- Email: [email protected]
Bioelectrical impedance analysis during pregnancy and neonatal birth weight
European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
Bioelectrical impedance analysis during pregnancy and neonatal birth weight Fabio Ghezzia,*, Massimo Franchia, Debora Balestreria, Barbara Lischettia, Maria Cristina Meleb, Salvatore Albericoc, Pierfrancesco Bolisa a
Department of Obstetrics and Gynecology, University of Insubria, Viale Borri 57, 21100 Varese, Italy b Institute of Biological Chemistry, Catholic University of Sacred Heart, Rome, Italy c Department of Obstetrics and Gynecology, University of Trieste, I.R.C.C.S. Burgo Garofalo, Trieste, Italy Received 5 January 2001; received in revised form 19 February 2001; accepted 26 February 2001
Abstract Objective: To generate reference ranges for bioelectrical impedance indices throughout pregnancy and to investigate whether a relationship exists between these indices and the neonatal birth weight. Study design: Pregnant women with a singleton gestation, gestational age lower than 12 weeks, and absence of medical diseases before pregnancy were enrolled. Patients with pregnancy complications, such as hypertensive disorders, diabetes, and antiphospholipides syndrome were excluded. Antrophometric maternal parameters and bioelectrical impedance measurements were performed during the ®rst, second, third trimester of pregnancy, at delivery and 60 days after delivery. Height2/resistance (cm2/O) and height2/reactance (cm2/O) were utilized to estimate the total and extracellular body water amounts, respectively. Spearman rank correlations and cox proportional hazard modelling were used for statistical purposes. Results: 169 patients completed all measurements. Total and extracellular water amounts signi®cantly increase as pregnancy advances and return to the pre-pregnancy values within 60 days after delivery. After adjustment for gestational age at delivery, fetal sex, and smoking habits, height2/resistance at 25 weeks (hazard 1:04, 95% con®dence interval (CI) 1.02±1.06, P < 0:005), height2/resistance at 30 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/reactance at 20 weeks (hazard 1:03; 95% CI 1.01±1.05, P < 0:005), and height2/ reactance at 25 weeks (hazard 1:03, 95% CI 1.01±1.04, P < 0:01) were found to be independent predictors of birth weight. Conclusion: We have provided reference ranges for bioimpedance analysis during pregnancy, an easy, fast and non invasive method to estimate the body water composition during pregnancy. Bioelectrical impedance indices during the second trimester of pregnancy are independently related to the birth weight. # 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Body water composition; Bioelectrical impedance analysis; Pregnancy; Birth weight
1. Introduction Pregnancy is one of the conditions during which dramatic changes in the body composition occur in a short period of time [1]. The increase of the body weight is not only the consequence of body fat deposition occurring physiologically throughout pregnancy but is also due to the increase of the total body water amount [2,3]. Although, the foetus, placenta and amniotic ¯uid represent the most important contributors, an expansion of the plasma volume, which reaches the highest values in the second trimester of
*
Corresponding author. Tel.: 39-0348-5649669; fax: 39-0332-278323. E-mail address: [email protected] (F. Ghezzi).
pregnancy, participates to the total body water increase during pregnancy [4,5]. Cumulative evidence has shown that a relationship exists between the total body water amount and pregnancy outcome [6±9]. Rosso et al. reported that, in the absence of an adaptation of the maternal cardiovascular system as pregnancy advances, there is an increased risk of foetal growth restriction and pregnancy induced hypertension [8]. Indeed, Valensise et al. reported that women affected by pre-eclampsia had a lower total body water composition than healthy pregnant women and that this was particularly evident in the second trimester [9]. Since, the increase of the body weight during pregnancy is not speci®c to assess the increase of the total body water, several methods to evaluate the increase of the lean mass during pregnancy have been proposed [9±13]. Invasive
0301-2115/01/$ ± see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 1 - 2 1 1 5 ( 0 1 ) 0 0 3 3 0 - X
172
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
techniques based on isotope dilution methods with deuterium or oxygen-18 have been previously reported either in animals and in humans [10±13]. Recently, non invasive techniques based on tetrapolar bioelectrical impedence analysis have been tested in a variety of conditions, such as malnutrition, renal insuf®ciency, obesity, diabetes, and cirrhosis [14±17]. However, only few studies including a limited number of patients have investigated the role of bioelectrical impedance analysis during pregnancy [18±20]. The purpose of this study was to explore the longitudinal changes of the total body water and the extracellular water amounts, assessed by bioelectrical impedance analysis, during gestation and puerperium in physiological pregnancies and to explore whether the body water composition in¯uences the neonatal birth weight. 2. Subjects and methods 2.1. Subjects Pregnant women admitted to the outpatients clinic of our departments between January 1998 and December 1999 were enrolled in the study. Inclusion criteria were: (1) singleton pregnancy, (2) gestational age lower than 12 weeks at the ®rst visit, (3) known gestational age, and (4) written informed consent. Pre-existing maternal diseases (i.e. diabetes, hypertensive disorders) and conditions known to be a risk factor for an unfavourable pregnancy outcome (i.e. antiphospholipides syndrome) were considered exclusion criteria. Patients were enrolled and than excluded in the presence of at least one of the following conditions: (1) foetal growth restriction; (2) intrauterine fetal demise; (3) pregnancy induced hypertension and pre-eclampsia; (4) gestational diabetes; (5) foetal chromosomal abnormalities. Gestational age was calculated on a reliable recollection of the last menstrual period and con®rmed at ultrasound within 15 weeks of gestation. Speci®c forms were created to record patient characteristics and the ®ndings of each visit during pregnancy, at delivery and 60 days after delivery. This study was approved by the Ethical Research Committee of Insubria University. 2.2. Experimental design The ®rst evaluation consisted of the measurement of the patient height and body weight with a stadiometer and a calibrated scale, respectively. Gestational age was sonographically estimated and the body mass index (BMI) calculated. Subsequently, every patient was evaluated ®ve times during pregnancy, immediately after delivery, and 60 days after delivery. The evaluations performed during pregnancy were conducted at 15±17 weeks, 20±22 weeks, 25±27 weeks, 30±32 weeks, and 35±37 weeks, respectively. Obesity was de®ned as a BMI greater than 29. At each visit, the increment of the body weight was recorded and the total
impedance was performed. The hematocrit was evaluated at 15±17 weeks of gestation, 20±22 weeks, 30±32 weeks, and post-partum. A questionnaire was given to each patient to assess the dietary habits before and during pregnancy to exclude malnourished conditions. 2.3. Bioelectrical impedance Determination of resistance (O) and reactance (O) were made with a tetrapolar impedance analyzer (Bia Quantum; Farmades, Rome, Italy). The women, clothed but without shoes and socks, lay supine on a table made of non conducting materials, with the limbs abducted from the body and the legs separated one from the other in a straightened position. Current injector electrodes were positioned in the middle of the dorsal surfaces of the right hand and right feet, proximal to the metacarpal±phalangeal and metatarsal±phalangeal joints, respectively. Detector electrodes were placed medially on the posterior side of the right wrist, between the distal prominence of the radius and the ulna and between the medial and lateral malleoli at the right ankle. An excitation current of 800 mA, alternative current, at 50 kHz was introduced into the patients at the distal electrodes of the hand and foot; the voltage drop across the pregnant women was detected with the proximal electrodes. The total time for the examination was approximately 3 min. Height2/resistance (cm2/O) and height2/reactance (cm2/O) (Bioelectrical impedance indices) were utilized to estimate the total body water and extracellular water amounts as previously reported by Lukasky and Bolonchuk [21] and Segal et al. [22]. 2.4. Statistical analysis Statistical analysis was performed with SPSS package (SPSS inc., Chicago, IL) and with Epistat 4.0 (Epistat Services, Richardson, TX). Analysis of variance was used to compare continuous variables, Tukey's test was utilized for post-hoc analysis. Spearman rank correlation was used to correlate bioelectrical impedance indices and both the BMI and the hematocrit. Cox proportion hazard modelling was used to assess the independent contribution of a number of variables on the birth weight. P < 0:05 was considered for entry and removal of variables into the model. Signi®cance was considered to be achieved when P < 0:05. 3. Results 212 patients were enrolled into the study. Of these, 43 patients were excluded for the following reasons, 24 were lost to follow up during pregnancy, 4 were obese, 3 had gestational diabetes, 2 had an intrauterine foetal demise, 3 developed pre-eclampsia, and 7 had foetal growth restriction. The clinical characteristics of the study population are presented in Table 1. Table 2 shows the ponderal and bioelectrical values during pregnancy, the ®rst day after
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176 Table 1 Patient characteristicsa Characteristic
(n 169)
Age (y) Body weight before pregnancy (kg) Height (cm) Smoking habits (n) BMI before pregnancy (kg/m2) Time of delivery (week) Birth weight (gr)
29 4.2 58.2 10.4 161.5 15.8 20 (11.8%) 21.9 2.8 39.2 1.9 3161.4 606.4
a
Data are presented as mean standard deviation or number (%).
delivery and 60 days after delivery. Table 3 shows the mean S:D: of height2/resistance and cm2/reactance throughout gestation. Figs. 1 and 2 show the 5th, 50th, and 95th percentiles during and after pregnancy of height2/resistance and height2/reactance, respectively. No correlation was found between hematocrit and total body water in the ®rst (r 0:10; P 0:2), in the second (r 0:19; P 0:07), and in the third (r 0:01; P 0:9) trimester of pregnancy. A signi®cant correlation was found between the body weight
173
gain from the pre-pregnancy value and total body water at each measurement during pregnancy (15 weeks: r 0:45, P < 0:001; 20 weeks: r 0:49, P < 0:001; 25 weeks: r 0:46, P < 0:001; 30 weeks: r 0:54, P < 0:001; 35 weeks: r 0:59, P < 0:001). Cox proportional hazard modelling demonstrated that gestational age at delivery (hazard 1:54, 95% con®dence interval (CI) 1.35±1.69, P < 0:005), smoking habits (hazard 1:61, 95% CI 1.03±2.5, P < 0:05), male gender of the newborn (hazard 1:35, 95% CI 1.23±1.96, P < 0:05), height2/resistance at 25 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/resistance at 30 weeks (hazard 1:02, 95% CI 1.01±1.04, P < 0:005), height2/reactance at 20 weeks (hazard 1:002, 95% CI 1.001±1.004, P < 0:05), and height2/reactance at 25 weeks (hazard 1:02, 95% CI 1.01±1.04, P < 0:05), were signi®cant predictors of the infant birth weight. Cox proportional hazard modelling revealed that, after adjustment for gestational age at delivery, foetal sex, and smoking habits height2/ resistance at 25 weeks (hazard 1:04, 95% CI 1.02±1.06, P < 0:005), height2/resistance at 30 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/reactance at 20 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), and height2/ reactance at 25 weeks (hazard 1:03, 95% CI 1.01±1.04, P < 0:01) remained independent predictors of birth weight. Non correlation was found between height2/resistance at 15 and 35 weeks and the infant birth weight. Similarly, no correlation was found between height2/reactance at 15, 30, and 35 weeks and birth weight. 4. Discussion
Fig. 1. Height2/resistance (cm2/O) during gestation. The lines represent the 10th, 50th and 90th percentiles.
Fig. 2. Height2/reactance (cm2/O) during gestation. The lines represent the 10th, 50th and 90th percentiles.
A MEDLINE search of the literature from 1966 to November 2000 revealed that the present study is the largest study evaluating the longitudinal changes of bioelectrical impedance indices during pregnancy and puerperium. In addition, this study explored for the ®rst time the independent relationship existing between bioelectrical impedance indices measured throughout pregnancy and birth weight. The ®rst observation of the present study is that the total body water and the extracellular water signi®cantly increase as pregnancy advances. This is in keeping with the study by Lukaski et al. who found that the most important variation of the total body water occurs during the second and third trimester of gestation [19]. Although, the measurement of bioelectrical impedance indices is an indirect evaluation of the total body water and extracellular water amounts, a number of studies [13,14,18] have demonstrated a strong correlation between the estimations of the body water composition using isotope dilution techniques and those using bioelectrical impedance indices. The evaluation of bioelectrical indices is an easy and fast method to assess the maternal hemodynamic adaptation to pregnancy [9,19]. Since, this phenomenon begins early in pregnancy the
174
Weight (kg) Weight gain (kg) Resistance (O) Reactance (O) Hemoglobin (gr/dl) Hematocrit a
<15 weeks (a)
20±22 weeks (b)
25±27 weeks (c)
30±32 weeks (d)
35±37 weeks (e)
Post-partum (f)
60 days after delivery (g)
60 1.4 585 67 12.9 37.4
63 4.5 561 63 11.7 34.2
65.1 6.8 545 62 ± ±
67.1 8.8 539 63 11.5 34.2
68.4 11.2 506 59 ± ±
67.5 8.2 515 56 10.7 32.7
63.0 3.1 574 67 ± ±
10.7 b,c,d,e,f,g 1.9 b,c,d,e,f,g 77 c,d,e,f 13 c,e,f 0.9 b,d,f 2 b,d,f
10.8 a,d,e 2.2 a,c,d,e,f,g 70 e,f,g 12 e,f 2.4 a,f 5.9 a,f
10.9 a,e 2.8 a,b 70 a,e,g 12 a,f
11.2 a,b,g 3.1 a,b,e,g 75 a,e,g 20 e,f 2.6 a 5.4 a
11.4 a,b,g 3.5 a,b,d,g 69 a,b,c,d 12 a,b,d,g
13a,g 3.9 a,b,g 76 a,b 11 a,b,c,d,g 1.4 a,b 6.6 a,b
14.8 a,d,e,f 3.7 a,b,d,e,f 82 b,c,d 14 e,f
Data are presented as mean standard deviation; letters show the values that are significantly different (P < 0:05, Tukey's test).
Table 3 Bioimpedance indices during and after pregnancya
2
2
Height /resistance (cm /O) Height2/reactance (cm2/O) a
<15 weeks (a)
20±22 weeks (b)
25±27 weeks (c)
30±32 weeks (d)
35±37 weeks (e)
Post-partum (f)
46 9.6 c,d,e,f 409.6 131 c,d,e,f
47.8 9.1 e,f 428.4 92.3 e,f
49.3 9.7 a,e,f 436.7 94.6 f
50.2 9.9 a,e,f 439 97.6 a,e,f,g
53.7 11.1 a,b,c,d,g 474 108.6a,e,f
53 10.6a,b,c,d,g 46.8 9.5 e,f 489 112a,b,c,d,g 402 89 c,d,e
Data are presented as mean standard deviation; letters show the values that are significantly different (P < 0:05, Tukey's test).
60 days after delivery (g)
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
Table 2 Ponderal, bioelectrical, and hematologic changes during and after pregnancya
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
possibility to identify patients at risk for hypertensive disorders by bioimpedance analysis might prompt the physician to monitor more closely the patient. Moreover, Valensise et al. reported that both the extracellular and intracellular water amounts are signi®cantly lower in the second and third trimester of pregnancy in women affected by pregnancy induced hypertension in comparison to healthy pregnant women [9]. On the contrary, these authors did not ®nd changes of the hematocrit values in the second and third trimester between healthy and pre-eclamptic women. This is supported by the present study in which no correlation was found between bioelectrical impedance indices and hematocrit values throughout gestation. Further prospective studies are required to explore the predictive values and accuracy of bioelectrical impedance indices in pathologic pregnancies. The second observation of the present study is that the reduction of the body resistance in the second trimester of pregnancy is independently associated with the birth weight. This ®nding suggests that the maternal hemodynamic adaptation occurring in the second trimester of pregnancy in¯uences the fetal and neonatal well being. This observation is supported by Langhoff-Roos et al. who found an association between the increase in lean body mass during pregnancy and birth weight [7]. Similarly, Villar et al. noted a signi®cant correlation between the fat-free mass deposition in the second half of gestation and the infant birth weight [23]. It has been reported in humans and in animals that plasma volume, the major contributor to the total body water, is correlated with birth weight [24]. Moreover, a lack of plasma volume expansion during pregnancy has been observed to be associated with poor outcome and low birth weight [24,25]. It has been speculated that the correlation existing between fat-free mass and foetal growth might be mediated by ¯uid retention including plasma volume which in turn in¯uences the cardiac output and, ultimately, the uterine blood ¯ow. The novel ®nding of the present study is that both the total body water and extracellular water amounts are predictors of the birth weight only in the second trimester of pregnancy and not later in gestation. This might be explained by the physiologic modi®cations of the cardiovascular system occurring during pregnancy. Indeed, the most important changes of the cardiovascular system during gestation occur in the ®rst and second trimester of gestation [4]. Furthermore, serial hemodynamic investigations throughout normal pregnancy by thoracic electrical bioimpedance in 50 healthy pregnant women showed that the mean cardiac output reaches the highest level in the second trimester and subsequently dramatically decreases [19]. Finally, the present study clearly demonstrated that the maternal hemodynamic condition returns completely to the pre-pregnancy status within 60 days after delivery in healthy uncomplicated pregnancies. This agrees with Lukasky et al. [18] who found no difference in resistance and reactance between the pre-pregnancy and post-partum periods. It remains to be explored whether in the presence of certain
175
pregnancy complications (i.e. pre-eclampsia) the return to normality presents a different pattern. In conclusion, the bioimpedance analysis during pregnancy is an easy, fast and non invasive method to estimate the body water composition during pregnancy. Reference ranges are now available for bioelectrical impedance indices throughout gestation which can form the basis for further studies investigating pathologic conditions, such as hypertensive disorders, diabetes, and foetal growth restriction.
References [1] Hytten FE, Thomas AM. Maternal physiological adjustements. In: Maternal Nutrition and Course of Pregnancy. National Academy of Science, Washington, DC, 1970. [2] McCartney CP, Pottinger RE, Harrod JP. Alternations in body composition during pregnancy. Am J Obstet Gynecol 1959;77: 1038±53. [3] Hutchinson D, Plentl A, Taylor HC. The total body water and the water turnover in pregnancy studied with deuterium oxide as isotopic tracer. J Clin Invest 1954;33:235±41. [4] Hytten F. Blood volume changes in normal pregnancy. Clin Haematol 1985;14:601±12. [5] Pirani BBK, Campbell DM, MacGillivray I. Plasma volume in normal first pregnancy. J Obstet Gynaecol Br Commonwealth 1973;80:884±7. [6] Duffus GM, MacGillivray I, Dennis JK. The relationship between baby weight and changes in maternal weight, and proteins and urinary oestriol excretion. J Obstet Gynaecol Br Commonwealth 1971;78:97±104. [7] Langhoff-Ross J, Lindmark G, Gebre-Medin M. Maternal fat stores and fat accretion during pregnancy in relation to infant birthweight. Br J Obstet Gynaecol 1987;94:1170±7. [8] Rosso P, Donoso E, Braun S, Espinoza R, Fernadez C, Salas SP. Maternal hemodynamic adjustements in idiopathic fetal growth retardation. Gynecol Obstet Invest 1993;35:162±5. [9] Valensise H, Andreoli A, Lello S, Magnani F, Romanini C, De Lorenzo A. Multifrequency bioelectrical analysis in women with a normal and hypertensive pregnancy. Am J Clin Nutr 2000;72: 780±3. [10] Walser M, Seldin DW, Grollman A. An evaluation of radiosulfate for the determination of the volume of extracellular fluid in man and dogs. J Clin Invest 1953;32:299±311. [11] Wong WW, Cochran WJ, Klish WJ, O'Brien Smith E, Lee LS, Kleim PD. In vivo isotope-fractionation factors and the measurement of deuterium and oxygen-18 dilution spaces from plasma, urine, saliva, respiratory water vapors, and carbon dioxide. Am J Clin Nutr 1988;47:1±6. [12] Lukaski HC, Johnson PE. A simple, inexpensive method of determining total body water using a tracer dose of D2O and infrared absorption of biological fluids. Am J Clin Nutr 1985;41:363±70. [13] Fjeld CR, Freundt-Thurne J, Scoeller DA. Total body water measured by 18O dilution and bioelectrical impedance in well and malnourished children. Pediatr Res 1990;27:98±102. [14] Kushner RF, Kunigk A, Alspaugh M, Andronis PT, Leitch CA, Schoeller DA. Validation of bioelectrical impedance analysis as a measurement of changes in body composition in obesity. Am J Clin Nutr 1990;52:219±23. [15] Tedner B, Lins LE. Fluid volume monitoring with electrical impedance technique during hemodialisis. Artificial Org 1984;8: 66±71.
176
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
[16] Zillikens MC, van de Berg JWO, Wilson JHP, Swart GR. Wholebody and segmental bioelectrical impedance analysis in patients with cirrhosis of the liver: changes after treatment of ascites. Am J Clin Nutr 1992;55:621±5. [17] Guglielmi FW, Contento F, Laddaga L, Panella C, Francavilla A. Bioelectric impedance analysis: experience with male patients with cirrhosis. Hepatology 1991;13:892±5. [18] Lukaski HC, Siders WA, Nielsen EJ, Hall CB. Total body water in pregnancy: assessment by using bioelectrical impedance. Am J Clin Nutr 1994;59:578±85. [19] van Oppen AC, van der Tweel I, Alsbach GP, Heethaar RM, Bruinse HW. A longitudinal study of maternal hemodynamics during normal pregnancy. Obstet Gynecol 1996;88:40±6. [20] Morita H, Takeuchi K, Funakoshi T, Mizutori M, Maruo T. Potential use of bioelectrical impedance analysis in the assessment of edema in pregnancy. Clin Exp Obstet Gynecol 1999;26:151±4.
[21] Lukaski H, Bolonchuk W. Estimation of body fluid volumes using tetrapolar bioelectrical impedance measurements. Aviat Space Environ Med 1988;59:1163±9. [22] Segal K, Wang J, Pierson R, Van Itallie T. Estimation of body water distribution by bioelectrical impedance. Fed Proc 1987;46: 1334. [23] Villar J, Cogswell M, Kestler E, Castillo P, Menendez R, Repke J. Effect of fat and free mass deposition during pregnancy on birth weight. Am J Obstet Gynecol 1992;167:1344±52. [24] Mardones-Santander F, Salazar G, Rosso P, Villaroel C. Maternal body composition near term and birth weight. Obstet Gynecol 1998;91:873±7. [25] Lederman SA, Paxton A, Heymsfield SB, Wang J, Thornton J, Pierson Jr RN. Maternal body fat and water during pregnancy: do they raise infant birth weight? Am J Obstet Gynecol 1999;180: 235±40.
Bioelectrical impedance analysis during pregnancy and neonatal birth weight Fabio Ghezzia,*, Massimo Franchia, Debora Balestreria, Barbara Lischettia, Maria Cristina Meleb, Salvatore Albericoc, Pierfrancesco Bolisa a
Department of Obstetrics and Gynecology, University of Insubria, Viale Borri 57, 21100 Varese, Italy b Institute of Biological Chemistry, Catholic University of Sacred Heart, Rome, Italy c Department of Obstetrics and Gynecology, University of Trieste, I.R.C.C.S. Burgo Garofalo, Trieste, Italy Received 5 January 2001; received in revised form 19 February 2001; accepted 26 February 2001
Abstract Objective: To generate reference ranges for bioelectrical impedance indices throughout pregnancy and to investigate whether a relationship exists between these indices and the neonatal birth weight. Study design: Pregnant women with a singleton gestation, gestational age lower than 12 weeks, and absence of medical diseases before pregnancy were enrolled. Patients with pregnancy complications, such as hypertensive disorders, diabetes, and antiphospholipides syndrome were excluded. Antrophometric maternal parameters and bioelectrical impedance measurements were performed during the ®rst, second, third trimester of pregnancy, at delivery and 60 days after delivery. Height2/resistance (cm2/O) and height2/reactance (cm2/O) were utilized to estimate the total and extracellular body water amounts, respectively. Spearman rank correlations and cox proportional hazard modelling were used for statistical purposes. Results: 169 patients completed all measurements. Total and extracellular water amounts signi®cantly increase as pregnancy advances and return to the pre-pregnancy values within 60 days after delivery. After adjustment for gestational age at delivery, fetal sex, and smoking habits, height2/resistance at 25 weeks (hazard 1:04, 95% con®dence interval (CI) 1.02±1.06, P < 0:005), height2/resistance at 30 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/reactance at 20 weeks (hazard 1:03; 95% CI 1.01±1.05, P < 0:005), and height2/ reactance at 25 weeks (hazard 1:03, 95% CI 1.01±1.04, P < 0:01) were found to be independent predictors of birth weight. Conclusion: We have provided reference ranges for bioimpedance analysis during pregnancy, an easy, fast and non invasive method to estimate the body water composition during pregnancy. Bioelectrical impedance indices during the second trimester of pregnancy are independently related to the birth weight. # 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Body water composition; Bioelectrical impedance analysis; Pregnancy; Birth weight
1. Introduction Pregnancy is one of the conditions during which dramatic changes in the body composition occur in a short period of time [1]. The increase of the body weight is not only the consequence of body fat deposition occurring physiologically throughout pregnancy but is also due to the increase of the total body water amount [2,3]. Although, the foetus, placenta and amniotic ¯uid represent the most important contributors, an expansion of the plasma volume, which reaches the highest values in the second trimester of
*
Corresponding author. Tel.: 39-0348-5649669; fax: 39-0332-278323. E-mail address: [email protected] (F. Ghezzi).
pregnancy, participates to the total body water increase during pregnancy [4,5]. Cumulative evidence has shown that a relationship exists between the total body water amount and pregnancy outcome [6±9]. Rosso et al. reported that, in the absence of an adaptation of the maternal cardiovascular system as pregnancy advances, there is an increased risk of foetal growth restriction and pregnancy induced hypertension [8]. Indeed, Valensise et al. reported that women affected by pre-eclampsia had a lower total body water composition than healthy pregnant women and that this was particularly evident in the second trimester [9]. Since, the increase of the body weight during pregnancy is not speci®c to assess the increase of the total body water, several methods to evaluate the increase of the lean mass during pregnancy have been proposed [9±13]. Invasive
0301-2115/01/$ ± see front matter # 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 0 1 - 2 1 1 5 ( 0 1 ) 0 0 3 3 0 - X
172
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
techniques based on isotope dilution methods with deuterium or oxygen-18 have been previously reported either in animals and in humans [10±13]. Recently, non invasive techniques based on tetrapolar bioelectrical impedence analysis have been tested in a variety of conditions, such as malnutrition, renal insuf®ciency, obesity, diabetes, and cirrhosis [14±17]. However, only few studies including a limited number of patients have investigated the role of bioelectrical impedance analysis during pregnancy [18±20]. The purpose of this study was to explore the longitudinal changes of the total body water and the extracellular water amounts, assessed by bioelectrical impedance analysis, during gestation and puerperium in physiological pregnancies and to explore whether the body water composition in¯uences the neonatal birth weight. 2. Subjects and methods 2.1. Subjects Pregnant women admitted to the outpatients clinic of our departments between January 1998 and December 1999 were enrolled in the study. Inclusion criteria were: (1) singleton pregnancy, (2) gestational age lower than 12 weeks at the ®rst visit, (3) known gestational age, and (4) written informed consent. Pre-existing maternal diseases (i.e. diabetes, hypertensive disorders) and conditions known to be a risk factor for an unfavourable pregnancy outcome (i.e. antiphospholipides syndrome) were considered exclusion criteria. Patients were enrolled and than excluded in the presence of at least one of the following conditions: (1) foetal growth restriction; (2) intrauterine fetal demise; (3) pregnancy induced hypertension and pre-eclampsia; (4) gestational diabetes; (5) foetal chromosomal abnormalities. Gestational age was calculated on a reliable recollection of the last menstrual period and con®rmed at ultrasound within 15 weeks of gestation. Speci®c forms were created to record patient characteristics and the ®ndings of each visit during pregnancy, at delivery and 60 days after delivery. This study was approved by the Ethical Research Committee of Insubria University. 2.2. Experimental design The ®rst evaluation consisted of the measurement of the patient height and body weight with a stadiometer and a calibrated scale, respectively. Gestational age was sonographically estimated and the body mass index (BMI) calculated. Subsequently, every patient was evaluated ®ve times during pregnancy, immediately after delivery, and 60 days after delivery. The evaluations performed during pregnancy were conducted at 15±17 weeks, 20±22 weeks, 25±27 weeks, 30±32 weeks, and 35±37 weeks, respectively. Obesity was de®ned as a BMI greater than 29. At each visit, the increment of the body weight was recorded and the total
impedance was performed. The hematocrit was evaluated at 15±17 weeks of gestation, 20±22 weeks, 30±32 weeks, and post-partum. A questionnaire was given to each patient to assess the dietary habits before and during pregnancy to exclude malnourished conditions. 2.3. Bioelectrical impedance Determination of resistance (O) and reactance (O) were made with a tetrapolar impedance analyzer (Bia Quantum; Farmades, Rome, Italy). The women, clothed but without shoes and socks, lay supine on a table made of non conducting materials, with the limbs abducted from the body and the legs separated one from the other in a straightened position. Current injector electrodes were positioned in the middle of the dorsal surfaces of the right hand and right feet, proximal to the metacarpal±phalangeal and metatarsal±phalangeal joints, respectively. Detector electrodes were placed medially on the posterior side of the right wrist, between the distal prominence of the radius and the ulna and between the medial and lateral malleoli at the right ankle. An excitation current of 800 mA, alternative current, at 50 kHz was introduced into the patients at the distal electrodes of the hand and foot; the voltage drop across the pregnant women was detected with the proximal electrodes. The total time for the examination was approximately 3 min. Height2/resistance (cm2/O) and height2/reactance (cm2/O) (Bioelectrical impedance indices) were utilized to estimate the total body water and extracellular water amounts as previously reported by Lukasky and Bolonchuk [21] and Segal et al. [22]. 2.4. Statistical analysis Statistical analysis was performed with SPSS package (SPSS inc., Chicago, IL) and with Epistat 4.0 (Epistat Services, Richardson, TX). Analysis of variance was used to compare continuous variables, Tukey's test was utilized for post-hoc analysis. Spearman rank correlation was used to correlate bioelectrical impedance indices and both the BMI and the hematocrit. Cox proportion hazard modelling was used to assess the independent contribution of a number of variables on the birth weight. P < 0:05 was considered for entry and removal of variables into the model. Signi®cance was considered to be achieved when P < 0:05. 3. Results 212 patients were enrolled into the study. Of these, 43 patients were excluded for the following reasons, 24 were lost to follow up during pregnancy, 4 were obese, 3 had gestational diabetes, 2 had an intrauterine foetal demise, 3 developed pre-eclampsia, and 7 had foetal growth restriction. The clinical characteristics of the study population are presented in Table 1. Table 2 shows the ponderal and bioelectrical values during pregnancy, the ®rst day after
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176 Table 1 Patient characteristicsa Characteristic
(n 169)
Age (y) Body weight before pregnancy (kg) Height (cm) Smoking habits (n) BMI before pregnancy (kg/m2) Time of delivery (week) Birth weight (gr)
29 4.2 58.2 10.4 161.5 15.8 20 (11.8%) 21.9 2.8 39.2 1.9 3161.4 606.4
a
Data are presented as mean standard deviation or number (%).
delivery and 60 days after delivery. Table 3 shows the mean S:D: of height2/resistance and cm2/reactance throughout gestation. Figs. 1 and 2 show the 5th, 50th, and 95th percentiles during and after pregnancy of height2/resistance and height2/reactance, respectively. No correlation was found between hematocrit and total body water in the ®rst (r 0:10; P 0:2), in the second (r 0:19; P 0:07), and in the third (r 0:01; P 0:9) trimester of pregnancy. A signi®cant correlation was found between the body weight
173
gain from the pre-pregnancy value and total body water at each measurement during pregnancy (15 weeks: r 0:45, P < 0:001; 20 weeks: r 0:49, P < 0:001; 25 weeks: r 0:46, P < 0:001; 30 weeks: r 0:54, P < 0:001; 35 weeks: r 0:59, P < 0:001). Cox proportional hazard modelling demonstrated that gestational age at delivery (hazard 1:54, 95% con®dence interval (CI) 1.35±1.69, P < 0:005), smoking habits (hazard 1:61, 95% CI 1.03±2.5, P < 0:05), male gender of the newborn (hazard 1:35, 95% CI 1.23±1.96, P < 0:05), height2/resistance at 25 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/resistance at 30 weeks (hazard 1:02, 95% CI 1.01±1.04, P < 0:005), height2/reactance at 20 weeks (hazard 1:002, 95% CI 1.001±1.004, P < 0:05), and height2/reactance at 25 weeks (hazard 1:02, 95% CI 1.01±1.04, P < 0:05), were signi®cant predictors of the infant birth weight. Cox proportional hazard modelling revealed that, after adjustment for gestational age at delivery, foetal sex, and smoking habits height2/ resistance at 25 weeks (hazard 1:04, 95% CI 1.02±1.06, P < 0:005), height2/resistance at 30 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), height2/reactance at 20 weeks (hazard 1:03, 95% CI 1.01±1.05, P < 0:005), and height2/ reactance at 25 weeks (hazard 1:03, 95% CI 1.01±1.04, P < 0:01) remained independent predictors of birth weight. Non correlation was found between height2/resistance at 15 and 35 weeks and the infant birth weight. Similarly, no correlation was found between height2/reactance at 15, 30, and 35 weeks and birth weight. 4. Discussion
Fig. 1. Height2/resistance (cm2/O) during gestation. The lines represent the 10th, 50th and 90th percentiles.
Fig. 2. Height2/reactance (cm2/O) during gestation. The lines represent the 10th, 50th and 90th percentiles.
A MEDLINE search of the literature from 1966 to November 2000 revealed that the present study is the largest study evaluating the longitudinal changes of bioelectrical impedance indices during pregnancy and puerperium. In addition, this study explored for the ®rst time the independent relationship existing between bioelectrical impedance indices measured throughout pregnancy and birth weight. The ®rst observation of the present study is that the total body water and the extracellular water signi®cantly increase as pregnancy advances. This is in keeping with the study by Lukaski et al. who found that the most important variation of the total body water occurs during the second and third trimester of gestation [19]. Although, the measurement of bioelectrical impedance indices is an indirect evaluation of the total body water and extracellular water amounts, a number of studies [13,14,18] have demonstrated a strong correlation between the estimations of the body water composition using isotope dilution techniques and those using bioelectrical impedance indices. The evaluation of bioelectrical indices is an easy and fast method to assess the maternal hemodynamic adaptation to pregnancy [9,19]. Since, this phenomenon begins early in pregnancy the
174
Weight (kg) Weight gain (kg) Resistance (O) Reactance (O) Hemoglobin (gr/dl) Hematocrit a
<15 weeks (a)
20±22 weeks (b)
25±27 weeks (c)
30±32 weeks (d)
35±37 weeks (e)
Post-partum (f)
60 days after delivery (g)
60 1.4 585 67 12.9 37.4
63 4.5 561 63 11.7 34.2
65.1 6.8 545 62 ± ±
67.1 8.8 539 63 11.5 34.2
68.4 11.2 506 59 ± ±
67.5 8.2 515 56 10.7 32.7
63.0 3.1 574 67 ± ±
10.7 b,c,d,e,f,g 1.9 b,c,d,e,f,g 77 c,d,e,f 13 c,e,f 0.9 b,d,f 2 b,d,f
10.8 a,d,e 2.2 a,c,d,e,f,g 70 e,f,g 12 e,f 2.4 a,f 5.9 a,f
10.9 a,e 2.8 a,b 70 a,e,g 12 a,f
11.2 a,b,g 3.1 a,b,e,g 75 a,e,g 20 e,f 2.6 a 5.4 a
11.4 a,b,g 3.5 a,b,d,g 69 a,b,c,d 12 a,b,d,g
13a,g 3.9 a,b,g 76 a,b 11 a,b,c,d,g 1.4 a,b 6.6 a,b
14.8 a,d,e,f 3.7 a,b,d,e,f 82 b,c,d 14 e,f
Data are presented as mean standard deviation; letters show the values that are significantly different (P < 0:05, Tukey's test).
Table 3 Bioimpedance indices during and after pregnancya
2
2
Height /resistance (cm /O) Height2/reactance (cm2/O) a
<15 weeks (a)
20±22 weeks (b)
25±27 weeks (c)
30±32 weeks (d)
35±37 weeks (e)
Post-partum (f)
46 9.6 c,d,e,f 409.6 131 c,d,e,f
47.8 9.1 e,f 428.4 92.3 e,f
49.3 9.7 a,e,f 436.7 94.6 f
50.2 9.9 a,e,f 439 97.6 a,e,f,g
53.7 11.1 a,b,c,d,g 474 108.6a,e,f
53 10.6a,b,c,d,g 46.8 9.5 e,f 489 112a,b,c,d,g 402 89 c,d,e
Data are presented as mean standard deviation; letters show the values that are significantly different (P < 0:05, Tukey's test).
60 days after delivery (g)
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
Table 2 Ponderal, bioelectrical, and hematologic changes during and after pregnancya
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
possibility to identify patients at risk for hypertensive disorders by bioimpedance analysis might prompt the physician to monitor more closely the patient. Moreover, Valensise et al. reported that both the extracellular and intracellular water amounts are signi®cantly lower in the second and third trimester of pregnancy in women affected by pregnancy induced hypertension in comparison to healthy pregnant women [9]. On the contrary, these authors did not ®nd changes of the hematocrit values in the second and third trimester between healthy and pre-eclamptic women. This is supported by the present study in which no correlation was found between bioelectrical impedance indices and hematocrit values throughout gestation. Further prospective studies are required to explore the predictive values and accuracy of bioelectrical impedance indices in pathologic pregnancies. The second observation of the present study is that the reduction of the body resistance in the second trimester of pregnancy is independently associated with the birth weight. This ®nding suggests that the maternal hemodynamic adaptation occurring in the second trimester of pregnancy in¯uences the fetal and neonatal well being. This observation is supported by Langhoff-Roos et al. who found an association between the increase in lean body mass during pregnancy and birth weight [7]. Similarly, Villar et al. noted a signi®cant correlation between the fat-free mass deposition in the second half of gestation and the infant birth weight [23]. It has been reported in humans and in animals that plasma volume, the major contributor to the total body water, is correlated with birth weight [24]. Moreover, a lack of plasma volume expansion during pregnancy has been observed to be associated with poor outcome and low birth weight [24,25]. It has been speculated that the correlation existing between fat-free mass and foetal growth might be mediated by ¯uid retention including plasma volume which in turn in¯uences the cardiac output and, ultimately, the uterine blood ¯ow. The novel ®nding of the present study is that both the total body water and extracellular water amounts are predictors of the birth weight only in the second trimester of pregnancy and not later in gestation. This might be explained by the physiologic modi®cations of the cardiovascular system occurring during pregnancy. Indeed, the most important changes of the cardiovascular system during gestation occur in the ®rst and second trimester of gestation [4]. Furthermore, serial hemodynamic investigations throughout normal pregnancy by thoracic electrical bioimpedance in 50 healthy pregnant women showed that the mean cardiac output reaches the highest level in the second trimester and subsequently dramatically decreases [19]. Finally, the present study clearly demonstrated that the maternal hemodynamic condition returns completely to the pre-pregnancy status within 60 days after delivery in healthy uncomplicated pregnancies. This agrees with Lukasky et al. [18] who found no difference in resistance and reactance between the pre-pregnancy and post-partum periods. It remains to be explored whether in the presence of certain
175
pregnancy complications (i.e. pre-eclampsia) the return to normality presents a different pattern. In conclusion, the bioimpedance analysis during pregnancy is an easy, fast and non invasive method to estimate the body water composition during pregnancy. Reference ranges are now available for bioelectrical impedance indices throughout gestation which can form the basis for further studies investigating pathologic conditions, such as hypertensive disorders, diabetes, and foetal growth restriction.
References [1] Hytten FE, Thomas AM. Maternal physiological adjustements. In: Maternal Nutrition and Course of Pregnancy. National Academy of Science, Washington, DC, 1970. [2] McCartney CP, Pottinger RE, Harrod JP. Alternations in body composition during pregnancy. Am J Obstet Gynecol 1959;77: 1038±53. [3] Hutchinson D, Plentl A, Taylor HC. The total body water and the water turnover in pregnancy studied with deuterium oxide as isotopic tracer. J Clin Invest 1954;33:235±41. [4] Hytten F. Blood volume changes in normal pregnancy. Clin Haematol 1985;14:601±12. [5] Pirani BBK, Campbell DM, MacGillivray I. Plasma volume in normal first pregnancy. J Obstet Gynaecol Br Commonwealth 1973;80:884±7. [6] Duffus GM, MacGillivray I, Dennis JK. The relationship between baby weight and changes in maternal weight, and proteins and urinary oestriol excretion. J Obstet Gynaecol Br Commonwealth 1971;78:97±104. [7] Langhoff-Ross J, Lindmark G, Gebre-Medin M. Maternal fat stores and fat accretion during pregnancy in relation to infant birthweight. Br J Obstet Gynaecol 1987;94:1170±7. [8] Rosso P, Donoso E, Braun S, Espinoza R, Fernadez C, Salas SP. Maternal hemodynamic adjustements in idiopathic fetal growth retardation. Gynecol Obstet Invest 1993;35:162±5. [9] Valensise H, Andreoli A, Lello S, Magnani F, Romanini C, De Lorenzo A. Multifrequency bioelectrical analysis in women with a normal and hypertensive pregnancy. Am J Clin Nutr 2000;72: 780±3. [10] Walser M, Seldin DW, Grollman A. An evaluation of radiosulfate for the determination of the volume of extracellular fluid in man and dogs. J Clin Invest 1953;32:299±311. [11] Wong WW, Cochran WJ, Klish WJ, O'Brien Smith E, Lee LS, Kleim PD. In vivo isotope-fractionation factors and the measurement of deuterium and oxygen-18 dilution spaces from plasma, urine, saliva, respiratory water vapors, and carbon dioxide. Am J Clin Nutr 1988;47:1±6. [12] Lukaski HC, Johnson PE. A simple, inexpensive method of determining total body water using a tracer dose of D2O and infrared absorption of biological fluids. Am J Clin Nutr 1985;41:363±70. [13] Fjeld CR, Freundt-Thurne J, Scoeller DA. Total body water measured by 18O dilution and bioelectrical impedance in well and malnourished children. Pediatr Res 1990;27:98±102. [14] Kushner RF, Kunigk A, Alspaugh M, Andronis PT, Leitch CA, Schoeller DA. Validation of bioelectrical impedance analysis as a measurement of changes in body composition in obesity. Am J Clin Nutr 1990;52:219±23. [15] Tedner B, Lins LE. Fluid volume monitoring with electrical impedance technique during hemodialisis. Artificial Org 1984;8: 66±71.
176
F. Ghezzi et al. / European Journal of Obstetrics & Gynecology and Reproductive Biology 98 (2001) 171±176
[16] Zillikens MC, van de Berg JWO, Wilson JHP, Swart GR. Wholebody and segmental bioelectrical impedance analysis in patients with cirrhosis of the liver: changes after treatment of ascites. Am J Clin Nutr 1992;55:621±5. [17] Guglielmi FW, Contento F, Laddaga L, Panella C, Francavilla A. Bioelectric impedance analysis: experience with male patients with cirrhosis. Hepatology 1991;13:892±5. [18] Lukaski HC, Siders WA, Nielsen EJ, Hall CB. Total body water in pregnancy: assessment by using bioelectrical impedance. Am J Clin Nutr 1994;59:578±85. [19] van Oppen AC, van der Tweel I, Alsbach GP, Heethaar RM, Bruinse HW. A longitudinal study of maternal hemodynamics during normal pregnancy. Obstet Gynecol 1996;88:40±6. [20] Morita H, Takeuchi K, Funakoshi T, Mizutori M, Maruo T. Potential use of bioelectrical impedance analysis in the assessment of edema in pregnancy. Clin Exp Obstet Gynecol 1999;26:151±4.
[21] Lukaski H, Bolonchuk W. Estimation of body fluid volumes using tetrapolar bioelectrical impedance measurements. Aviat Space Environ Med 1988;59:1163±9. [22] Segal K, Wang J, Pierson R, Van Itallie T. Estimation of body water distribution by bioelectrical impedance. Fed Proc 1987;46: 1334. [23] Villar J, Cogswell M, Kestler E, Castillo P, Menendez R, Repke J. Effect of fat and free mass deposition during pregnancy on birth weight. Am J Obstet Gynecol 1992;167:1344±52. [24] Mardones-Santander F, Salazar G, Rosso P, Villaroel C. Maternal body composition near term and birth weight. Obstet Gynecol 1998;91:873±7. [25] Lederman SA, Paxton A, Heymsfield SB, Wang J, Thornton J, Pierson Jr RN. Maternal body fat and water during pregnancy: do they raise infant birth weight? Am J Obstet Gynecol 1999;180: 235±40.