Acta PRdiatr Scand 80: 682-681, 1991
Quantitation of Muscles and Fat by Ultrasonography: A Useful Method in the Assessment of Malnutrition in Children E.-K. KOSKELO,’ L. M. KIVISAARI,* U. M. SAARINEN’ and M. A. SIIMES’ From the ‘Children’s Hospital and the ‘Department of Diagnostic Radiology, University of Helsinki, Helsinki, Finland ABSTRACT. Koskelo, E.-K., Kivisaari, L. M., Saarinen, U. M. and Siimes, M. A. (Children’s Hospital and the Department of Diagnostic Radiology, University of Helsinki, Helsinki, Finland). Quantitation of muscles and fat by ultrasonography: A useful method in the assessment of malnutrition in children. Acta Paediatr Scand 80: 682,1991. We used ultrasonography to measure muscles in the arms and thighs of 16 children with malignant diseases. Thicknesses of transverse sections of the brachial biceps muscle and the femoral quadriceps muscle were measured by ultrasound at the midpoint of the right arm and thigh. These two measures had a linear correlation (r=0.76).The ultrasound measurements did not differ from those obtained by the CT scan which was used as a reference standard. The reproducibility of the measurements was good, the coefficient of variation was 2.4% for the midarm muscles and 2.8% for the midthigh muscles. We conclude that the ultrasound method combined with simple anthropometric measurements is helpful in the assessment of nutritional status of children with potential malnutrition. Key words: anthropometqv, mulnutrifion,muscle wasting, ultrasonography
Malnutrition and muscle wasting are problems occurring in multiple pediatric diseases, of which those of the gastrointestinal tract are probably most common. Severe malnutrition may be obvious without any measurements. However, in modem pediatrics quantitative measures are required for follow-up purposes, and the detection of even milder conditions of malnutrition is important. Detection of milder degrees of protein-energy malnutrition may not be easy by the methods clinically available. The basic anthropometric measurements do not detect marginal changes in the nutritional status. The biochemical parameters may, on the other hand, be unreliable; the disease condition itself, e.g., infections or malignant diseases, may lead to a decrease in serum albumin, prealbumin and transferrin levels that is unrelated to the nutritional status. The purpose of the present study was to evaluate ultrasonography as a noninvasive clinical method in the assessment of muscle and fat thickness in children. Computerized tomography (CT scan) was used as a reference standard. Our target group of patients consisted of children with cancer, based on the fact that modem aggressive anticancer therapy is often associated with the development of proteinenergy malnutrition. SUBJECTS AND METHODS Our series comprised 16 children with malignancies, with a mean age of 9.8 years (range 3-19 years). Ten were boys and six were girls. The weight for height (1) of these patients ranged from - 21 % to 22 %. The diagnoses included neuroblastoma ( n=2), Wilms’ tumor (n= 3), rhabdomyosarcoma ( n=2), osteosarcoma (n= 2), other sarcoma (n=2), non-Hodgkin’s lymphoma (n=2), Hodgkin’s disease (n= l), germ cell tumor (n= l ) , and malignant histiocytosis (n= 1). The time from the initial diagnosis ranged from 0 months to 10.9 years (mean 2.2
+
Acta Paediatr Scand 80
Quantitation of muscles by ultrasonography 683
Fig. 1. Measurements of the biceps ( a )and the quadriceps (6)muscles and the subcutaneous fat at the midlimb (a, midarm; b, midthigh) levels by ultrasound (left) and CT scan (right). Closed triangle and arrow, bone-muscle interface; open triangle and arrow, muscle-fat interface; open square and arrow, fat-skin interface. Broken lines between the symbols indicate the thicknesses measured. The built-in software measuring line of the cross-sectional muscle area is shown on the CT scans.
years). Informed consent was obtained from both parents and children for the two extra X-ray exposures needed for the CT scans. In order to determine the reproducibility of the method, the ultrasound measurements were performed on two consecutive days on nine additional oncological patients aged 4 to 25 years (mean 9.7 years). Measurement of muscle thickness by ultrasound and CT scan. The thicknesses (ultrasound and CT scan) and cross-sectional areas (CT scan) of the brachial biceps muscles and the femoral quadriceps muscles were measured with the patients lying in the supine position on the scanner platform. The midpoint of the right arm was determined (2) and marked on the anterior surface where the brachial biceps muscle then was measured. The measurements on the femoral quadriceps muscle were made midway between the distal border of the patella and the superior spine of the anterior iliac crest. A Pie Medical real time scanner, model 400, equipped with a 5.0 MHz linear transducer, display screen, freeze-frame capacity, and electronic calipers (Pie Data Medical BV, Maastricht, the Netherlands) was used for the ultrasound measurements. Compression of the tissues was avoided by using a standoff pad in addition to the normal transmission gel. Transverse sections of the muscles were focused at right angles and the muscle thicknesses
684 E.-K. Koskelo et al.
Acta Paediatr Scand 80
(Fig. 1) were measured from the frozen images with the electronic calipers. The measurements were made in duplicate. After the ultrasound measurement, the patient was moved into the scanner gantry and the X-ray beam was lined up parallel to the markings on the limbs. A CT scan was performed at 125 kV, 4 sec scanning time, and 8 mm slice thickness. One slice for each limb was taken. The CT scanner window was set for each image so that the interfaces between the different tissues were maximally sharp (window level 450-600). The muscle diameters corresponding to those measured by ultrasound were then measured from the tissue interfaces (Fig. 1). The crosssectional areas of the midarm and midthigh muscles were determined with a built-in software program (Fig. 1). During the measurements, the leg or the arm was placed in relaxed position on a cardboard splint, in order to standardize the rotation of the limb. The ultrasonographic measurements and calculations on the CT images were performed by the author (E. K. K.). Reproducibility of ultrasound measurements. The reproducibility of the ultrasound method was tested by comparing two measurements performed on two consecutive days by using the same technique and the same set-up as above. Measurement of adipose tissue. The thickness of the adipose tissue was additionally measured at the sites marked on the arm and the thigh with both ultrasound and CT scan. In the ultrasound measurement, a sharp delineation of the skin was obtained with the help of a standoff pad. Skin thickness was excluded from the fat diameter (Fig. 1). Anthropometric methods. The circumferences of both the arm and the thigh were measured with a cloth tape at the sites of the ultrasound and CT scan measurements. Statistical methods. The Microsoft program Systat (Systat Inc., Evanston, IL, USA) was used for the statistical analyses. The tests included simple linear regression and correlation analysis. The methodological error of ultrasound was expressed as the coefficient of variation: CV (Yo) = SD x 100 /X CV = coefficient of variation; SD = standard deviation; X = mean.
RESULTS The thickness of the brachial biceps muscles, as measured by the ultrasound method, ranged from 0.6 to 2.7 cm (mean +- SEM, 1.520.3 cm). The thickness of the femoral quadriceps muscles ranged from 1.4 to 4.0 cm (mean 2 SEM, 2.7 20.8 cm). The thicknesses of the biceps muscles had a linear correlation with those of the quadriceps muscles ( ~ 0 . 7 6p
Quantitation of Muscles and Fat by Ultrasonography: A Useful Method in the Assessment of Malnutrition in Children E.-K. KOSKELO,’ L. M. KIVISAARI,* U. M. SAARINEN’ and M. A. SIIMES’ From the ‘Children’s Hospital and the ‘Department of Diagnostic Radiology, University of Helsinki, Helsinki, Finland ABSTRACT. Koskelo, E.-K., Kivisaari, L. M., Saarinen, U. M. and Siimes, M. A. (Children’s Hospital and the Department of Diagnostic Radiology, University of Helsinki, Helsinki, Finland). Quantitation of muscles and fat by ultrasonography: A useful method in the assessment of malnutrition in children. Acta Paediatr Scand 80: 682,1991. We used ultrasonography to measure muscles in the arms and thighs of 16 children with malignant diseases. Thicknesses of transverse sections of the brachial biceps muscle and the femoral quadriceps muscle were measured by ultrasound at the midpoint of the right arm and thigh. These two measures had a linear correlation (r=0.76).The ultrasound measurements did not differ from those obtained by the CT scan which was used as a reference standard. The reproducibility of the measurements was good, the coefficient of variation was 2.4% for the midarm muscles and 2.8% for the midthigh muscles. We conclude that the ultrasound method combined with simple anthropometric measurements is helpful in the assessment of nutritional status of children with potential malnutrition. Key words: anthropometqv, mulnutrifion,muscle wasting, ultrasonography
Malnutrition and muscle wasting are problems occurring in multiple pediatric diseases, of which those of the gastrointestinal tract are probably most common. Severe malnutrition may be obvious without any measurements. However, in modem pediatrics quantitative measures are required for follow-up purposes, and the detection of even milder conditions of malnutrition is important. Detection of milder degrees of protein-energy malnutrition may not be easy by the methods clinically available. The basic anthropometric measurements do not detect marginal changes in the nutritional status. The biochemical parameters may, on the other hand, be unreliable; the disease condition itself, e.g., infections or malignant diseases, may lead to a decrease in serum albumin, prealbumin and transferrin levels that is unrelated to the nutritional status. The purpose of the present study was to evaluate ultrasonography as a noninvasive clinical method in the assessment of muscle and fat thickness in children. Computerized tomography (CT scan) was used as a reference standard. Our target group of patients consisted of children with cancer, based on the fact that modem aggressive anticancer therapy is often associated with the development of proteinenergy malnutrition. SUBJECTS AND METHODS Our series comprised 16 children with malignancies, with a mean age of 9.8 years (range 3-19 years). Ten were boys and six were girls. The weight for height (1) of these patients ranged from - 21 % to 22 %. The diagnoses included neuroblastoma ( n=2), Wilms’ tumor (n= 3), rhabdomyosarcoma ( n=2), osteosarcoma (n= 2), other sarcoma (n=2), non-Hodgkin’s lymphoma (n=2), Hodgkin’s disease (n= l), germ cell tumor (n= l ) , and malignant histiocytosis (n= 1). The time from the initial diagnosis ranged from 0 months to 10.9 years (mean 2.2
+
Acta Paediatr Scand 80
Quantitation of muscles by ultrasonography 683
Fig. 1. Measurements of the biceps ( a )and the quadriceps (6)muscles and the subcutaneous fat at the midlimb (a, midarm; b, midthigh) levels by ultrasound (left) and CT scan (right). Closed triangle and arrow, bone-muscle interface; open triangle and arrow, muscle-fat interface; open square and arrow, fat-skin interface. Broken lines between the symbols indicate the thicknesses measured. The built-in software measuring line of the cross-sectional muscle area is shown on the CT scans.
years). Informed consent was obtained from both parents and children for the two extra X-ray exposures needed for the CT scans. In order to determine the reproducibility of the method, the ultrasound measurements were performed on two consecutive days on nine additional oncological patients aged 4 to 25 years (mean 9.7 years). Measurement of muscle thickness by ultrasound and CT scan. The thicknesses (ultrasound and CT scan) and cross-sectional areas (CT scan) of the brachial biceps muscles and the femoral quadriceps muscles were measured with the patients lying in the supine position on the scanner platform. The midpoint of the right arm was determined (2) and marked on the anterior surface where the brachial biceps muscle then was measured. The measurements on the femoral quadriceps muscle were made midway between the distal border of the patella and the superior spine of the anterior iliac crest. A Pie Medical real time scanner, model 400, equipped with a 5.0 MHz linear transducer, display screen, freeze-frame capacity, and electronic calipers (Pie Data Medical BV, Maastricht, the Netherlands) was used for the ultrasound measurements. Compression of the tissues was avoided by using a standoff pad in addition to the normal transmission gel. Transverse sections of the muscles were focused at right angles and the muscle thicknesses
684 E.-K. Koskelo et al.
Acta Paediatr Scand 80
(Fig. 1) were measured from the frozen images with the electronic calipers. The measurements were made in duplicate. After the ultrasound measurement, the patient was moved into the scanner gantry and the X-ray beam was lined up parallel to the markings on the limbs. A CT scan was performed at 125 kV, 4 sec scanning time, and 8 mm slice thickness. One slice for each limb was taken. The CT scanner window was set for each image so that the interfaces between the different tissues were maximally sharp (window level 450-600). The muscle diameters corresponding to those measured by ultrasound were then measured from the tissue interfaces (Fig. 1). The crosssectional areas of the midarm and midthigh muscles were determined with a built-in software program (Fig. 1). During the measurements, the leg or the arm was placed in relaxed position on a cardboard splint, in order to standardize the rotation of the limb. The ultrasonographic measurements and calculations on the CT images were performed by the author (E. K. K.). Reproducibility of ultrasound measurements. The reproducibility of the ultrasound method was tested by comparing two measurements performed on two consecutive days by using the same technique and the same set-up as above. Measurement of adipose tissue. The thickness of the adipose tissue was additionally measured at the sites marked on the arm and the thigh with both ultrasound and CT scan. In the ultrasound measurement, a sharp delineation of the skin was obtained with the help of a standoff pad. Skin thickness was excluded from the fat diameter (Fig. 1). Anthropometric methods. The circumferences of both the arm and the thigh were measured with a cloth tape at the sites of the ultrasound and CT scan measurements. Statistical methods. The Microsoft program Systat (Systat Inc., Evanston, IL, USA) was used for the statistical analyses. The tests included simple linear regression and correlation analysis. The methodological error of ultrasound was expressed as the coefficient of variation: CV (Yo) = SD x 100 /X CV = coefficient of variation; SD = standard deviation; X = mean.
RESULTS The thickness of the brachial biceps muscles, as measured by the ultrasound method, ranged from 0.6 to 2.7 cm (mean +- SEM, 1.520.3 cm). The thickness of the femoral quadriceps muscles ranged from 1.4 to 4.0 cm (mean 2 SEM, 2.7 20.8 cm). The thicknesses of the biceps muscles had a linear correlation with those of the quadriceps muscles ( ~ 0 . 7 6p
