaccine ranged from 93 to 99% (1·93±0·13 µg P/ml). The phosphorus determination of NIST SRM 3139 phosphorus spectrometric solution differed by 3·0% from the certified phosphorus value (10·00 mg P/ml).
Introduction Haemophilus influenzae type b conjugate vaccines are used to prevent bacterial meningitis in humans.1–6 Haemophilus influenzae type b conjugate vaccines have been approved for routine use in children 2 months of age and older for prevention of bacterial meningitis caused by Haemophilus influenzae type b (Hib). Prior to the routine administration of Haemophilus influenzae type b conjugate vaccines, Hib was the leading cause of infant meningitis in the United States.1–5 Recent studies have shown that the use of conjugate Haemophilus influenzae type b vaccines has resulted in a dramatic decline in meningitis among children.7,8 In the United States, from 1987 to 1995, there has been a *Presented in part at NRC-CNRC meeting, The Future of Spectroscopy: From Astronomy to Biology, SainteAdele, Quebec, Canada, September 1994. To whom correspondence should be addressed: Laura A. Swartz, Center for Biologics Evaluation and Research, 1401 Rockville Pike, Rockville, MD 20852-1448, U.S.A. 1045–1056/00/040227+05 $35.00/0
96% decrease in the incidence of invasive Hib disease among children <5 years old.9 The polysaccharide, polyribosyl-ribitol phosphate (PRP), which surrounds Hib is the major virulence factor.1 The concentration of polysaccharide in these vaccines can be calculated from the concentration of phosphorus measured in the laboratory. Phosphorus is present in the PRP group of Haemophilus influenzae type b conjugate vaccine. The repeating unit of PRP is 3-B-D ribose[1-1]ribitol-5-phosphate.10 The measured phosphorus (g P per 0·5 ml dose) is converted to PRP (g PRP per 0·5 ml dose) based on the stoichiometric ratios. The method currently used to determine phosphorus content in Hib conjugate vaccines is the colorimetric molybdate method.11 In this method, phosphate present in the vaccine reacts with ammonium molybdate to form a phosphomolybdate complex. As this complex is reduced, a blue colour develops and the absorbance is measured with a visible spectrometer set at 820 nm. The amount of phosphorus contained in the vaccine
228
L. A. Swartz et al.
is determined by comparison of the sample absorbance with the absorbances obtained from a series of phosphorus standard solutions.11,12 With the colorimetric molybdate method, as with any colorimetric procedure, there is the possibility that chemical components of the matrix will interfere with the assay. Also, the colorimetric molybdate method is a time-consuming method that involves the preparation of a series of reagents and several additional steps including sample digestion. In this study inductively coupled plasma-atomic emission spectroscopy (ICP-AES) was investigated as an alternative analytical technique. The sample is aspirated into an argon plasma, which is a partially ionized gas having a temperature ranging from 6000 to 10 000K. As the sample traverses the plasma, desolvation, vaporization, atomization and ionization processes take place in turn and at di#erent locations within the plasma.13 As a result of the high temperatures achieved by the plasma, e#icient excitation of the analyte atoms occurs, as well as ionization of the atoms and subsequent excitation of the resulting ions. The wavelength of the light emitted when the atom or ion returns to a more stable configuration is specific for each element. The emission intensity of the element of interest is measured at one of its characteristic wavelengths and then this result is compared with the emission intensities obtained from a series of standard solutions.
Materials and methods Reagents
Ammonium phosphate, dibasic, crystal ((NH4)2HPO4) and ultrapure nitric acid (HNO3) were obtained from the J.T. Baker Chemical Co. (Philipsburg, NJ, U.S.A.). Reagent Grade water, obtained from a Milli-RO4/ MilliQ system (Millipore, Bedford, MA, U.S.A.) was used in all solutions. Reference materials and samples
Phosphorus Spectrometric Solution, National Institute of Standards and Technology (NIST), Standard Reference Material (SRM) 3139 (NIST, Department of Commerce, Gaithersburg, MD, U.S.A.) was used as a reference standard for phosphorus. This SRM has a certified content of 10·00 mg/ml P in 0·8% (v/v) hydrochloric acid (HCl).
Table 1. ICP-AES Instrumental Parameters Spectrometer Wavelength RF power Observation Height Aspiration Rate Plasma gas flow Auxiliary gas flow Nebulizer gas flow Nebulizer type Torch attachment
Perkin-Elmer Model 6500 213·618 nm 1·11 kW 14 mm 1 ml/min 18 l/min 0·50 l/min 1·0 l/min Cross-Flow Torch Extension
The samples used in this study were commercially prepared single dose lyophilized Haemophilus b polysaccharide conjugate vaccines. Instrumentation
Emission measurements were made with a Perkin-Elmer Inductively Coupled Plasma-Atomic Emission Spectrometer (ICP-AES), Model 6500 (Perkin-Elmer, Norwalk, CT, U.S.A.). The instrumental parameters are listed in Table 1. All measurements were made using the phosphorus line of 213·618 nm. Calibration
A stock standard solution of 5·00 mg/ml phosphorus in 5% (v/v) HNO3 was prepared by dissolving 2·132 g of ammonium phosphate in deionized water. Then 5 ml of concentrated HNO3 was added and the solution was brought to volume with deionized water in a 100 ml volumetric flask. A calibration curve was constructed by diluting the stock solution (1:100) to create a working standard solution. This solution was then used to make standards for the calibration curve by adding various volumes to separate 10 ml volumetric flasks. A five point standard curve, ranging from 0·50 to 5·00 g/ml phosphorus was used for analysis. Procedure
Four or more vials containing lyophilized Haemophilus b polysaccharide conjugate vaccine samples were reconstituted to the appropriate volume with deionized water and pooled. The samples tested were all from one manufacturer. A working phosphorus control solution containing 100 g/ml of phosphorus was prepared by diluting the NIST SRM (1:100) with deionized water. To obtain a 2·00 g/ml phosphorus control solution, the working phosphorus control solution was diluted
Determination of Phosphorus in Haemophilus influenzae
(1:50) by pipetting 0·2 ml of this solution into duplicate 10 ml volumetric flasks. To evaluate analyte recovery, duplicate samples of the phosphorus control and the vaccine were spiked with 2·00 g/ml phosphorus. The spike was added to evaluate the possibility of suppression or enhancement of the intensity signal caused by differences in sample matrices.14 The ‘‘spiked’’ sample was read by ICP-AES in the same manner as the other samples and standard solutions. The sample was corrected for enhancement or suppression in proportion to the percent recovery of the ‘‘spike’’ added to the sample. The concentration of phosphorus found in the sample was multiplied by the following correction factor:
229
Table 2. Determination of phosphorus in the NIST control* No. of tests
Av. g P/mlSD
CV (%)
% Error†
22
1·940·13
6·7
3·0
*Spiked results, average of NIST phosphorus spectrometric standard, SRM 3139, measurements taken on 11 di#erent test dates. †% relative error from the 10·00 mg P/ml certified NIST spectrometric standard (the 10·00 mg P/ml standard was diluted to 2·00 g P/ml for the analysis).
Known amount P standard added to ‘‘spiked’’ sample
Table 3. Mean spike recovery, ICP-AES Method for 2·00 g P/ml spike added to haemophilus b polysaccharide conjugate vaccine
(Total assayed P in ‘‘spiked’’ sample)(assayed P in unspiked’’ sample)
Sample
Samples, controls, standards, and blanks were aspirated directly into the ICP-AES. Intensity readings were recorded in triplicate after a 20 s read delay in order to allow for plasma stabilization. A calibration curve was constructed as a plot of intensity versus phosphorus concentration in the standards. Linear regression yielded an equation from which amounts of phosphorus in the samples were calculated.
Results The ICP-AES instrumental parameters are listed in Table 1. The limit of detection for phosphorus, using ICP-AES at wavelength 213·62 nm, was determined to be 0·08 g P/ml. This was accomplished by calculating the concentration of phosphorus equal to three times the standard deviation of the background signal of ten matrix blanks.15 The vaccine samples were found to contain phosphorus within the manufacturer’s specification of 0·67 to 1·01 g P/0·5 ml dose (1·34 to 2·02 g P/ml) which is approximately sixteen times higher than the calculated limit of detection for phosphorus. The emission intensity readings were linear over the range of the calibration standards employed (0·50 to 5·00 g/ml P). The average r2 value was calculated and found to be greater than 0·99. The accuracy of the ICP-AES method was established by the analysis of the phosphorus content of the NIST control. The results are listed in Table 2. The phosphorus concentration results for the NIST
A B C D E F G H I
No. of determinations
% recovery
4 4 8 3 4 5 2 3 4
95 99 93 94 95 97 97 95 95
control (Table 2) di#ered from the certified value of 10·00 mg/ml by 3·0%. In order to further establish accuracy, the Hib conjugate vaccine was spiked with a known amount of phosphorus. The vaccine samples that were tested consisted of di#erent lots from one vaccine manufacturer. The tests were run on many di#erent test dates. The percent recovery of the phosphorus spike ranged from 93 to 99%. This is summarized in Table 3. The concentration of phosphorus in the vaccines was measured by both the molybdate method and the ICP-AES method. The results are compared in Table 4. The vaccine samples that were tested for this study consisted of di#erent lots from one vaccine manufacturer. The tests conducted using the molybdate method were performed by the vaccine manufacturer. This table shows that comparable results were obtained by the ICP-AES method. Percent variation between the two methods, relative to the molybdate method, ranged from 2·2 to 10·6% (Table 4).
230
L. A. Swartz et al.
Table 4. Comparison of phosphorus results for haemophilus b polysaccharide conjugate vaccine (from one manufacturer) by the ICP-AES method and the molybdate method Av. g P/ml Sample A B C D E F G H I
Molybdate method* 1·82 1·70 1·80 1·64 1·50 1·56 1·78 1·70 1·84
ICP-AES method† 1·680·26 1·620·18 1·840·28 1·600·10 1·540·22 1·460·44 1·840·54 1·520·28 1·680·20
(n=5)§ (n=4) (n=8) (n=3) (n=4) (n=5) (n=5) (n=3) (n=4)
Percent‡ di#erence 7·7 4·7 2·2 2·4 2·7 6·4 3·4 10·6 8·7
*Results determined by vaccine manufacturer. †Results calculated with spike correction; testing was conducted on many di#erent test dates. ‡Percent di#erence between ICP-AES and molybdate results relative to the molybdate results. §n is the number of determinations.
To establish precision for the ICP-AES method, the phosphorus concentration in the NIST control was determined in duplicate on eleven di#erent test dates. The %CV was calculated to be 3·0. This is summarized in Table 2. Discussion By using a NIST control and also by spiking the samples, the ICP-AES method was shown to demonstrate both analytical accuracy and precision when determining phosphorous concentrations at the low levels typically found with Hib conjugate vaccine preparations (Tables 2 and 3). The percent di#erence between the molybdate and ICP-AES methods (Table 4) shows that the ICP-AES provides comparable results for phosphorus determination in the vaccine matrix. Traditional colorimetric procedures often have the potential for being complicated as well as timeconsuming. These methods generally involve the preparation of a series of reagents. Not only is it time-consuming to prepare many reagents, but a time factor is also required for complex formation and full colour development. Reagent stability is another concern. To have an accurate analysis, it is critical that reagents remain stable not only during colour complex formation but also long enough to allow for sample absorbance readings. Selection of a suitable method of analysis requires that an analyst consider not only potential timeconsuming processes, but also whether or not
sample pretreatment is necessary. In the current molybdate method, sample must be digested with perchloric acid prior to complex formation and analysis. Complete digestion of a polysaccharide sample with perchloric acid can take many hours. Also, it is during sample pretreatment stages, such as digestion, that there is an increased potential for analyte loss. Analyte loss may occur by thermal vaporization, adsorption of analyte to the digestion vessel or more often during transfer of sample from the digestion vessel to the dilution flask or sampling container. Loss of analyte inevitably translates into a loss of analytical accuracy and precision. The ICP-AES method o#ers a viable alternative to the molybdate method for phosphorous determination in Hib conjugate vaccines. This method avoids most of the pitfalls commonly encountered with colorimetric procedures. Several advantages are apparent with ICP methodology, namely, ease of analysis, greater sample throughput and the relative freedom from analytical interferences. Sample pretreatment simply involves dilution of sample with deionized water prior to actual analysis. Sample digestion is not necessary, thus avoiding the use of hazardous chemicals such as perchloric acid.16 In addition, there is a lower potential for analyte loss since there is less sample manipulation. Since the sample is aspirated directly into the argon plasma, there is no requirement to have an incubation period in order to allow for complete complex formation and colour development. Other than a short 20 s time interval, in order to allow for
Determination of Phosphorus in Haemophilus influenzae
sample stabilization within the plasma, signal intensity readings are almost instantaneous. Additionally, due to the high temperature of the plasma (6000 to 10 000K) chemical interferences are significantly reduced. Organic compounds are destroyed in the plasma. Thus, other chemical compounds normally found in vaccines, such as phenol, thimerosal, formaldehyde, and 2-phenoxyethanol, are not expected to interfere. The recovery studies performed for the development of the ICP-AES method show that the protein carrier in the conjugate portion of the vaccine does not interfere with the phosphorus determination. The recovery studies show (Table 3) a range from 93 to 99% for the phosphorus recoveries and, therefore, no significant interference from the conjugate protein. Interferences from inorganic elements are avoided by selecting a wavelength that is specific for the element of interest. The ICP-AES method, thus, is a viable alternative to the molybdate method for phosphorous determination in lyophilized Hib conjugate vaccines. Further studies will be necessary to evaluate the method for determining phosphorus in Hib conjugate vaccines adsorbed to aluminium adjuvants. 2000 United States Government References 1. Centers for Disease Control. Haemophilus b conjugate vaccines for the prevention of Haemophilus influenzae type b disease among infants and children two months of age and older: recommendations of the Immunization Practices Advisory Committee (ACIP). MMWR 1991; 40(RR-1): 1–7. 2. West DJ et al. Safety and immunogenicity of a bivalent Haemophilus influenzae type b/hepatitis B vaccine in healthy infants. Pediatr Infect Dis J 1997; 16: 593–599. 3. Plumb JE, Yost SE. Molecular size characterization of Haemophilus influenzae type b polysaccharideprotein conjugate vaccines. Vaccine 1996; 14: 399–404.
231
4. Adams WG et al. Decline of childhood Haemophilus influenzae type b (Hib) disease in the Hib vaccine era. JAMA 1993; 269: 221–226. 5. Anderson EL, Decker MD, Englund JA et al. Interchangeability of Conjugated Haemophilus influenzae Type b Vaccines in Infants. JAMA 1995; 273(11): 849–853. 6. Booy R, Hodgson S, Carpenter L et al. E#icacy of Haemophilus influenzae type b conjugate vaccine PRP-T. The Lancet 1994; 344: 362–366. 7. Rosenstein NE, Perkins BE. Update on Haemophilus influenzae serotype type b and meningococcal vaccines. Pediatr Clin N Am 2000; 47(2): 337–352. 8. Schuchat A, Robinson K et al. Bacterial meningitis in the United Sates in 1995. Active Surveillance Team. N Engl J Med 1997; 337(14): 970–976. 9. Lee AK, Crutcher JM. Oklahoma notes decline in Haemophilus influenzae: invasive Haemophilus influenzae disease among children aged <5 years— Oklahoma, 1990–1997. J Okla State Med Assoc 1999; 92(6): 276–277. 10. Anderson PW, Pichichero ME et al. Vaccines consisting of periodate-cleaved oligosaccharides from the capsule of Haemophilus influenzae type b coupled to a protein carrier: structural and temporal requirements for priming in the human infant. J Immunol 1986; 137(4): 1181–1116. 11. Chen PS, Toribara TY, Warner H. Microdetermination of Phosphorus. Anal Chem 1956; 28: 1756–1758. 12. Skoog D. Principles of Instrumental Analysis, 3rd Edn. Saunders College Publishing, 1985: p. 210. 13. Boss CB, Fredeen KJ. Concepts, Instrumentation, and Techniques in Inductively Coupled Plasma Atomic Emission Spectrometry. Norwalk, CT, U.S.A., The Perkin Elmer Corporation, 1989: pp. 2.5–2.7. 14. Rains TC. Determination of Aluminum, Barium, Calcium, Lead, Magnesium, and Silver in Ferrous Alloys by Atomic Emission and Absorption Spectrometry. In: Javier-Son A (ed.) New Analytical Techniques for Trace Constituents of Metallic and Metal-Bearing Ores, ASTM STP 747. Philadelphia, Pennsylvania, U.S.A., American Society of Testing and Materials, 1981: pp. 43–54. 15. U.S. Pharmacopeia National Formulary. The United Sates Pharmacopeial Convention, Inc. 1995: p. 1983. 16. The Merck Index, 12th Edn, 1996: pp. 7299.
Received for publication 6 December 1999; accepted 24 November 2000