Geriatric nutritional risk index and mortality from all-cause, cancer, and non-cancer in US cancer survivors: NHANES 2001-2018

Abstract

Background: Malnutrition is strongly correlated with worsened treatment outcomes, reduced standard of living, and heightened mortality rates among individuals with cancer. Our research explores how the Geriatric Nutritional Risk Index (GNRI), a measure of nutritional status, relates to all-cause mortality, cancer-specific, and non-cancer mortality among middle-aged and older adult cancer patients.

Methods: We enrolled 3,253 participants aged 40 and above who were diagnosed with cancer. The data was obtained from the National Health and Nutrition Examination Survey (NHANES) dataset covering the period from 2001 to 2018, with a median follow-up duration of 83 months. According to the GNRI levels, patients in the study were classified into two distinct groups: the group with a low GNRI (<98) and the group with a high GNRI (≥ 98). We conducted a Kaplan-Meier survival analysis to assess how survival rates vary with different nutritional conditions. Multivariable Cox regression analyses were performed to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for all-cause mortality, as well as cancer-specific and non-cancer-related mortality. Restricted cubic spline (RCS) analyses and subgroup evaluations were performed to augment the robustness and validity of our findings.

Results: A total of 1,171 deaths were documented, with 383 attributed to cancer, and 788 from other causes. After adjusting for potential confounders, the analysis demonstrated that, within a specified range, an elevation in the GNRI is inversely associated with mortality from all causes, cancer-specific, and non-cancer causes. Moreover, Kaplan-Meier survival curves for all-cause, cancer-specific, and non-cancer mortality distinctly showed a more pronounced decrease in survival rates among individuals in the low GNRI group (<98). Notably, the restricted cubic spline regression model (RCS) revealed statistically significant non-linear associations between GNRI scores and mortality rates. The P-values were ≤0.001 for both all-cause and non-cancer mortality, and 0.024 for cancer-specific mortality.

Conclusion: Our study conclusively demonstrated a robust correlation between GNRI scores and mortality rates among cancer patients, encompassing all-cause mortality as well as specific mortality related to both cancerous and non-cancerous causes. The GNRI may be a valuable prognostic tool for predicting cancer mortality outcomes, offering insights that may inform nutritional management and influence the clinical treatment strategies for cancer survivors.

Keywords: NHANES; cancer survivors; geriatric nutritional risk index; mortality; older adults.

Figure 1

Flowchart of the participant’s selection from NHANES 2001-2018.

Figure 2

Kaplan-Meier survival rates illustrating mortality among US cancer patients categorized by different Geriatric nutritional risk index (GNRI) groups: (A) all cause mortality; (B) cancer mortality; (C) non-cancer mortality. In the Kaplan-Meier curves, the population is stratified into twor groups (GNRI <98, and GNRI ≥98), and statistical analysis is conducted using the long-rank test.

Figure 3

Nonlinear relationship between GNRI and mortality in cancer patients. (A) all cause mortality; (B) cancer mortality; (C) non-cancer mortality. Statistical adjustments were made for age, gender, marital status, ethnicity, education level, Body Mass Index (BMI), smoking status, drinking status, energy intake, physical activity, the ratio of family income to poverty (PIR), diabetes, hypertension, coronary heart disease (CHD), stroke, and tumor types.

Figure 4

The figure show the association between GNRI and risk of mortality in different participants for (A) all cause mortality; (B) cancer-specific mortality; and (C) non-cancer mortality. The analysis was adjusted for age, gender, marital status, education level, ethnicity, drinking status, smoking status, Body Mass Index (BMI), and physical activity, hypertension, diabetes, stroke, coronary heart disease (CHD), and tumor types.