Objectives: The gestational weight gain chart published by Rosso and Mardones (RM) was used by the Ministry of Health in Argentina from the 1990s until 2009, when it was replaced by the Calvo et al. chart (CEA). This study compared the diagnostic capability of the RM and CEA charts for identifying mothers at risk of delivering babies with signs of intrauterine growth impairment (Birth length<50 cm, Birth weight <2500 g, Birth weight <3000 g and Birth weight >4000 g).

Methods: Data from pregnant women and neonates studied in Santiago, Chile (n = 27,600), as well as a sub-sample of 11,465 term healthy singleton cases were utilized. Frequency distributions of women and their 95% confidence intervals were calculated for both charts at the first and third trimester of pregnancy. Sensitivity, specificity, positive and negative predictive values of undesirable outcomes of the RM and CEA charts were compared.

Results: The CEA chart classified as either underweight or obese a smaller number of women than the RM chart. The proportion of neonates with signs of intrauterine growth impairment was similar in both charts, however the number of affected neonates was higher in the women considered to be either underweight or obese by the RM chart. Higher sensitivity values for undesirable outcomes were displayed by the RM chart. The sum of sensitivity and specificity was also higher for the RM chart, meaning a greater diagnostic accuracy; this sum was comparable only for birth weight <2500 g. The RM chart was found to have higher positive and negative predictive values for undesirable outcomes than the CEA chart.

Conclusion: The RM chart has a better diagnostic capability than the CEA chart to identify mothers at risk of delivering babies with signs of intrauterine growth impairment.

Keywords: gestational weight gain, growth charts

RM, Rosso and Mardones; CEA, Calvo et al; BMI, body mass index; BW, birth weight; GA, gestational age

Maternal body mass index (BMI) at conception and gestational weight gain significantly influence birth weight (BW).¹ Both, obese and underweight women have a higher probability of delivering babies either large or small for gestational age, respectively. Thus, nutritional counseling before and during pregnancy may have a significant impact on neonatal outcomes.² Based on data from a population of Chilean gravidas and using BMI as an indicator of maternal nutritional status, we developed a chart, named Rosso-Mardones (RM), to monitor adequacy of maternal nutritional status throughout pregnancy.³ This chart was used by the Chilean National Health Systems until 2004, when a study by Atalah et al.⁴ proposed a wider “normalcy” area for at term maternal BMI.⁴

Currently, most Latin American countries are using either the original RM chart or the modification of this instrument proposed by Atalah et al.⁵ One of the exceptions is Argentina, where in 2009 the Ministry of Health replaced the RM chart by a chart proposed by Calvo et al. (CEA). In the United States and in many countries of the Northern Hemisphere, the U.S. Institute of Medicine (IOM) guidelines have been used since the nineties.⁶ Thus, a global consensus on desirable gestational weight gain for a given maternal weight/height in early pregnancy is still lacking.⁵

The RM chart has shown a greater sensitivity to identify women at nutritional risk of delivering babies with signs of fetal growth impairment than both the US IOM recommendations and the Atalah chart.⁷ Consequently, a percentage of women misdiagnosed as “normal” failed to receive adequate nutritional counseling regarding their desirable gestational weight gains.        

In the present study we compared the diagnostic capability of the RM chart and the maternal weight gains chart used by the Argentinian Ministry of Health (CEA chart)^8–10 for identifying mothers at risk of delivering babies with signs of intrauterine growth impairment.

The study was performed using a data base containing information from deliveries at the maternity unit of Hospital Dr. Sotero del Rio, in the South-East Public Health Service of Santiago, Chile.¹¹ which serves a mixed population of low-income and middle-income families.¹² Genetic studies have shown that in the Chilean population 44 per cent of the genes are indigenous American, 52 percent European and 4 per cent African.¹³ Data analysis contributing to this work comply with the ethical standards of the relevant national guidelines on human experimentation provided by the Chilean National Fund for Research in Science and Technology (FONDECYT) and with the World Medical Association's Declaration of Helsinki. The study was approved by the Institutional Ethics Committee (Project ID 12-331).

The general study group included all individual clinical records with singleton live births during the recruitment period from January 22, 2000, to April 24, 2004. From an initial number of 28,898 cases evaluated, a total sample of 27,600 cases with complete anthropometric and health information was considered. 1,298 cases were excluded due to maternal height ⩽130 cm or ⩾200 cm or lacked the required maternal weight data at recruitment. In addition, a sub-sample of 11 466 healthy pregnancies were selected to ascertain the effect of maternal nutritional status on fetal growth excluding the potential effect of confounding factors. Inclusion criteria for the sub-sample included: maternal age over 18 years old, term deliveries (39–41 weeks of gestation), without medical and obstetrical conditions with known effect on fetal growth, and absence of alcohol or drugs consumption. Term deliveries were defined as those with 39-41 weeks of gestation considering that births before 39 weeks and over 41 weeks of gestational age (GA) are associated with adverse child health outcomes.^14–16 Moreover, births before 39 weeks are associated with chronic diseases in adults.¹⁷ The 39–41 weeks GA category constitutes the ideal and typical duration of pregnancy.¹⁸

Nutritional classification of the women at the beginning and at the end of pregnancy was performed utilizing both the RM and CEA charts. The BMI cut-off points for the two charts at week 10 are the following: underweight category for the RM chart < 21.15 and for the CEA chart <20.00; normal category for the RM chart 21.15-24.49 and the CEA chart 20.00-27.2; overweight category for the RM chart 24.50-26.73 and the CEA chart 27.30-35.70; obese category for the RM chart > 26.73 and the CEA chart > 35.70. At week 40 are the following: underweight category for the RM chart < 26.55 and for the CEA chart <24.60; normal category for the RM chart 26.55-28.90 and the CEA chart 24.60-32.59; overweight category for the RM chart 28.91-30.03 and the CEA chart 32.60-39.60; obese category for the RM chart > 30.03 and the CEA chart > 39.60.

Considering that those extremes of BW, i.e. <2500 g, <3000 g, and >4000 g, are related to increased mortality and morbidity risks during the first month and first year of life, neonates at risk were defined with those values.^14,19–21 In addition, we utilized BL <50 cm as a proxy for shortness at birth, which has been associated with obesity, high blood pressure, insulin resistance, and lower school achievement scores in Chilean children.^22–25

Maternal anthropometry was measured with beam scales (Condor^®, Santiago, Chile) at prenatal health visits and the maternity unit, following a standard protocol.¹¹

Pre-pregnancy weight was reported by patients. The start point of nutritional status classification was 10 weeks of gestation for both gestational weight gain charts. Considering the modest weight change typically observed during the first weeks of pregnancy, chart comparison would not be affected by the use of similar pre-pregnancy weight to estimate its value at the beginning of gestation. Instructions for using the CEA chart have been published in 2009, 2012 and 2013.^9,26,27

Maternal blood pressure was measured utilizing calibrated sphygmomanometers, after seating and resting for at least 15 min. Blood samples were obtained at first trimester for blood group classification, rhesus status, glycaemia and rapid plasma reagin test for syphilis. A urine sample was also included for urine sediment analysis. A second syphilis test, as well as a glucose tolerance test, were performed at week 28 of gestation. Pre-eclampsia was diagnosed by the presence of hypertension, edema, and proteinuria, confirmed by laboratory analysis of a 24-hour urine collection (more than 300 mg/day). Plasma urea concentration and creatinine clearance were also measured. Appropriate high-risk care was assured for all women with significant pathology.

Gestational age was based on the date of the last menstrual period and confirmed by fetal biometry using trans-vaginal ultrasound (first trimester) or trans-abdominal ultrasound (second trimester), performed by trained personnel. Ultrasound equipment included: Voluson 730 PRO (GE Healthcare, Chalfont St. Giles, UK); Acuson 120XP (Acuson Inc., Mountain view, CA, USA).

Mothers and neonates were evaluated and diagnosed by the attending physicians. After delivery, infants were dried and weighed using an electronic self-calibrating scale Tanita 1583 (Tanita Corporation, Arlington Heights, IL, USA, accurate to 10 g), or Seca 345 (Secacorp, Hamburg, Germany, accurate to 20 g). A custom-made neonatometer was used to measure the crown-heel length to the nearest 1 mm.

Proportions of pregnant women in each nutritional category with 95% confidence intervals (95%CI) were calculated at the beginning and the end of pregnancy for both charts. McNemar’s χ² test was used to investigate the potential differences in the proportions of the four main target events, BL <50 cm, BW <2500 g, BW <3000 g and BW >4000 g, at each maternal nutritional category of the RM and CEA charts. Sensitivity and specificity values for both charts were obtained using the four target events: BL <50 cm, BW <2500 g and BW <3000 g for underweight women and BW >4000 g for obese women. Positive and negative predictive values (PPV and NPV) were also estimated. A p value < 0.05 was considered statistically significant. All analyses were performed using Phyton 3.0 for Windows.

The total sample and the sub-sample of healthy pregnant women did not differ in mean ± SD values of demographic and anthropometric characteristics (Table 1). Nevertheless, the sub-sample of healthy pregnant women had significantly lower proportions of neonates with signs of intrauterine growth impairment: BW<2500 g, BW <3000 g, and BL <50 cm, than the total sample: 6.50 (95%CI 6.23-6.82) versus 0.61 (95%CI 0.47-0.75); 9.24 (95%CI 8.72-9.78) versus 15.12 (95%CI 14.70-15.55) and 34.11 (95%CI 33.24-34.98) versus 46.91 (95%CI 46.32-47.49). In addition, the sub-sample of healthy pregnant women had a significantly higher proportion of BW >4000 g than the total sample: 11.74 (95%CI 11.15-12.33) versus 9.14 (95%CI 8.80-9.48).

The number of women classified as underweight and obese in the sub-sample of 11,465 healthy pregnant cases was higher for the RM chart than the CEA chart. The number of women included in the normal and overweight categories, both at the beginning and the end of pregnancy, was higher for the CEA chart than for the RM chart (Table 2).

Percentages of neonates with BL <50 cm, BW <2500, BW <3000 g and BW >4000 g observed in each maternal nutrition category were similar for the RM and the CEA charts, both at the beginning and the end of pregnancy in the total sample of 27,600 women. None of the differences observed between the proportions were statistically significant (Table 3). However, the absolute numbers of newborns with signs of intrauterine growth impairment were higher in women categorized as underweight or obese by the RM chart than those classified by the CEA chart.

Sensitivity values were higher for the RM chart for all the performed comparisons (Table 4). Although the specificity values were higher for the CEA chart, the sum of sensitivity and specificity values was higher for the RM chart. PPV and NPV were also higher for the RM chart in all comparisons (Table 4).

Similar to the RM chart, the CEA chart was developed using weight for height values, i.e., BMI, for the entire gestation.^3,8 However, the CEA and the RM charts differ significantly with respect to the criteria applied to define the limits of the area of “adequate BMI” or “normal” maternal nutrition status. In the RM chart the area of normal BMI throughout pregnancy was based on obstetric outcomes from 1,745 healthy adult women and their singleton babies delivered at 39–41 weeks of gestation.³ In contrast, the CEA chart used just percentile values, with extremely low or extremely high BMI values defining underweight or overweight mothers.⁸ This approach led to a definition of a normal area proportionally much larger than in the RM chart.^3,8 Consequently, a percentage of underweight and obese women were considered nutritionally “normal” despite being at risk of an impaired fetal growth.

The percentages of newborns with length <50 cm, weight <2500 g or <3000 g and weight >4000 g diagnosed by the two charts were similar. However, the absolute number of newborns with signs of intrauterine growth impairment were higher in women considered to be underweight or obese by the RM chart than the CEA chart. The latter is consistent with the fact that sensitivity values, were higher in the RM chart than in the CEA chart. The higher sensitivity of the RM chart than the CEA chart means that in a higher extent women presenting the diagnosis of underweight or obese truly deliver BW <2500 g, BW <3000 g or BW >4000 g, respectively.²⁸

Thus, the CEA chart failed to identify women who might had benefited from nutritional counseling during gestation. The public health relevance of this finding is stressed by the fact that babies who have suffered intrauterine growth impairment have a higher incidence of neonatal morbidity and mortality and the consequent expenses in health care.²⁹

The RM chart also performs better than the CEA chart in almost all outcomes when values of the sum of sensitivity and specificity are compared. The only exception was BW < 2500 g when the sum of those indicators in week 10 had a value of 1 meaning that for this outcome the test was not useful, but the expected gain in diagnostic certainty increases as the values of this sum increases.³⁰ In addition, all PPV and NPV values were higher in the RM chart, thus indicating a higher accuracy in disease diagnosis.³¹ From an epidemiological point of view, the adequate recognition of a subset of pregnant women associated with a higher number of at-risk neonates with altered BW and BL using a diagnostic instrument such as the RM chart is preferable.

Since experimental studies concerning maternal nutrition have many practical difficulties, most comparisons among pregnant women cohorts have been done without randomization. From this perspective, present results have the limitation that they were obtained in an observational study. However, the intergroup comparison was made using a high number of subjects from the same population of women. Although the population selected for this study is from Chile, anthropometric information for pregnant women 10-49 years old has been reported as similar in Argentina with data from two important public maternity hospitals located in the capital cities of both countries -Buenos Aires and Santiago-, showing that mean and SD values for height were the same, presenting 1.56 ± 0.06 m; obesity was slightly higher in Chile.³²

To better assess the influence of maternal nutrition on neonatal anthropometry, we analyzed a subsample of healthy-term mothers. Interestingly, it produced a new nutritional distribution of women, probably due to the exclusion of women with preterm deliveries and other gestational morbidities. Thus, while the proportion of women classified as undernourished by both gestational weight charts was higher in this sub-sample of healthy women, the proportion of women with obesity was lower. Notably, this strategy allows us to identify underweight during pregnancy as an important factor to predict risky BL and BW newborns in a clearer way than in the total sample.

Prenatal activities to promote adequate fetal growth in women diagnosed as underweight and obese are primary aims for health personnel in Latin American countries. Those activities include weight gain monitoring, quality and quantity of diet surveillance, nutritional education, and iron and folic acid supplementation.^26,27 An increase in the frequency of babies with BW <3000 g and BL <50 cm has been observed in Chile after 2005, the year when the RM Chart was replaced by the AEA Chart,³³ suggesting a causal relationship. The discontinuation of the RM guidelines has probably meant a decreased ability to diagnose and intervene in subsets of pregnant women at risk of having neonates with altered size at birth.

The selection of weight gain charts for pregnant women must be based in fetal outcomes as the key parameter for defining “normal” areas of maternal BMI at the beginning and at the end of gestation. As shown in this and previous studies, to underestimate the impact of a maternal low BMI or an excessive BMI determines significant differences in the risk of delivering babies with either low or excessive birth weight. In turn, birth weight <2500 g and <3000 g and height <50 cm at birth, or birth weight >4000 g implies a higher risk of morbidity and mortality.^19–21 In addition, altered fetal growth is associated with a higher risk of developing chronic diseases such obesity, arterial hypertension, and diabetes later in life.^19,34

The author RC-C was funded by Project code LPR23-17, year 2023, Universidad Tecnológica Metropolitana.

None.

The authors declare that they have no competing interests.

1. Rosso P. A new chart to monitor weight gain during gestation. American J Clin Nutr. 1985;41(3): 644–6
2. Gluckman P, Hanson M, Seng CY, et al. Pre–conception maternal body composition and gestational weight gain. In: Gluckman P, Hanson M, Seng CY, editors. Nutrition and Lifestyle for Pregnancy and Breastfeeding. Oxford: Oxford University Press; 2014:216–220.
3. Mardones F, Rosso P. A weight gain chart for pregnant women designed in Chile. Matern Child Nutr. 2005;1(2):77–90.
4. Atalah E, Castillo C, Castro R, et al. Proposal of a new standard for the nutritional assessment of pregnant women. Rev Med Chil. 1997;125(12):1429–1436.
5. Scott C, Andersen CT, Valdez N, et al. No global consensus: a cross–sectional survey of maternal weight policies. BMC Pregnancy Childbirth. 2024;14:167.
6. National Research Council. Determining optimal weight gain. In: Yaktine AL, Rasmussen KM, editors. Weight Gain During Pregnancy: Reexamining the Guidelines. Washington DC: Institute of Medicine, National Academy Press; 2009:241–262.
7. Mardones F, Rosso P, Erazo Á, et al. Comparison of three gestational weight gain guidelines under use in Latin America. Front Pediatr. 2021;9:744760.
8. Calvo EB, López LB, Balmaceda Ydel V, et al. Reference charts for weight gain and body mass index during pregnancy obtained from a healthy cohort. J Matern Fetal Neonatal Med. 2009;22(1):36–42.
9. Abeyá E, Calvo E, Durán P, et al, editors. Evaluation of the nutritional status of girls, boys and pregnant women. Buenos Aires, Argentina: Ministry of Health of the Argentine Nation; 2009. 144 p.
10. Calvo E, López L, Balmaceda Y, et al. Weight gain and food selection patterns during pregnancy and their impact on the newborn]. Buenos Aires, Argentina: Ministry of Health of the Argentine Nation; 2013:13–58.
11. ​Mardones F, García–Huidobro T, Ralph C, et al. Combined influence of preconception body mass index and gestational weight gain on fetal growth. Rev Med Chil. 2011;139(6):710–716.
12. Mardones F, Arnaiz P, Barja S, et al. Nutritional status, metabolic syndrome and insulin resistance in children from Santiago (Chile). Nutr Hosp. 2013;28(6):1999– 2005.
13. Fuentes M, Pulgar I, Gallo C, et al. Gene geography of Chile: regional distribution of American, European and African genetic contributions. Rev Med Chil. 2014;142(3):281–289. 
14. Thomson AM. VII. The implications of fetal growth. In: Barron SL, Thomson AM, editors. Obstetrical epidemiology. London: Academic Press; 1983:130–131.
15. Raju TNK, Higgins RD, Stark AR, et al. Optimizing care and outcome for late–preterm (near–term) infants: a summary of the workshop sponsored by the National Institute of Child Health and Human Development. 2006;118(3):1207–1214.
16. Delnord M, Zeitlin J. Epidemiology of late preterm and early term births – An international perspective. Semin Fetal Neonatal Med. 2019;24(1):3–10.
17. Kajantie E, Strang–Karlsson S, Evensen KAI, et al. Adult outcomes of being late preterm or early term – What do we know? Semin Fetal Neonatal 2019;24(1):66–83.
18. Pan American Health Organization and Texas Children’s Hospital, editors. Interventions Based on Evidence. 2nd edn. Washington, DC: Pan American Health Organization. 2010. 87 p.
19. Blencowe H, Krasevec J, de Onis M, et al. National, regional, and worldwide estimates of low birthweight in 2015, with trends from 2000: a systematic analysis. Lancet Glob Health. 2019;7(7):e849–e860.
20. Serrano CV, Puffer RR. Utilization of hospital birth weights and mortality as indicators of health problems in infancy. Bull Pan Am Health Organ. 1974;8(4):325–346.
21. Puffer RR, Serrano CV. Birth weight distribution among live births. In: Puffer RR, Serrano CV, editors. Patterns of birth weight. (Scientific Publication No. 504). Washington DC: Pan American Health Organization; 1987:9–44.
22. Mardones F, Villarroel L, Karzulovic L, et al. Association of perinatal factors and obesity in 6–8 years old Chilean children. Int J Epidemiol. 2008;37(4):902–910.
23. Mardones F, Villarroel L, Arnaiz P, et al. Prenatal growth and metabolic syndrome components among Chilean children. J Dev Orig Health Dis. 2012;3(4):237–244.
24. Villarroel L, Karzulovic L, Manzi J, et al. Association of perinatal factors and school performance in primary school Chilean children. J Dev Orig Health Dis. 2013 ;4(3):232–238.
25. Mardones F, Arnaiz P, Pacheco P, et al. Associations of prenatal growth with metabolic syndrome, insulin resistance and nutritional status in Chilean children. Biomed Res Int. 2014;2014:472017.
26. Suarez MM, López LB. Argentinian BMI charts for gestational age. In: Suarez MM, López LB, editors. Healthy alimentation. Buenos Aires, Argentina: Ed. Akadia; 2012:66–67.
27. National Office of Maternity and Infancy, Minstry of Health, Argentina. Recomendations for the practice of the preconceptional, prenatal and puerperal health check–ups. Buenos Aires, Argentina: Ministry of Health; 2013:39–42.
28. MacMahon B, Pugh TF. Sensitivity and specificity. In: MacMahon B, Pugh TF, editors. Epidemiology. Boston: Little, Brown and Company; 1970:261–262.
29. Rose G. Sick individuals and sick populations. Int J Epidemiol. 2001;30(3):427–432.
30. Connell FA, Koepsell TD. Measures of gain in certainty from a diagnostic test. Am J Epidemiol. 1985;121(5):744–753.
31. Mallett S, Halligan S, Thompson M, et al. Interpreting diagnostic accuracy studies for patient care. 2012;345:e3999.
32. Mardones F, Grandi C, Moratorio X. [Fetal life and healthy growth. Healthy growth. Between malnutrition and obesity in the Southern Cone. In: Uauy R, Carmuega S, editors. Buenos Aires, Argentina: Danone Civil Association for Nutrition, Health and Quality of Life; 2012:153–174.
33. Mardones F, Acuña J. Evolution of perinatal outcomes and sociodemographic variables in Chile (1996–2017). J Dev Orig Health Dis. 2021;12(5):788–797.
34. Barker DJP. Past obstacles and future promise. In: Gluckman P, Hanson M, editors. Developmental origins of health and disease. Cambridge: Cambridge University Press; 2006:481–495.