Page 14 - Predicting survival in patients with spinal bone metastasesL

CHAPTER I
in nature and well tolerated. The downside is that the effects take several days to weeks to become apparent, making it inferior to surgery in the acute setting of severe neurological deficits. Also, radiotherapy is not indicated in the case of spinal instability, as this would require surgical fixation for proper management. Surgical techniques range from minimal invasive options to en-bloc resection of the affected spinal segments. Vertebroplasty, kyphoplasty or percutaneous pedicle screw fixation can also be combined with radiofrequency ablation (RFA) in order to achieve better local control. Open techniques include decompressive surgery and corpectomies, which should always be followed by fixation of the spinal column in order to ensure stability. Even though these techniques are highly effective in achieving local control and alleviating or avoiding neurologic symptoms, the complication rate is high (20-33%)21 and recovery from major surgery can be especially challenging in this fragile patient population. Therefore careful patient selection with a focus on extent of disease, life expectancy and patient preference is needed in order to avoid overtreatment.
PROGNOSTIC MODELS
Prediction of survival has become an integral part of selecting the appropriate treatment for patients suffering from symptomatic spinal bone metastases (SBM). Depending on the symptoms, patients with a short expected survival are most likely to benefit from short radiotherapy regimens, best supportive care or minimally invasive surgery. Patients with a relatively long expected survival could in some cases benefit from high dose radiotherapy or – in the case of radioresistant tumors and biomechanical instability – more extensive surgical interventions. Several studies have shown that the estimation of survival by clinicians in terminally ill patients is inaccurate and have suggested the use of prognostication models in order to prevent exposure of patients to unnecessarily extensive treatments.22-24 Models to aid in the decision making process have been developed by Tokuhashi, Tomita and Van der Linden, amongst others.25-27 These models encompass prognostic factors such as primary tumor type, amount and location of spinal metastases, presence of visceral, brain and extraspinal bone metastases, functional status and neurological status. However, clinical applicability seems limited due to over- or under treatment and especially when predicting short survival the existing models fall short.28,29
12