Abstract
Aim
To use the Hospital Frailty Risk Score (HFRS) to investigate the impact of frailty on complication rates and healthcare resource utilization in patients who underwent endovascular treatment of ruptured intracranial aneurysms (IAs).
Methods
A retrospective cohort study was performed using the 2016–2019 National Inpatient Sample database. All adult patients (≥18 years) undergoing endovascular treatment for IAs after subarachnoid hemorrhage were identified using ICD-10-CM codes. Patients were categorized into frailty cohorts: low (HFRS <5), intermediate (HFRS 5–15) and high (HFRS >15). Patient demographics, adverse events, length of stay (LOS), discharge disposition, and total cost of admission were assessed. Multivariate logistic regression analysis was used to identify independent predictors of prolonged LOS, increased cost, and non-routine discharge.
Results
Of the 33 840 patients identified, 7940 (23.5%) were found to be low, 20 075 (59.3%) intermediate and 5825 (17.2%) high frailty by HFRS criteria. The rate of encountering any adverse event was significantly greater in the higher frailty cohorts (low: 59.9%; intermediate: 92.4%; high: 99.2%, p<0.001). There was a stepwise increase in mean LOS (low: 11.7±8.2 days; intermediate: 18.7±14.1 days; high: 26.6±20.1 days, p<0.001), mean total hospital cost (low: $62 888±37 757; intermediate: $99 670±63 446; high: $134 937±80 331, p<0.001), and non-routine discharge (low: 17.3%; intermediate: 44.4%; high: 69.4%, p<0.001) with increasing frailty. On multivariate regression analysis, a similar stepwise impact was found in prolonged LOS (intermediate: OR 2.38, p<0.001; high: OR 4.49, p<0.001)], total hospital cost (intermediate: OR 2.15, p<0.001; high: OR 3.62, p<0.001), and non-routine discharge (intermediate: OR 2.13, p<0.001; high: OR 4.17, p<0.001).
Conclusions
Our study found that greater frailty as defined by the HFRS was associated with increased complications, LOS, total costs, and non-routine discharge.
Keywords: aneurysm, complication, economics
Introduction
Ruptured intracranial aneurysms (IAs) account for a majority of non-traumatic subarachnoid hemorrhage (SAH).1 In recent years the majority of ruptured IAs have been treated using an endovascular approach. Despite continued advancements in this technology and neurocritical care, aneurysmal SAH remains a devastating disease with a mortality rate in excess of 50% and only 20–33% of survivors making a full recovery.2–10 Given these variable outcomes and the high resource utilization associated with inpatient care, the identification of those most at risk for poor outcomes is desirable.
Frailty is an evolving concept that reflects an individual’s increased vulnerability to poor health outcomes. Its utility as a predictive tool is increasingly recognized in general medicine and surgery. Its relevance to neurovascular disease remains poorly studied but of great potential interest. For example, it has been reported that 25–45% of patients with aneurysmal SAH are frail and may be at increased risk for poor outcomes.11–14 Operationally, there are various ways to define frailty without widespread consensus on a ‘best’ methodology. Two commonly used tools are the modified 5-item Frailty Index (mFI-5) and modified 11-item Frailty Index (mFI-11).15 16
More recently there have been efforts to systematically identify frailty using variables that are routinely collected in national databases.17 18 One example is the Hospital Frailty Risk Score (HFRS), which is based on a broad set of ICD-10 diagnostic codes over-represented in frailty including codes for volume depletion, chronic pulmonary disease and heart failure.17 An increasing number of studies have shown the utility of the HFRS to predict mortality and poor health outcomes across a range of medical conditions and surgical procedures including pneumonia, acute myocardial infarction, COVID-19 hospitalizations, and revision total hip and knee arthroplasty.15 19–22 Other recent studies have explored the HFRS as a predictive tool in stroke.23 To date, the HFRS has not been explored in aneurysmal SAH.
The aim of this study was to use the HFRS to investigate the impact of frailty on complication rates and healthcare resource utilization in patients who underwent endovascular treatment of ruptured IAs.
Methods
Data source and patient population
The Healthcare Cost and Utilization Project’s National Inpatient Sample (NIS) database is a stratified discharge database representing 20% of all inpatient admissions from community hospitals in the USA. It is the largest all-payer healthcare database in the USA, containing over 7 million hospital admissions (approximately 35 million hospitalizations, weighted) per year. A retrospective study was performed using years 2016–2019 of the NIS for all adult (≥18 years old) patients undergoing endovascular treatment for a ruptured IA. Institutional Review Board (IRB) exemption was obtained prior to initiation of the study.
Within the NIS, the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis and procedural coding system (PCS) was used to identify patients with ruptured aneurysm, as well as their previously diagnosed comorbidities and the surgical interventions they underwent. The ICD-10-CM codes I60.X, I61.X were used (see online supplemental table 1). These identified patients were then cross-matched using ICD-10-PCS codes for endovascular treatment (see online supplemental table 2).
neurintsurg-2021-018484supp001.pdf (209.7KB, pdf)
Hospital Frailty Risk Score (HFRS)
Frailty was defined using the HFRS, a validated ICD-10 coding algorithm developed by Gilbert et al17 and used in multiple studies.19–25 The coding algorithm is derived from a cluster of frail individuals with 109 over-represented ICD-10 codes assigned a number of points ranging from 0.1 to 7.1 reflecting the strength of association to the frail cluster.17 These points are added together for a final frailty risk score.17 All ICD-10 codes for the admission were entered into the calculation. In addition to presenting frailty as a continuous variable, we then stratified the frailty score into low (HFRS <5), intermediate (HFRS 5–15), and high (HFRS >15), as previously described.17 A flow chart demonstrating the criteria used to identify the cohorts is shown in online supplemental figure 1.
Data collection
Patient demographics such as age, sex, race, median household income quartile, and insurance provider were all collected from the NIS database. Additional hospital characteristics such as bed size volume, region (Northeast, South, Midwest, and West), and type (rural, urban teaching, and urban non-teaching) were also collected. Elixhauser comorbidities were used to evaluate the incidence of congestive heart failure (CHF), cardiac arrhythmia, valvular disease, hypertension, pulmonary circulation disorders, peripheral vascular disorders, paralysis, other neurological disorders, chronic pulmonary disease, diabetes, hypothyroidism, renal failure, liver disease, peptic ulcer disease excluding bleeding, coagulopathy, obesity, weight loss, fluid and electrolyte disorders, anemia (blood loss or deficiency), and alcohol abuse. In addition, patient admission comorbidities, affective disorders, smoking, and long-term aspirin or anticoagulant use were assessed (online supplemental table 1). Data on aneurysm location were collected including anterior (carotid siphon bifurcation, middle cerebral aneurysm, anterior communication) and posterior (posterior communicating, basilar, vertebral artery) location. Procedural data collected included ventriculoperitoneal shunt, central venous catheter, arterial monitoring, fluoroscopy of cerebral arteries with contrast, vasospasm treatment, video-electroencephalogram monitoring, and unspecified procedures such as external ventricular drain (online supplemental table 2).
Data regarding postoperative adverse events for each patient were collected by indexing additional diagnoses from the NIS database. Complications included in the analysis were urinary tract infection, post-procedural fever, pressure ulcer, sepsis, cardiogenic shock, cerebral infarction, syndrome of inappropriate antidiuretic hormone secretion, hypo-osmolar hyponatremia, pulmonary embolism, phlebitis, acute deep vein thrombosis, periprocedural stroke, acute myocardial infarction, acute respiratory failure, acute kidney injury, mechanical ventilation, tracheostomy, and gastrostomy (see online supplemental tables 1 and 2). In addition, postoperative outcome measures such as hospital length of stay (LOS), total cost of hospital admission, and discharge disposition were also assessed between the two cohorts. Discharge disposition was classified as routine (home, self-care, home healthcare), non-routine (patient sent to short-term hospital, skilled nursing facility, intermediate care facility), and other (leaving against medical advice, died in hospital, unknown destination). All-payer inpatient cost-to-charge ratios were used to convert total hospital charge to total cost of hospital services.
Statistical analysis
Discharge-level weights provided by the Agency for Healthcare Research and Quality were used to calculate national estimates. Parametric data were expressed as mean±SD and compared using a two-way independent t-test. Non-parametric data were expressed as median (IQR) and compared using the Mann–Whitney U test. Nominal data were compared with the χ2 test. For our primary hypothesis, weighted univariate and multivariate logistic regressions were fitted with extended postoperative hospital LOS (as defined by LOS greater than the 75th percentile for the entire cohort), increased hospital costs (as defined by total costs greater than the 75th percentile for the entire cohort), and non-routine discharge as the dependent variables. There were no patients with missing information on LOS. Patients with missing total cost variable (n=64) were excluded from this portion of the analysis. Patients with ‘other’ discharge were also excluded from this portion of the analysis in order to dichotomize routine versus non-routine discharge. Backward stepwise multivariate logistic regression analysis was used to select variables in the final model, using 0.1 as entry and stay criteria. Based on our primary aim and in view of the plausibility for confounding, age and female sex were forced into the models. A p value of <0.05 was deemed to be statistically significant. Statistical analysis was performed using R Studio Version 3.6.2 (RStudio Inc, Boston, Massachusetts, USA).
Results
Patient demographics and hospital characteristics
A total of 33 840 patients were included, of which 7940 (23.5%) were classified as low frailty, 20 075 (59.3%) as intermediate frailty, and 5825 (17.2%) as high frailty (table 1). Age increased significantly with increasing HFRS score (low: 54.1 years; intermediate: 57.5 years; high: 58.6 years, p<0.001) (table 1). Healthcare coverage also varied significantly between the frailty cohorts, with the low frailty cohort having a higher percentage of privately insured patients (low: 45.8%; intermediate: 36.7%; high: 35.5%) and patients in the high frailty cohort having a higher percentage of Medicare patients (low: 21.3%; intermediate: 32.8%; high: 35.7%, p<0.001) (table 1). While treating hospital region varied significantly between cohorts (p<0.001), race, household income quartile, weekend admission rate, and hospital bed size and type were all similar among the frailty cohorts (table 1).
Table 1.
Patient demographics and hospital characteristics
| Variables | Low (n=7940) |
Intermediate (n=20 075) |
High (n=5825) |
P value |
| Age (years) | ||||
| Mean±SD | 54.1±13.6 | 57.5±14.2 | 58.6±13.7 | <0.001 |
| Female (%) | 63.0 | 67.2 | 69.9 | <0.001 |
| Race (%) | 0.386 | |||
| White | 56.8 | 57.9 | 55.1 | |
| Black | 18.9 | 17.4 | 17.6 | |
| Hispanic | 13.7 | 13.8 | 16.1 | |
| Other | 10.5 | 10.9 | 11.2 | |
| Income quartile (%) | 0.158 | |||
| 0–25th | 30.5 | 29.7 | 30.8 | |
| 26–50th | 26.7 | 24.1 | 23.7 | |
| 51–75th | 22.3 | 25.7 | 25.2 | |
| 76–100th | 20.4 | 20.5 | 20.3 | |
| Admitted during weekend (%) | 24.2 | 28.3 | 28.7 | 0.005 |
| Healthcare coverage (%) | <0.001 | |||
| Medicare | 21.3 | 32.8 | 35.7 | |
| Medicaid | 18.8 | 20.0 | 19.6 | |
| Private insurance | 45.8 | 36.7 | 35.5 | |
| Other | 14.1 | 10.5 | 9.2 | |
| Hospital demographics | ||||
| Hospital region (%) | <0.001 | |||
| Northeast | 17.8 | 19.4 | 15.1 | |
| Midwest | 17.5 | 18.3 | 22.9 | |
| South | 42.8 | 40.0 | 36.7 | |
| West | 21.9 | 22.3 | 25.3 | |
| Hospital bed size (%) | 0.626 | |||
| Small | 5.0 | 4.9 | 5.4 | |
| Medium | 18.1 | 16.7 | 18.3 | |
| Large | 77.0 | 78.4 | 76.3 | |
| Hospital type (%) | 0.618 | |||
| Rural | 0.3 | 0.1 | 0.3 | |
| Urban non-teaching | 6.4 | 6.3 | 5.9 | |
| Urban teaching | 93.3 | 93.5 | 93.8 | |
Patient comorbidities, aneurysm locations and other surgical/intensive care procedures
The average HFRS score for each cohort was 2.7±1.6, 9.7±2.8, 18.7±3.0 for the low, intermediate, and high frailty cohorts, respectively (table 2). On comparison of the admission comorbidities and clinical presentation between frailty cohorts, seizure (p<0.001), neurological deficit associated with SAH (p=0.002), coma (p<0.001), aphasia (p<0.001), dysphagia (p<0.001), and hypertensive chronic kidney disease (p<0.001) were notably all significantly increased with increasing frailty score (table 2). IA location also varied across the frailty cohorts, with the greatest proportion of IAs located in the anterior communicating artery (low: 26.9%; intermediate: 27.4%; high: 27.8%, p<0.001) (table 2). In general, significantly more procedures were done as the frailty score increased across the cohorts (table 2). For example, the proportion of patients who underwent external ventricular drainage (low: 25.3%; intermediate: 51.8%; high: 60.3%, p<0.001) and subsequent ventriculoperitoneal shunt placement (low: 3.6%; intermediate: 10.6%; high: 15.8%, p<0.001) increased along with the HFRS score (table 2).
Table 2.
Patient comorbidities, aneurysm location, and surgical/intensive care procedures
| Variables | Low (n=7940) |
Intermediate (n=20 075) |
High (n=5825) |
P value |
| Frailty risk score | ||||
| Mean±SD | 2.7±1.6 | 9.7±2.8 | 18.7±3.0 | <0.001 |
| Comorbidities (%) | ||||
| Seizure | 1.3 | 8.4 | 16.2 | <0.001 |
| Neurological deficit associated with subarachnoid hemorrhage | 0.1 | 0.6 | 1.0 | 0.002 |
| Coma | 0.9 | 5.5 | 7.9 | <0.001 |
| Aphasia | 1.8 | 9.4 | 21.8 | <0.001 |
| Dysphagia | 2.6 | 15.7 | 34.9 | <0.001 |
| Cranial nerve palsy | 3.6 | 4.0 | 4.5 | 0.527 |
| Diplopia | 2.1 | 2.3 | 1.8 | 0.581 |
| Hypertensive chronic kidney disease | 0.4 | 3.6 | 6.2 | <0.001 |
| Personal history of transient ischemic attack | 3.4 | 3.9 | 3.4 | 0.633 |
| Headache | 4.8 | 3.2 | 3.7 | 0.010 |
| Affective disorder | 12.8 | 16.3 | 17.1 | 0.002 |
| Smoking | 24.8 | 23.5 | 20.3 | 0.018 |
| Congestive heart failure | 3.1 | 8.1 | 11.1 | <0.001 |
| Cardiac arrhythmias | 12.7 | 23.0 | 30.9 | <0.001 |
| Valvular disease | 1.3 | 3.4 | 3.9 | <0.001 |
| Pulmonary circulation disorders | 1.1 | 3.3 | 4.3 | <0.001 |
| Peripheral vascular disorders | 4.7 | 5.9 | 6.1 | 0.156 |
| Hypertension, combined | 63.5 | 71.2 | 73.6 | <0.001 |
| Paralysis | 0.9 | 15.6 | 53.6 | <0.001 |
| Other neurological disorders | 14.2 | 43.9 | 64.0 | <0.001 |
| Chronic pulmonary disease | 10.1 | 14.4 | 13.2 | <0.001 |
| Diabetes, uncomplicated | 7.5 | 6.8 | 6.0 | 0.321 |
| Renal failure | 1.0 | 5.0 | 8.5 | <0.001 |
| Hypothyroidism | 6.4 | 8.1 | 7.6 | 0.076 |
| Liver disease | 1.8 | 3.5 | 3.0 | 0.003 |
| Peptic ulcer disease | 0.2 | 0.3 | 1.4 | 0.017 |
| Coagulopathy | 4.3 | 7.6 | 8.3 | <0.001 |
| Obesity | 12.5 | 14.6 | 12.6 | 0.074 |
| Weight loss | 2.7 | 9.4 | 16.3 | <0.001 |
| Fluid and electrolyte disorders | 33.2 | 70.0 | 86.6 | <0.001 |
| Deficiency anemias | 2.1 | 2.9 | 4.1 | 0.009 |
| Alcohol abuse | 4.1 | 6.9 | 8.3 | <0.001 |
| Current aspirin use | 10.5 | 10.5 | 9.4 | 0.554 |
| Current anticoagulant use | 2.3 | 2.9 | 2.3 | 0.301 |
| Aneurysm location (%) | <0.001 | |||
| Anterior | ||||
| Carotid siphon bifurcation | 4.1 | 3.0 | 3.3 | |
| Middle cerebral artery | 6.2 | 7.3 | 10.9 | |
| Anterior communicating artery | 26.9 | 27.4 | 27.8 | |
| Posterior | ||||
| Posterior communicating artery | 21.4 | 19.6 | 19.9 | |
| Basilar artery | 5.5 | 6.5 | 4.3 | |
| Vertebral artery | 0.8 | 1.9 | 2.3 | |
| Unspecified | 35.1 | 34.2 | 31.5 | |
| Procedures (%) | ||||
| External ventricular drain | 25.3 | 51.8 | 60.3 | <0.001 |
| Ventriculo-peritoneal shunt | 3.6 | 10.6 | 15.8 | <0.001 |
| Central venous catheter placement | 15.2 | 33.5 | 45.6 | <0.001 |
| Arterial monitoring | 6.4 | 10.6 | 13.6 | <0.001 |
| Fluoroscopy of cerebral arteries with contrast | 52.8 | 55.7 | 63.3 | <0.001 |
| Vasospasm treatment | 2.5 | 6.3 | 10.7 | <0.001 |
| Video-EEG monitoring | 1.4 | 6.0 | 10.0 | <0.001 |
EEG, electroencephalogram
Adverse events and postoperative outcomes
Overall, the proportion of patients who experienced any adverse event (AE) significantly increased with increasing frailty (low: 59.9%; intermediate: 92.4%; high: 99.2%, p<0.001) (table 3). With regard to specific AEs, the most common were vasospasm (low: 13.0%; intermediate: 37.9%; high: 55.3%, p<0.001), hydrocephalus (low: 34.1%; intermediate: 60.5%; high: 69.4%, p<0.001), cerebral infarction (low: 2.2%; intermediate: 11.5%; high: 22.4%, p<0.001), acute respiratory failure (low: 14.2%; intermediate: 47.3%; high: 66.4%, p<0.001), and mechanical ventilation (low: 16.2%; intermediate: 47.5%; high: 66.1%, p<0.001) (table 3). All adverse events increased in prevalence as the frailty score increased (table 3). Postoperative inpatient outcomes varied significantly among the frailty cohorts. Mean LOS (low: 11.7±8.2; intermediate: 18.7±14.1; high: 26.6±20.1, p<0.001), total cost of admission (low: $62 888±37 757; intermediate: $99 670±63 446; high: $134 937±80 331, p<0.001), and rate of non-routine discharge (low: 17.3%; intermediate: 44.4%; high: 69.4%, p<0.001) were all significantly increased with increasing frailty (table 3).
Table 3.
Adverse events and postoperative outcomes
| Variables | Low (n=7940) |
Intermediate (n=20 075) |
High (n=5825) |
P value |
| Adverse events (%) | ||||
| Vasospasm | 13.0 | 37.9 | 55.3 | <0.001 |
| Hydrocephalus | 34.1 | 60.5 | 69.4 | <0.001 |
| Urinary tract infection | 1.3 | 16.4 | 46.5 | <0.001 |
| Post-procedural fever | 0.5 | 0.2 | 0.5 | 0.108 |
| Pressure ulcer | 0.0 | 1.4 | 4.0 | <0.001 |
| Sepsis | 0.4 | 6.6 | 13.8 | <0.001 |
| Cardiogenic shock | 0.3 | 1.4 | 1.5 | 0.002 |
| Cerebral infarction | 2.2 | 11.5 | 22.4 | <0.001 |
| Syndrome of inappropriate antidiuretic hormone secretion | 3.5 | 4.2 | 4.1 | 0.421 |
| Hypo-osmolar hyponatremia | 15.9 | 30.5 | 33.4 | <0.001 |
| Pulmonary embolism | 0.4 | 1.8 | 2.6 | <0.001 |
| Phlebitis | 0.3 | 0.8 | 1.1 | 0.045 |
| Acute deep vein thrombosis | 2.0 | 7.8 | 13.6 | <0.001 |
| Periprocedural stroke | 0.4 | 1.6 | 1.7 | 0.001 |
| Acute myocardial infarction | 1.3 | 3.2 | 3.8 | <0.001 |
| Acute respiratory failure | 14.2 | 47.3 | 66.4 | <0.001 |
| Acute kidney injury | 1.5 | 10.2 | 17.9 | <0.001 |
| Mechanical ventilation | 16.2 | 47.5 | 66.1 | <0.001 |
| Tracheostomy | 2.1 | 11.4 | 22.3 | <0.001 |
| Gastrostomy | 3.0 | 14.1 | 31.0 | <0.001 |
| Any adverse event | 59.9 | 92.4 | 99.2 | <0.001 |
| Number of adverse events | <0.001 | |||
| 0 | 48.7 | 14.4 | 3.3 | |
| 1 | 27.5 | 18.0 | 9.2 | |
| >1 | 23.8 | 67.6 | 87.6 | |
| Mortality (%) | 6.4 | 15.3 | 10.8 | <0.001 |
| Length of stay (days) | ||||
| Mean±SD | 11.7±8.2 | 18.7±14.1 | 26.6±20.1 | <0.001 |
| Median (IQR) | 11 (7–15) | 16 (11–23) | 22 (16–31) | <0.001 |
| Total cost of admission ($) | ||||
| Mean±SD | 62 888±37 757 | 99 670±63 446 | 134 937±80 331 | <0.001 |
| Median (IQR) | 55 247 (39 050–75 934) | 83 358 (56 981–122 835) | 112 881 (82 426–164 897) | <0.001 |
| Disposition (%) | <0.001 | |||
| Routine | 75.1 | 39.8 | 19.6 | |
| Non-routine | 17.3 | 44.4 | 69.4 | |
| Other | 7.6 | 15.7 | 11.1 |
Logistic multivariate regression analyses
Increasing frailty was significantly associated with greater odds of experiencing a prolonged LOS (intermediate: OR 2.38 (95% CI 1.84 to 3.09), p<0.001; high: OR 4.49 (95% CI 3.36 to 6.00), p<0.001) (table 4). Similarly, increasing frailty was significantly associated with an increased cost of hospital admission (intermediate: OR 2.15 (95% CI 1.66 to 2.77), p<0.001; high: OR 3.62 (95% CI 2.71 to 4.83), p<0.001) (table 4). Finally, the odds of experiencing a non-routine discharge were found to increase with advancing HFRS (intermediate: OR 2.13 (95% CI 1.79 to 2.55), p<0.001; high: OR 4.17 (95% CI 3.26 to 5.33), p<0.001) (table 4). The complete univariate and multivariate data are available in online supplemental tables 3–5.
Table 4.
Logistic multivariate regression analyses on prolonged LOS, increased cost, and non-routine discharge disposition
| Variables | Prolonged LOS | Increased cost | Non-routine discharge | |||
| Frailty risk category | OR (95% CI) | P value | OR (95% CI) | P value | OR (95% CI) | P value |
| Low | Reference | |||||
| Intermediate | 2.38 (1.84 to 3.09) | <0.001 | 2.15 (1.66 to 2.77) | <0.001 | 2.13 (1.79 to 2.55) | <0.001 |
| High | 4.49 (3.36 to 6.00) | <0.001 | 3.62 (2.71 to 4.83) | <0.001 | 4.17 (3.26 to 5.33) | <0.001 |
| Age | 0.99 (0.99 to 0.99) | 0.038 | 0.99 (0.98 to 0.99) | 0.003 | 1.04 (1.03 to 1.05) | <0.001 |
| Female sex | 0.91 (0.79 to 1.04) | 0.172 | 0.86 (0.74 to 0.99) | 0.035 | 0.76 (0.65 to 0.88) | <0.001 |
| Aphasia | 1.63 (1.32 to 2.01) | <0.001 | 1.35 (1.09 to 1.66) | 0.005 | 1.68 (1.35 to 2.11) | <0.001 |
| Dysphagia | 3.00 (2.55 to 3.52) | <0.001 | 2.04 (1.72 to 2.41) | <0.001 | 3.95 (3.20 to 4.87) | <0.001 |
| Headache | 0.72 (0.49 to 1.07) | 0.101 | 0.51 (0.32 to 0.83) | 0.006 | 0.50 (0.34 to 0.73) | <0.001 |
| Aneurysm location | ||||||
| Anterior | Reference | |||||
| Posterior | 1.02 (0.87 to 1.20) | 0.824 | 1.05 (0.89 to 1.23) | 0.587 | 0.85 (0.72 to 0.99) | 0.041 |
| Unspecified | 1.00 (0.85 to 1.17) | 0.954 | 1.10 (0.93 to 1.30) | 0.254 | 1.30 (1.11 to 1.53) | <0.001 |
| Number of adverse events | ||||||
| 0 | Reference | |||||
| 1 | 4.20 (2.61 to 6.76) | <0.001 | 2.81 (1.87 to 4.24) | <0.001 | 1.89 (1.52 to 2.36) | <0.001 |
| >1 | 19.75 (12.73 to 30.65) | <0.001 | 14.40 (9.99 to 20.73) | <0.001 | 6.74 (5.55 to 8.18) | <0.001 |
LOS, length of stay.
Discussion
This retrospective national database study of 33 840 patients who underwent endovascular treatment of a ruptured IA shows that increasing frailty (as measured by HFRS) was associated with increased LOS, rates of hospital AEs, rates of non-routine discharges, and total costs. Further, on multivariate analysis, we found that intermediate and high frailty independently predicted prolonged LOS, increased cost, and non-routine discharge.
Frailty is increasingly recognized as a predisposing factor for adverse health outcomes and increased healthcare resource utilization.26 27 Preliminary efforts have begun to assess the impact of frailty on patients with aneurysmal SAH. In a retrospective analysis of 217 patients who underwent surgical clipping of ruptured IAs, McIntyre et al found that ~26% of patients were frail, defined as mFI score ≥2.11 Similarly, in a single-institution study of 173 elderly patients (≥60 years) who underwent surgical clipping of a ruptured IA, Yue et al found frailty, as defined by the presence of severe anemia, hypoalbuminemia, or low body mass index, to be common affecting ~45% of their patient cohort.14 In our study, nearly 75% of patients undergoing endovascular treatment for a ruptured IA were considered to have intermediate (59.3%) or high (17.2%) frailty, as defined by the HFRS.
While there is a paucity of studies directly assessing frailty and outcomes, prior studies have attempted to identify the relationship between frailty and complications after management of ruptured IAs. In the retrospective study of 217 patients with aSAH by McIntyre et al, the authors found on multivariate regression analysis that frail patients were significantly more likely to experience a complication during their hospital stay (OR 2.6).11 In a retrospective NIS database study of 5353 patients undergoing treatment for ruptured aneurysms between 2012 and 2015, Chotai et al found on multivariate analysis that the odds of experiencing any inpatient complication significantly increased as the neurovascular comorbidity index (NCI) score, a value closely related to frailty, progressed from 1 (OR 1.13) to 4 (OR 1.58) and to 7 (OR 2.05).28 Similarly, our study of patients undergoing endovascular treatment of ruptured IAs found that, as the severity of frailty increased, the rate of perioperative AEs increased by more than fivefold from the low to the high frailty cohort. Further studies assessing frailty in ruptured IAs are needed to aid in patient risk stratification and patient-centered management strategies.
While there is a paucity of studies directly assessing the impact of frailty with hospital LOS and non-routine discharges in patients with ruptured IA, there have been a few that evaluated the impact of comorbidities and other factors closely associated with frailty. In a retrospective NIS database study of 19 034 patients who underwent clipping or coiling of a ruptured or unruptured IA between 2002 and 2006, Hoh et al demonstrated that older age as well as diabetes, coagulopathy, CHF, renal failure, peripheral vascular disorders, and neurological or electrolyte disorders were each independently associated with longer hospitalizations.29 Of these identified comorbidities, four are included in the commonly used 11-item mFI,30 three are included in the 5-item mFI,31 and five are included in the HFRS.17
However, not all studies have reported results similar to the findings outlined above. For example, in a retrospective observational database study of 203 patients treated for a ruptured IA between 2012 and 2017, Hammer et al found that all analyzed comorbidities including hypertension, diabetes, hypothyroidism, cholesterinemia, and smoking did not have a significant impact on LOS in the intensive care unit.32 Analogous to the aforementioned studies, our study found that mean hospital LOS more than doubled when comparing low to high HFRS cohorts. Furthermore, on multivariate analysis, we also found that both intermediate and high frailty were significant independent predictors of prolonged LOS when compared with patients with low baseline frailty. With respect to discharge disposition, in the study by McIntyre et al the authors showed on multivariate analysis that frail patients were significantly less likely to be discharged to home (OR 0.32).11 Our study also found that rates of non-routine discharge were nearly four times higher in patients with high frailty than in those with low frailty. As prolonged LOS and non-routine discharges have repeatedly been shown to increase healthcare resource utilization and be associated with inferior outcomes in patients undergoing various neurosurgical procedures, further studies should be done to better define the relationship between frailty and LOS following treatment of ruptured IAs.
While few studies have examined the impact of frailty on the cost of hospital admission in patients undergoing treatment for IAs, there have been efforts to characterize the effect of frailty-associated factors on hospital admission cost within this patient population. In the NIS database study of patients treated for a ruptured IA by Chotai et al, the authors found on regression analysis that higher NCI scores were associated with a greater cost of hospital admission.28 Consistent with these findings, the NIS database study of patients with ruptured or unruptured IA by Hoh et al demonstrated that frailty-associated comorbidities (eg, diabetes, CHF, coagulopathy, electrolyte disorders, neurological disorders, renal failure, and peripheral vascular disorders) and older age were each independently associated with greater hospital costs.29 Similarly, our study found that the mean cost of hospital admission more than doubled between the low frailty and the high frailty cohort. These data, although limited, demonstrate how frailty may be contributing to the soaring healthcare costs in the USA.
This study has a number of limitations inherent to all administrative databases, including the NIS. First, the analysis is retrospective, potentially limiting the interpretation of our results. The data are also available only by ICD-10-CM codes, which may contain coding and reporting biases. Second, data may be misclassified or incomplete. Third, information regarding preoperative factors such as the size of the IA prior to rupture is unavailable, which may have an unrecognized effect on our results. Also, there may be additional confounding variables present which are unavailable or have not been measured. Finally, given the NIS has information specific to a single inpatient admission, we cannot comment on long-term functional outcomes or the durability of treatment. Despite these limitations, this study uses one of the largest inpatient databases in the USA to provide unique insights into the impact of frailty on postoperative outcomes and healthcare resource utilization for patients undergoing endovascular treatment of ruptured IAs.
Conclusion
Our study is the first to use the HFRS to assess the impact of frailty on patients who underwent endovascular treatment for ruptured IAs. We found that greater frailty was associated with increased complications, prolonged hospital LOS, higher total costs, and non-routine discharge. These results highlight the need for further studies that examine the impact of frailty on outcomes following endovascular treatment for a ruptured IA, as it may improve patient care and reduce healthcare resource utilization.
neurintsurg-2021-018484supp002.pdf (400.3KB, pdf)
Footnotes
Twitter: @AjayMalhotraRad
Contributors: ABK: Study conception and design, acquisition of data, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, statistical analysis. AAE: Study conception and design, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content. DR, MS, JS, BCR, JH, JA, NS: Acquisition of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content. RH: Study conception and design, critical revision of the manuscript for important intellectual content, study supervision. AM: Study conception and design, critical revision of the manuscript for important intellectual content, study supervision. CM: Study conception and design, critical revision of the manuscript for important intellectual content, study supervision, guarantor of the study.
Funding: The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests: None declared.
Provenance and peer review: Not commissioned; externally peer reviewed.
Supplemental material: This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplemental information. Not applicable.
Ethics statements
Patient consent for publication
Not applicable.
References
- 1.Muehlschlegel S. Subarachnoid hemorrhage. Continuum 2018;24:1623–57. 10.1212/CON.0000000000000679 [DOI] [PubMed] [Google Scholar]
- 2.Broderick J, Connolly S, Feldmann E, et al. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Stroke 2007;38:2001–23. 10.1161/STROKEAHA.107.183689 [DOI] [PubMed] [Google Scholar]
- 3.Broderick JP, Brott T, Tomsick T, et al. Intracerebral hemorrhage more than twice as common as subarachnoid hemorrhage. J Neurosurg 1993;78:188–91. 10.3171/jns.1993.78.2.0188 [DOI] [PubMed] [Google Scholar]
- 4.Brown RD, Broderick JP. Unruptured intracranial aneurysms: epidemiology, natural history, management options, and familial screening. Lancet Neurol 2014;13:393–404. 10.1016/S1474-4422(14)70015-8 [DOI] [PubMed] [Google Scholar]
- 5.Hop JW, Rinkel GJ, Algra A, et al. Case-fatality rates and functional outcome after subarachnoid hemorrhage: a systematic review. Stroke 1997;28:660–4. 10.1161/01.STR.28.3.660 [DOI] [PubMed] [Google Scholar]
- 6.Karimy JK, Reeves BC, Kahle KT. Targeting TLR4-dependent inflammation in post-hemorrhagic brain injury. Expert Opin Ther Targets 2020;24:525–33. 10.1080/14728222.2020.1752182 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Long B, Koyfman A, Runyon MS. Subarachnoid hemorrhage: updates in diagnosis and management. Emerg Med Clin North Am 2017;35:803–24. 10.1016/j.emc.2017.07.001 [DOI] [PubMed] [Google Scholar]
- 8.Sarti C, Tuomilehto J, Salomaa V, et al. Epidemiology of subarachnoid hemorrhage in Finland from 1983 to 1985. Stroke 1991;22:848–53. 10.1161/01.STR.22.7.848 [DOI] [PubMed] [Google Scholar]
- 9.Vlak MH, Algra A, Brandenburg R, et al. Prevalence of unruptured intracranial aneurysms, with emphasis on sex, age, comorbidity, country, and time period: a systematic review and meta-analysis. Lancet Neurol 2011;10:626–36. 10.1016/S1474-4422(11)70109-0 [DOI] [PubMed] [Google Scholar]
- 10.Williams LN, Brown RD. Management of unruptured intracranial aneurysms. Neurol Clin Pract 2013;3:99–108. 10.1212/CPJ.0b013e31828d9f6b [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.McIntyre MK, Gandhi C, Long A, et al. Age predicts outcomes better than frailty following aneurysmal subarachnoid hemorrhage: a retrospective cohort analysis. Clin Neurol Neurosurg 2019;187:105558. 10.1016/j.clineuro.2019.105558 [DOI] [PubMed] [Google Scholar]
- 12.Panayi AC, Orkaby AR, Sakthivel D, et al. Impact of frailty on outcomes in surgical patients: a systematic review and meta-analysis. Am J Surg 2019;218:393–400. 10.1016/j.amjsurg.2018.11.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Vermeiren S, Vella-Azzopardi R, Beckwée D, et al. Frailty and the prediction of negative health outcomes: a meta-analysis. J Am Med Dir Assoc 2016;17:1163.e1–17. 10.1016/j.jamda.2016.09.010 [DOI] [PubMed] [Google Scholar]
- 14.Yue Q, Liu Y, Leng B, et al. A prognostic model for early post-treatment outcome of elderly patients with aneurysmal subarachnoid hemorrhage. World Neurosurg 2016;95:253–61. 10.1016/j.wneu.2016.08.020 [DOI] [PubMed] [Google Scholar]
- 15.McAlister F, van Walraven C. External validation of the Hospital Frailty Risk Score and comparison with the Hospital-patient One-year Mortality Risk Score to predict outcomes in elderly hospitalised patients: a retrospective cohort study. BMJ Qual Saf 2019;28:284–8. 10.1136/bmjqs-2018-008661 [DOI] [PubMed] [Google Scholar]
- 16.Shebeshi DS, Dolja-Gore X, Byles J. Validation of hospital frailty risk score to predict hospital use in older people: evidence from the Australian Longitudinal Study on Women’s Health. Arch Gerontol Geriatr 2021;92:104282. 10.1016/j.archger.2020.104282 [DOI] [PubMed] [Google Scholar]
- 17.Gilbert T, Neuburger J, Kraindler J, et al. Development and validation of a Hospital Frailty Risk Score focusing on older people in acute care settings using electronic hospital records: an observational study. Lancet 2018;391:1775–82. 10.1016/S0140-6736(18)30668-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Theou O, Brothers TD, Mitnitski A, et al. Operationalization of frailty using eight commonly used scales and comparison of their ability to predict all-cause mortality. J Am Geriatr Soc 2013;61:1537–51. 10.1111/jgs.12420 [DOI] [PubMed] [Google Scholar]
- 19.Gunnarsdottir GM, Helgadottir S, Einarsson SG, et al. Validation of the hospital frailty risk score in older surgical patients: a population-based retrospective cohort study. Acta Anaesthesiol Scand 2021;65:1033–42. 10.1111/aas.13837 [DOI] [PubMed] [Google Scholar]
- 20.Kwok CS, Zieroth S, Van Spall HGC, et al. The Hospital Frailty Risk Score and its association with in-hospital mortality, cost, length of stay and discharge location in patients with heart failure. Int J Cardiol 2020;300:184–90. 10.1016/j.ijcard.2019.09.064 [DOI] [PubMed] [Google Scholar]
- 21.Meyer M, Parik L, Leiß F, et al. Hospital Frailty Risk Score predicts adverse events in primary total hip and knee arthroplasty. J Arthroplasty 2020;35:3498–504. 10.1016/j.arth.2020.06.087 [DOI] [PubMed] [Google Scholar]
- 22.Ramos-Rincon J-M, Moreno-Perez O, Pinargote-Celorio H, et al. Clinical Frailty Score vs Hospital Frailty Risk Score for predicting mortality and other adverse outcome in hospitalised patients with COVID-19: Spanish case series. Int J Clin Pract 2021;75:e14599. 10.1111/ijcp.14599 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Kilkenny MF, Phan HT, Lindley RI, et al. Utility of the hospital frailty risk score derived from administrative data and the association with stroke outcomes. Stroke 2021;52:2874–81. 10.1161/STROKEAHA.120.033648 [DOI] [PubMed] [Google Scholar]
- 24.Acosta CJ, Goldberg D, Amin S. Evaluating the impact of frailty on periprocedural adverse events and mortality among patients with Gi bleeding. Gastrointest Endosc 2021;94:517–25. 10.1016/j.gie.2021.03.021 [DOI] [PubMed] [Google Scholar]
- 25.Aitken SJ, Lujic S, Randall DA, et al. Predicting outcomes in older patients undergoing vascular surgery using the hospital frailty risk score. Br J Surg 2020. doi: 10.1002/bjs.12043. [Epub ahead of print: 09 Oct 2020]. [DOI] [PubMed] [Google Scholar]
- 26.Clegg A, Young J, Iliffe S, et al. Frailty in elderly people. Lancet 2013;381:752–62. 10.1016/S0140-6736(12)62167-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.van den Berg MEL, Castellote JM, Mahillo-Fernandez I, et al. Incidence of spinal cord injury worldwide: a systematic review. Neuroepidemiology 2010;34:184–92. discussion 92. 10.1159/000279335 [DOI] [PubMed] [Google Scholar]
- 28.Chotai S, Patel PD, Liles C, et al. Impact of neurovascular comorbidities and complications on outcomes after procedural management of intracranial aneurysm: Part 2, ruptured intracranial aneurysm. World Neurosurg 2021;146:e270–312. 10.1016/j.wneu.2020.10.091 [DOI] [PubMed] [Google Scholar]
- 29.Hoh BL, Chi Y-Y, Lawson MF, et al. Length of stay and total hospital charges of clipping versus coiling for ruptured and unruptured adult cerebral aneurysms in the nationwide inpatient sample database 2002 to 2006. Stroke 2010;41:337–42. 10.1161/STROKEAHA.109.569269 [DOI] [PubMed] [Google Scholar]
- 30.Tracy BM, Adams MA, Schenker ML, et al. The 5 and 11 factor modified frailty indices are equally effective at outcome prediction using TQIP. J Surg Res 2020;255:456–62. 10.1016/j.jss.2020.05.090 [DOI] [PubMed] [Google Scholar]
- 31.Weaver DJ, Malik AT, Jain N, et al. The modified 5-Item Frailty Index: a concise and useful tool for assessing the impact of frailty on postoperative morbidity following elective posterior lumbar fusions. World Neurosurg 2019. doi: 10.1016/j.wneu.2018.12.168. [Epub ahead of print: 11 Jan 2019]. [DOI] [PubMed] [Google Scholar]
- 32.Hammer A, Ranaie G, Erbguth F, et al. Impact of complications and comorbidities on the intensive care length of stay after aneurysmal subarachnoid haemorrhage. Sci Rep 2020;10:6228. 10.1038/s41598-020-63298-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
neurintsurg-2021-018484supp001.pdf (209.7KB, pdf)
neurintsurg-2021-018484supp002.pdf (400.3KB, pdf)
Data Availability Statement
All data relevant to the study are included in the article or uploaded as supplemental information. Not applicable.