Water, Water, Everywhere, But Please Don't Give IV
Problem: The treatment of severe hypernatremia can be challenging, especially in patients with preexisting conditions that may seem to limit therapeutic options. Such a situation recently resulted in an ill-conceived decision to give sterile water for injection IV to an elderly patient who had been admitted to an ICU with pneumonia, CHF, respiratory failure, and severe hypernatremia. The physician did not want the patient to receive any further infusions containing sodium. But the patient also was severely hyperglycemic. The physician's concern with giving sodium or dextrose to a patient with CHF and a high blood sugar led to an order to change the patient's peripheral IV to "free water" at 100 mL/hr.
"Free water" refers to water not associated with organic or inorganic ions. Because hypernatremia usually results from a deficit of "free water," it's likely that the physician intended to replace this loss when he wrote the order. Water can be replaced orally, however, it should never be given IV without additives to normalize tonicity, or hemolysis may occur.
Just before writing the order, the physician had contacted a pharmacist to ask if "large bags of sterile water for injection" were available. The pharmacist checked the computer and told the physician that sterile water for injection (used for compounding parenteral nutrition solutions) was available. When he received the order, he entered it into the computer and a label printed for a 2,000 mL bag of the solution. A pharmacy intern retrieved a bag from the sterile compounding area, placed the label on the back of the bag, and dispensed it to ICU.
The nurse began the infusion without question because she was aware of the patient's hypernatremia and overheard the physician ask the pharmacist if bags of sterile water were available. She failed to see a red statement on the bag stating "Pharmacy Bulk Package, Not For Direct Infusion" because the pharmacy label was on the opposite side of the bag. Another nurse noticed the statement later and the infusion was stopped, but not before 550 mL had infused. The patient experienced a hemolytic reaction, acute renal failure, and died.
Safe Practice Recommendations
Alert practitioners to the danger of infusing sterile water without appropriate additives. Clinicians should have a clear understanding of the physiology behind infusing hypotonic, isotonic and hypertonic solutions in context of the patient's blood electrolyte levels. They also should recognize that treatment of severe hypernatremia generally consists of infusions that contain sodium to reduce blood levels slowly. Too rapid correction of hypernatremia may lead to cerebral edema, seizures and possibly death.
Develop protocols to guide safe and effective treatment of hypernatremia. If there are concerns about using dextrose solutions, elevated blood sugars can be treated with insulin. If there are concerns about fluid volume, patients can be given diuretics. If an order for sterile water is received, it should trigger an immediate call to the physician and referral to the facility's peer review process.
In the pharmacy, never allow IV compounding products to leave the sterile compounding area. Segregate these solutions and store them with warnings that they should never leave the pharmacy. The pharmacy computer should flash an alert, "Use Only as a Diluent," when these products are entered, and sterile water for injection should never appear as a choice in prescriber order entry systems. While it did not help prevent the error cited above, use of sterile water for injection in two-liter (or larger) containers for IV compounding could alert staff to its intended use. The difference in size also will reduce the risk of confusion with other one-liter IV solutions. Finally, be sure to place labels on the front of IV bags without obscuring important information.
The hospital involved in this error has asked manufacturers of sterile water for injection to place a warning label on both sides of the container. ISMP also found that current labeling on sterile water products is inconsistent among the various manufacturers. Some containers boldly state: "Pharmacy bulk package. Not for direct infusion" within a red border, while others simply state: "For drug diluent use only." USP also requires a statement that these products are not suitable for intravascular injection without first being made approximately isotonic by addition of a suitable solute. However, this warning blends in with other label text and is not seen easily (see Figure 1). We are aware of additional cases of direct injection of sterile water, so we've asked FDA and manufacturers to place stronger, more easily recognizable warnings on all large volume parenteral containers of sterile water for injection.