Note · 10 November 2026

How Long Bacteriostatic Water Lasts: 28 Days From First Puncture

Opened bacteriostatic water carries a 28-day beyond-use date, counted from first puncture: the default that FDA guidance sets for any opened multiple-dose container and that CDC attributes to USP General Chapter 797. Unopened, it runs to the printed expiry. The figure rests on preservative challenge testing. It is not a measured limit of the water and does not cover anything dissolved in it.

The answer, and what it applies to

The answer is 28 days, and it belongs to the container rather than to the water. FDA guidance on package type terms from October 2018, which covers diluents as well as drugs, defines a multiple-dose container as one that has met, or is excluded by regulation from, antimicrobial effectiveness testing requirements, and sets its beyond-use date, once opened or needle-punctured, at 28 days unless the manufacturer's label specifies otherwise. CDC injection-safety guidance, last updated in January 2024, attributes the same rule to USP General Chapter 797, adds that an unopened vial is discarded at the manufacturer's expiration date, and holds that the beyond-use date never exceeds the original expiration date.

The bacteriostatic water label prints no 28. The Hospira package insert, NDA 018802, contains no in-use period at all, and the FDA guidance explains why: such containers do not normally carry a discard statement because 28 days is assumed, and any other period is to be printed, with supporting data. The figure quoted for bacteriostatic water is a default for a class of container, not a measurement on this water, and neither document calls it a stability limit.

Unopened, the label gives a storage condition of 20 to 25 degrees C, a lot number and expiration date on each vial, and no general shelf life. Pages that put a number on the unopened period give one to three years; the one examined in full cited nothing, and the figure is logged as untraced.

What the label records

Bacteriostatic Water for Injection, USP is the official monograph title; this note says bacteriostatic water. The Hospira insert describes sterile, nonpyrogenic water for injection with 0.9 per cent (9 mg/mL) or 1.1 per cent (11 mg/mL) benzyl alcohol added as a bacteriostatic preservative, in a container from which repeated withdrawals may be made. The pH is 5.7, within 4.5 to 7.0. The indication is limited to diluting or dissolving drugs as their own manufacturers instruct, and the insert opens with a warning that it is not for use in neonates.

Its precautions direct that reconstituted solutions not be stored unless the manufacturer of the dissolved drug directs otherwise, so the water's label hands the storage period of a solution to that manufacturer and claims none for itself.

Sterile water for injection is the contrast. Hospira's insert under NDA 018801 describes water with no bacteriostat, antimicrobial agent or added buffer, supplied only in single-dose containers whose unused portion is discarded. Under the 2018 guidance such a container is not required to meet antimicrobial effectiveness testing. The two waters differ by a preservative, and with it by package type and the testing that type requires.

What the 28 days was tested against

FDA documents state what the 28 days rests on, though neither derives the number. A 2020 FDA veterinary guidance on in-use stability describes the rule for human products: multiple-dose injectables for humans carry a 28-day in-use period unless otherwise labelled, when supported by successful antimicrobial effectiveness testing under USP General Chapter 51. The 2018 guidance adds that the test determines whether a product prevents microbial growth if contaminated during use, and that its results support the labelled beyond-use date.

Moser and Meyer described the test in a 2011 review. Five organisms, among them a yeast and a mould, are added separately at 100,000 to 1,000,000 colony-forming units per mL and held at 20 to 25 degrees C whatever the product's own storage condition. For injections the USP requires bacteria to fall by at least 1 log by day 7 and 3 log by day 14, with no increase at day 28; yeasts and moulds need only show no increase from the starting count at days 7, 14 and 28. The last reading falls on day 28.

The pharmacopoeias disagree on the bar. The same review records European criteria that test bacterial reduction at 6 and 24 hours and Japanese criteria that begin at day 14. On the USP criteria a product passing at day 14 may still hold up to a thousandth of its challenge, 100 to 1,000 organisms per mL on the stated inoculum, and fungi need not be reduced at all. Bacteriostatic is the accurate word, and nothing in the test measures chemistry.

The 0.9 per cent is a nominal content. Moser and Meyer note that preservative levels are assayed across shelf life, the lower limit set by challenging formulations at 100, 75 and 50 per cent of the labelled amount. Lot certificates can report the preservative as measured: a published bacteriostatic water lot certificate lists benzyl alcohol at 0.91 per cent against a specification of 0.8 to 1.0 per cent. A certificate describes a lot when tested, not a vial after its first puncture.

Bacteriostatic is not sterile

CDC's guidance states that the preservative has no effect on viruses and does not give complete protection against bacterial contamination. Marumo and colleagues reported in 2004 that hospital isolates of Serratia marcescens in heparin-saline preserved with 1 per cent benzyl alcohol were still viable after 24 and 72 hours at 4 degrees C, from inocula as small as 1 and 10 colony-forming units per mL. That was in vitro, in preserved saline rather than bacteriostatic water.

Surveys of vials in use measured contamination in practice. Longfield and colleagues reported in 1984 on 1,223 weekly cultures from 863 multidose vials at one hospital over three months: none was culture-positive, median use was 9.5 days, 13 per cent of vials stayed in use beyond 30 days, and the authors' existing guideline was to discard vials when empty or at expiry, with no fixed in-use period, a policy that differs from the 28-day default. Mattner and Gastmeier surveyed 227 vials in a 1,300-bed tertiary hospital on one day in November 2001 and reported in 2004 a contamination rate of 0.9 per cent (95 per cent confidence interval 0.3 to 2.1), one vial and one spike growing Staphylococcus epidermidis, and found an opening date on only half the vials. Neither survey measured benzyl alcohol or the chemistry of the contents.

Why the label excludes newborns

The neonatal warning traces to 1982. The Morbidity and Mortality Weekly Report of 11 June that year recorded sixteen deaths of pre-term neonates, reported to the FDA by two centres and thought to be caused by benzyl alcohol; the infants' central catheters had been flushed periodically each day with bacteriostatic normal saline containing 9 mg/mL benzyl alcohol. Ten occurred at one institution over six months and six at the other over sixteen months. The illness included metabolic acidosis progressing to respiratory distress and gasping respirations, both hospitals reported no similar deaths after preservative-free flushes were substituted, and the FDA advised against benzyl alcohol flushes and preserved diluents for newborns.

The 1982 report's editorial note proposed that in premature infants an underdeveloped glycine conjugation pathway let benzoic acid, the oxidation product of benzyl alcohol, accumulate and cause the acidosis, and cautioned that many of the same clinical features occur in neonates seriously ill from other causes. The reporting groups published separately: Brown and colleagues in a Lancet letter of 29 May 1982, and Gershanik and colleagues in the New England Journal of Medicine of 25 November 1982. Neither is indexed with an abstract, and the counts here are the MMWR's. The Hospira insert now contraindicates benzyl alcohol solutions in neonates and restricts water for their medicines to preservative-free sterile water.

The water's date is not the solution's date

Some peptide guide pages extend the 28 days to the solution, giving a reconstituted peptide up to 28 days in bacteriostatic water and 48 to 72 hours in sterile water. Of three pages examined for this note, the one giving those figures cited nothing and the other two disagreed with it. PubMed searches run on 10 November 2026 for bacteriostatic water with peptide in titles and abstracts, and for benzyl alcohol with reconstitution, peptide and stability, each returned zero records. No primary source was found for a 28-day period applying to reconstituted peptides, and the claim is logged as untraced.

Where a period exists for a dissolved product, it was set for that product. The EGRIFTA WR kit packs a 30 mL bottle of bacteriostatic water, approved under the same application as the Hospira insert, NDA 018802, beside lyophilised tesamorelin. Its patient information gives the water 28 days after first use; its prescribing information gives the mixed tesamorelin vial 7 days at 20 to 25 degrees C, not refrigerated or frozen. The label does not say what limits the 7 days, but it sets the two periods separately, in one document, for one diluent.

Weck and colleagues measured the two questions separately in 2005. In vitro, alteplase at 1 mg/mL in sterile water or in bacteriostatic water with 0.9 per cent benzyl alcohol, in duplicate glass vials at 37 degrees C, kept 91 and 90 per cent of its day-0 clot-lysis activity respectively at day 7, the last protein measurement; antimicrobial effectiveness testing from day 7 to day 28 failed the sterile-water solution and passed the bacteriostatic one.

Benzyl alcohol and the dissolved protein

The preservative also becomes an ingredient of whatever the water dissolves. Meyer and colleagues' 2007 review of licensed parenteral products found phenol and benzyl alcohol the two most frequent preservatives in peptide and protein products, and listed effect on the active ingredient among the key criteria for choosing one.

Zhang and colleagues reported in 2004 that benzyl alcohol accelerated aggregation of recombinant human interleukin-1 receptor antagonist in solution at 37 degrees C, measured as loss of native monomer, through weak hydrophobic binding that shifted the protein toward partially unfolded, aggregation-prone species; sucrose partly inhibited it. Roy and colleagues reported in 2005 that lyophilised formulations of the same protein aggregated more when reconstituted with 0.9 per cent benzyl alcohol than with water, to a degree set by how much native secondary structure freeze-drying had preserved, but that benzyl alcohol did not accelerate aggregation during room-temperature storage after reconstitution.

The two papers place the effect at different points, and both measured one protein in vitro. Neither establishes whether a given research peptide is affected, in either direction, or bears on a 28-day period. The 28 days records how long a preserved container was challenged. How long a solution stays intact is a separate measurement, made for each product or not made at all.

Disclosure. Vialog is published by the team behind Disguised Alpha, which sells the product linked in this note. The note follows the same editorial policy as every other page on this site.

References

  1. US Food and Drug Administration (CDER, CBER, CDRH and Office of Combination Products). Selection of the Appropriate Package Type Terms and Recommendations for Labeling Injectable Medical Products Packaged in Multiple-Dose, Single-Dose, and Single-Patient-Use Containers for Human Use. Guidance for Industry, October 2018. Section III, definitions (scope includes diluents; multiple-dose container defined by antimicrobial effectiveness testing; beyond-use date of 28 days for an opened or entered multiple-dose container unless otherwise specified by the manufacturer on the label) and Section IV, discard statements. Retrieved 10 November 2026. View on fda.gov
  2. Centers for Disease Control and Prevention. Preventing Unsafe Injection Practices. Injection Safety, clinical safety page, key points for multi-dose vials, attributing the 28-day discard rule to United States Pharmacopeia General Chapter 797 (page reference 14: USP 797, 1 November 2023). Page last updated 9 January 2024. Retrieved 10 November 2026. View on cdc.gov
  3. BACTERIOSTATIC WATER for Injection, USP, multiple-dose vials (benzyl alcohol 9 mg/mL in 30 mL plastic vials, 11 mg/mL in 20 mL glass vials): package insert LAB-0977-2.0, revised 08/2019. Hospira, Inc. NDA 018802. DailyMed SPL setid 87d6e9dc-fe3b-4593-ac9a-d7493d1959c7, version 20, effective 22 May 2026. Companion label: STERILE WATER for Injection, USP, single-dose vials, package insert LAB-1292-4.0, revised 03/2026, Hospira, Inc., NDA 018801, DailyMed SPL setid 2b9ea23a-f9e9-48a8-25b7-519e3bbd6828, version 10, effective 18 March 2026. Retrieved 10 November 2026. View on dailymed.nlm.nih.gov
  4. US Food and Drug Administration, Center for Veterinary Medicine. Guidance for Industry #242: In-Use Stability Studies and Associated Labeling Statements for Multiple-Dose Injectable Animal Drug Products. November 2020. Section I (28-day in-use period for human multiple-dose injectables when supported by antimicrobial effectiveness testing under USP chapter 51). Retrieved 10 November 2026. View on fda.gov
  5. Moser CL, Meyer BK. Comparison of compendial antimicrobial effectiveness tests: a review. AAPS PharmSciTech. 2011;12(1):222-6. doi:10.1208/s12249-010-9575-9. PMID 21221871 View on pubmed.ncbi.nlm.nih.gov
  6. Marumo K, Taguchi K, Oniki H, Endoh M, Sekine H. Experimental confirmation of Serratia marcescens contamination in multiple-dose vials of heparin-saline solution. J Infect Chemother. 2004;10(5):288-92. doi:10.1007/s10156-004-0342-2. PMID 16163464 View on pubmed.ncbi.nlm.nih.gov
  7. Longfield R, Longfield J, Smith LP, Hyams KC, Strohmer ME. Multidose medication vial sterility: an in-use study and a review of the literature. Infect Control. 1984;5(4):165-9. doi:10.1017/s0195941700059154. PMID 6562087 View on pubmed.ncbi.nlm.nih.gov
  8. Mattner F, Gastmeier P. Bacterial contamination of multiple-dose vials: a prevalence study. Am J Infect Control. 2004;32(1):12-6. doi:10.1016/j.ajic.2003.06.004. PMID 14755229 View on pubmed.ncbi.nlm.nih.gov
  9. Centers for Disease Control (CDC). Neonatal deaths associated with use of benzyl alcohol: United States. MMWR Morb Mortal Wkly Rep. 1982;31(22):290-1. PMID 6810084. Full text read at cdc.gov/mmwr/preview/mmwrhtml/00001109.htm, retrieved 10 November 2026. View on pubmed.ncbi.nlm.nih.gov
  10. Brown WJ, Buist NR, Gipson HT, Huston RK, Kennaway NG. Fatal benzyl alcohol poisoning in a neonatal intensive care unit. Lancet. 1982;1(8283):1250. doi:10.1016/s0140-6736(82)92377-7. PMID 6123006 View on pubmed.ncbi.nlm.nih.gov
  11. Gershanik J, Boecler B, Ensley H, McCloskey S, George W. The gasping syndrome and benzyl alcohol poisoning. N Engl J Med. 1982;307(22):1384-8. doi:10.1056/NEJM198211253072206. PMID 7133084 View on pubmed.ncbi.nlm.nih.gov
  12. EGRIFTA WR (tesamorelin) for injection, co-packaged with Bacteriostatic Water for Injection, USP (30 mL multiple-dose bottle, NDA 018802): full prescribing information and patient information. Theratechnologies Inc. BLA 022505. DailyMed SPL setid 839334d3-8c1d-4c26-9036-2ab524a6ea75, version 2, effective 29 July 2026. Sections 2.2 and 16 and the patient information consulted. Retrieved 10 November 2026. View on dailymed.nlm.nih.gov
  13. Weck S, Cheung S, Hiraoka-Sutow M, Patapoff T, Semba CP. Alteplase as a catheter locking solution: in vitro evaluation of biochemical stability and antimicrobial properties. J Vasc Interv Radiol. 2005;16(3):379-83. doi:10.1097/01.RVI.0000148154.30967.27. PMID 15758134 View on pubmed.ncbi.nlm.nih.gov
  14. Meyer BK, Ni A, Hu B, Shi L. Antimicrobial preservative use in parenteral products: past and present. J Pharm Sci. 2007;96(12):3155-67. doi:10.1002/jps.20976. PMID 17722087 View on pubmed.ncbi.nlm.nih.gov
  15. 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. J Pharm Sci. 2004;93(12):3076-89. doi:10.1002/jps.20219. PMID 15514986 View on pubmed.ncbi.nlm.nih.gov
  16. 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):382-96. doi:10.1002/jps.20258. PMID 15614819 View on pubmed.ncbi.nlm.nih.gov