A multivariate analysis of prognostic factors for health-related quality of life in patients with surgically managed meningioma

Abstract

The objective of this study was to examine the prognostic significance of health-related quality of life (HQOL) parameters combined with baseline clinical factors in patients undergoing neurosurgery for treatment of meningioma. A total of 147 patients (61 male, 86 female; mean age 43 years, range 5–77 years) who underwent resection of a meningioma between January 2002 and December 2004 were studied. HQOL was evaluated using a modified questionnaire based on the World Health Organization Quality of Life-100 Scale and the Karnofsky Performance Scale. The relationships between HQOL and clinical history, radiological findings, extent of resection, histological grade and recurrence were investigated using multivariate analysis. The mean HQOL score was 73.94 ± 1.79 for preoperative patients with meningioma, 84.88 ± 2.14 for postoperative patients, and 91.13 ± 1.61 for healthy controls. HQOL for patients with meningioma was significantly lower than that for normal controls (P < 0.001), and postoperative patients had a more satisfactory HQOL than preoperative (P < 0.05). Cox proportional hazards analysis showed that significant predictors of health-related quality of life were tumor size, extent of surgical excision, and histologic grade. Multivariate backward logistic regression yielded the regression equation HQOL = 119.1097 – 1.5002X₃ – 8.6650X₆ – 10.4210X₇ (R = 0.7466; where X₃ is tumor size, X₆ is extent of surgical excision, and X₇ is the histologic grade of the tumor). This equation can be used preoperatively to predict the HQOL of meningioma patients after neurosurgery. A specialized HQOL questionnaire for patients with meningioma provides useful information when planning the operative procedure, and may make it more likely that patients have a satisfactory HQOL after surgery.

Introduction

Meningioma is a common brain tumor for which surgical excision is the often the treatment of choice. Survival rates for patients with meningioma have increased substantially in the past 30 years to the point where attention is now focused on the burdens of morbidity associated with the short- and long-term effects of treatment.¹ Measurements of health status and health-related quality of life (HQOL) can make useful contributions in this regard.² HQOL is a multidimensional construct covering the patient’s perception of the impact of disease and treatment on his or her function in various aspects of life, including the physical, psychological and social domains.[3], [4]

Baseline patient and disease characteristics are well-established predictors of outcome in tumor patients. In particular, baseline HQOL correlates with clinical outcome of tumor treatment, and parameters such as good global HQOL, good physical functioning, minimal pain, and minimal fatigue are predictors of better clinical outcome.[5], [6] Several groups have investigated the prognostic significance of baseline HQOL parameters and baseline clinical characteristics for clinical outcome.[7], [8], [9], [10] In these studies, a variety of unspecialized HQOL instruments were used to evaluate global HQOL after meningioma resection. However, it is not sufficient to describe the relationship between meningioma and HQOL in global terms. Rather, identification of preoperative factors influencing the HQOL of meningioma patients and development of a predictive equation for postoperative QOL would be more valuable.

Due to the many factors involved, and the complex interactions between them, postoperative HQOL in patients undergoing brain surgery is difficult to predict. Hence, using HQOL to assess the success of an operation and other relevant factors synthetically may be very useful. In a previous study, we assessed HQOL in patients with meningioma using the modified Quality of Life Questionnaire,¹¹ a validated, comprehensive, multidimensional self-report instrument. We also determined the prognostic significance of HQOL parameters in the presence of predefined clinical prognostic factors for time-to-event endpoints (overall survival, time to progressive disease, and time to treatment failure). In the present study, we aimed to investigate the prognostic significance of HQOL for predicting postoperative outcome.

Unlike an epidemiologic study, we focused primarily on the clinical features of most concern to neurosurgeons, including inherent tumor characteristics and surgical factors. We wished to investigate the relationships among inherent, intervening (surgical) factors and the HQOL of patients, how they impact on HQOL, and how the HQOL of patients with meningioma can be evaluated and improved.