What it is
Bacteriostatic water is clean, germ-free (sterile) water made for injections. It has a small amount of benzyl alcohol added, usually 0.9%. The benzyl alcohol is a preservative. It stops bacteria from growing in the bottle. That is what “bacteriostatic” means. Because of this, it comes in multiple-dose containers. More than one dose can be drawn from the same bottle.
It is not a medicine on its own. Its FDA-approved label gives it only one job. It is used to mix or dissolve other injectable drugs. This must follow the instructions from the maker of that other drug. The label also lists what it is not for. It must not be used in newborns. It is not for replacing body fluids. It is not for spinal or epidural procedures (injections near the spine).
It is not the same as plain sterile water for injection. Plain sterile water has no preservative. It is meant to be used only once.
Bacteriostatic Water for Injection, USP is Water for Injection containing 0.9% (9 mg/mL) or 1.1% (11 mg/mL) benzyl alcohol, pH 5.7 (4.5 to 7.0), supplied in multiple-dose containers. It contains no other solute, so it is hypotonic, and the label states that intravenous administration without a solute may cause haemolysis and that the preparation must be made approximately isotonic before use.
Benzyl alcohol (phenylmethanol, C7H8O, 108.14 g/mol) is an aromatic alcohol used as an antimicrobial preservative in parenteral products; FDA defines a multiple-dose container as one that has met antimicrobial effectiveness testing requirements. Its disposition explains the neonatal warning. Benzyl alcohol is normally oxidised rapidly to benzoic acid, conjugated with glycine in the liver and excreted in urine as hippuric acid. In premature infants that conjugation pathway may be immature, so benzoic acid can accumulate and cause metabolic acidosis. The label’s clinical pharmacology section is otherwise about water itself, noting that the small volumes involved are unlikely to affect fluid balance except possibly in very small infants.
Who made it and when
No single person invented bacteriostatic water. It is a standard product. Its recipe is set by the United States Pharmacopeia (the official US book of drug standards). Several companies sell it.
FDA records show Hospira’s Bacteriostatic Water for Injection was approved on 27 October 1982. Hospira is now part of Pfizer. Other companies repackage and sell the same product under Hospira’s approval.
The timing matters. In 1982, the same year it was approved, US doctors reported a group of deaths in premature babies. These deaths were traced to benzyl alcohol in solutions used to flush their tubes. This is why today’s labels warn against use in newborns.
Drugs@FDA lists NDA 018802 (Hospira) for Bacteriostatic Water for Injection in a plastic container, with an original approval date of 27 October 1982; the current Hospira label on DailyMed (revised August 2019) covers 0.9% benzyl alcohol in plastic vials and 1.1% in glass vials. DailyMed also carries repackager labels that cite the same NDA.
The neonatal history is documented in two 1982 reports. MMWR of 11 June 1982 described 16 deaths of pre-term infants under 2,500 g whose central catheters had been flushed daily with bacteriostatic normal saline containing 9 mg/mL benzyl alcohol, at two centres (New Orleans and Portland). Estimated benzyl alcohol intake ranged from 99 to 405 mg/kg/day; urine showed benzoate and hippurate, and serum benzoic acid was elevated. The illness featured metabolic acidosis progressing to respiratory distress and gasping respirations. Gershanik and colleagues described the “gasping syndrome” in the New England Journal of Medicine that November. Both centres reported no further similar deaths after switching to preservative-free flush solutions.
What the data say
There are no studies testing how well bacteriostatic water works. None would make sense. It is only used to mix other drugs. It does not treat anything. So the usual questions about whether a drug works do not apply.
Instead, there is safety information on the FDA label and in medical reports. The label lists problems that can come from the water, from the drug mixed into it, or from the injection itself. These include fever and soreness. They also include a pus-filled lump (abscess), and tissue damage or infection where the needle went in. Swelling or clots in the vein can also happen. The most serious known harm is to newborns. In them, benzyl alcohol has been linked to deaths. The label also says that giving about 9 mL to a 6 kg baby could change blood pressure. It also says that putting it into a vein without adding anything to it can damage red blood cells.
The preservative does not make a finished product sterile. In September 2026 a South Carolina compounding pharmacy recalled 16 injectable products. For 11 of them the FDA’s report says sterility could not be assured, partly because they were made with bacteriostatic water.
There are no efficacy trials of Bacteriostatic Water for Injection, because it is a vehicle rather than an active treatment and has no therapeutic claim to test. The safety record comes from the label and from post-marketing and case literature.
Label adverse reactions are attributed to the solution, added drugs, or the technique of reconstitution or administration: febrile response, local tenderness, abscess, tissue necrosis or infection at the injection site, venous thrombosis or phlebitis extending from the site, and extravasation. For benzyl alcohol, the label cites experimental data estimating that up to 30 mL may be given intravenously to adults, while an estimated 9 mL to a 6 kg infant or neonate is potentially capable of producing blood pressure changes. The 1982 MMWR note gives the underlying animal data: an intravenous LD50 of roughly 300 mg/kg in rats under rapid infusion and a serum half-life of about 1.5 hours in adult dogs. Haemolysis with intravenous use of a hypotonic vehicle is a physical property of the solution, shared with plain sterile water.
Benzyl alcohol inhibits microbial growth in the container; it does not sterilise a preparation made with the water. In a Class II recall initiated 2026-09-10 (FDA enforcement report event 99867, recalls D-0017-2027 to D-0032-2027), Carolina Infusion, LLC (Ridgeland, South Carolina) withdrew 16 compounded injectables nationwide, including semaglutide/cyanocobalamin, tirzepatide/cyanocobalamin and Tri-Mix preparations, and for 11 of them the stated reason was “Lack of Assurance of Sterility: Related to formulation using bacteriostatic water; Related to processing equipment used in compounded product.”
Regulatory picture
Bacteriostatic water is an FDA-approved prescription product. It is approved for one use only. That use is mixing other injectable drugs, following those drugs’ own instructions. Mixing a drug into it does not make that drug approved. Its approval says nothing about unapproved substances.
The label does not say when to throw the bottle away after it is first opened. The 28-day limit people often mention comes from FDA guidance. That guidance sets 28 days as the default limit for an opened multiple-dose bottle. This applies unless the maker’s label says something different.
Pharmacy rules for mixing bulk drug ingredients (compounding) do not apply here. This is a finished, approved supply product.
In 2026 the FDA sent warning letters to three firms that sold research peptides and also sold bacteriostatic water. The FDA’s view is that selling the water next to the peptides shows it is meant to be injected with them. In that setting the FDA treats the water itself as an unapproved new drug. One letter (March 2026) went to Gram Peptides; two more (August 2026) went to Tex Peptides and Peptide Partners.
Bacteriostatic Water for Injection is marketed under NDA 018802 and related applications as a prescription (Rx only) pharmaceutical aid. Its sole labelled indication is diluting or dissolving drugs for intravenous, intramuscular or subcutaneous injection according to the instructions of the manufacturer of the drug to be administered. The compounding field is recorded as not applicable: it is a finished approved product, not a bulk drug substance nominated for the 503A list.
FDA’s October 2018 guidance on package-type terms defines a multiple-dose container as a container of sterile medication for parenteral administration that has met antimicrobial effectiveness testing requirements, generally holding 30 mL or less. It states that the beyond-use date for an opened or entered multiple-dose container is 28 days unless otherwise specified by the manufacturer on the label, and that multiple-dose containers therefore do not normally carry a discard statement. The neonatal contraindication traces to FDA’s 1982 recommendation that benzyl-alcohol-preserved flush solutions and diluents not be used for newborns.
Intended use: FDA’s 2026 warning letters treat bacteriostatic water sold alongside research peptides as a drug in its own right. The letter to Gram Peptides (2026-03-31, ref 721806) reasons that listing Bacteriostatic Water for Injection alongside peptide products, which are drugs, shows the firm intends the two to be used together for injection, so the water itself is a drug, citing sections 201(g)(1), 201(p), 505(a) and 301(d) of the FD&C Act. The letters to TXP Innovations LLC dba Tex Peptides (2026-08-24, ref 735067; “Bacteriostatic Water 0.9% Benzyl Alcohol (10 ml)”) and Peptide Partners LLC (2026-08-24, ref 735063; “Reconstitution Solution (BAC)”) list the water as an unapproved new drug on the same reasoning, the Tex Peptides letter adding that marketing it as a solvent for reconstituting the firm’s products provides the means to prepare an injectable drug for human administration.
Supply: FDA’s drug shortage entry for Sterile Water Injection (status “Currently in Shortage”, first posted 11/23/2021, revised 10/29/2026) lists Hospira’s Bacteriostatic Water for Injection in plastic container, 1 mL (NDC 0409-3977-03), as “Available”.
