Impact of oral nutrition supplementation on outcomes of esophageal cancer patients treated with chemotherapy: A retrospective cohort study with propensity score matching

OBJECTIVE: There is a lack of evidence regarding the outcomes of oral nutrition supplementation (ONS) in patients with esophageal cancer (EC) who received chemotherapy treatment. The aim of this study was to perform a retrospective cohort study by comparing an adequate ONS group with a control group. MATERIALS AND METHODS: The study was performed in the Oncology Department of West China Hospital of Sichuan University. Patients at nutritional risk were identified from March 2016 to June 2019, and divided into an ONS group and a control group. To control for potential confounding variables, the propensity score method with matching was carried out. The main outcomes were length of stay (LOS) and hospitalization cost. Secondary outcomes included the incidence of pulmonary infection and myelosuppression. RESULTS: Out of 5,316 hospitalizations, a one-to-one matched sample was created (N = 229). The pathological tumor, node, metastasis (pTNM) stage of patients ranged from II to IV. A total of 69 patients received ONS, and 160 patients did not receive ONS. The incidence of myelosuppression in the ONS group and the control group was 4.3 vs. 17.4% (P = 0.014), respectively. However, ONS was associated with a 2 days increase in LOS, from 7 to 9 days (P < 0.000) and a hospitalization cost increase of $731, from $1134 to $1865 (P = 0.005). No statistical differences were observed in the incidence of pulmonary infection between the two groups. Further subgroup analysis based on body mass index (BMI) showed that at BMI ≤ 18.5 kg/m(2), the incidence of myelosuppression in the ONS group was lower than that in the control group (3.0 vs. 20.8%, P = 0.022). At BMI > 18.5 kg/m(2), no statistical differences were observed in the incidence of myelosuppression between the two groups. CONCLUSION: Although ONS increases hospitalization cost and LOS, it may be associated with reduced myelosuppression incidence, especially for patients with a BMI ≤ 18.5 kg/m(2).