Introduction

Malnutrition is an imbalance of nutrition, with or without inflammation, leading to a change in body composition and decreased function [1, 2]. Malnutrition is found in 15–60% of adult patients, depending on the patient population and criteria for definition [3]. Surgical patients are at risk for malnutrition for multiple reasons including the underlying pathologic process, the inflammatory nature of surgical intervention, and decreased intake due to nil per os status and delayed return of bowel function [4].

Multiple factors contribute to malnutrition, including inadequate intake, poor absorption, inflammatory conditions, and chronic disease states. Inflammation increases the risk for malnutrition and diminishes the response to intervention resulting in increased mortality [2]. Critical illness and injury leads to a catabolic state with inflammation contributing to alterations in body composition and stress metabolism [3]. The acute phase inflammatory response elevates a patient’s resting energy expenditure and increases nitrogen excretion, which results in increased energy and protein requirements. If nutritional intake is inadequate, this can lead to loss of lean tissue and adverse outcomes including increased length of hospital stay and mortality. Without inflammation, advanced malnutrition can usually be treated with appropriate nutritional supplementation. However, if inflammation is present, the treatment is more challenging and response to nutritional supplementation more limited [3].

Previous data have shown that malnourished patients have longer hospital stays, more complications, and higher mortality than patients with adequate nutrition [3]. Preexisting malnutrition or weight loss is associated with various complications [5]. Providing adequate nutrition improves outcomes including physical function, quality of life and hospital readmission rates [3].

Malnutrition is a concern in many low-resource settings. Rwanda is a low-income country in East Africa. With a population of 11.6 million, it has the highest population density in sub-Saharan Africa [6]. Almost half (44.9%) of the population lives below the national poverty line [6]. In 2015, 32% of the population of Rwanda was undernourished, compared with 18% in sub-Saharan Africa and 26% in low-income countries [6].

Kigali University Teaching Hospital (Centre Hospitalier Universitaire de Kigali, CHUK) is a 565-bed teaching and referral hospital in Kigali, Rwanda. The general surgery department has 48 inpatient beds. There are an estimated 140 general surgery operations each month with 70% of operations urgent or emergent [7, 8]. The acute care surgery service manages and treats all general surgery emergency patients including those presenting with trauma and non-trauma emergencies. The aim of this study was to characterize the nutritional status of acute care surgery patients in Rwanda, determine risk factors for malnutrition, and describe outcomes in patients with malnutrition.

Methods

This was a prospective, observational study of all ACS patients admitted to CHUK over a 6-month time period (January–June, 2017). All patients admitted to the ACS service were enrolled in the study after providing informed consent.

Admission details included demographics, vital signs, height, weight, history of weight loss, symptoms, admitting diagnosis, and laboratory investigations. Details on the hospital course included American Society of Anesthesiologists (ASA) score, operation performed, ICU admission, initiation of enteral feeding, and provision of parenteral nutrition. Outcomes included length of hospital stay, in-hospital mortality, complications, intensive care unit (ICU) admission, and hospital discharge disposition. Complications were assessed according to National Surgical Quality Improvement Program (NSQIP) definitions modified to the local setting [9].

Patients were assessed for malnutrition by trained data collectors within 24 h of admission using American Society for Parenteral and Enteral Nutrition (ASPEN) and Subjective Global Assessment (SGA). Patients were assessed on a weekly basis (±2 days) using ASPEN guidelines. Per ASPEN guidelines, [3] patients were classified as having malnutrition if they have 2 or more of the following: insufficient energy intake, weight loss, loss of muscle mass, loss of subcutaneous fat, localized or generalized fluid accumulation. Insufficient energy intake on hospital admission included patients nil per os or without intake for 5 days. During hospital stay, insufficient intake was defined as less than 25% of energy intake. On admission, weight loss was defined as loss of 3 or more kilograms in the past 6 months. During the hospital stay, weight loss was defined as any decrease in weight since hospital admission. Localized or generalized fluid accumulation was defined as ascites, ankle, or sacral edema greater than or equal to 1+. Patients were also assessed for malnutrition on hospital admission using the SGA [10]. Malnutrition was defined as SGA B or C. We used both the ASPEN and SGA measurement to assess their correlation in assessing malnutrition in patients from low- and middle-income countries.

Data were entered in Research electronic data capture (RedCap) database [11] and analyzed using STATA/SE (version 13.0, College Station, TX). Categorical variables were reported as frequencies and percentages. Continuous variables were reported as medians and interquartile ranges. Correlation between malnutrition on hospital admission using ASPEN and SGA was assessed using kappa statistic and area under the receiver operator characteristic (AUROC). Agreement was defined as very good if AUROC > 0.9, good if AUROC 0.8–0.9, fair if AUROC 0.7–0.8, and poor if AUROC < 0.7.

Analyses of risk factors for malnutrition on hospital admission and mortality were performed. Univariate analysis of categorical variables was performed using Chi-squared test, whereas continuous variables were analyzed using Wilcoxon rank-sum test. Factors with a p value < 0.1 on univariate analysis were entered into a multivariate regression model. Factors with p value < 0.05 on multivariate analysis were considered statistically significant.

This study was approved by the University of Rwanda IRB, CHUK IRB and University of Minnesota IRB. Informed consent was obtained from all participants.

Results

Patient characteristics

Over a 6-month time period, there were 289 acute care surgery patients enrolled in the study. Ten patients had insufficient data on malnutrition or mortality and were excluded from further analysis. We included 279 patients in the final analysis (Table 1).

Table 1 Patient characteristics and outcomes (N = 279)
Full size table

Median age was 38 years (IQR 26, 54), and 202 (72%) patients were male. Most (n = 133, 56%) patients were from Kigali. The most common diagnoses were trauma (n = 83, 30%), intestinal obstruction (n = 76, 27%), and peritonitis (n = 49, 18%). Most (n = 201, 73%) patients underwent an operation. In-hospital mortality was 3%, and most (n = 246, 89%) patients were discharged to home. The median length of hospital stay was 6 days (IQR 4, 11).

Malnutrition on hospital admission and weekly

On hospital admission, 99 (35%) patients had malnutrition using ASPEN and 76 (27%) patients had malnutrition using SGA, with 23 (8%) severely malnourished (Table 2). There was 79.6% agreement between ASPEN and SGA for malnutrition with a kappa statistic of 0.529. There was fair agreement between malnutrition based on ASPEN versus SGA with a AUROC of 0.7508.

Table 2 Nutrition characteristics of acute care surgery patients on hospital admission (N = 279)
Full size table

After 1, 2, 3, and 4 weeks of hospital stay, 41%, 37%, 50%, and 43% of patients were malnourished, respectively (Table 3).

Table 3 Weekly malnutrition in acute care surgery patients using American Society for Parenteral and Enteral Nutrition criteria
Full size table

Factors associated with malnutrition on hospital admission

On univariate analysis, factors associated with malnutrition on hospital admission included age, symptom duration, tachycardia, trauma, intestinal obstruction, peritonitis, and nasogastric tube placement (Table 4). On multivariate analysis, factors associated with malnutrition on hospital admission included symptom duration greater than 5 days [odds ratio (OR) 3.78, 95% confidence interval (CI) 1.91, 7.50, p value < 0.001], age greater than 40 (OR 3.15, 95% CI 1.68, 5.91, p value < 0.001), intestinal obstruction (OR 2.33, 95% CI 1.01, 5.38, p value 0.046), nasogastric tube placement (OR 2.16, 95% CI 1.02, 4.59, p value 0.045), and tachycardia (OR 2.10, 95% CI 1.13, 3.89, p value 0.019) (Table 5).

Table 4 Characteristics and outcomes of acute care surgery patients with malnutrition on hospital admission using American Society for Parenteral and Enteral Nutrition criteria (N = 279)
Full size table
Table 5 Multivariate analysis of factors associated with malnutrition on hospital admission in ACS patients (N = 264)
Full size table

Outcomes associated with malnutrition

Patients with malnutrition more likely to be admitted to the ICU (10% vs. 4%, p = 0.041). The length of hospital stay was longer in patients with malnutrition on hospital admission (6 days versus 5 days, p = 0.044).

Overall mortality was 3%, with higher mortality seen in patients with malnutrition (8% vs. 0.6%, p = 0.004). On univariate analysis, factors on hospital admission associated with mortality included gender, human immunodeficiency virus (HIV) infection, symptom duration, hypotension, and malnutrition (Table 6). After controlling for other factors on multivariate analysis, malnutrition remained associated with increased odds of mortality (Table 7).

Table 6 Characteristics of survivors and non-survivors at ACS admission
Full size table
Table 7 Multivariate analysis of factors associated with mortality in Rwanda ACS patients
Full size table

Discussion

In this study, we described the demographic parameters of patients who were managed by the Acute Care Surgery service at CHUK. We also described the nutrition status of these patients at admission and at 1 week intervals until the 4th week of admission. We described the relationship between malnutrition and the length of hospital stay, and mortality rate.

In our study, the malnutrition rate at admission was 35% and this is similar to what was reported in other studies (23–39%), though the measurement tools were different [12, 13]. There was fair agreement between ASPEN and SGA in determining malnutrition on hospital admission. Multiple assessment tools have been developed to identify patients at risk for and with malnutrition. These tools include anthropometry, diet factors, illness severity, physical variables, psychology, and symptoms [3]. Implementing and utilizing these tools can be challenging due to variability in resources. Demographic and Health Surveys have included biomarkers such as hemoglobin, transferrin, and retinol-binding protein in low-resource settings as a mechanism to assess micronutrient deficiencies [14].

Decision about when to start feeding after the operation is made by the surgical team. At CHUK, there are limited supplies of total parenteral nutrition and it is expensive. For patients who are allowed to start oral feeding, the family members and caregivers provide meal to the patient. One study in Malawi showed a lack of in-service training or nutrition protocols for critically ill patients contributes to a poor knowledge of nutrient supplementation in health care providers [15]. In Rwanda, a dietician is available for nutritional counseling and social workers assist in nutrient support when the family is unable to provide meals. Overall, the rate of malnutrition in acute care surgery patients did not change during the hospital stay.

The majority if the patients who were managed by the ACS service at CHUK during the study period had trauma or intestinal obstruction. Patients with intestinal obstruction were more likely to have malnutrition because they are initially managed by a nasogastric tube draining while waiting for surgery and they remain nil per os. Another factor is that the mean time from the onset of the signs and symptoms to the time of surgery was 5 days.

Benoist found that age greater that 70 years and sepsis are the risk factors for malnutrition in oncologic patients. In our study, patients older than 40 years had a higher risk of malnutrition compared to those younger than 40 years. Sepsis and advanced age have both been shown to be risk factors for malnutrition [4].

In this study, patients with malnutrition had increased length of hospital stay (6 vs. 5 days) and increased mortality rate (8% vs. 1%). Similar studies that looked at the impact of malnutrition on the postoperative outcome in surgical patients found that patients with malnutrition had higher risk of complications including increased length of hospital stay, increased odds of hospital readmission, and increased risk of mortality, compared to the patients whose nutritional status was normal [4, 12, 13, 16,17,18].

There are some limitations to this study. Due to patient’s functional status, weight and height data were limited as patients were unable to stand on the scale. Energy intake was estimated based on feedback from patients and their families. From this information, we estimated the content consumed. However, meals were individually prepared by family members as the hospital does not provide meal to patients. Therefore, the content in each meal is difficult to quantify.

Though the tools we used in measuring the nutrition status are validated in high income countries, we acknowledge that they may not be valid in low- and middle-income countries because of the variability in context and the inability to adequately measure all the parameters required to determine the nutrition status. More studies should assess the validity if the nutrition assessment tools in low- and middle-income countries.

Conclusions

Acute care surgical patients present to the hospital at high-risk for malnutrition. Efforts are needed to ensure that surgical patients receive adequate nutrition support to improve patient outcomes and minimize complications.