SpineUniverse Case Study Library

Adult Idiopathic Thoracolumbar Kyphoscoliosis

MIS or Open?


A 66-year-old female presented with a history of chronic low back pain, left thigh pain, and a progressive left-sided lumbar hump. She was "hunched over" forward and to the right.


The physical exam revealed that the patient was pitched forward 20 cm and to the right 13 cm. She had a 47º left apex thoracolumbar curve, as well as 75º kyphosis.

Prior Treatment

Previously, the patient had had a pain pump placed by a pain anesthesiologist.

Pre-treatment Images


fig1 Roh MIS Scoliosis Pre-op ClinicalsFigure 1: Clinical pictures showing left curve.


fig2 Roh MIS Scoliosis Pre-op Clinicals 2Figure 2: Clinical pictures demonstrating 75º kyphosis. The patient is hunched forward and leans to the right.


fig3 Roh MIS Scoliosis Pre-op Clinicals 3Figure 3: Note the left-sided lumbar lump.

Figure 4 Roh MIS Scoliosis Pre-op X-ray 

Figure 4: X-ray showing 47º thoracolumbar curve.

 Figure 5 Roh MIS Scoliosis Pre-op Lat X-ray

Figure 5: Lateral x-ray showing 75º kyphosis.


The patient was diagnosed with adult idiopathic thoracolumbar kyphoscoliosis.

Suggest Treatment

Indicate how you would treat this patient by completing the following brief survey. Your response will be added to our survey results below.

Selected Treatment

The patient underwent a two-stage surgery.

The first stage was L1-L5 XLIF.

Two days later, the patient had T10-S1 percutaneous instrumented spinal fusion with an L5-S1 TLIF and facet fusions at T10-L1.

Post-stage 1 Image

 fig6 Roh MIS Scoliosis Post-stage 1 X-rayFigure 6: Standing x-ray after Stage 1.


fig7 Roh MIS Scoliosis Post-stage 1 Lat X-rayFigure 7: Standing lateral x-ray after Stage 1.

Stage 2 Intraoperative Images

 fig8 Roh MIS Scoliosis Intraop Rod InsertionFigure 8: Rod insertion


fig9 Roh MIS Scoliosis Intraop MIS IncisionsFigure 9: MIS incisions

Post-treatment Images

 fig10 Roh MIS Scoliosis Pre-op to Post-op AP X-rayFigure 10: Pre-operative x-ray (left) compared to post-operative x-ray (right) showing reduction of thoracolumbar curve to 6º.


fig11 Roh MIS Scoliosis Pre-op to Post-op Lat X-rayFigure 11: Pre-operative lateral x-ray (left) compared to post-operative lateral x-ray (right). Note the decrease in kyphosis-from 75º to 20º.


fig12 Roh MIS Scoliosis Pre-op to Post-op ClinicalsFigure 12: Pre-operative clinical picture (left) compared to post-operative clinical picture (right). Note the decreased hump.


The patient's thoracolumbar curve was reduced from 47 ºto 6º. The kyphosis was reduced from 75º to 20º.

Case Discussion

This is a complex deformity that requires correction of both the sagittal and coronal planes. The XLIF procedure corrected both of these deformities quite well.

Without MRI, my concern is that restoration of the coronoal and sagittal deformity only will in part address the stenosis. Part of her sagittal plane imbalance may in fact be stenosis-related.

The jury is still out on the amount of distal fixation required for these minimally invasive deformity constructs. It has become the standard of care with open degenerative deformity to add iliac fixation, which is not present in this case. In addition, there appears there was no attempt at a posterior fusion through the entire construct.

It has been our experience that the most difficult area to obtain arthrodesis is at the thoracolumbar junction, and without an interbody arthrodesis, this will be the likely region of potential pseudoarthrosis. Psuedoarthrosis can lead to implant failure at the thoracolumbar region and will likely lead to recurrence of sagittal plane imbalance. It also appears that the immediate post-operative lateral full length spine did not completely address the sagittal plane imbalance.

Community Case Discussion (3 comments)

SpineUniverse invites spine professionals to share their thoughts on this case.

I too am concerned about the lack of fusion potential at the TL junction and above. Additionally, this construct puts a great deal of force on the sacral fixation and, even with exceptional medial convergence of the sacral screws and bicortical fixation at S1 and L5 there is a substantial risk of hardware failure and psuedoarthrosis at L5-S1. The facets have already proved themselves incompetent due to the deformity and the sagittal balance is still slightly positive. I would have added sacropelvic fixation. Additionally, I would have strongly considered performing small SPOs or Ponte's at T12 an T11 to better flatten the TL junction and added a little kyphosis to the top of the rod at the cephalad end, which for me would have been T9. There is some evidence that this might have a protective effect from junctional kyphosis. Percutaneous pedicle screw fixation of the lumbar spine after lateral interbody fusion is becoming common-place for the treatment of degenerative scoliosis (which is what I think this is). However, minimally invasive placement of hardware should not trump fusion technique.... In my opinion.

I would be concerned that the sacral fixation and/or proximal adjacent segment is at high risk for failure because of 3 issues. 1) Sacral screws not in tricortical buttress or bicortical. 2)No iliac screws. 3) Lumbar lordosis is minimal and far less than patient's pelvic inclination would necessitate to achieve sustained global sagittal balance with minimal stresses. Attending to these issues as well as adequate neural decompression, particularly in the foramina, is what seems to lead to the best results, regardless of the approach, MIS or open. I am starting to see severe rapid adjacent segment disease in patients following XLIFs with minimal lordosis.

It seems like the procedure resulted in improvement of patient's sagittal and coronal alignments. But I am afraid that the correction may only be short-term, and may breakdown above, and maybe below (without iliac screws-- which could be placed in MIS fashion). First of, as mentioned in other comments, the alignment is not ideal (much better than pre-op). The XLIFs although provides great anterior column support, and some indirect decompression by increasing foraminal volume, when one has either severe stenosis, or flatback, simply relying on instrumenting, without performing open decompression (for moderate or severe stenosis), or without performing SPO (for flatback) will likely yield an okay to good looking xray, but likely not so happy patient in the long-run.

Although one can use hyperlordotic type anteriolateral cages to improve sagittal alignment in a patient with flatback, most of the time, it is not enough, and open procedure may still be necessary. In this case, the XLIFs, did not really improve much of the lumbar sagittal balance -- the patient still has a flatback (it helped patient improve from slight kyphosis to a flatback).

Also in these long-construct deformity MIS cases, at times, in attempts to make MIS rod insertion and reduction easier, one may inadvertently cheat and "under bend/contour" the rod, so as to make for easier rod introduction into the screw extenders.

With regards to the top of the construct, given the extent of patients pre-operative kyphosis, stopping at T10 is probably "too short." I would have gone higher. There will be increased stress at the top of the construct given patient's pre-operative posture, which the surgery did not do much to correct.

Patient's pre-operative kyphosis doesn't seem to be structural, but as previously mentioned by Dr. Schwender, could be related to lumbar stenosis. That sort of curvature will correct significantly under anesthesia, especially when patient is prone. The procedure more or less holds patient at where she passively corrects to under anestheia. Even with SPOs, which would better have an impact of kyphosis correction and improvement of global sagittal balance, I would still have gone above T10.

I also agree that there is increased stress on the sacral fixation. Various options for these type long constructs includes a) TLIF at L5-S1, which usually introduces greater stresses on the sacral fixation, b) Iliac screws, which helps decrease stress on the sacral fixation, c) AxiaLIF which only in select patient may be used as the bolt provides more rigid fixation of the L5-S1 segment and helps protect the screws, however this fixation I believe is still inferior to bi-iliac screws, and should not suffice for a decompressive procedure at the L5-S1 segment if there is stenosis or symptomatic herniated disc present.

It is good to think MIS, and aim at trying to reduce trauma to patients during surgeries, however we should not hesitate to open if need be to achieve better patient outcome (if case warrants).

It's easy to evaluate, critic and comment on another's cases, but great kudos to Dr. Roh on tackling this very challenging case, and improving patient's alignment.


Get new patient cases delivered to your inbox

Sign up for our healthcare professional eNewsletter, SpineMonitor.
Sign Up!