Lumbar spinal stenosis is an abnormal narrowing of the spinal canal that causes impingement of the
nerves traveling through the lower back into the legs. Symptoms may include pain, weakness or
numbness in legs, calves or buttocks. Walking, standing and extending the lower back can aggravate
symptoms. Lumbar spinal stenosis is a frequent age-related spinal disorder with an estimated
prevalence of 20-50% among the geriatric segment of the population and is the leading cause of
spinal surgery among the elderly. According to the Danish national spine registry DaneSpine, 3/4 of
patients can expect considerable pain relief 1 year after surgery and 2/3 will experience
improvements in quality of life. About 1/4 - 1/3 of patients do not feel a clear improvement.
The variation in surgical outcome often makes it difficult to communicate a reliable prognosis to the
patient. The main purpose of this thesis was to establish if reliable individualized estimates of
outcome could be computed preoperatively through predictive algorithms modelled on existing
registry data.
In study 1, we found that the predictive performance of the Swedish decision support Dialogue
Support did not generalize well to Danish patient samples. While AUC values were comparable
with the results reported by the authors of the Dialogue Support, both calibration plots and
performance metrics revealed a low ability to correctly identify unfavourable outcomes (true
negatives).
In study 2, seven different machine learning algorithms were applied to Danish spine data. On
average, they performed nearly equally well but variation was found across outcome measures. The
EQ-5D and VAS back models performed almost equally well, while the ODI and VAS leg models
were less convincing. VAS leg and Return to work models exhibited the largest differences in
performance between algorithms. MARS and deep learning performed consistently well.
In study 3, non-operated LSS patients were matched with operated patients to ensure equivalent
baseline characteristics. Both groups were diagnosed with MRI-confirmed LSS by spine surgeons.
The outcome was compared at 1 year following their initial consultation with a spine surgeon.
Although both groups improved on average, differences were in favour of the operated patients
whether measured as mean improvement or proportions reaching a minimal clinical important
difference. Less than half of the non-operated achieved MCID on EQ-5D, VAS back/legs compared to
2/3 of the operated.
In study 4, we developed a decision support tool PROPOSE capable of predicting personalized
outcome at 1 year following surgery. It was validated by surgeons in a clinical consultation setting.
The predictive performance of PROPOSE with respect to EQ-5D, VAS back and ODI was fair to
excellent, while VAS leg was less predictive in comparison. The tool could prove useful as a mean to
let patients and surgeons engage in discussions on likely outcome and support clinical decisionmaking.
Before considering its use in clinical practice it should however be thoroughly validated on
external data sources independent of the data used in development.
| 