Bacteriology, drug stability and exchange of percutaneous delivery systems and antibacterial filters in long-term intrathecal infusion of opioid drugs and bupivacaine in refractory' pain

OBJECTIVE: To provide a basis for recommendations on the exchange of containers (syringes and cassettes) and antibacterial filters, and for choice of administration device in patients with "refractory" pain treated with long-term percutaneous intrathecal (IT) infusions of opioid (morphine or buprenorphine) and bupivacaine mixtures. DESIGN: Prospective, cohort, nonrandomized control trial-case series, with consecutive sample, no standard criterion, and cost-benefit analysis. SETTING: Tertiary care center, institutional practice as well as hospitalized and ambulatory care. PATIENTS: Eighty-nine (51 women and 38 men); 81 with malignant pain and 8 with benign "refractory" pain. INTERVENTIONS: (a) The chemical stability of the drugs in the containers during 30 days. (b) The results of bacteriologic culture of the residual volumes of the analgesic mixtures from used and reused (1-16 times) syringes (n = 135) and cassettes (n = 258), and of 5 ml of sterile isotonic saline filtered through the used Millipore filters (n = 149). The bacteriologic samples from the 89 patients were taken after 1-40 (median = 7), 1-86 (median = 20), and 5-78 (median = 31) days of IT treatment, respectively. MAIN OUTCOME MEASURES: Chemical stability: buprenorphine and bupivacaine concentrations-liquid chromatography; morphine concentrations--gas chromatography. Bacteriologic cultures: standard laboratory procedures. The hypothesis (repeated use of the infusion systems and their exchange once a month does not significantly affect drug concentrations or increase the infection risk) was elaborated before data collection began. RESULTS: The bupivacaine-opioid mixtures were found to be chemically stable within 3-10% of the original doses up to 30 days. Seventeen cultures (from five syringes, six cassettes, and six filters) in 13 patients (having no signs of meningeal infection) were found to be colonized with Staphylococcus aureus (n = 4), coagulase negative staphylococci (n = 7), viridans streptococci (n = 3), Neisseria sp (n = 4), Corynebacterium sp (n = 4), Enterobacter sp (n = 2), Klebsiella sp (n = 3), gram-negative bacilli (n = 1), and yeasts (n = 2). The place of IT treatment, its duration, and patient-related infection risk factors (age; malignancy; diabetes; presence of a colostomy, pyelostomy, or indwelling urinary catheter; and the presence and location of infection foci) were not related to the results of the cultures. However, 9 of the 17 positive cultures came from patients with skin ulcers, a notable incidence. The positive cultures had no connection with the cultured item, its in-use duration, the number of times of reuse, the analgesic drugs used, their concentrations or the presence of preservatives (sodium metabisulfite and sodium edetate), or the antiseptic agent (70% ethanol or 0.5% chlorhexidine gluconate) used during bacteriologic sampling. The bacterial growth was sparse in 14 and massive in 3 of the 17 positive cultures. One item (filter) from one patient with meningitis was sterile. CONCLUSIONS: In our population, exchange of the infusion systems when they are empty (within 1 month) and of the antibacterial filters once a month does not appear to affect the concentrations of, or increase the infection risk from, the opioid-bupivacaine mixtures. The risk of bacterial contamination/colonization of the syringes from syringe drivers does not seem to be higher than that of cassettes from external portable pumps.