Bacteriostatic Water

1. Studied for

   Preservative identity and the multiple-entry container

   The ingredient that makes this water re-enterable is benzyl alcohol, listed on the product's own label at 0.9 percent (9 mg/mL) or 1.1 percent (11 mg/mL). Bacteriostatic means the preservative holds microbial growth in check rather than killing microbes outright, so the liquid is not a sterilant. The label describes the product as supplied in a container from which repeated withdrawals may be made, and gives its pH as 5.7 within a range of 4.5 to 7.0. This is a regulatory labeling fact about the container and the preservative in it. It is not a stability or potency finding about anything dissolved in the water. a regulatory product label, read verbatim; not an experiment of any kind Preclinical strong Stated verbatim in the product's own regulatory label, or documented in the clinical literature in people. On this page 'strong' attaches to labeling facts and to a documented toxicity, never to a benefit claimed for the material. Open for the full finding and the paper it came from.

   In the literature

   Preservative identity and the multiple-entry container

   The ingredient that makes this water re-enterable is benzyl alcohol, listed on the product's own label at 0.9 percent (9 mg/mL) or 1.1 percent (11 mg/mL). Bacteriostatic means the preservative holds microbial growth in check rather than killing microbes outright, so the liquid is not a sterilant. The label describes the product as supplied in a container from which repeated withdrawals may be made, and gives its pH as 5.7 within a range of 4.5 to 7.0. This is a regulatory labeling fact about the container and the preservative in it. It is not a stability or potency finding about anything dissolved in the water.^{References 1}

   Modela regulatory product label, read verbatim; not an experiment of any kind

   Confidence: strong

   Stated verbatim in the product's own regulatory label, or documented in the clinical literature in people. On this page 'strong' attaches to labeling facts and to a documented toxicity, never to a benefit claimed for the material.

   Source

   Bacteriostatic Water for Injection, USP prescribing information, Hospira, Inc.; DailyMed SPL setid 87d6e9dc-fe3b-4593-ac9a-d7493d1959c7 (label information updated May 22, 2026; label text revised 08/2019)

   Source

   How this citation was checked

   The DailyMed label page was fetched live on 2026-07-28 (HTTP 200) and the text read back verbatim from its DESCRIPTION and HOW SUPPLIED sections: 'a sterile, nonpyrogenic preparation of water for injection containing 0.9% (9 mg/mL) or 1.1% (11 mg/mL) of benzyl alcohol added as a bacteriostatic preservative', 'It is supplied in a multiple-dose container from which repeated withdrawals may be made to dilute or dissolve drugs for injection', 'The pH is 5.7 (4.5 to 7.0)', 'WARNING: NOT FOR USE IN NEONATES' and 'Store at 20 to 25 degrees C (68 to 77 degrees F). [See USP Controlled Room Temperature.]'. The packager was confirmed on the same page as Hospira, Inc. TWO DELIBERATE OMISSIONS. The label's parenteral-use sentence is a true labeling quotation and is the sentence most readable as human-use framing on a research-use page, so it is not reproduced. And the source pass deleted its companion citation to a USP general chapter, together with the in-use figure that chapter was carrying, because the chapter text could not be re-resolved from this environment (usp.org and uspnf.com both returned HTTP 403) and because usp.org's own chapter page dates the chapter official November 1, 2023, contradicting the attribution the figure had been carried under. No in-use window appears anywhere in this dossier as a result.

2. Studied for

   Documented toxicity of the preservative

   Benzyl alcohol toxicity in newborn infants is documented in the clinical literature, and the product's own label reflects it. The originating report is a 1982 paper in the New England Journal of Medicine; its full title is in the reference list. The current label states that benzyl alcohol, the preservative in this product, has been associated with toxicity in neonates, directs that only preservative-free sterile water be used where water is needed for preparing or diluting medications for newborns, and carries the heading NOT FOR USE IN NEONATES. Separately, a mouse study using intraperitoneal administration found the lethal amount at four hours was the same for adult and newborn mice, measured comparable blood levels of benzyl alcohol and its breakdown product in both age groups, and concluded the acute toxicity comes from the alcohol itself rather than from that breakdown product. Stated precisely: that animal study did not show newborn animals were more sensitive. The restriction rests on the human clinical literature and on the label, not on the mouse data. a published clinical report in newborn infants, plus a separate mouse study, plus the product's regulatory label Preclinical strong Stated verbatim in the product's own regulatory label, or documented in the clinical literature in people. On this page 'strong' attaches to labeling facts and to a documented toxicity, never to a benefit claimed for the material. Open for the full finding and the paper it came from.

   In the literature

   Documented toxicity of the preservative

   Benzyl alcohol toxicity in newborn infants is documented in the clinical literature, and the product's own label reflects it. The originating report is a 1982 paper in the New England Journal of Medicine; its full title is in the reference list. The current label states that benzyl alcohol, the preservative in this product, has been associated with toxicity in neonates, directs that only preservative-free sterile water be used where water is needed for preparing or diluting medications for newborns, and carries the heading NOT FOR USE IN NEONATES. Separately, a mouse study using intraperitoneal administration found the lethal amount at four hours was the same for adult and newborn mice, measured comparable blood levels of benzyl alcohol and its breakdown product in both age groups, and concluded the acute toxicity comes from the alcohol itself rather than from that breakdown product. Stated precisely: that animal study did not show newborn animals were more sensitive. The restriction rests on the human clinical literature and on the label, not on the mouse data.^{References 14, 15, 1}

   Modela published clinical report in newborn infants, plus a separate mouse study, plus the product's regulatory label

   Confidence: strong

   Stated verbatim in the product's own regulatory label, or documented in the clinical literature in people. On this page 'strong' attaches to labeling facts and to a documented toxicity, never to a benefit claimed for the material.

   Source

   Gershanik J, Boecler B, Ensley H, McCloskey S, George W. The gasping syndrome and benzyl alcohol poisoning. N Engl J Med. 1982;307(22):1384-1388. PMID 7133084, DOI 10.1056/NEJM198211253072206; McCloskey SE, Gershanik JJ, Lertora JJ, White L, George WJ. Toxicity of benzyl alcohol in adult and neonatal mice. J Pharm Sci. 1986;75(7):702-705. PMID 3761172, DOI 10.1002/jps.2600750718; Bacteriostatic Water for Injection, USP prescribing information, DailyMed SPL setid 87d6e9dc-fe3b-4593-ac9a-d7493d1959c7

   PubMed 7133084 PubMed 3761172 Source

   How this citation was checked

   The 1982 paper was re-resolved two ways on 2026-07-28: Crossref for 10.1056/NEJM198211253072206 returned first author Juan Gershanik, New England Journal of Medicine, 1982, volume 307, issue 22, pages 1384-1388, and PubMed esummary for PMID 7133084 matched exactly (1982 Nov 25). PubMed carries no abstract for it and the publisher page is not retrievable, so nothing beyond its confirmed title and subject is claimed from it. The mouse study was re-resolved twice: Crossref for 10.1002/jps.2600750718 returned first author Sharon E. McCloskey, Journal of Pharmaceutical Sciences, 1986, 75(7):702-705, which corrected an earlier 'Fundam Appl Toxicol' attribution; PubMed esummary for PMID 3761172 matched. Its abstract confirms intraperitoneal administration to adult and neonatal CD-1 mice, the same lethal figure at four hours 'for both age groups', plasma levels of benzyl alcohol and benzaldehyde 'comparable in both neonatal and mature animals', and acute toxicity 'due to the alcohol itself and not to its metabolite, benzaldehyde'. Every numeric value from that study is deliberately omitted. A draft sentence claiming animal support for greater newborn sensitivity was deleted at the source, because its own citation contradicts it.

3. Studied for

   Contamination found in real multiple-entry vials

   Two hospital surveys checked vials that were actually in use, and both found contamination at a low but real rate. A German study collected 227 opened multiple-entry vials on a single day and found one vial and one spike contaminated with Staphylococcus epidermidis, a rate of 0.9 percent (95 percent confidence interval 0.3 to 2.1). An Austrian study collected 96 vials over four months and found 4 of them (4.17 percent) were not sterile, and reported that three of those four were vials that contained a preservative. Read together, these real-world observations say something plain: a preservative reduces the risk that comes with entering a container repeatedly, but it does not remove it. in hospital surveys of vials in real use (no people were studied, only the vials) Preclinical moderate Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people. Open for the full finding and the paper it came from.

   In the literature

   Contamination found in real multiple-entry vials

   Two hospital surveys checked vials that were actually in use, and both found contamination at a low but real rate. A German study collected 227 opened multiple-entry vials on a single day and found one vial and one spike contaminated with Staphylococcus epidermidis, a rate of 0.9 percent (95 percent confidence interval 0.3 to 2.1). An Austrian study collected 96 vials over four months and found 4 of them (4.17 percent) were not sterile, and reported that three of those four were vials that contained a preservative. Read together, these real-world observations say something plain: a preservative reduces the risk that comes with entering a container repeatedly, but it does not remove it.^{References 4, 5}

   Modelin hospital surveys of vials in real use (no people were studied, only the vials)

   Confidence: moderate

   Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people.

   Source

   Mattner F, Gastmeier P. Bacterial contamination of multiple-dose vials: a prevalence study. Am J Infect Control. 2004;32(1):12-16. PMID 14755229, DOI 10.1016/j.ajic.2003.06.004; Nogler-Semenitz E, Lass-Florl C, Nogler M, Speer G, Dierich MP. Bacterial contamination of solutions for parenteral administration for single- and multiple-dose vials after multiple use in the hospital. Wien Med Wochenschr. 2007;157(15-16):398-401. PMID 17922089, DOI 10.1007/s10354-007-0423-9

   PubMed 14755229 PubMed 17922089

   How this citation was checked

   Both re-resolved twice on 2026-07-28. Crossref: 10.1016/j.ajic.2003.06.004 returns first author Frauke Mattner, American Journal of Infection Control, 2004, 32(1):12-16; 10.1007/s10354-007-0423-9 returns first author Elisabeth Nogler-Semenitz, Wiener Medizinische Wochenschrift, 2007, 157(15-16):398-401. PubMed esummary matched both. The Mattner abstract states verbatim 'Of the 227 vials available, 1 vial and 1 spike were contaminated with Staphylococcus epidermidis (contamination rate 0.9%; 95% CI, 0.3-2.1).' The Nogler-Semenitz abstract states verbatim 'Four of 96 (4.17%) vials were not sterile' and 'Three of the samples were MDVs containing a preservative.' A 5.6 percent contamination figure carried by an earlier draft appears in neither abstract and stays deleted, as does the reversed claim that preservative-free products were over-represented among contaminated samples, which the Austrian abstract contradicts.

4. Studied for

   The structural mechanism behind preservative-driven clumping

   Two laboratories, working on different proteins, described the same underlying mechanism. In interferon alpha-2a, more benzyl alcohol meant more clumping, and the temperature at which the protein began clumping fell in a straight line as the preservative concentration rose; nuclear magnetic resonance and hydrogen-exchange measurements showed the protein was not falling apart entirely but coming loose in one specific region, the same region a computer model had predicted would be the trouble spot. In a separate study on the model protein alpha-chymotrypsinogen A, rehydrating in buffer containing benzyl alcohol produced large amounts of insoluble clumps within 24 hours (more than 10 percent), and attaching a few molecules of a large polyethylene glycol chain to the protein prevented that completely, while a small polyethylene glycol chain did nothing. The authors concluded the large chains work by physically shielding sticky surfaces on the protein, not by making it more stable. in purified proteins in the lab, from two independent research groups (NMR, hydrogen exchange, fluorescence, clumping assays) Preclinical moderate Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people. Open for the full finding and the paper it came from.

   In the literature

   The structural mechanism behind preservative-driven clumping

   Two laboratories, working on different proteins, described the same underlying mechanism. In interferon alpha-2a, more benzyl alcohol meant more clumping, and the temperature at which the protein began clumping fell in a straight line as the preservative concentration rose; nuclear magnetic resonance and hydrogen-exchange measurements showed the protein was not falling apart entirely but coming loose in one specific region, the same region a computer model had predicted would be the trouble spot. In a separate study on the model protein alpha-chymotrypsinogen A, rehydrating in buffer containing benzyl alcohol produced large amounts of insoluble clumps within 24 hours (more than 10 percent), and attaching a few molecules of a large polyethylene glycol chain to the protein prevented that completely, while a small polyethylene glycol chain did nothing. The authors concluded the large chains work by physically shielding sticky surfaces on the protein, not by making it more stable.^{References 7, 8}

   Modelin purified proteins in the lab, from two independent research groups (NMR, hydrogen exchange, fluorescence, clumping assays)

   Confidence: moderate

   Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people.

   Source

   Bis RL, Singh SM, Cabello-Villegas J, Mallela KMG. Role of benzyl alcohol in the unfolding and aggregation of interferon alpha-2a. J Pharm Sci. 2015;104(2):407-415. PMID 25100180, PMCID PMC4312256, DOI 10.1002/jps.24105; Rodriguez-Martinez JA, Rivera-Rivera I, Griebenow K. Prevention of benzyl alcohol-induced aggregation of chymotrypsinogen by PEGylation. J Pharm Pharmacol. 2011;63(6):800-805. PMID 21585378, PMCID PMC3121166, DOI 10.1111/j.2042-7158.2011.01288.x

   PubMed 25100180 PubMed 21585378

   How this citation was checked

   Both re-resolved twice on 2026-07-28. Crossref: 10.1002/jps.24105 returns first author Regina L. Bis, Journal of Pharmaceutical Sciences, 2015, 104(2):407-415; 10.1111/j.2042-7158.2011.01288.x returns first author Jose A. Rodriguez-Martinez, Journal of Pharmacy and Pharmacology, 2011, 63(6):800-805. PubMed esummary matched both. The Bis abstract states verbatim 'With increasing concentration of BA, the apparent aggregation temperature of IFNA2 linearly decreased', that NMR chemical shifts and hydrogen exchange rates identified an intermediate 'which correlated with an aggregation hot-spot predicted by computational methods', and that benzyl alcohol acts 'by partial unfolding rather than global unfolding of the entire protein'. The Rodriguez-Martinez abstract states verbatim 'copious amounts of buffer-insoluble aggregates formed within 24 h (>10%)', that aggregation 'was completely prevented when two or five molecules of PEG with a molecular weight of 5000 Da were attached' while 'two or four molecules of bound 700 Da PEG were completely inefficient', and attributes the effect to 'shielding of exposed hydrophobic protein surface area' rather than to increased thermodynamic stability. Confidence is moderate and not strong because both are laboratory bench work on isolated proteins; strong is reserved on this page for human evidence or verbatim regulatory labeling.

5. Studied for

   Choosing a preservative when a formulation has to survive repeated entry

   Two separate laboratories screened preservatives for multi-entry antibody formulations and both landed on benzyl alcohol as a comparatively gentle option. The first compared benzyl alcohol, phenol and m-cresol against a reference IgG1 antibody and found all three loosened the antibody's structure, with phenol causing the most damage, then m-cresol, then benzyl alcohol; after four weeks of warm storage, phenol and m-cresol increased the count of tiny particles in the solution while benzyl alcohol did not, and none of the three measurably changed how much of the antibody remained in its single-molecule form. The second, an earlier study on a humanized antibody, screened six preservatives and found the antibody most stable with the two parabens, compatible with benzyl alcohol and chlorobutanol at low concentrations, and incompatible with phenol and m-cresol, concluding that as a single preservative benzyl alcohol was promising. Both results belong to the specific antibody tested and do not transfer to peptides. in purified antibodies in the lab, from two independent research groups (hydrogen exchange-mass spectrometry, calorimetry, size-exclusion chromatography, particle counting, statistical experimental design) Preclinical moderate Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people. Open for the full finding and the paper it came from.

   In the literature

   Choosing a preservative when a formulation has to survive repeated entry

   Two separate laboratories screened preservatives for multi-entry antibody formulations and both landed on benzyl alcohol as a comparatively gentle option. The first compared benzyl alcohol, phenol and m-cresol against a reference IgG1 antibody and found all three loosened the antibody's structure, with phenol causing the most damage, then m-cresol, then benzyl alcohol; after four weeks of warm storage, phenol and m-cresol increased the count of tiny particles in the solution while benzyl alcohol did not, and none of the three measurably changed how much of the antibody remained in its single-molecule form. The second, an earlier study on a humanized antibody, screened six preservatives and found the antibody most stable with the two parabens, compatible with benzyl alcohol and chlorobutanol at low concentrations, and incompatible with phenol and m-cresol, concluding that as a single preservative benzyl alcohol was promising. Both results belong to the specific antibody tested and do not transfer to peptides.^{References 12, 13}

   Modelin purified antibodies in the lab, from two independent research groups (hydrogen exchange-mass spectrometry, calorimetry, size-exclusion chromatography, particle counting, statistical experimental design)

   Confidence: moderate

   Two or more independent studies agree, or the same thing has been observed repeatedly in real use. Still laboratory bench work or survey work, never a controlled result in people.

   Source

   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. Int J Pharm. 2023;648:123600. PMID 37967687, DOI 10.1016/j.ijpharm.2023.123600; Gupta S, Kaisheva E. Development of a multidose formulation for a humanized monoclonal antibody using experimental design techniques. AAPS PharmSci. 2003;5(2):E8. PMID 12866935, PMCID PMC2751516, DOI 10.1208/ps050208

   PubMed 37967687 PubMed 12866935

   How this citation was checked

   Both re-resolved twice on 2026-07-28. Crossref: 10.1016/j.ijpharm.2023.123600 returns first author Sachini P. Karunaratne, International Journal of Pharmaceutics, 2023, volume 648, article 123600; 10.1208/ps050208 returns first author Supriya Gupta, AAPS PharmSci, 2003, volume 5, issue 2. PubMed esummary matched both (Karunaratne 2023 Dec 15; Gupta 2003, 5(2):E8). The Karunaratne abstract names exactly three preservatives and no parabens, which corrected an earlier mis-attribution, and states verbatim 'The addition of phenol caused the greatest conformational destabilization, followed by m-cresol and benzyl alcohol', the four-week 40 degrees C subvisible particle result 'in the presence of phenol and m-cresol but not in the presence of benzyl alcohol', and 'neither a significant change in the monomeric content nor an accumulation of soluble aggregate'. The Gupta abstract names all six screened preservatives, states 'The protein was most stable in the presence of methylparaben and propylparaben, and was compatible with benzyl alcohol and chlorobutanol at low concentrations. Phenol and m-cresol were not compatible with the protein', and concludes verbatim 'as an individual preservative, benzyl alcohol was promising.'

6. Studied for

   What the preservative does to a protein it dissolves

   In a laboratory study of a freeze-dried protein (a recombinant human interleukin-1 receptor antagonist), rehydrating the dried powder with benzyl alcohol caused more clumping of the protein than rehydrating it with plain water. How much extra clumping happened depended heavily on how well the protein had kept its natural shape during freeze-drying: infrared measurements showed that a high starting protein concentration and the sugar sucrose protected that shape, while table salt damaged it. Once the sample was already back in solution and sitting at room temperature, benzyl alcohol did not make the clumping any worse over time. This is the single clearest reason not to assume a preserved diluent is inert toward whatever is dissolved in it. It is evidence in one protein and should not be carried across to any specific compound. in a purified protein in the lab (freeze-dried recombinant protein, infrared spectroscopy and clumping assays) Preclinical preliminary A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference. Open for the full finding and the paper it came from.

   In the literature

   What the preservative does to a protein it dissolves

   In a laboratory study of a freeze-dried protein (a recombinant human interleukin-1 receptor antagonist), rehydrating the dried powder with benzyl alcohol caused more clumping of the protein than rehydrating it with plain water. How much extra clumping happened depended heavily on how well the protein had kept its natural shape during freeze-drying: infrared measurements showed that a high starting protein concentration and the sugar sucrose protected that shape, while table salt damaged it. Once the sample was already back in solution and sitting at room temperature, benzyl alcohol did not make the clumping any worse over time. This is the single clearest reason not to assume a preserved diluent is inert toward whatever is dissolved in it. It is evidence in one protein and should not be carried across to any specific compound.^{References 6}

   Modelin a purified protein in the lab (freeze-dried recombinant protein, infrared spectroscopy and clumping assays)

   Confidence: preliminary

   A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference.

   Source

   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. J Pharm Sci. 2005;94(2):382-396. PMID 15614819, DOI 10.1002/jps.20258

   PubMed 15614819

   How this citation was checked

   Re-resolved twice on 2026-07-28. Crossref for 10.1002/jps.20258 returns first author Shouvik Roy, Journal of Pharmaceutical Sciences, 2005, 94(2):382-396. PubMed esummary for PMID 15614819 matched, with coauthors Jung R, Kerwin BA, Randolph TW, Carpenter JF as written. The retrieved abstract states verbatim 'Reconstitution of the dried solid with 0.9% (w/v) benzyl alcohol caused a greater degree of protein aggregation than reconstitution with water', that the extent 'was strongly modulated by the degree of retention of native rhIL-1ra secondary structure during lyophilization', that 'high initial solution protein concentration and the stabilizer sucrose minimized structural perturbation' while 'NaCl was destabilizing', and that 'During storage of the reconstituted lyophilized samples at room temperature, benzyl alcohol did not accelerate aggregation of rhIL-1ra.' Preliminary rather than moderate because it is a single laboratory study.

7. Studied for

   Peptide evidence, which points the other way from the protein evidence

   Two published studies looked at benzyl alcohol against an actual peptide rather than a protein, and got the opposite result from the protein work above. Working with a fatty-acid-modified 31-residue peptide, the first study found that m-cresol and phenol latched onto the peptide and changed its shape, while benzyl alcohol did not shift the peptide's fluorescence signal, showed no measurable binding energy, and caused no shape change at all. Benzyl alcohol actually improved that peptide's physical stability and solubility when it was stressed with salt. The same paper reported that these results did not carry over to two closely related peptides, which is the point worth keeping. The second study used nuclear magnetic resonance on the same peptide and again found no clumping and no detectable interaction with benzyl alcohol, while m-cresol produced insoluble clumps containing a quarter of the peptide by weight after 24 hours at room temperature. The honest summary is that preservative behavior is specific to the individual molecule and cannot be generalized from proteins, or even from one peptide to a similar one. in purified synthetic peptides in the lab (solution biophysics, calorimetry and NMR) Preclinical preliminary A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference. Open for the full finding and the paper it came from.

   In the literature

   Peptide evidence, which points the other way from the protein evidence

   Two published studies looked at benzyl alcohol against an actual peptide rather than a protein, and got the opposite result from the protein work above. Working with a fatty-acid-modified 31-residue peptide, the first study found that m-cresol and phenol latched onto the peptide and changed its shape, while benzyl alcohol did not shift the peptide's fluorescence signal, showed no measurable binding energy, and caused no shape change at all. Benzyl alcohol actually improved that peptide's physical stability and solubility when it was stressed with salt. The same paper reported that these results did not carry over to two closely related peptides, which is the point worth keeping. The second study used nuclear magnetic resonance on the same peptide and again found no clumping and no detectable interaction with benzyl alcohol, while m-cresol produced insoluble clumps containing a quarter of the peptide by weight after 24 hours at room temperature. The honest summary is that preservative behavior is specific to the individual molecule and cannot be generalized from proteins, or even from one peptide to a similar one.^{References 10, 11}

   Modelin purified synthetic peptides in the lab (solution biophysics, calorimetry and NMR)

   Confidence: preliminary

   A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference.

   Source

   D'Addio SM, Su Y, Yin D, Zhang J, Kemp E, Gindy ME. Antimicrobial Excipient-Induced Reversible Association of Therapeutic Peptides in Parenteral Formulations. J Pharm Sci. 2021;110(2):850-859. PMID 32980392, DOI 10.1016/j.xphs.2020.09.027; Li M, Falk BT, Lu X, Schroder R, McCoy M, Xu W, Yin DH, Gindy ME, D'Addio SM, Su Y. Molecular Mechanism of Antimicrobial Excipient-Induced Aggregation in Parenteral Formulations of Peptide Therapeutics. Mol Pharm. 2022;19(9):3267-3278. PMID 35917158, DOI 10.1021/acs.molpharmaceut.2c00449

   PubMed 32980392 PubMed 35917158

   How this citation was checked

   Both re-resolved twice on 2026-07-28. Crossref: 10.1016/j.xphs.2020.09.027 returns first author Suzanne M. D'Addio, Journal of Pharmaceutical Sciences, 2021, 110(2):850-859; 10.1021/acs.molpharmaceut.2c00449 returns first author Mingyue Li, Molecular Pharmaceutics, 2022, 19(9):3267-3278. PubMed esummary matched both. The D'Addio abstract names 'a non-insulin, linear, palmitoylated 31 amino acid peptide', states verbatim that 'benzyl alcohol does not impact tryptophan fluorescence, demonstrate any interaction enthalpy, or induce conformational changes', that 'These same trends did not hold true for the other palmitoylated peptides evaluated', and that 'the presence of benzyl alcohol does increase the physical stability and solubility of the linear, 31 amino acid peptide under salt stress.' The Li abstract states verbatim that 'The addition of benzyl alcohol does not induce aggregation of acyl-peptide A and has no chemical shift perturbation in 1H-1H NOESY spectra', and that 1% (w/v) m-cresol 'results in insoluble aggregates composed of 25% (w/w) peptides after a 24-hour incubation at room temperature.' Confidence is preliminary rather than moderate because the two papers share authors (D'Addio, Su, Gindy, Yin) and examine the same peptide, so they are one line of evidence rather than two independent ones.

8. Studied for

   How benzyl alcohol compares with other preservatives

   One laboratory compared five common preservatives head to head in a single test protein (cytochrome c) and found that all five caused clumping, in this order from worst to mildest: m-cresol, then phenol, then benzyl alcohol, then phenoxyethanol, then chlorobutanol. All five worked the same way, by loosening one particular region of the protein, and the more a preservative loosened that region the more clumping it caused. Stabilizing that one region reduced the clumping caused by every one of them. The useful observation is the ranking within a single test system: benzyl alcohol sat in the middle, gentler than m-cresol and phenol, harsher than chlorobutanol. in a single purified protein in the lab (cytochrome c) Preclinical preliminary A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference. Open for the full finding and the paper it came from.

   In the literature

   How benzyl alcohol compares with other preservatives

   One laboratory compared five common preservatives head to head in a single test protein (cytochrome c) and found that all five caused clumping, in this order from worst to mildest: m-cresol, then phenol, then benzyl alcohol, then phenoxyethanol, then chlorobutanol. All five worked the same way, by loosening one particular region of the protein, and the more a preservative loosened that region the more clumping it caused. Stabilizing that one region reduced the clumping caused by every one of them. The useful observation is the ranking within a single test system: benzyl alcohol sat in the middle, gentler than m-cresol and phenol, harsher than chlorobutanol.^{References 9}

   Modelin a single purified protein in the lab (cytochrome c)

   Confidence: preliminary

   A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference.

   Source

   Hutchings RL, Singh SM, Cabello-Villegas J, Mallela KMG. Effect of antimicrobial preservatives on partial protein unfolding and aggregation. J Pharm Sci. 2013;102(2):365-376. PMID 23169345, PMCID PMC3990441, DOI 10.1002/jps.23362

   PubMed 23169345

   How this citation was checked

   Re-resolved three ways on 2026-07-28. Crossref for 10.1002/jps.23362 returns first author Regina L. Hutchings, Journal of Pharmaceutical Sciences, 2013, 102(2):365-376. PubMed esummary for PMID 23169345 matched, and the abstract states the rank order verbatim: 'All APs induced protein aggregation in the order m-cresol > phenol > benzyl alcohol > phenoxyethanol > chlorobutanol', along with the partial-unfolding hot-spot mechanism and the correlation between destabilization and aggregation. Because that abstract names cytochrome c only as the subject of the group's PREVIOUS work, the PMC full text (PMC3990441) was fetched through NCBI efetch and confirms this study's own model protein in its Methods and Discussion: 'Therefore, we used a model protein, cytochrome c (Cyt c; Fig. 1)' and 'Here, we used a model protein Cyt c to examine the effect of five APs.'

9. Studied for

   How preservative effectiveness varies with conditions

   An early laboratory study measured how quickly benzyl alcohol kills microbes and showed the answer is not a fixed number. Effectiveness tracked the concentration of the preservative, the temperature, and the pH of the solution. The work was done against five test organisms (the mold Aspergillus niger, the yeast Candida albicans, and the bacteria Escherichia coli, Pseudomonas aeruginosa and Staphylococcus aureus), and reported its results as kill-rate and temperature-sensitivity parameters rather than as a single potency figure. The takeaway is that antimicrobial performance is a property of the conditions, not of the preservative alone. in laboratory cultures of five test microbes Preclinical preliminary A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference. Open for the full finding and the paper it came from.

   In the literature

   How preservative effectiveness varies with conditions

   An early laboratory study measured how quickly benzyl alcohol kills microbes and showed the answer is not a fixed number. Effectiveness tracked the concentration of the preservative, the temperature, and the pH of the solution. The work was done against five test organisms (the mold Aspergillus niger, the yeast Candida albicans, and the bacteria Escherichia coli, Pseudomonas aeruginosa and Staphylococcus aureus), and reported its results as kill-rate and temperature-sensitivity parameters rather than as a single potency figure. The takeaway is that antimicrobial performance is a property of the conditions, not of the preservative alone.^{References 3}

   Modelin laboratory cultures of five test microbes

   Confidence: preliminary

   A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference.

   Source

   Karabit MS, Juneskans OT, Lundgren P. Studies on the evaluation of preservative efficacy--II. The determination of antimicrobial characteristics of benzylalcohol. J Clin Hosp Pharm. 1986;11(4):281-289. PMID 3531242

   PubMed 3531242

   How this citation was checked

   NCBI E-utilities esummary and efetch for PMID 3531242 on 2026-07-28 returned first author Karabit MS, Journal of Clinical and Hospital Pharmacy, 1986 Aug, volume 11, issue 4, pages 281-9, which corrected an earlier 'Int J Pharm' attribution. The retrieved abstract names the D-value, activation energy, temperature coefficient and concentration exponent as the parameters used, names pH, temperature and preservative concentration as the factors investigated, and lists exactly the five organisms stated above. No DOI is asserted for this 1986 record because none was confirmed.

10. Studied for

    How benzyl alcohol acts at the bacterial membrane

    In one laboratory study, benzyl alcohol made bacterial membranes more fluid and destabilized their structure, and it also switched off membrane proteins including a pump that bacteria use to expel disinfectant. The work was done in cultures of the bacterium Methylobacterium and in artificial membrane bubbles, using Raman and fluorescence spectroscopy. An important limit: the study was about benzyl alcohol acting as a booster for a separate disinfectant, benzalkonium chloride, at alcohol concentrations that were not lethal on their own. It was not a study of how a preserved diluent holds microbes in check, and it does not establish that this membrane effect is the reason preservation works. in bacterial cultures and artificial membrane bubbles in the lab (Methylobacterium; giant vesicles) Preclinical preliminary A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference. Open for the full finding and the paper it came from.

    In the literature

    How benzyl alcohol acts at the bacterial membrane

    In one laboratory study, benzyl alcohol made bacterial membranes more fluid and destabilized their structure, and it also switched off membrane proteins including a pump that bacteria use to expel disinfectant. The work was done in cultures of the bacterium Methylobacterium and in artificial membrane bubbles, using Raman and fluorescence spectroscopy. An important limit: the study was about benzyl alcohol acting as a booster for a separate disinfectant, benzalkonium chloride, at alcohol concentrations that were not lethal on their own. It was not a study of how a preserved diluent holds microbes in check, and it does not establish that this membrane effect is the reason preservation works.^{References 2}

    Modelin bacterial cultures and artificial membrane bubbles in the lab (Methylobacterium; giant vesicles)

    Confidence: preliminary

    A single laboratory study, an early mechanistic observation, or a result whose interpretation is the authors' own inference.

    Source

    Yano T, Miyahara Y, Morii N, Okano T, Kubota H. Pentanol and Benzyl Alcohol Attack Bacterial Surface Structures Differently. Appl Environ Microbiol. 2016;82(1):402-408. PMID 26519389, PMCID PMC4702632, DOI 10.1128/AEM.02515-15

    PubMed 26519389

    How this citation was checked

    Re-resolved twice on 2026-07-28. Crossref returned DOI 10.1128/AEM.02515-15 as first author Takehisa Yano, Applied and Environmental Microbiology, 2016, volume 82, issue 1, pages 402-408, title matching. NCBI E-utilities for PMID 26519389 returned the same journal, year, volume 82, issue 1, pages 402-8. The retrieved abstract states verbatim 'BzA increased the membrane fluidity and destabilized the structures', that pentanol and benzyl alcohol 'inactivate bacterial membrane proteins, including an efflux pump for BAC transportation', and that all of this was measured for 'the combination of BAC with particular alcohols at nonlethal concentrations', which is the potentiation context named in the finding. An earlier draft's claim that membrane partitioning is the basis of this material's bacteriostatic behavior was deleted at the source, because the study makes no such claim.

2 strong / 3 moderate / 5 preliminaryRated by how many independent labs and how many kinds of experiment found the same thing.