The 5-Item Modified Frailty Index Is Predictive of Severe Adverse Events in Patients Undergoing Surgery for Adult Spinal Deformity

Mitsuru Yagi, MD, PhD; Takehiro Michikawa, MD, PhD; Naobumi Hosogane, MD, PhD; Nobuyuki Fujita, MD, PhD; Eijiro Okada, MD, PhD; Satoshi Suzuki, MD, PhD; Osahiko Tsuji, MD, PhD; Narihito Nagoshi, MD, PhD; Takashi Asazuma, MD, PhD; Takashi Tsuji, MD, PhD; Masaya Nakamura, MD, PhD; Morio Matsumoto, MD, PhD; Kota Watanabe, MD, PhD


Spine. 2019;44(18):E1083-E1091. 

In This Article

Abstract and Introduction


Study Design: A retrospective review of 281 consecutive cases of adult spine deformity (ASD) surgery (age 55 +/- 19 yrs, 91% female, follow-up 4.3 +/- 1.9 yrs) from a multicenter database.

Objective: To compare the value and predictive ability of the 5-item modified frailty index (mFI-5) to the conventional 11-item modified frailty index (mFI-11) for severe adverse events (SAEs).

Summary of Background Data: Several recent studies have described associations between frailty and surgical complications. However, the predictive power and usefulness of the mFI-5 have not been proven.

Methods: SAEs were defined as: Clavien-Dindo grade >3, reoperation required, deterioration of motor function at discharge, or new motor deficit within 2 years. The patients' frailty was categorized by the mFI-5 and mFI-11 (robust, prefrail, or frail). Spearman's rho was used to assess correlation between the mFI-5 and mFI-11. Univariate and multivariate Poisson regression analyses were conducted to analyze the relative risk of mFI-5 and mFI-11 as a predictor for SAEs in ASD surgery. Age, sex, and baseline sagittal alignment (Schwab-SRS classification subcategories) were used to adjust the baseline variance of the patients.

Results: Of the 281 patients, 63 (22%) had developed SAE at 2 years. The weighted Kappa ratio between the mFI-5 and mFI-11 was 0.87, indicating excellent concordance across ASD surgery. Frailty was associated with increased total complications, perioperative complications, implant-related complications, and SAEs. Adjusted and unadjusted models showed similar c-statistics for mFI-5 and mFI-11 and a strong predictive ability for SAEs in ASD surgery. As the mFI-5 increased from 0 to ≥2, the rate of SAEs increased from 17% to 63% (P < 0.01), and the relative risk was 2.2 (95% CI: 1.3–3.7).

Conclusion: The mFI-5 and the mFI-11 were equally effective predictors of SEA development in ASD surgery. The evaluation of patient frailty using mFI-5 may help surgeons optimize procedures and counsel patients.

Level of Evidence: 4


Although adult spinal deformity (ASD) is heterogeneous, with a wide range of ages, physical functions, types of curvature, and disease conditions, it is widely recognized that corrective spine surgery for this condition improves patient clinical outcomes.[1–7] However, ASD surgery usually requires a large dissection, multilevel instrumentation and fusion, and osteotomy, and is associated with blood loss, and therefore, it carries a substantial risk for severe adverse effects (SAEs).[2–7] As the population is rapidly aging in modern countries, more elderly patients will undergo surgery for ASD. Elderly ASD patients often have multiple comorbidities, poor cardiopulmonary function, and loss of musculoskeletal condition due to limited activity.[8–12] Hence, the influence of comorbidities and frailty on the development of surgical complications has drawn increasing attention, although many of these relationships are still controversial.[13–20] Worley et al[21] reviewed 11,982 ASD patients from the Nationwide Inpatient Sample database and reported that age and comorbidities are significant risk factors for postoperative complications (age >65 yrs [odds ratio; 1.09], congestive heart failure [odds ratio; 1.62]). In contrast, Daubs et al[22] reviewed 46 surgically treated elderly (>60 yrs) ASD patients and concluded that comorbidity had no association with complication rates or outcomes. Puvanesarajah et al[23] reviewed 2293 surgically treated elderly (>65 yrs) ASD patients and reported that the reoperation rate was comparable to that in younger patients, but comorbidities such as osteoporosis and smoking increased the likelihood of reoperation in the elderly population.

Recent studies have demonstrated that frailty is a useful predictor of SAEs when considering surgical treatment for elderly patients.[13–20] A consensus conference in December of 2012 led by the International Association of Gerontology and Geriatrics and the World Health Organization defined frailty as "a medical syndrome with multiple causes and contributors that is characterized by diminished strength, endurance, and reduced physiologic function that increases an individual's vulnerability for developing increased dependency and/or death."[24] Flexman et al[14] stated that frailty is an important predictor of major postoperative complications after correction surgery for degenerative spine conditions in 53,080 adults. Although frailty has been shown to be a predictor of both poor clinical outcomes and increased perioperative complications in ASD surgery, standardized approaches for measuring and assessing patient frailty remain challenging.[15,25–27] Recently, several assessment tools were reported, including the Canadian Study of Health and Aging Frailty Index, the Charlson Comorbidity Index, and the 11-item modified frailty index (mFI-11) (Supplement Table 1,[17,25,28–30] These measurement tools can pose some challenges in clinical practice, such as measurement subjectivity (i.e., grip strength, gait speed), neurological examinations, patient reported outcomes (PRO), a high resource requirement (i.e., biomarkers, radiologic sarcopenia measurements). The mFI-11 is a deficit accumulation model with a scoring system based on an individual's comorbidities, activities of daily living, and physical characteristics, using variables from the National Surgical Quality Improvement Program (NSQIP) dataset.[17] Although comorbidities and frailty can negatively influence the clinical outcomes and complication rates in spine surgery, the results are inconsistent.[14–16,18,19] Furthermore, the evaluation of comorbidities and physical function of patients often requires the acquisition of multiple PROs, physical and pulmonary function tests, and neurological examinations. Therefore, it can be difficult to comprehensively asses a patient's comorbidities and frailty in typical clinical settings.

The 5-item mFI (mFI-5) is a recently established index that was developed to address growing concerns about substantial missing data in the NSQIP when using the mFI-11.[31–33] The mFI-5 is an abbreviated 5-item index of functional status and comorbidities condensed from the mFI-11.[31–33] It is designed to be more suitable to apply in daily practice. However, the predictive power and usefulness of the mFI-5 have not been proven. Therefore, the goal of this study was to compare the mFI-5 to the conventional mFI-11 in terms of value and predictive ability for SAEs in ASD surgery.