The role of weight loss and fat free mass as indicators of malnutrition
In 2015, the European Society for Clinical Nutrition and Metabolism (ESPEN) proposed a new set of criteria to diagnose malnutrition:
- BMI < 18.5
- OR >10% weight loss (no timeframe specified) or >5% weight loss in the last 3 months AND BMI <20 (70 years) or <22 (>70 years)
- OR >10% weight loss (no timeframe specified) or >5% weight loss in the last 3 months AND FFM Index <15 kg/m2 for women and <17 kg/m2 for men.
These ESPEN criteria are stricter than the old definition, which was based on BMI and unintentional weight loss only. According to the ESPEN 2015 criteria, a patient with a BMI >22 can only be classified malnourished if there is unintentional weight loss in combination with a low FFM Index. For overweight patients, this would mean that they would only be classified malnourished once their FFM Index passes a critical point, even if they have (already) lost a considerable amount of weight.
What to do in case of large weight loss and normal BMI ? FFMI?
Concerns have been raised whether the new ESPEN definition for malnutrition correctly classifies malnutrition in patients with normal weight or overweight and concomitant weight loss, as they do not necessarily meet the criteria for low FFMI. The aim of this study summarized here is to assess the association between critical weight loss and one-year mortality in hospitalized patients, stratified by BMI and FFMI subgroups.
Survival analysis in 769 hospital patients
In a group of hospital inpatients and outpatients, we determined the association between weight loss (CWL), BMI and FFMI and one-year mortality. Out of 769 patients (51% inpatients, 49% outpatients, mean age 60±16 years, 55% male), 270 (35%) had experienced weight loss. Overall, weight loss was associated with an increased one-year mortality rate compared to no weight loss. In BMI and FFMI subcategories, weight loss was predictive for mortality only in patients with low FFMI. For example: in patients with BMI ≥ 20/22, FFMI was crucial in the observed association with mortality (weight loss + normal BMI + low FFMI, HR 2.69 (1.29-5.65); weight loss + normal BMI + normal FFMI, HR 1.38 (0.84-2.27)) (figure).
Figure: Cox regression analysis on the association between patients with critical weight loss compared to patients with no critical weight loss and one-year mortality in total cohort and stratified by BMI and FFMI subgroups.
This study provides insight in the association between weight loss and mortality, stratified by BMI and FFMI subgroups. The results show that one-year mortality is higher in patients with weight loss compared to patients with no weight loss. Further analyses suggest that depletion of protein reserves plays a crucial role, because patients with low FFMI showed a significant association between weight loss and one-year mortality whereas patients with a normal FFMI did not show this association. The influence of low FFMI was also observed in subgroups of patients with normal BMI, stressing the importance of FFMI measurements in the assessment of malnutrition. Due to small subgroups, re-evaluation in larger datasets is warranted.
Viewpoint of the Dutch Malnutrition Steering Group
The Malnutrition Steering Group has not yet adopted the ESPEN 2015 malnutrition criteria. Until more research and validation data are available, the Malnutrition Steering Group will continue using the standard basic set of risk indicators for malnutrition:
– BMI <18.5 (18-69 years) and BMI <20 (≥70 years) and/or
– Unintentional weight loss >10% in 6 months and/or
– Unintentional weight loss >5% in one month
In addition, the measurement of the lean body mass in the diagnostic work-up of the nutritional status is strongly recommended and encouraged by the Dutch Malnutrition Steering Group as it reflects the amount of muscle mass, which cannot be properly estimated just with weight.