Surgical treatment of thoraco-lumbar kyphosis (TLK) associated with low pelvic incidence.

PURPOSE: Thoraco-lumbar kyphosis (TLK) is poorly described in the literature and its surgical treatment remains equivocal for patients with low pelvic incidence. The aim of the study was to identify which surgical correction would yield the best functional results as measured by the Oswestry score.

MATERIALS AND METHODS: This is a retrospective study including patients described as a type 1 of Roussouly's classification with a thoraco-lumbar kyphosis. Fifty-six patients with degenerative spinal disease were included, 42 (75%) with scoliosis and 14 (25%) without. Patients had a median age of 56 years (49-63), and there were 6 (11%) men. The primary outcome was the functional Oswestry disability index (ODI).

RESULTS: At last follow-up, the median ODI was 15 (Q1-Q3: 6-23). In the degenerative spinal disease group, the median ODI was 7 (3.5-20) at last follow-up. Ten (84%) patients were classified as Roussouly's type 1 after surgery. In the degenerative scoliosis group, the ODI was 17 (8-23) in patients classified as Roussouly's type 1 and 20 (7.5-25) in patients classified as Roussouly's type 2 with no significant difference between these groups (p = 1). There were two patients classified as Roussouly's type 3a and their ODI at last follow-up were of 60 and 50.

CONCLUSION: It seems that keeping the physiological morphology is the treatment of choice. For patients with degenerative scoliosis, reducing the kyphosis could work. We do not recommend to increase the lordosis and obtain a type 3 morphology with an anteverted pelvis because of the risk of PJK and poor functional results.