Preventing medication errors in cancer chemotherapy.

Recommendations for preventing medication errors in cancer chemotherapy are made. Before a health care provider is granted privileges to prescribe, dispense, or administer antineoplastic agents, he or she should undergo a tailored educational program and possibly testing or certification. Appropriate reference materials should be developed. Each institution should develop a dose-verification process with as many independent checks as possible. A detailed checklist covering prescribing, transcribing, dispensing, and administration should be used. Oral orders are not acceptable. All doses should be calculated independently by the physician, the pharmacist, and the nurse. Dosage limits should be established and a review process set up for doses that exceed the limits. These limits should be entered into pharmacy computer systems, listed on preprinted order forms, stated on the product packaging, placed in strategic locations in the institution, and communicated to employees. The prescribing vocabulary must be standardized. Acronyms, abbreviations, and brand names must be avoided and steps taken to avoid other sources of confusion in the written orders, such as trailing zeros. Preprinted antineoplastic drug order forms containing checklists can help avoid errors. Manufacturers should be encouraged to avoid or eliminate ambiguities in drug names and dosing information. Patients must be educated about all aspects of their cancer chemotherapy, as patients represent a last line of defense against errors. An interdisciplinary team at each practice site should review every medication error reported. Pharmacists should be involved at all sites where antineoplastic agents are dispensed. Although it may not be possible to eliminate all medication errors in cancer chemotherapy, the risk can be minimized through specific steps. Because of their training and experience, pharmacists should take the lead in this effort.

Ondansetron use in a major university teaching hospital.

A drug-use evaluation of ondansetron was conducted. Literature-based criteria for the use of ondansetron were developed by an oncology specialty resident and a clinical pharmacy specialist in oncology. Orders for ondansetron written in the hospital or in the outpatient oncology clinic were reviewed over a nine-week period. After an interval of education, the use of ondansetron was re-evaluated. The data collected included the indication for use, dosage, route of administration, and, for inpatient orders during the first evaluation, therapeutic outcome. In the first evaluation, 100 orders were reviewed. Some 87% of the orders for inpatients and 80% of the outpatient orders met the criteria for therapeutic indication. Criteria for dosage and administration were met by 76% of the inpatient orders and 83% of the outpatient orders. The inpatient orders met the criteria for therapeutic outcome in 93% of cases. In the follow-up evaluation, 50 orders were reviewed. Some 96% of the inpatient orders and 100% of the outpatient orders met the criteria for therapeutic indication. Dosage and administration criteria were met by 92% of the inpatient orders but only 24% of the outpatient orders. A considerable potential for cost avoidance was identified. An evaluation of ondansetron use led to general improvements in prescribing patterns, but continued monitoring is necessary to reduce the use of inappropriate dosages and routes of administration for outpatients.

Stability and compatibility of fluorouracil and mannitol during simulated Y-site administration.

The stability and compatibility of fluorouracil admixtures with mannitol during simulated Y-site administration was studied. Fluorouracil injection 50 mg/mL was diluted with 5% dextrose injection, 0.9% sodium chloride injection, and 5% dextrose and 0.45% sodium chloride injection to final concentrations of 1 and 2 mg/mL. Combinations of fluorouracil admixtures with 20% mannitol injection were made using equal volumes in glass test tubes; immediately after mixing and at one, two, and four hours, the samples were examined for visual incompatibilities. Duplicate combinations of fluorouracil admixtures with 20% mannitol injection were made using equal volumes in plastic syringes; immediately after mixing with internal standard in glass test tubes and at 2, 4, 8, and 24 hours, samples were removed for chemical analysis. A high-performance liquid chromatographic assay was used to determine fluorouracil concentrations. No evidence of precipitation, color change, or haze was observed. During the 24-hour study, fluorouracil concentrations remained within 6% of initial concentrations for all combinations with mannitol. Fluorouracil 1 and 2 mg/mL in 5% dextrose injection, 0.9% sodium chloride injection, and 5% dextrose and 0.45% sodium chloride injection was chemically stable and visually compatible when combined with 20% mannitol injection during simulated Y-site administration.

Carboplatin: a new cisplatin analog.

Carboplatin, a new antineoplastic agent with a spectrum of antitumor activity similar to cisplatin, has shown appreciable activity in patients with ovarian carcinoma, head and neck cancer, and small-cell lung cancer. This platinum complex is less nephrotoxic, ototoxic, and neurotoxic than cisplatin. Myelosuppression may be severe and dose-limiting. Carboplatin distributes into a volume approximating total body water, and is slowly bound to plasma proteins; its elimination is a biphasic process. Renal clearance of free platinum from carboplatin correlates highly with creatinine clearance in patients with normal or impaired renal function. The recommended iv dose of carboplatin as a single agent in previously untreated patients is 400-500 mg/m2; dosage must be reduced in patients with decreased renal function, low initial platelet count, or extensive prior chemotherapy or radiation therapy. Carboplatin will be most useful in patients with decreased renal function and those who cannot tolerate high-volume hydration regimens. Patients at higher risk for development of cisplatin-related ototoxicity or neurotoxicity (e.g., patients expected to receive cumulative cisplatin doses exceeding 600-800 mg/m2) may be ideal candidates for carboplatin as initial therapy. Large-scale comparative trials are needed before carboplatin can be recommended as a replacement for cisplatin.

Liquid-chromatographic monitoring of cytosine arabinoside and its metabolite, uracil arabinoside, in serum.

This is a "high-performance" liquid-chromatographic method for quantifying the antileukemic drug cytosine arabinoside (cytarabine; 1-beta-D-arabinofuranosylcytosine; Ara-C), with a structural analog, 5-methylcytidine, as the internal standard. We used a C18 reversed-phase column and ammonium acetate (0.5 mol/L, pH 6.5) as the mobile phase, monitoring the column effluent at 280 nm. Tetrahydrouridine was present in the sample-collection tubes to inhibit conversion of cytosine arabinoside to uracil arabinoside. The standard curve is linear to 100 mg/L. Analytical recovery is 98%. Coefficients of variation for within-run and between-run imprecision were 2.0% and 4.3% at 20 mg/L and 2.7% and 2.7% at 80 mg/L, respectively. Assay sensitivity was limited by the amount of endogenous material in each patient's serum, making assay of a pre-infusion sample necessary for accurate calculations. In a trial patient population, the assay was shown to have potential for the detection of toxic concentrations in patients receiving high doses of Ara-C.

Topical mechlorethamine in the treatment of mycosis fungoides. Uniformity of application and potential for environmental contamination.

Topical mechlorethamine hydrochloride is commonly used in the treatment of mycosis fungoides and has been formulated in both aqueous and ointment vehicles. Two concerns regarding the topical application of mechlorethamine hydrochloride relate to the adequacy of skin coverage that can be attained by the patient and the extent to which others in the patient's household might be exposed to the drug. In this study six patients applied either aqueous or ointment vehicles containing a fluorescent dye. Subsequent examination of the skin under a Wood's lamp revealed a significant percentage of body surface area to be missed during application; several areas were noted to be missed most commonly. These observations have led to specific alterations in instructions given to patients regarding drug application. Examination of the surrounding environment showed minimal evidence of contamination.