Assessment of surgery-related morbidity in patients with diffuse gliomas

Abstract

The work presented in this thesis focuses on assessment of surgery-related morbidity in patients with diffuse gliomas – the most common primary malignant brain tumour in adults. Diffuse gliomas can be subdivided into low-grade (grade II) and high-grade (grade III-IV) gliomas according to the World Health Organization (WHO) classification, where higher grades have worse prognosis. Gliomas exhibit a heterogeneous clinical course, but are for many patients associated with a high symptom burden and short life expectancy. Extensive surgical resections may potentially prolong survival, but there is a risk of new or worsened neurological deficits, postoperative complications, and/or reduced quality of life after treatment. Thus, potential survival benefits must be weighed against potential risks of surgery in a given patient. To make evidence-based clinical treatment decisions, reliable research data regarding risks and benefits of surgery are needed. In the first study in this thesis, we investigated if retrospective reviews of medical records provide a realistic impression of the incidence of postoperative complications and surgically acquired new and/or worsened neurological deficits by conducting a patient validation. We found all major postoperative complications to be registered in the medical journal within 30 days, but a considerable proportion new and/or worsened neurological deficits reported by the patients at 30 days were unregistered in the medical journal at time of discharge. In particular, cognitive deficits were largely underestimated in the medical journal compared to patient-reported data. In the second study, the accuracy of the operating neurosurgeons’ predictions about patients’ functional levels after brain tumour surgery were assessed. We found the operating neurosurgeons to be overly optimistic regarding their patients’ postoperative functional level, especially when it comes to their ability to perform normal activities at 30 days. Overestimation of functional levels was more common in patients with postoperative complications. To assess surgery-related morbidity in a more standardized and more patient centered way than traditional outcome measures, patient-reported health-related quality of life (HRQoL) was used as an outcome measure in the last two studies. By first exploring longitudinal and cumulative HRQoL in a descriptive and hypothesis-generating study, we assessed the quality of survival the 1st year after diagnosis of glioblastoma, which is the most common and most aggressive glioma subtype. We found progression-free survival to be a surrogate marker for quality of survival, and that quality of survival was associated with gross total resection (i.e. surgical resection of all visible tumour tissue). Finally, we developed a brain atlas for depicting and assessing the importance of tumour location on pre- and postoperative HRQoL in patients with high-grade glioma. The preliminary findings showed that the impact of tumour location on overall perioperative HRQoL was less than frequently believed, and the distinction between critical and less critical brain regions was more unclear according to the patients than often judged by surgeons. Based on the work in this thesis, we have experienced that it is challenging to define and measure surgery-related morbidity in brain tumour patients. The overall results and discussion highlight the need for standardized reporting, use of common definitions, development of a feasible and sensitive patient-reported HRQoL-questionnaire, and development of evidence-based tools for surgical decision making.