Objectives: The aims were to assess the prevalence of malnutrition and to validate bioelectrical impedance analysis (BIA) against whole-body DXA for the assessment of body composition in patients with SSc. Methods: Malnutrition was defined as BMI <18.5 kg/m 2 or unintentional weight loss >10% in combination with a fat-free mass index (FFMI) <15 kg/m 2 for women or <17 kg/m 2 for men or BMI <20.0 kg/m 2 (age <70 years) or <22 kg/m 2 (age >70 years). Body composition was assessed in 72 patients with whole-body DXA (Hologic, Discovery A) and BIA (Bodystat Quadscan 400). The manufacturer's equation and the Geneva equation were used to estimate FFM and fat mass. The agreement between BIA and whole-body DXA was assessed with Bland-Altman analysis and intraclass correlation coefficient. Results: Malnutrition was found in 8.3% (n = 6) and low FFMI in 20.8% (n = 15) of patients. The mean difference in FFM between BIA and DXA applying the Geneva equation was 0.02 ( s . d . 2.4) kg, intraclass correlation coefficient 0.97 (95% CI: 0.95, 0.98). Limits of agreement were ±4.6 kg. The manufacturer's equation was less adequate to predict FFM. Conclusion: This study shows a relatively low prevalence of malnutrition in comparison with other studies, but a high prevalence of low FFMI, underlining the necessity of measuring body composition in SSc patients with a standardized and validated method. A good validity of BIA in determining FFM was found at a group level, while at an individual level the FFM may vary by 4.6 kg.
Objectives: The aims were to assess the prevalence of malnutrition and to validate bioelectrical impedance analysis (BIA) against whole-body DXA for the assessment of body composition in patients with SSc. Methods:Malnutrition was defined as BMI <18.5 kg/m 2 or unintentional weight loss >10% in combination with a fat-free mass index (FFMI) <15 kg/m 2 for women or <17 kg/m 2 for men or BMI <20.0 kg/m 2 (age <70 years) or <22 kg/m 2 (age >70 years). Body composition was assessed in 72 patients with whole-body DXA (Hologic, Discovery A) and BIA (Bodystat Quadscan 400). The manufacturer's equation and the Geneva equation were used to estimate FFM and fat mass. The agreement between BIA and whole-body DXA was assessed with Bland-Altman analysis and intraclass correlation coefficient. Results:Malnutrition was found in 8.3% (n = 6) and low FFMI in 20.8% (n = 15) of patients. The mean difference in FFM between BIA and DXA applying the Geneva equation was 0.02 ( s . d . 2.4) kg, intraclass correlation coefficient 0.97 (95% CI: 0.95, 0.98). Limits of agreement were ±4.6 kg. The manufacturer's equation was less adequate to predict FFM. Conclusion: This study shows a relatively low prevalence of malnutrition in comparison with other studies, but a high prevalence of low FFMI, underlining the necessity of measuring body composition in SSc patients with a standardized and validated method. A good validity of BIA in determining FFM was found at a group level, while at an individual level the FFM may vary by 4.6 kg.