A quarter of intrathecal pumps for non-cancer pain came out, mostly for infection

Design

retrospective single-centre cohort, mean follow-up 80 ± 48.8 months

Population

215 patients with chronic non-cancer pain implanted 2009-2022; mean age 63.2 years, 184 with persistent spinal pain syndrome type 2

Primary outcome

device explant and the complications leading to it

Effect

explant rate 26.1%; device-related infection in 9.3% of patients and 5.9% of device surgeries; loss of analgesic efficacy in 5.6%; no demographic or comorbidity predictors

The companion question to the cancer data is what happens over years in patients with chronic non-cancer pain, who live long enough for device complications to accumulate. This retrospective series followed 215 patients implanted between 2009 and 2022 at a tertiary academic practice for a mean of 80 months. Average age at implantation was 63, and 184 of the 215 had persistent spinal pain syndrome type 2 - chronic back pain after previous lumbar surgery.

The explant rate was 26.1%. Device-related infection was the leading cause, affecting 9.3% of patients and 5.9% of device surgeries, and infections occurred much earlier after implantation than other reasons for removal. Loss of analgesic efficacy accounted for explant in 5.6% of patients, which the authors note is significantly lower than the explant rate for habituation with spinal cord stimulation in a comparable population. No demographic or comorbidity factor predicted explant.

Two things follow. The first is a consent figure: over nearly seven years, one device in four came out, and roughly one patient in eleven had an infection serious enough to require removal. That is a different conversation from the one implied by describing a pump as a long-term solution. The second is more encouraging - tolerance was not the main failure mode, which distinguishes intrathecal delivery from spinal cord stimulation and is an argument for it in a patient who has already failed stimulation for that reason.

• Consent for an explant rate around a quarter over long follow-up, and an infection-related removal rate near one in eleven
• Concentrate infection prevention effort on the implantation and early revision period - that is when infective explants occurred
• Do not expect demographic or comorbidity screening to identify who will need explant; nothing predicted it here
• Where a patient has lost benefit from spinal cord stimulation through habituation, note that loss of efficacy was a less common failure mode with intrathecal delivery
• Track your own unit's explant and infection rates - this is a single-centre figure to benchmark against, not a standard

Why it matters

It gives the long-term failure profile of a device usually discussed in terms of its benefits.

Don't overread it

A single-centre retrospective series - rates reflect that centre's technique, patient selection and infection control, and the stimulation comparison is not randomised.

The statistics, in plain English

A 26.1% explant rate across a mean 80 months is a cumulative figure, not an annual one, so it should be presented to a patient with the time frame attached. The comparison with spinal cord stimulation habituation is drawn against a separate population at the same centre rather than within a randomised comparison, so it indicates a difference rather than establishing one. That no demographic or comorbidity factor predicted explant is a genuinely useful null: it means there is no screening shortcut here, and 215 patients is enough to have found a strong predictor had one existed.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.