A Clinical Evaluation of the Accuracy of an Intrathecal Drug Delivery Device

Implantable intrathecal drug delivery (ITDD) devices are used to treat severe pain and spasticity refractory to conventional medical management. Although off-label medications and drug admixtures are commonly used in clinical practice and recommended by international guidelines, manufacturers state that this practice can result in device failure. The impact of off-label drugs and drug combinations on pump accuracy has hitherto never been assessed.

A multinational, three-center, retrospective review of patient records was undertaken. The inclusion criterion was the presence of an ITDD device implantation in adult patients, with the pump in situ for the expected battery lifespan. Residual drug volumes at each refill, drug mixtures and concentrations, and rate and flow pattern of the pump (simple or flex) were recorded. A normalized flow rate ratio was calculated (actual to theoretical flow rate). The impact of nonapproved drugs, battery life, pump size, and flow program on drug delivery accuracy was assessed.

Data from 1402 pump refills were collected (73 patients). The overall mean accuracy ratio was 0.995 (95% CI = 0.986-1.004). The ratio for approved drug status was 0.990 vs 0.997 in nonapproved, with a difference of -0.007 (-0.032 to 0.017). At the tenth centile for remaining battery life (14 months), the ratio was 0.983 vs 1.009 for the 90th centile (69 months), with a difference of -0.026 (-0.038 to -0.014). The ratio for flex administration was 0.982 vs 1.006 for simple, with a difference of -0.024 (-0.040 to -0.008). For pump size of 40 mL, the ratio was 0.975 vs 1.010 for 20 mL, with a difference of -0.035 (-0.063 to -0.008). The 95% prediction interval for individual refill ratios was ±0.15.

In a clinical setting, the ITDD pumps retained high levels of accuracy and acceptable precision across their lifespan despite using unapproved drugs or admixtures and under various flow modes and rates.