To evaluate the pain relief effect of locally injected corticosteroid treatment in women with long-lasting low back pain beginning in pregnancy. We hypothesize that the insertion of the sacrospinous ligament on the ischial spine would be a source of pain and therefore be a target for therapy.

Primary outcome measure is reported pain intensity on visual analogue scale and secondary outcome measures number of pain-drawing locations, pain-provoking test results and tests of function.

Pregnancy related low back pain is a global problem. In most women low back pain induced in pregnancy disappears during the first six months after delivery. However, one of five women with pain during pregnancy from both sacroiliac regions and the symphysis still experienced disabling daily back pain two years after childbirth, which corresponded to 8% of the total study population. Hence, long-lasting low back pain with onset during pregnancy has to be considered a major public health problem, with high impact on the individual, family and society. Despite this, treatment and care is directed to general pain relief methods.

Precise localization of the site of pain release is fundamental in the search for an effective treatment. To date, the source of such pregnancy related low back pain is uncertain. However, the pelvic ligaments or their insertions have been proposed a source of pain and in particular the sacrospinous/sacrotuberous ligament has been indicated in this respect in pregnant and non-pregnant women.

Injection treatment with slow-release corticosteroid has shown a positive effect on pain conditions where the pain is thought to derive from collagen tissues.

We hypothesize that the insertion of the sacrospinous ligament on the ischial spine could be a source of pain in women with long-lasting low back pain beginning in pregnancy and might therefore be a target for therapy.

The women included will be randomized according to a computer generated random allocation sequence, with block size of four, concealed from the investigators until study closure.

A physiotherapist will perform the assessments at baseline and follow up four weeks after treatment. The assessment will consist of a questionnaire and a clinical examination of the back and pelvis.

Participants will be randomized to receive an injection treatment of either a compound of 1 ml triamcinolone 20mg/ml (Lederspan), Meda AB, Solna, Sweden) and 1 ml lidocaine hydrochloride 10mg/ml (Xylocain), Astra Zeneca, Södertälje, Sweden) or 0.99 ml saline solution 9mg/ml, 1 ml lidocaine hydrochloride and 0.01 ml fat emulsion (Intralipid), Fresenius Kabi, Uppsala, Sweden), the latter to make the solution opalescent as Lederspan.

Not the participant, the physician who will give the injection or the assessing physiotherapist will have information about what treatment that will be given.

Inclusion Criteria:

• reported ongoing pain in the sacral region with onset during pregnancy with six months to seven years duration after delivery
• reported pain intensity at present between 30 and 50 mm on a 100 mm horizontal visual analogue scale (VAS)
• at least one positive pain provocation test out of three and pain elicited on internal palpation at the ischial spine at least unilaterally.

Exclusion Criteria:

• on-going low back pain with onset before pregnancy
• previous back surgery
• positive straight leg-raising test
• loss of tendinous reflex in the legs