Colorectal cancer is the most common cancer in Japan. It is the third leading cause of cancer-related deaths in men and the leading cause of cancer-related deaths in women. Surgical resection is an established curative treatment option for colorectal cancer that contributes to improved prognosis and quality of life (1,2). With the rapidly aging global population, the rate of surgeries performed on elderly colorectal cancer patients is also growing incrementally.

The benefits of laparoscopic colorectal resections over open surgery have been well described in the general population, regardless of age (3,4). In addition, elective laparoscopic resection for elderly colorectal cancer patients has been demonstrated as feasible and safe (5). Laparoscopic colorectal resections in older adults have thus become increasingly common; however, there is a direct correlation between age and risk of morbidity and mortality (6,7). Elderly patients have accompanying co-morbidities, such as cardiovascular and pulmonary issues, that place them at a significantly higher risk of postoperative mortality, morbidity, and a longer hospital stay after surgery (8,9). Therefore, it is important to assess the level of physical independence of frail elderly patients.

Frailty is defined as an objective measure of increased vulnerability and decreased physiological reserves resulting from the age-associated accumulation of physiological deficits in multiple systems.

To assess frailty, various frailty predictive models have been developed, such as the Physiological and Operative Scoring System for Enumeration of Morbidity and Mortality, Estimation of Physiologic Ability and Surgical Stress Score, and Comprehensive Geriatric Assessment; however, these models are often complex or time-consuming (10-12). The Canadian Study of Health and Aging introduced a 70-item frailty index (CSHA-FI) based on the accumulated deficits (symptoms, signs, functional impairments, and laboratory abnormalities) (13). Subsequently, a modified frailty index (mFI) was derived by matching the CSHA-FI to the 11 variables collected by the American College of Surgeons National Surgical Quality Improvement Program (14). This 11-item mFI (mFI-11) has been proven to reflect frailty adequately and is a valid predictor of postoperative outcomes in several surgical populations (15-17). In 2015, an even simpler and more time-efficient five-item mFI (mFI-5) was developed. The mFI-5 has proven equally as predictive of 30-day outcomes as the mFI-11 (18-20). However, to our knowledge, the predictive ability of the mFI-5 has not been shown in a cohort consisting of elderly patients who have undergone laparoscopic colorectal surgery exclusively. Our study aimed to clarify the accuracy of the mFI-5 score on predicting postoperative outcomes in elderly patients undergoing laparoscopic colorectal surgery.

Patient population. A total of 107 patients, aged over 80 years, who underwent laparoscopic surgery for colorectal cancer at Nagasaki University Hospital between 2011 and 2018 were included in this study. Data were collected retrospectively from their hospital charts. This study was conducted in accordance with the 1975 Declaration of Helsinki and was approved by the clinical research ethics committee of Nagasaki University Hospital (approval number: 19102141). Acquisition of informed consent from patients was waived owing to the retrospective nature of this study.

Five-item modified frailty index. The mFI-5 was assessed based on the following co-morbidities: congestive heart failure; diabetes mellitus; chronic obstructive pulmonary disease, or diagnosis of pneumonia within 30 days; functional status (partially dependent or completely dependent); and hypertension requiring medication (Table I). The mFI-5 score was calculated by adding the number of frailty variables present (scored 1 point per variable), resulting in a score range from 0 to 5 points. The patients were grouped into three categories corresponding to the mFI-5 scores of 0, 1, and 2+.

Perioperative variables and outcomes. Patient demographics and perioperative and postoperative outcomes, including age, sex, body mass index (BMI), American Society of Anesthesiology (ASA) classification, functional status, tumor location, surgical approach, operation time, bleeding, reoperation, duration of hospital stay, discharge status, and postoperative complications, were collected and compared between the three mFI-5 groups. Postoperative complications were evaluated according to the Clavien–Dindo classification (21).

Statistical analysis. Data were collected as medians for continuous variables and frequencies for categorical variables. Patient demographics and perioperative and postoperative outcomes across the mFI-5 groups were compared using univariate analyses. Categorical variables were compared using the chi-square test, and continuous variables were compared using the Mann–Whitney U-test. Statistical significance was defined as a p-value <0.05. All statistical analyses were performed using the SPSS statistical software (version 22.0; SPSS Inc., Chicago, IL, USA).

Patient demographics. The results of the demographic analysis are presented in Table II. The median age was 83 years (range=80-99 years), and the median BMI was 21.4 (range=14.9-31.9) kg/m². Males represented 44.9% of the patients in the sample. Of the enrolled patients, 29.0% were diagnosed with rectal cancer, while 71.0% were diagnosed with colon cancer. The pathological stages of the 107 patients were as follows: pStage 0, 4.7%; pStage I, 22.4%; pStage II, 34.6%; pStage III, 30.8%; and pStage IV, 6.5%. Three patients (2.8%) underwent emergency colorectal surgery, and in nine cases (8.4%), it was necessary to convert from laparoscopic to open surgery.

Univariate analysis of demographics and perioperative factors by mFI-5 group. Of the 107 patients, 33.6% were classified as mFI-5=0; 41.1% were classified as mFI-5=1; and 25.3% were classified as mFI-5=2+. We compared the three groups’ demographics and perioperative factors (Table III). There were no significant differences in age, sex, BMI, or tumor location. ASA classification correlated with the mFI-5 score (p<0.001). Higher mFI-5 scores were associated with a significantly higher rate of congestive heart failure, diabetes mellitus, functional dependence, and hypertension; however, there was no significant trend for chronic obstructive pulmonary disease (COPD). Moreover, the three groups showed no significant differences regarding other co-morbidities.

Univariate analysis of outcomes by mFI-5 group. We compared the postoperative outcomes of the three groups (Table IV). Forty-three patients (40.2%) had postoperative complications, with no significant differences among the three groups. Postoperative ileus was the most common complication in Groups 0 and 1, whereas wound infection and pneumonia were more common in Group 2+. Thirteen patients (12.1%) had Clavien–Dindo classification grade III or higher complications. As the mFI-5 scores increased, the percentage of complications of Clavien–Dindo classification ≥ grade III was higher, with significant differences (p=0.02).

The duration of postoperative hospital stay was longer in Group 2+ (p=0.039). In mFI-5 Group 1, 61.1% were discharged home, and in mFI-5 Group 2+, 59.2% were transferred to another hospital.

This study focused on the correlation between frailty and postoperative complications in elderly patients. With the rapidly aging population, the number of surgeries for elderly colorectal cancer is increasing. For elderly patients, it is necessary to develop a treatment strategy based on a comprehensive assessment of not only age but also physical function, mental function, cognitive function, co-morbidities, social environment, and nutritional status. It is also important to adequately assess their preoperative condition using predictive tools for adverse events and prognosis, and to plan preoperative preparation and postoperative management while considering perioperative risks. There are a variety of frailty assessments, some of which have been reported to be associated with complications and prognosis (22,23). However, owing to the complexity of the evaluation methods and the large number of items per index, these are not widely used. Therefore, to determine whether it is possible to evaluate frailty in a simpler and more convenient manner, we focused on the mFI-5. In addition to being time-saving, the mFI-5 shows inter-rater reliability due to its objective nature, making it an attractive tool.

In the existing literature, Weaver et al. (20) concluded that the mFI-5 score was associated with postoperative complications in elective posterior lumbar fusion. Simon et al. (24) found that the mFI-5 predicted higher morbidity and mortality in elderly patients needing emergency colorectal surgery. Therefore, we hypothesized that the mFI-5 could be useful for predicting postoperative outcomes in elderly patients undergoing laparoscopic colorectal surgery.

In this study, we examined the association between the mFI-5 score and postoperative outcomes of laparoscopic colorectal surgery in elderly patients. Of the 107 patients, a mFI-5 score of 1 was the most common, accounting for approximately 40% of the cohort, and the highest score was 3. In the univariate analysis of preoperative factors, COPD and pneumonia were the only co-morbidities that were not significantly different among the five mFI-5 score items, reflecting the small number of patients. From this analysis, a significant between-group difference was found in the ASA classification, suggesting that the mFI-5 is useful for assessing the general clinical status and surgical risk factors.

In terms of postoperative factors, patients with higher mFI-5 scores also had higher rates of postoperative complications of Clavien–Dindo classification grade III or higher. In particular, the mFI-5 Group 2+ had a higher incidence of pneumonia, wound infection, and intraperitoneal bleeding, with longer hospital stays and higher rates of hospital transfer. These results suggest that patients with higher preoperative frailty require prolonged postoperative care and rehabilitation.

There were no significant differences in the 5-year survival rates or disease-specific 5-year survival rates among the three groups (Median observation period: 810 days; 5-year survival rates: Group 0, 62.1%; Group 1, 64.7%; and Group 2+, 59.7%; Disease-specific 5-year survival rates: Group 0, 87.7%; Group 1, 82.8%; and Group 2+, 79.8%). These results show that even patients with a high mFI-5 score have a good disease-specific prognosis. Thus, it is necessary to develop treatment strategies, including surgical intervention, based on adequate risk management, which includes predictive tools such as the mFI-5.

The results of this study demonstrate that for laparoscopic colorectal resection in elderly patients, the mFI-5 is a useful and simple scale for predicting the occurrence of severe postoperative complications and prolonged postoperative treatment. Therefore, when the preoperative mFI-5 score is high, aggressive therapeutic interventions should be performed during the waiting period before surgery to maximally improve postoperative recovery. Examples include respiratory and physical rehabilitation, preoperative oral care, and intensified diabetes therapies. It is also important to provide early release and rehabilitation after surgery to prevent postoperative complications.

The mFI-5 is a simple tool with only five items. However, because it is based on the presence or absence of the included co-morbidities, it does not adequately reflect the severity of each disease. Additionally, the mFI-5 is limited in that it only evaluates physical frailty and does not assess cognitive, mental, psychological, and social frailty.

In conclusion, the mFI-5 is an objective tool that can accurately predict postoperative complications in elderly patients undergoing colorectal cancer surgery. In the future, it is necessary to develop a simple tool that can assess frailty in a composite manner, by including psychological, cognitive, and social dimensions.

The Authors have no conflicts of interest to declare in relation to this study.

SO participated in the study concept and design, collected data, performed a literature review, and drafted the manuscript. YI, TA, SI, TA, AS, KK, MH, KK, and SE reviewed the manuscript.

The Authors thank their colleagues in the Department of Surgery, Graduate School of Biomedical Sciences, Nagasaki University for their kind cooperation and support.

The Authors received no funding for this work.