How Often Can You Have Epidural Steroid Injections?

Written by Dr. Mar\u00eda Alejandra De La Pe\u00f1a

Updated September 2026 ·8 min read

There is no single legislated number. In practice most clinicians work to a limit of around three to four epidural steroid injections in a twelve-month period, spaced at least two weeks apart and usually longer.

More useful than the number is understanding what the limit is protecting against, and what should happen when injections stop working.

What actually sets the limit

Cumulative steroid exposure. Each injection delivers a meaningful dose — typically 40 to 80 mg for an interlaminar approach and around half that for transforaminal. Repeated often enough, the exposure begins to resemble systemic steroid treatment, which is the context in which bone density loss and adrenal suppression arise.

Dr. María Alejandra De La Peña draws that distinction explicitly: the long-term complications of steroids are “usually more common with patients that have systemic steroids, meaning oral steroids” rather than with spaced injections. Spacing is what keeps the two categories apart.

Diminishing returns. An injection that helps less than the one before is information about the diagnosis, not a reason to try again sooner.

The glucose burden. In diabetic patients each injection brings three to five days of raised blood glucose. Frequent injections mean a substantial share of the year spent in rebound hyperglycaemia.

How far apart

Two weeks is the usual minimum, and there is a reason it is not shorter: the steroid has not finished working before then.

Relief in the first hours comes from local anaesthetic. The steroid effect appears over roughly three to fourteen days and plateaus after that. Repeating at one week means repeating before you know whether the first injection worked.

Where a series is planned in advance, three to four weeks between injections is common. Where an injection has worked well, the next is given when symptoms return rather than on a schedule.

Is a series of three standard?

Not automatically, and it should not be booked as a package before the first result is known.

A series makes sense where each injection produces partial, additive improvement. It makes no sense where the first produced nothing — two more of the same are unlikely to differ.

The more useful question after a failed injection is whether the approach was right. Faculty teaching is explicit that the route follows the symptom pattern: interlaminar where there is back and leg pain, transforaminal where the patient says “I only have sciatica”, caudal where previous fusion has produced adhesions. A poor result sometimes means the target was wrong rather than the treatment.

When injections stop working

Three reasonable next steps, in rough order:

1. Re-examine the correlation. De La Peña's governing rule is that “we always have to correlate images with symptoms, not just images.” A scan showing several abnormalities does not tell you which one hurts; the dermatomal pattern does.
2. Reconsider the approach. A different route may reach a target the first did not.
3. Move on. Continuing to inject a patient who is not responding is not conservative management. Other procedures, rehabilitation and surgical opinion all exist.

She is also candid about the post-surgical group in particular: a patient may have “good initial response… but over time, with the degeneration, he will have recurrence.” Recurrence in a degenerative spine is the expected course, not a treatment failure — and saying so in advance is what keeps expectations intact.

How long each injection should last

Duration depends partly on which steroid was used, and that choice is made for safety rather than longevity.

Particulate steroids such as triamcinolone are depot preparations and last longer. Non-particulate dexamethasone, preferred for transforaminal injections because of the arteries in the foramen, has “a shorter duration of action”.

So a transforaminal injection wearing off sooner than a previous interlaminar one may reflect a deliberate safety tradeoff rather than a worse result. It is worth asking which steroid was used.

Questions worth asking

1. Which approach was used, and why that one for my symptom pattern?
2. Which steroid, particulate or non-particulate, and what duration should I expect?
3. At what point do we decide this is working? Name the date.
4. What is the plan if it does not — a different approach, or a different treatment?
5. How many are we willing to do in a year, and what happens when we reach that?