Bacteriostatic Water for Injection

General Characteristics and Composition

Bacteriostatic Water for Injection (BWFI, USP) is a distinct pharmacopeial monograph, not merely "water with a preservative added": it is sterile water for injection to which benzyl alcohol has been added at a concentration of 0.9% w/v (9 mg/mL) as the sole bacteriostatic preservative. It is a clear, colorless liquid free of suspended particles; per USP its pH falls within the range 4.5–7.0, and the purity requirements for the water itself (sterility, absence of pyrogens/endotoxins, control of foreign particulates) match the same standards applied to non-preserved sterile water. Formally these are three distinct USP monographs: Water for Injection (the starting water, mainly for further manufacturing), Sterile Water for Injection (unpreserved, for single-dose withdrawal only), and Bacteriostatic Water for Injection proper. Confusion among them is common, even though in essence they differ only in the presence or absence of benzyl alcohol. The requirement to add a preservative to multidose parenteral forms reflects long-standing practice: as soon as a container is intended for repeated needle punctures, it requires protection against microorganisms that may be introduced with each withdrawal; the history and regulatory practice of preserving injectable products has been systematized in a dedicated review ^[1].

Commercially, BWFI is supplied mainly in rubber-stoppered vials of 10, 20, or 30 mL. Bacteriostatic water should be distinguished from bacteriostatic saline (Bacteriostatic 0.9% Sodium Chloride), which likewise contains 0.9% benzyl alcohol but additionally 0.9% sodium chloride for isotonicity. The choice between them for laboratory reconstitution is determined primarily by the chemistry of the specific peptide, not by the properties of the preservative; BWFI remains the standard diluent for reconstituting lyophilizates precisely because of the convenience of multiple withdrawals from a single vial.

Benzyl Alcohol as a Preservative

Benzyl alcohol (C₆H₅CH₂OH; CAS (Chemical Abstracts Service number) 100-51-6; molar mass 108.14 g/mol) is an aromatic alcohol, a colorless, oily liquid with a faint, pleasant odor. The molecule combines a hydrophilic hydroxyl group with a hydrophobic aromatic ring, giving it moderate lipophilicity (log P ≈ 1.0–1.1). By this measure it is markedly less lipophilic than other common parenteral preservatives such as phenol or m-cresol ^[3]. Its water solubility is approximately 40 mg/mL (≈4%) at room temperature, so the working concentration of 9 mg/mL lies well below the saturation limit. The substance also occurs naturally (in jasmine and ylang-ylang essential oils) and is widely used outside pharmacy, in cosmetics and the food industry, which is why its toxicological profile has historically been assessed together with its related metabolites, benzoic acid and sodium benzoate ^[2].

The antimicrobial action of benzyl alcohol is membranotropic in nature: the molecule partitions into the lipid bilayer of microorganisms, increases its fluidity, and disrupts membrane barrier integrity and permeability, thereby inhibiting bacterial cell growth and division; the same membrane effect underlies the substance's weak local-anesthetic properties (mechanistically it is related to the group of membrane-active anesthetics, though considerably weaker than specialized agents of that class). At a concentration of about 0.9% the action is predominantly bacteriostatic, that is, it suppresses proliferation without necessarily killing existing cells; higher concentrations can act bactericidally.

Spectrum and Kinetics of Antimicrobial Action

Preservative efficacy is not a binary "works / doesn't work" property but a function of time and the type of microorganism. According to efficacy-testing data summarized in review ^[3] (D-value: the time required for the microbial count to decrease tenfold), benzyl alcohol at a concentration of about 1% shows differing rates of action depending on the organism:

Against gram-negative bacteria the preservative acts quickly, whereas suppressing yeasts and molds takes orders of magnitude longer, a typical pattern for membrane-active preservatives and one reason why pharmacopeial Antimicrobial Effectiveness Testing evaluates the full spectrum of organisms separately.

Benzyl alcohol is only one of several preservatives historically used in multidose injectable formulations, including formulations of protein and peptide drugs (others include m-cresol and phenol, typical for insulins, as well as chlorobutanol, benzalkonium chloride, and parabens in certain formulations); the choice of a specific agent depends on compatibility with the active molecule, regulatory tradition for the given product class, and the required antimicrobial potency ^[3]. This distinction matters both microbiologically and from the standpoint of compatibility with protein molecules. The comparative destabilizing effects of different preservatives are discussed in more detail in the "Stability and Compatibility" section below.

Metabolically, benzyl alcohol is oxidized (sequentially via benzaldehyde) to benzoic acid, which is conjugated in the liver with glycine by the enzyme glycine N-acyltransferase to form hippuric acid, which is excreted by the kidneys. This detoxification pathway is well characterized and efficient in adults, but, as discussed below, its immaturity in neonates is a key factor in the substance's toxicological profile ^[2].

Bacteriostatic Water versus Sterile Water

The fundamental difference between the two diluents is the presence of a preservative, and it is this that determines their different use cases:

Sterile water for injection contains no antimicrobial additives: once the container is punctured it is unprotected against contamination, so it is intended for single use and any remainder is discarded. Bacteriostatic water, owing to its 0.9% benzyl alcohol, suppresses microbial growth during repeated needle punctures of the stopper, so multiple withdrawals can be made from a single vial over a limited period. In aseptic-compounding pharmaceutical practice, for multidose containers lacking dedicated stability data, a typical "default" window of about 28 days after first puncture is often used as a guideline. This is a general rule of aseptic technique, not a dosing instruction. At the same time, the preservative is not universally inert toward protein molecules (see "Stability and Compatibility"): for aggregation-sensitive research proteins and antibodies, unpreserved sterile water is often deliberately chosen instead, despite the loss of multi-withdrawal convenience.

Reconstitution Chemistry (Concentration Calculation, RUO)

A lyophilized peptide is obtained by freeze-drying the solvent out of solution (freezing → vacuum sublimation of ice → desorption of residual moisture). The result is an amorphous, porous "cake," which often contains bulking agents (e.g., mannitol as a structure former) and buffer salts ^[4]. The choice of excipients is linked to the glass transition temperature (Tg') and collapse temperature (Tc) of the concentrated solution: exceeding these thresholds during drying destroys the cake's porous structure and impairs the rate of subsequent reconstitution. Protein protection during freeze-drying is explained by two mechanisms: "water replacement" (sugars form hydrogen bonds with the protein surface in place of the lost hydration shell) and "vitrification" (an amorphous glassy matrix restricts molecular mobility and slows degradation) ^[4].

Reconstitution consists of rehydrating the cake: the diluent is introduced slowly down the vial wall, after which the contents are gently swirled, not shaken. Vigorous agitation creates foam and an air-liquid interface at which the protein can partially unfold and aggregate (discussed in more detail below). Dissolution rate is limited by diffusion of the diluent into the porous cake, so complete dissolution takes anywhere from seconds to a few minutes.

The resulting concentration is a simple dilution calculation, C = m/V:

It is the volume of diluent added, not the mass of substance, that determines the final concentration; this is purely a laboratory calculation for labeling the solution, not a dosing instruction. The contribution of benzyl alcohol itself to osmolality is small and easily estimated by direct calculation: 9 mg/mL corresponds to a molar concentration of 9 ÷ 108.14 ≈ 0.083 mol/L, so, since the substance does not dissociate, its osmotic contribution is ≈83 mOsm/L, substantially below the physiological range (≈275–295 mOsm/kg); BWFI itself is thus noticeably hypotonic and has practically no effect on the solubility of most peptides.

Stability and Compatibility

The stability of a reconstituted solution is governed by a combination of physical (unfolding, aggregation, precipitation, surface adsorption) and chemical (hydrolysis, oxidation, deamidation) degradation pathways, the minimization of which is a central task in developing formulations for protein drugs. The presence of a preservative adds yet another variable, and the conflict between the microbiological need for a preservative and the chemical sensitivity of the protein molecule is recognized as one of the classic challenges in protein drug formulation.

Case Study: rhIL-1Ra (From Observation to Mechanism)

This conflict is documented in the greatest detail in the case of recombinant human interleukin-1 receptor antagonist (rhIL-1Ra), a series of studies by a single research group that traces the problem from observation to molecular mechanism:

• Observation. In reconstituted lyophilized rhIL-1Ra formulations, the presence of 0.9% benzyl alcohol accelerated aggregate formation compared with reconstitution without the preservative; the effect intensified with storage time and temperature ^[5].
• Binding mechanism. Spectroscopic analysis showed that benzyl alcohol binds weakly to the protein through hydrophobic interactions, shifting the equilibrium toward partially unfolded, aggregation-prone states, without a sharp change in the overall free energy of unfolding. Sucrose partially counteracted this through "preferential exclusion", the preferential accumulation of sugar outside the protein's hydration shell, which stabilizes the native conformation ^[6].
• Structural localization. NMR analysis refined this, showing that 0.9% benzyl alcohol causes chemical shifts in the amide resonances of residues 90–97 of rhIL-1Ra, indicating that these surface residues are the ones involved in preservative binding; hydrogen-deuterium exchange changed minimally in the process, pointing to a localized surface interaction rather than global destabilization of the backbone ^[7].

Is Benzyl Alcohol the "Worst" Preservative for Proteins?

Comparative studies do not give a definitive answer, but they agree on one point: benzyl alcohol is usually not the most destabilizing option. In review ^[3], the degree of preservative binding to a model peptide correlated with lipophilicity (log P): m-cresol (≈1.98) > phenol (≈1.48) > benzyl alcohol (≈1.05). In a separate study of a monoclonal antibody using hydrogen-exchange mass spectrometry, destabilization of the CH2 domain ranked as phenol > m-cresol > benzyl alcohol, and subvisible particle counts increased on storage for phenol and m-cresol but not for benzyl alcohol ^[8]. The ranking between phenol and m-cresol varies depending on the protein and methodology, but in both studies benzyl alcohol ends up at the "milder" end of the spectrum. This does not negate the effect on rhIL-1Ra documented above; rather, it means that sensitivity to a preservative is a property of the specific molecule, not a universal constant of the preservative itself.

Practical takeaway: short, stable peptides generally tolerate BWFI well, whereas structurally complex or aggregation-prone proteins should be tested empirically (SEC, DLS, particle monitoring) or unpreserved water should be chosen from the outset. Reconstituted solutions are stored cold (2–8 °C), freeze-thaw cycles are avoided (ice crystallization creates a local zone of increased solute concentration, an additional stress factor for sensitive proteins), and the solution is protected from light where necessary.

Safety Data from the Literature

This section concerns the toxicological profile of benzyl alcohol itself as a chemical substance. It has no bearing on the safety or efficacy of the peptides under research and is included solely because the preservative is an inseparable component of the diluent. A joint safety assessment of benzyl alcohol and its metabolites, benzoic acid and sodium benzoate, has been systematized by the Cosmetic Ingredient Review expert panel: in adults, the substance is efficiently metabolized via the pathway described above, acute toxicity at preservative-range concentrations is low, no consistent signal of genotoxicity or carcinogenicity has been found, although isolated cases of contact skin sensitization with topical application have been reported ^[2]. These conclusions apply to the adult, metabolically mature organism.

Vulnerability in Neonates: From Clinical Observation to Outcome Statistics

A critical exception to this generally favorable picture is neonates, especially preterm infants: because the glycine-conjugating system is immature, benzoic acid accumulates instead of being rapidly excreted as hippurate, causing metabolic acidosis. The classic description is the "gasping syndrome," first systematically described in 1982 as a cluster of fatal cases in a neonatal intensive care unit: progressive metabolic acidosis, persistent gasping respiration, CNS depression, renal and hepatic failure, cardiovascular collapse, and, in blood and cerebrospinal fluid, increased concentrations of benzyl alcohol and its metabolites, which confirmed the causal link ^[7].

Subsequent epidemiological studies turned this observation into a quantitative evidence base. A retrospective cohort study of very-low-birth-weight infants compared infants who received benzyl-alcohol-preserved catheter flush solutions with infants from the period after the practice was discontinued: among infants weighing under 1000 g, mortality fell from 80.7% to 45.7%, and the incidence of severe (grade III-IV) intraventricular hemorrhage fell from 46% to 19% ^[8]. An analysis of long-term outcomes among survivors showed even more striking figures: the incidence of cerebral palsy fell from 50% to 2.4%, and the combined proportion of children with cerebral palsy and developmental delay fell from 53.9% to 11.9%.

These data provided the basis for regulatory warnings regarding benzyl-alcohol-preserved solutions in neonatology, which are still reflected in the labeling of commercial BWFI products today. For the purposes of this material, the more important conclusion is different: BWFI is a research-use-only (RUO) reagent, not intended for administration to humans or animals. The clinical data presented above describe precisely the scenario the product is meant to avoid.

Handling and Storage

Vials of bacteriostatic water are stored at controlled room temperature (approximately 20–25 °C per the USP definition) and protected from freezing: ice crystallization can mechanically damage the glass container or cause the solution to become cloudy after thawing. Before every use, not only upon first opening, clarity is checked visually: liquid that is cloudy or contains visible particles is not used.

Work is performed under aseptic conditions: the stopper is wiped with an antiseptic swab (typically 70% isopropyl alcohol) before each puncture, and the number of punctures is limited where possible: repeated punctures with a wide-bore needle progressively shear off small fragments of the rubber stopper ("coring"), which can enter the solution as foreign particulates. An opened vial is used within a limited period from the time of first puncture (see the guideline above regarding the usage window for multidose containers), and reconstituted peptide solutions are labeled with the preparation date and concentration and stored according to the requirements of the specific peptide.

The product is intended exclusively for laboratory and research use (RUO); it is not a drug and is not intended for the diagnosis, treatment, or administration to humans or animals.

References

1. Meyer BK, Ni A, Hu B, Shi L. "Antimicrobial preservative use in parenteral products: Past and present" Journal of Pharmaceutical Sciences (2007). doi:10.1002/jps.20976
2. Cosmetic Ingredient Review Expert Panel (Nair B). "Final Report on the Safety Assessment of Benzyl Alcohol, Benzoic Acid, and Sodium Benzoate" International Journal of Toxicology (2001). doi:10.1080/10915810152630729
3. Stroppel L, Schultz-Fademrecht T, Cebulla M, Blech M, Marhöfer RJ, Selzer PM, Garidel P. "Antimicrobial Preservatives for Protein and Peptide Formulations: An Overview" Pharmaceutics (2023) 15(2):563. doi:10.3390/pharmaceutics15020563
4. Wang W. "Lyophilization and development of solid protein pharmaceuticals" International Journal of Pharmaceutics (2000). doi:10.1016/S0378-5173(00)00423-3
5. Roy S, Jung R, Kerwin BA, Randolph TW, Carpenter JF. "Effects of Benzyl Alcohol on Aggregation of Recombinant Human Interleukin-1-Receptor Antagonist in Reconstituted Lyophilized Formulations" Journal of Pharmaceutical Sciences (2005) 94(2):382-396. doi:10.1002/jps.20258
6. Zhang Y, Roy S, Jones LS, Krishnan S, Kerwin BA, Chang BS, Manning MC, Randolph TW, Carpenter JF. "Mechanism for Benzyl Alcohol-Induced Aggregation of Recombinant Human Interleukin-1 Receptor Antagonist in Aqueous Solution" Journal of Pharmaceutical Sciences (2004) 93(12):3076-3089. doi:10.1002/jps.20219
7. Alford JR, Fowler AC, Wuttke DS, Kerwin BA, Latypov RF, Carpenter JF, Randolph TW. "Effect of Benzyl Alcohol on Recombinant Human Interleukin-1 Receptor Antagonist Structure and Hydrogen-Deuterium Exchange" Journal of Pharmaceutical Sciences (2011) 100(10):4215-4224. doi:10.1002/jps.22601
8. Karunaratne SP, Jolliffe MC, Trayton I, Shanmugam RK, Darton NJ, Weis DD. "Interaction between preservatives and a monoclonal antibody in support of multidose formulation development" International Journal of Pharmaceutics (2023) 648:123600. doi:10.1016/j.ijpharm.2023.123600
9. Manning MC, Chou DK, Murphy BM, Payne RW, Katayama DS. "Stability of Protein Pharmaceuticals: An Update" Pharmaceutical Research (2010). doi:10.1007/s11095-009-0045-6
10. Frokjaer S, Otzen DE. "Protein drug stability: a formulation challenge" Nature Reviews Drug Discovery (2005). doi:10.1038/nrd1695
11. Gershanik J, Boecler B, Ensley H, McCloskey S, George W. "The Gasping Syndrome and Benzyl Alcohol Poisoning" New England Journal of Medicine (1982). doi:10.1056/NEJM198211253072206
12. Hiller JL, Benda GI, Rahatzad M, Allen JR, Culver DH, Carlson CV, Reynolds JW. "Benzyl Alcohol Toxicity: Impact on Mortality and Intraventricular Hemorrhage Among Very Low Birth Weight Infants" Pediatrics (1986) 77(4):500-506. doi:10.1542/peds.77.4.500
13. Benda GI, Hiller JL, Reynolds JW. "Benzyl Alcohol Toxicity: Impact on Neurologic Handicaps Among Surviving Very Low Birth Weight Infants" Pediatrics (1986) 77(4):507-512. doi:10.1542/peds.77.4.507