Frailty and postoperative outcomes in brain tumor patients: a systematic review subdivided by tumor etiology

Purpose Frailty has gained prominence in neurosurgical oncology, with more studies exploring its relationship to postoperative outcomes in brain tumor patients. As this body of literature continues to grow, concisely reviewing recent developments in the field is necessary. Here we provide a systematic review of frailty in brain tumor patients subdivided by tumor type, incorporating both modern frailty indices and traditional Karnofsky Performance Status (KPS) metrics. Methods Systematic literature review was performed using PRISMA guidelines. PubMed and Google Scholar were queried for articles related to frailty, KPS, and brain tumor outcomes. Only articles describing novel associations between frailty or KPS and primary intracranial tumors were included. Results After exclusion criteria, systematic review yielded 52 publications. Amongst malignant lesions, 16 studies focused on glioblastoma. Amongst benign tumors, 13 focused on meningiomas, and 6 focused on vestibular schwannomas. Seventeen studies grouped all brain tumor patients together. Seven studies incorporated both frailty indices and KPS into their analyses. Studies correlated frailty with various postoperative outcomes, including complications and mortality. Conclusion Our review identified several patterns of overall postsurgical outcomes reporting for patients with brain tumors and frailty. To date, reviews of frailty in patients with brain tumors have been largely limited to certain frailty indices, analyzing all patients together regardless of lesion etiology. Although this technique is beneficial in providing a general overview of frailty’s use for brain tumor patients, given each tumor pathology has its own unique etiology, this combined approach potentially neglects key nuances governing frailty’s use and prognostic value.


Introduction
An age-related syndrome of physiologic decline, frailty refers to an overall decreased state of health characterized by diminished reserves and resistance to stressors linked In the first major systematic review of frailty in brain tumor patients, Huq et al. note that relative to mFI, fewer studies employed the JHACG or HFS.Patients classified as frail by the JHACG or HFS were associated with higher mortality, complications, reoperation rates, length of stay (LOS), charges, costs, discharge disposition, and readmission rates [6,9].Given most frailty measures include a combination of history, physical examination, and determination of physical capabilities, the mFI-11 and subsequent mFI-5 were developed [4,5].Huq et al. highlight that most studies employing mFI-5 categorized patients into one of three frailty levels based on raw scores or percentages of the total frailty items [6].Some studies treated mFI-5 as a dichotomous variable, while others treated it as a continuous variable [6,[10][11][12].Studies utilizing the mFI-11 also placed patients into one of several categories based on integer or fractionated scores.The decision to group patients into two (frail vs. non-frail), three (non-frail vs. moderately frail vs. highly frail), or four (non-frail vs. low frailty vs. intermediately frail vs. highly frail) categories varied by study and cohort demographics [6].Ultimately, Huq et al. note both the mFI-5 and mFI-11 have been associated with mortality, complications, LOS, charges, discharge disposition, and readmission rates amongst brain tumor patients [6].The mFI-5 has been linked with operation time [13], and the mFI-11 has been linked with end-of-life care needs and reoperation rates [14].
Still, in reviewing various tumor types together, Huq et al. [6] potentially neglect key nuances governing frailty indices' inherent value.Their combined approach, while crucial to developing an overview, may underestimate frailty's use and prognostic value given each tumor pathology has a unique etiology.As the body of literature exploring neurosurgical outcomes in brain tumor patients continues to grow, we aim to capture some of these nuances by presenting the first review of frailty in brain tumor patients subdivided by tumor type, incorporating both modern frailty indices and traditional KPS metrics.Benign and malignant lesions will be reviewed separately, with a major focus placed on glioblastoma and meningioma given their relatively higher prevalence amongst all tumor types [15].

Methods
A systematic literature review was performed using PRISMA guidelines.PubMed and Google Scholar were queried for articles related to frailty, KPS, and brain tumor outcomes.The keywords "frailty" and "KPS" were used in combination with "glioma," "glioblastoma," "meningioma," "acoustic neuroma," and "vestibular schwannoma."Searches were performed combining the term "frailty" with each of the aforementioned tumor types.To maximize output, a search was also done combining the terms "frailty" and "brain tumor."Two authors (HQ, KP) screened abstracts and reviewed papers for study inclusion.A third author (JT) served as a reviewer/arbitrator to achieve consensus.Only articles that described novel associations between frailty or KPS and primary intracranial tumors were included.Studies examining frailty, including relevant review articles and meta-analyses, were included regardless of their frailty criteria; however, any studies that failed to define explicit frailty criteria were excluded.Those focusing on metastatic disease or radiotherapy were also generally excluded.Only studies written in English were included.References in the selected articles were reviewed to obtain a more holistic, longitudinal description of frailty in brain tumor patients.

Frailty across brain tumor types
Studies analyzing multiple tumor pathologies together have largely focused on understanding frailty's predictive value, finding that higher frailty was significantly associated with increased risk for non-home discharge [9,62], postoperative complications [9,[60][61][62], extended length of ICU and overall hospital stays [9,52,[60][61][62], mortality [54,55,60,62], readmission rates [54], all-payer hospital costs [61], and decreased rates of postoperative day one discharge [10].Although Bonney et al. report frail patients were not more likely to be readmitted than their non-frail counterparts following brain tumor resection, this may be influenced by the authors' decision to use the JHACG frailty index as opposed to the more established mFI [49].While Torres-Perez et al. report frailty scales identified vulnerable patients across tumor types with greater reliability than KPS alone, they note not every frailty scale significantly corresponded with postoperative outcomes [56].These findings highlight the importance of using more established standardized frailty indices to prognosticate outcomes.
Beyond the geriatric patient demographic, two major studies have explored frailty's relationship to postoperative outcomes in patients with GBM more generally.In a retrospective review, Botros et al. reported an increased odds of 30-day readmission with each 10-point decrease in KPS score and with each single-point increase in mFI-5 scores [16].Readmitted patients were also noted to have lower mean KPS scores relative to their non-readmitted counterparts [16].In another review, Klingenschmid et al. reported on pre and postoperative frailty using KPS and the Clinical Frailty Scale (CFS) [22,63].Both preoperative and postoperative KPS and CFS scores correlated, suggesting that the CFS may be equally reliable to KPS for patients with GBM [22].Higher scores on both scales not only significantly correlated with decreased OS, but also decreased survival by roughly the same percent per scale point [22].While age and KPS scores both predicted OS, the two variables only marginally correlated with each other, suggesting there may be distinct multiple frameworks to approach frailty [22].Still, Katiyar et al. argue that the mFI-11 is a better predictor than age of certain surgical outcomes, including hospital and ICU LOS, postoperative complications, in-hospital mortality, and psychosocial/financial difficulty post-discharge [14].perspectives on frailty's relevance across pathologies and demographic groups.

Frailty in patients with glioblastoma
Most literature concerning frailty and GBM has focused on the elderly, who tend to have poor prognoses with limited responses to treatment [66].Though Bruno et al. acknowledged age may factor into more aggressive tumor biology, they highlighted that elderly patients with poor clinical statuses due to comorbidities may not be able to tolerate surgery or adjuvant chemotherapy, making physicians more reluctant to offer aggressive treatments, in turn contributing to worse prognoses for this demographic [66].Lorimer et al.'s cross-sectional survey of UK based consultant neuro-oncologists further supports this theory [67].Another study of geriatric patients who underwent craniotomy for lobar GBM revealed that frailer patients were not only less likely to undergo surgical resection than their less frail counterparts, but also experience increased hospital stays, an increased overall risk of complications, and decreased overall survival (OS) [17].Frailty was quantified using the mFI and was associated with these outcomes independent of age, KPS, cardiovascular risk, and comorbid disease [17].At least four other studies also explored frailty's relationship to postoperative outcomes in elderly patients with GBM [19,24,26,29].Although each used different metrics to define frailty, all four found significant negative relationships between frailty and OS.In fact, Krezlin et al. reported that KPS improvement postoperatively did not reliably predict postoperative outcomes in patients with GBM unless they were already frail, suggesting that clinical and comorbidity consideration alongside KPS is critical to identifying patients for aggressive treatment and allowing for accurate prognostication [29].In contrast, Ikawa et al. found frailty predicted operative outcomes for younger patients with meningioma [34].Unlike previous studies, they utilized the mFI-5, applying it to a larger cohort of 8138 database patients, finding that mFI-5 ≥ 2 was a more significant risk factor than chronological age for poor outcomes, including mortality and complication rates, in patients under 65 [34].Their results highlight a need to examine frailty in meningioma more critically across age groups.
Another study defined early postoperative deterioration in meningioma patients as a 20-point or more drop in KPS [42].Unfavorable long-term functional autonomy and quality of life was defined as a postoperative KPS decrease of at least 20 points and overall quality of life below the 75th percentile of the examined population [42].Using a 34-point Frailty Index score, Tariciotti et al. demonstrated that preoperative frailty directly predicted both early postoperative deterioration, and long-term unfavorable outcomes along with other key tumor characteristics [42].However, they were unable to report whether this relationship is longstanding because of inherent frailty or from known longstanding medical comorbidities.
Still, studies indicate that higher mFI scores are independently associated with overall postoperative morbidity and mortality.Institutional and large database studies alike have linked mFI with non-routine discharge disposition, extended hospital LOS, readmission, and postoperative complication development, including life-threatening complications and mortality for patients with meningioma [32,33,37,41,42].Like those reporting on GBM and frailty, many of these studies in patients with meningioma treated frailty as a binary variable, using various cutoff scores to distinguish frail vs. non-frail patients.However, there does appear to be more standardization in the indices utilized by studies analyzing frailty in patients with meningioma than in patients with GBM.

Frailty in patients with other types of brain tumors
Though limited, several studies have explored frailty's relationship with postoperative outcomes in patients with vestibular schwannoma.Like studies in patients with meningioma and GBM, Nasrollahi et al. report that frail geriatric patients with vestibular schwannoma are more likely to experience increased readmission rates, LOS, and non-home discharges [47].The authors also report on higher postoperative infection, facial paralysis, urinary tract infection, hydrocephalus, and dysphagia rates specifically in geriatric patients with vestibular schwannoma [47].While Helal et al. suggest frail, elderly patients can safely undergo surgery for vestibular schwannoma, their small cohort was a limiting factor [46].
Ultimately, while studies have connected increased frailty in patients with GBM to malnutrition [50], poor follow-up care [23], and decreased temporal muscle thickness [25,27], many studies define frailty using incomparable metrics.This overall lack of standardization combined with GBM's progressive nature may be contributing to disputes regarding frailty's role in guiding care for geriatric patients with malignant primary brain tumors.Although only three studies have explored frailty's relationship to postoperative outcomes outside the geriatric demographic, frailty's relationship to OS, increased LOS, and readmissions is clear [14,22,30].

Frailty in patients with meningioma
Relative to patients with GBM, more studies have explored frailty's relationship to postoperative outcomes and clinical characteristics for patients with meningioma.Various measures of frailty have been deemed predictive of survival and morbidity amongst patients with meningioma [43].Many studies have also incorporated younger demographics, potentially enabling more meaningful longitudinal comparisons between studies across age groups.
In a study of elderly patients with primarily low-grade, skull-base meningiomas, age, sex, KPS, tumor size, tumor location, and frailty (defined as a combination of body mass index and serum albumin levels) were evaluated as risk factors for postoperative deterioration over the course of one year [38].Preoperative KPS scores, body mass index (BMI), and low serum albumin levels were each linked with poor prognostic factors [38].The BMI component was identified as a risk factor for KPS score deterioration in the immediate postoperative period [38], suggesting alternative frailty frameworks may offer more nuanced assessments of postoperative functional status than preoperative KPS scores can predict.Isobe et al. offered further insights, highlighting that both tumor location and serum albumin measures of frailty were risk factors for KPS score deterioration at discharge [35].These variables along with age and tumor size were significant risk factors for perioperative intracranial complications [35].
Other studies have approached frailty longitudinally, directly comparing the relative importance of specific clinical variables on outcomes in older vs. younger patient populations.Two out of three major frailty studies in meningioma suggest that differences in outcomes are unrelated to tumor biology or characteristics, arising instead from natural differences in KPS secondary to aging and medical comorbidities [31,39].While declining functional status in older patients predicted perioperative complications, the overall perioperative complication profile of older and younger patient demographics following resection were similar [39].
Overall, frailty studies in vestibular schwannoma demonstrate relatively more methodological standardization, with the vast majority utilizing the mFI-5 or mFI-11 to classify frailty [13,44,45,48].Patients were classified into one of three groups (non-frail, intermediate frailty, and frail) with raw mFI scores of 2 or 3 serving as the high frailty cutoff score.Both Casazza et al. [44] and Goshtasbi et al. [13] note that patients with higher mFI scores tended to have longer hospital LOS than those with lower scores.They also indicate that frailty bears little relationship with increased complication rates.In contrast, a large database study of patients with vestibular schwannoma suggests that increasing frailty is strongly associated with development of postoperative hemorrhagic or ischemic stroke and increased LOS, particularly amongst non-White patient demographics [45].While both Dicpinigaitis et al. [45] and Casazza et al. [44] utilize the mFI-11, they do utilize different cutoff scores to classify highly frail patients, potentially contributing to the differences reported by each.Ultimately, Tang et al. note that while mFI is significantly associated with perioperative outcomes, a customized frailty index for patients with vestibular schwannoma outperforms any mFI in predicting routine discharge [48].

Conclusion
Our comprehensive, systematic literature review identified several patterns of overall postsurgical outcomes reporting for patients with brain tumors.Amongst patients with GBM, frailty is associated with OS, extended hospital LOS, hospital readmission, complications, and post-discharge specialist or extra-familial care.Studies of patients with vestibular schwannoma and meningiomas reported relationships between frailty and postoperative complications, extended hospital LOS, readmission rates, non-home discharges, and mortality.A few meningioma-specific studies also reported relationships between frailty, unexpected reoperation, and KPS deterioration.While we report frailty is a more significant risk factor than age for poor outcomes in patients with meningioma, similar conclusions cannot be drawn about patients with other tumor types without further studies in younger patient demographics.In the future, it could be beneficial to use a tumor pathology-specific frailty index to incorporate clinical and functional status history alongside tumor genomics and radiographic features to better enable shared-decision making and long-term prognostication for patients with brain tumors.

Table 1
Studies Identified in Systematic Literature Review