SEARCH

SEARCH BY CITATION

Keywords:

  • symphysis-fundus height;
  • BMI;
  • overt morbidity;
  • pre-term delivery;
  • foetal growth;
  • Mozambique

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Our aim was to construct a new symphysis-fundus height (SFH) growth chart, based on Mozambican women with ultrasound-dated singleton pregnancy, who represent the largest obstetric cohort in a developing country followed for this purpose. Two antenatal clinics were chosen in the suburban area of Maputo City. A cohort of 904 consecutively recruited antenatal clients was followed until delivery. The growth of the SFH was measured every second to third week. Gestational age was determined by ultrasound at enrolment. Women with multiple pregnancy or with gestational age > 21 weeks at enrolment were excluded. The average number of antenatal SFH measurements per woman was 7.8 (SD 2.4). The drop out rate was 9.6%. Mean birthweight was 2909 g. Pre-term deliveries occurred in 15% and low birthweight deliveries (< 2500 g) in 16%. Using proper longitudinal methods, we constructed an FH growth chart and compared it with various previously published SFH charts, which showed the Mozambican chart to be 0–3 cm below the others. Nulliparous women were 0.5 cm below multiparous women. We did not find any difference in the SFH growth charts between women with or without overt morbidity. Women with a body mass index (BMI) < 19 and women with a BMI > 27 had approximately 1 cm lower and 1 cm higher readings, respectively, than women with normal BMI. The Mozambican SFH growth chart is an example of an elaborated growth chart for a well-defined population in alow-income country. It constitutes the basis for further studies to predict the small-for-gestational age newborn from anthropometrical data obtained by use of appropriate technology.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Interventions with the purpose to reduce perinatal morbidity and mortality rates in low-income countries should focus upon early detection and management of recognizable risk factors affecting maternal and foetal health. To achieve such interventions, there are few affordable approaches available. Various anthropometric methods, such as measurement of maternal mid-upper arm circumference (Liljestrand 1989; Walraven et al. 1995b), height (van Roosmalen & Brand 1992; Moller & Lindmark 1997; Osman et al. 2001) and symphysis-fundus distance (SFH) (Westin 1977; Persson et al. 1986; Bergstrom & Liljestrand 1989) have been introduced in antenatal care. There are appreciable differences between SFH growth charts in different countries according to method used and presumably because of ethnic and socio-economic differences (Ogunranti 1990; Walraven et al. 1995a). The reviewers' conclusion in the Cochrane database (one trial involving 1639 women was included) was that there is not enough evidence to evaluate the usefulness of SFH measurements during antenatal care (Neilson 2000).

There is a need for a standard SFH growth chart for a specific population in a specific region. At a state-of-the art conference in Sweden 1990, one conclusion was that such a standard growth chart should be based on a large and geographically well-defined population with ultrasound-dated pregnancies (Lindmark & Cnattingius 1992). Some studies have resulted in SFH growth charts from different affluent countries (Westin 1977; Persson et al. 1986; Jensen & Larsen 1991; Steingrimsdóttir et al. 1995; Kieler et al. 1996). Little recent information is, however, available from low-income countries (Kiserud 1986; Mathai et al. 1987; Munjanja et al. 1987; Pattinson 1988; Bergstrom & Liljestrand 1989; Medhat et al. 1991; Walraven et al. 1995a).

In Mozambique, substantial attention has been drawn to methods to improve perinatal outcome (Bergstrom & Liljestrand 1989; Axemo 1995). Some reports have been published on the use of SFH measurements in the country (Bergstrom & Libombo 1992; Bique Osman et al. 1993), but there is no SFH standard growth chart in the country. Therefore, the aim of this study was to follow the hitherto largest obstetric African cohort through pregnancy to establish a representative SFH growth chart.

Subjects and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study population

Two antenatal clinics in the semirural area of Maputo City were chosen for this study. The populations covered by these clinics were considered representative of the vast majority of the pregnant city population. In spite of being semirural, both areas offered conditions for follow-up and for localization of dropouts in order to minimize poor compliance. At each antenatal clinic there were always the same two to three midwives doing the symphysis-fundus measurements, a circumstance shown to be important to reduce interobserver variation (Pattinson & Theron 1989).

A pilot study of 50 women was conducted in order to give logistical information for the main study. During the period October 1993 to October 1994, women attending the two antenatal clinics were consecutively invited to participate in the cohort and were registered for antenatal care. They were estimated to have gestational ages of14–21 weeks by menstrual dates and were followed from enrolment to the first week postpartum. To enhance the accuracy of gestational age determination, ultrasound was used to measure the biparietal diameter (BPD) at enrolment. Women with foetal BPD > 50 mm and multiple pregnancies were excluded, resulting in 904 remaining women. All women were encouraged to return every2–3 weeks for measurement of the SFH. The SFH growth chart was drawn with 5, 50 and 95th percentiles per gestational week by a cubic regression model and compared with other SFH growth charts (Westin 1977; Belizán & Villar 1978; Persson et al. 1986; Mathai et al. 1987; Munjanja et al. 1987; Steingrimsdóttir et al. 1995; Hakansson et al. 1995).

For the SFH measurements, repeated special training for the observers (midwives) was arranged in order to comply with the norms suggested by Westin (1977). The women were asked to empty the bladder before measurement.

A subgroup without overt morbidity was selected from the cohort by excluding women with pre-term(< 37 completed gestational weeks) and post-term(> 42 completed gestational weeks) birth, vaginal bleeding after 20 weeks of gestation, placenta praevia, malformed or myomatous uterus, hypertension (140/90), severe systemic disease, anaemia (< 10 g/dL), syphilis (positive VDRL) or malaria (positive slide). Parturients with newborns with an Apgar score < 7 (at 1 and 10 min), malformation or significant neonatal disease at control on day 7 were also excluded. With these criteria 421 women were selected.

Previous SFH studies (Westin 1977; Munjanja et al. 1987; Steingrimsdóttir et al. 1995) have shown that heavy women have a higher SFH throughout pregnancy. In order to have a more correct predictive value of SFH measurements in overweight and underweight pregnant women, respectively, we made SFH graphs for three groups of pregnant women with low (< 19 corresponding to 10th percentile), normal and high (> 27 corresponding to 90th percentile) body mass index (BMI). The weight and height were measured at first antenatal visit. To assess the influence of nulliparity on the SFH graph, nulliparous and parous women were analysed separately.

Pre-term birth in the setting studied is associated with infection pathology (Bique Osman et al. 1993). Therefore, we constructed two separate SFH graphs for the women with pre-term vs. term and post-term birth. Besides the SFH growth chart for all women in the cohort we thus made separate SFH growth charts on the following subpopulations: (1) all cohort women vs. women without overt morbidity; (2) all cohort women vs. women with high and low BMI; (3) nulliparous vs. parous women and (4) term parturients vs. pre-term and post-term parturients. All enrolled women were encouraged to deliver at the Maputo Central Hospital.

Statistical methods

Only women with at least two SFH measurements between 15 and 42 weeks were included (n=817). Each measurement consists of a pair – day of pregnancy at measurement and SFH in centimetre. For each woman a straight line was drawn from the first measurement in time to the second to the third, and so on. Then for each day in the gestational interval the SFH values for all women were arranged, from the lowest to the highest, and the 5, 50 and 95th percentiles were calculated. After rounding off the values, smoothed continuous lines were fitted by a cubic regression model to get the three curves.

Each woman contributed 2–12 measurements. This method implies that the contribution from each woman to the curve is because of the length between the first and the last measurement and independent of the number of measurements.

We used Epi-Info version 6.02 (CDC, Atlanta, USA), SPSS version 10.1 (Scandinavia AB, Stockhom, Sweden) and Mathematica version 4.1 (© 2001 Wolfram Research Inc., USA) software.

The ethical committees at the Faculty of Medicine of the Universidade Eduardo Mondlane and at the Maputo Central Hospital approved the project.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The drop-out rate (SFH measurements < 2) was 9.6% (87/904). The mean number of SFH measurements per woman was 7.8 (SD=2.4). In all, 6544 SFH measurements were taken. Mean birthweight was 2909 g, low birthweight (< 2500 g) occurred at a rate of 16%. Pre-term deliveries occurred in 128 (15%) out of 832 women for whom gestational age at delivery was estimated. The proportion of nulliparous and parous women was 42 and 58%, respectively. The nutritional status was reflected in anthropometric characteristics (Table 1). Mean BMI was 22.9 (SD=8.2) and 6.1% had BMI < 19.0. The mean birthweight in women with BMI below 19 was significantly (P < 0.01) lower than in women with a normal BMI(19–26) and the mean birthweight in women with a BMI of 27 and above was significantly (P < 0.01) higher (Table 2). The SFH growth for 817 ultrasound-dated singleton Mozambican pregnancies from 20 to 41 gestational weeks, with 5, 50 and 95th percentiles, fitted by the cubic regression model is presented in Figure 1. The median SFH values from Figure 1 were tabulated together with corresponding values in 10 other studies for comparison (Table 3). Additional SFH growth charts were drawn for subpopulations taking into account the following features: (1) Overt morbidity (Figure 2): There was no difference between women with and without overt morbidity (n=421), except for those in 95th percentile at term. (2) Nulliparity (Figure 3): There were no differences in median SFH measurements between p–37 when parous women had approximately 0.5 cm higher median. (3) BMI (Figure 4): For women with a BMI > 27 the SFH growth graph lies approximately 1 cm above the graph for the whole study population while the graph for women with a BMI < 19 lies approximately 1 cm below. The SFH differences reflect the differences in birthweight in the different BMI groups (Table 1). (4) Gestational age at birth: Having a pre-term outcome did not influence the SFH growth chart for any category.

Table 1.  Anthropometric characteristics of the study population (n 904) Thumbnail image of
Table 2.  Mean birthweight (g) by body mass index (BMI) Thumbnail image of
image

Figure 1. Symphysis-fundus height growth chart based on 817 ultrasound-dated singleton Mozambican pregnancies with 5, 50 and 95th percentiles, fitted by a cubic regression model.

Download figure to PowerPoint

Table 3.  Comparison of standard symphysis-fundus height growth charts (cm) from various studies Thumbnail image of
image

Figure 2. Symphysis-fundus height growth chart based on 817 ultrasound-dated singleton Mozambican pregnancies (intact lines) and superimposed 421 selected `healthy' pregnancies (interrupted lines) with 5, 50 and 95th percentiles, fitted by a cubic regression model.

Download figure to PowerPoint

image

Figure 3. Symphysis-fundus height growth chart for nulliparous (n=337) (interrupted lines) and (n=480) (dotted line) women, fitted by a cubic regression model and with 5, 50 and 95th percentiles.

Download figure to PowerPoint

image

Figure 4. Symphysis-fundus height growth chart for BMI < 19 (dotted lines), and all women (intact lines). Fitted by cubic regression model and with 50th percentiles.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

To our knowledge, this is the largest obstetric cohort in a low-income country followed throughout pregnancy to establish the SFH growth pattern. The sample studied is considered representative at the community level with the exception of parity (Bique Osman et al. 1999). Ultrasound measurement of BPD for estimation of gestational age requires antenatal booking early enough to make BPD measurements meaningful. As multiparous women tend to register later, nulliparous women are over-represented in the sample. It may be argued that this bias would render the graph obtained non-representative of the background population. It is, however, obvious that proper dating by ultrasound in any pregnant population is dependent on early registration. Because the use of SFH graphs for diagnostic purposes requires optimal determination of gestational age, it would seem that the standard population defined is representative of a healthy group of women registering early for antenatal care. Our graph is based on ultrasound-dated pregnancies and the sample size is to our knowledge the largest one reported in those SFH studies which have been published from low-income countries (Table 3). In comparison with studies from Argentina, Zimbabwe, Tanzania and India, our graph differs both by being ultrasound-dated and by having a less selected study population.

Comparisons of the Mozambican SFH graph with other similar graphs (Table 3) indicate that an American graph is situated approximately 3 cm above and an Indian graph approximately 2 cm below. The neighbouring country, Zimbabwe, had a similar graph except at 40 gestational weeks, when the Zimbabwean women had an average of 36.7 cm. A comparable Swedish graph (Hakansson et al. 1995) is, however, almost identical.

The SFH growth chart from Cardiff (Calvert et al. 1982) has been recommended for use in developing countries (King & Cairas 1990). However, it fits rather badly in comparison with the curves from Tanzania and India (Table 3). The use of SFH growth charts based on measurements in the local population is recommended (Walraven et al. 1995a).

The differences between the SFH growth charts published from various countries (Table 3) may be because of different methods used to elaborate the charts, to ethnic, nutritional and social differences in the population samples drawn, and to sample size. For example, Westin and Steingrimsdottir have very strict criteria of inclusion in their studies. To use similar criteria in low-income countries may not be meaningful. It is, for instance, obvious that the burden of infectious diseases is much more significant in a deprived Mozambican population than in an affluent American one. Furthermore, many of these infections are never diagnosed or are subclinical and escape detection (Naeye 1981; Axemo 1995; Osman et al. 1995; Folgosa et al. 1997). Women suffering from them cannot be excluded and are thereby erroneously considered free from disease.

From our anthropometric data it appears that Mozambican women do not deviate significantly from other African populations concerning weight, height, BMI and arm circumference (Kiserud 1986; Munjanja et al. 1987; Walraven et al. 1995b; Bique Osman et al. 1999). The BMI (mean 22.9, SD 3.1) in our cohort was similar to that found in other African settings except for women from Egypt (Allen et al. 1994).

In our study, there was no difference in SFH growth chart if we excluded women having pre-term deliveries, or if we excluded women and their newborns without overt morbidity in accordance with other studies (Westin 1977; Steingrimsdóttir et al. 1995).

The SFH growth chart has been criticized for its low sensitivity (27%) in detecting small for gestational age (SGA) (Rosenberg et al. 1982; Persson et al. 1986) as well as for its poor specificity. Other authors (Westin 1977; Belizán & Villar 1978; Pattinson & Theron 1989) have found the sensitivity satisfactory (73–86%). The specificity, sensitivity, positive and negative predictive values of our SFH growth chart for detecting SGA will be analysed separately (Challis et al. unpublished observations).

Training of medical staff on how to measure SFH can improve quality of measurements (Johnsen & Jacobsen 1988). This has been questioned in another study (Calvert et al. 1982), which showed that experience did not reduce intraobserver variation. Our team of two to three midwives at each antenatal centre followed the plea for individualized antenatal care (Pattinson & Theron 1989) in order to reduce the interobserver variability. We agree with Steingrimsdóttir et al. (1995) on the need for adequate training and education, the importance of which is easily underestimated in `low-technology' methods. Differences in skills may be one of the main reasons for the divergent results reported in validity of the SFH method as a screening instrument to detect foetal growth disturbances.

In WHO's RH library, no. 42001, the editors conclude that the incorporation of an efficient symphysis-fundal measurement system into peripheral antenatal care systems, resulting in appropriate action being taken when required, presents a major challenge.

Our conclusion is that our findings indicate that ultrasound-based estimates of gestational age at early registration followed by frequent measurements of SFH in a Mozambican pregnant population allows for the construction of an SFH graph for local use. In low-income countries ultrasound is rarely available. The SFH growth chart is the most appropriate, simple and cost-effective tool we have for detection of the foetus being SGA or large for gestational age (LGA), multiple pregnancy and polyhydramnios. We consider the SFH growth chart an appropriate and affordable ingredient in antenatal care in countries with scarce human and material resources.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The study was made possible by grants from SAREC, the Sida department of Research Co-operation with Developing countries and FoU, Mid Sweden Research and Development Centre, Sweden. Without the fieldwork done by assistant medical officer and medical student Manuel Cotiro this study could not have been implemented.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Subjects and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    Allen LH, Lung'aho MS, Shaheen M, Harrison GG, Neumann C & Kirksey A (1994) Maternal body mass index and pregnancy outcome in the Nutrition Collaborative Research Support Program. European Journal of Clinical Nutrition 48 (Suppl. 3), S68S77.
  • 2
    Axemo P (1995) Pregnancy outcome in Mozambican women with special reference to intrauterine infection. Acta Universitatis Upsaliensis 570 (Ph D Thesis).
  • 3
    Belizán JM & Villar J (1978) Diagnosis of intrauterine growth retardation by a simple clinical method: measurement of uterine height. American Journal of Obstetrics and Gynecology 131, 643646.
  • 4
    Bergstrom S & Libombo A (1992) Puerperal measurement of the symphysis-fundus distance. Gynecological and Obstetric Investigation 34, 7678.
  • 5
    Bergstrom S & Liljestrand J (1989) Parturient fundal height and its relation to fetal weight. Journal of Tropical Pediatrics 35, 2730.
  • 6
    Bique Osman N, Challis K & Bergström S (2000) Maternal and fetal characteristics in an obstetric cohort at the community level in Mozambique. African Journal of Reproductive Health 4, 110119.
  • 7
    Bique Osman N, Ching C, Machungo F, Axemo P & Bergström S (1993) Adverse pregnancy outcome and its association with stillbirth and preterm birth in Mozambican parturient women. Gynecological and Obstetric Investigation 35, 108113.
  • 8
    Calvert JP, Crean EE, Newcombe RG & Pearson JE (1982) Antenatal Screening by measurement of symphysis-fundal height. British Medical Journal 285, 846849.
  • 9
    Folgosa E, Gonzalez C, Osman NB, Hagerstrand I, Bergstrom S & Ljungh A (1997) A case control study of chorioamniotic infection and histological chorioamnionitis in stillbirth. APMIS 105, 329336.
  • 10
    Hakansson A, Aberg A, Nyberg P & Schersten B (1995) A new symphysis-fundus height growth chart based on a well defined female population with ultrasound-dated singleton pregnancies. Acta Obstetrica Gynecologica Scandinavica 74, 682686.
  • 11
    Jensen O & Larsen S (1991) Evaluation of symphysis-fundus measurements and weighing during pregnancy. Acta Obstetrica Gynecologica Scandinavica 1, 1316.
  • 12
    Johnsen T, Jacobsen G & Knoff T (1988) The effect of practical training in obstetrics among medical students: symphysis-fundus height measurements. Medical Education 22, 438444.
  • 13
    Kieler H, Axelsson O, Hellberg D, Nilsson S & Valdenström U (1996) Serial measurements of symphysis-fundus height in women with ultrasonically dated pregnancies. Journal of Obstetrics and Gynaecology 16, 228229.
  • 14
    King M & Cairas J (1990) Primary Surgery, Vol. 1. Oxford University Press, Oxford.
  • 15
    Kiserud T (1986) Fundal height growth in rural Africa. Acta Obstetrica Gynecologica Scandinavica 65, 713715.
  • 16
    Liljestrand JBS (1989) Arm circumference as a nutritional predictor in antenatal care. Journal of Tropical Pediatrics 35, 281281.
  • 17
    Lindmark G & Cnattingius S (1992) The scientific basis of antenatal care routines: the state-of -the art. International Journal of Technol Assess Health Care 8 (Suppl. 1), 182188.
  • 18
    Mathai M, Jairaj P & Muthurathnam S (1987) Screening for light-for-gestational age infants: a comparison of three simple measurements. British Journal of Obstetrics and Gynaecology 94, 217221.
  • 19
    Medhat WM, Fahmy SI, Mortada MM et al. (1991) Construction of a local standard symphysis fundal height curves for monitoring intrauterine fetal growth. Journal of the Egyptian Public Health Association 66, 305331.
  • 20
    Moller B & Lindmark G (1997) Short stature: an obstetric risk factor? A comparison of two villages in Tanzania. Acta Obstetrica Gynecologica Scandinavica 76, 394397.
  • 21
    Munjanja SP, Masona D, Maxwell M & Mahomed K (1987) A symphysial-fundal height nomogram for central Africa. Central African Journal of Medicine 33, 2932.
  • 22
    Naeye RL (1981) Nutritional/nonnutritional interactions that affect the outcome of pregnancy. American Journal of Clinical Nutrition 34, 727731.
  • 23
    Neilson JP (2000) Symphysis-fundal height measurement in pregnancy. Cochrane Database Systematic Review 2, 11.
  • 24
    Ogunranti JO (1990) Fundal height in normal pregnant Nigerian women: anthropometric gravidogram. International Journal of Gynaecology and Obstetrics 33, 299305.
  • 25
    Osman NB, Folgosa E, Gonzales C & Bergstrom S (1995) Genital infections in the aetiology of late fetal death: an incident case-referent study. Journal of Tropical Pediatrics 41, 258266.
  • 26
    Osman NB, Challis K, Cotiro M, Nordahl G & Bergstrom S (2001) Perinatal outcome in an obstetric cohort of Mozambican women. Journal of Tropical Pediatrics 47, 3038.
  • 27
    Pattinson RC (1988) Antenatal detection of small-for-gestational-age babies. Choice of a symphysis-fundus growth curve. South African Medical Journal 74, 282283.
  • 28
    Pattinson RC & Theron GB (1989) Inter-observer variation in symphysis-fundus measurements. A plea for individualised antenatal care. South African Medical Journal 76, 621622.
  • 29
    Persson B, Stangenberg M, Lunell N, Brodin U & Holmberg N (1986) Prediction of size of infants at birth by measurement of symphysis fundus height. British Journal of Obstetrics and Gynaecology 93, 206211.
  • 30
    Van Roosmalen J & Brand R (1992) Maternal height and the outcome of labor in rural Tanzania. International Journal of Gynaecology and Obstetrics 37, 169177.
  • 31
    Rosenberg K, Grant J, Tweedie I, Aitchison T & Gallagher F (1982) Measurement of fundal height as a screening test for fetal growth retardation. British Journal of Obstetrics and Gynaecology 89, 447450.
  • 32
    Steingrimsdóttir T, Cnattingius S & Lindmark G (1995) Symphysis-fundus height: construction of a new Swedish reference curve, based on ultrasonicaly dated pregnancies. Acta Obstetrica Gynecologica Scandinavica 74, 346351.
  • 33
    Walraven GE, Mkanje RJ, Van Dongen PW, Van Roosmalen J & Dolmans WM (1995a) The development of a local symphysis-fundal height chart in a rural area of Tanzania. European Journal of Obstetrics Gynecology and Reproductive Biology 60, 149152.
  • 34
    Walraven GE, Mkanje RJ, Van Roosmalen J, Van Dongen PW, Van Asten HA & Dolmans WM (1995b) Single pre-delivery symphysis-fundal height measurement as a predictor of birthweight and multiple pregnancy. British Journal of Obstetrics and Gynaecology 102, 525529.
  • 35
    Westin B (1977) Gravidogram and fetal growth. Acta Obstetrica et Gynecologica Scandinavica 56, 273273.